7004_4623 010_System80_OS3_Job Control_Prog Guide_Nov91 7004 4623 010 System80 OS3 Job Control Prog Guide Nov91
7004_4623-010_System80_OS3_Job-Control_ProgGuide_Nov91 7004_4623-010_System80_OS3_Job-Control_ProgGuide_Nov91
User Manual: 7004_4623-010_System80_OS3_Job-Control_ProgGuide_Nov91
Open the PDF directly: View PDF 
.
Page Count: 334
| Download | |
| Open PDF In Browser | View PDF |
UNISYS
System 80
OS/3
Job Control
Programming
Guide
Copyright © 1991 Unisys Corporation
All rights reserved.
Unisys is a registered trademark of Unisys Corporation.
OS/3 Release 14
November 1991
Priced Item
Printed in U S America
7004 4623-010
C
The names, places, and/or events used in this publication are not intended to correspond to any individual, group,
or association existing, IMng, or otherwise. Any similarity or likeness of the names, places, and/or events with the
names of any individual, living or otherwise, or that of any group or association is purely coincidental and
unintentional.
NO WARRANTiES OF ANY NATURE ARE EXTENDED BY THIS DOCUMENT. Any product and related material
disclosed herein are only furnished pursuant and subject to the terms and conditions of a duly executed Program
Product Litense or Agreement to purchase or lease equipment. The only warranties made by Unisys, if any, with
respect to the products described in this document are set forth in such License or Agreement. Unisys cannot
accept financial or other responsibility that may be the result of your use of the information in this document or
software material, including direct, indirect, special, or consequential damages.
You should be very careful to ensure that the use of this information and/or software material complies with the
laws, rules, and regulations of the jurisdictions with respect to which it is used.
The information contained herein is subject to change without notice. Revisions may be issued to advise of such
changes and/or additions.
Correspondence regarding this publication should be forwarded to Unisys Corporation either by using the User
Reply Form at the back of this manual or by addressing remarks directly to Unisys Corporation, SPG East Coast
Systems Documentation Development, Tredyifrin Plant, 2476 Swedesford Road, P.O. Box 203, Paoli, PA,
19301-0203, U.S.A.
Product Information
Announcement
UNISYS
o New Release
o Revision
•
Update
o New Mail Code
Title
System 80 OS/3 Job Control Programming Guide
This announces the release of an update to this document
This guide is intended for the novice programmer with a basic knowledge of data processing but with little programming
experience, and the programmer whose experience is not on Unisys systems.
This update includes mode characteristic information for the 5073 magnetic tape. It also describes the SYSM64 parameter
which has been added to the ASM jproc, and the BUFMODE parameter which has been added to the DD job control statement.
All other changes are corrections, deletions, or expanded descriptions applicable to items present in the software prior to this
release. References to other Unisys manuals have been updated to reflect the changeover to the new 1 1-digit document
numbering system.
You can order the update only or the complete manual with all updates. To receive only the update, order 7004 4623-010. To
receive the complete manual, order 7004 4623-000.
To order additional copies of this document:
•
•
•
United States customers should call Unisys Direct at 1-800-442-1424.
All other customers should contact their Unisys Subsidiary Librarian.
Unisys personnel should use the Electronic Literature Ordering (ELO) system.
See important notice on the back of this sheet.
Announcement only:
Announcement and attachments:
SAB, SAE, MBOO
ECC3, MBO1, MBB1,
MBB3
System: System 80
Release 14
Date: November 1991
Part Number: 7004 4623-010
01
NOTICE
Please note that, in Release 14, the UP numbers of certain documents were changed to the new Unisys 11-digit document numbering system:
Old
Number
New
Number
Old
Number
New
Number
UP-8044
UP-8076
UP-8613
UP-8811
UP-8834
UP-8839
UP-8859
UP-8870
UP-8913
UP-8985
UP-9744
UP-9745
7004 4482-000
7004 5190-000
7004 4490-000
7004 4508-000
7004 4516-000
7004 5505-000
7004 5208-000
70044524-000
7004 4532-000
7004 5919-000
7004 4540-000
7004 4557-000
UP-9748
UP-9975
UP-9976
UP-9979
Up-9g82
UP-9985
UP-9986
UP-10003
UP-12443
UP-12649
UP-14207
UP-14208
7004 4565-000
7004 4581-000
7004 4599-000
7004 4607-000
7004 4615-000
7004 5224-000
7004 4623-000
7004 5232-000
7004 5240-000
7004 4631-000
7005 3434-000
7005 3442-000
In this update, some old UP numbers are still used in references to other documents; they Will be changed in the next revision of this document.
PAGE STATUS SUMMARY
ISSUE: 7004 4623-010
Part/Section
Page
Number
Update
Level
Cover
000
Title Page/Disclaimer
010
PSS
iii
010
About This Guide
v thru ix
000
Contents
xi thru xix
000
1
1 thru 13
000
2
1 thru 13
000
3
lthru23
000
4
1 thru 21
22
23 thru 50
000
010
000
5
1 thru 41
000
6
1 thru 58
59, 60
61 thru 78
000
010
000
7
1 thru 17
000
8
1 thru 15
000
9
1 thru 11
000
Appendix A
1 thru 9
000
Appendix B
1, 2
000
Appendix C
1 thru 9
10
11 thru 20
000
010
000
Index
I thru 13
000
Part/Section
Page
Number
Update
Level
Part/Section
Page
Number
Update
Level
User Reply Form
Back Cover
000
Unisys uses an 11-digit document numbering system. The suffix of the document number (1234 5678-xyz) indicates the document
level. The first digit of the suffix (x)
designates a revision level; the second digit (y) designates an update level. For example, the first release of a document has
a suffix of-COO. A suffix of -130 designates the
third update to revision 1.The third digit (z) is used to indicate errata for a particular level and is not reflected in the page status
summary.
7004 4623-010
Ill
C
UNISYS
System 80
OS/3
Job Control
Programming
Guide
Copyright © 1991 Unisys Corporation
All rights reserved.
Unisys is a registered trademark of Unisys Corporation.
OS/3 Release 14
April 1991
Priced Item
Printed in U S America
7004 4623-000
Ci
The names, places, and/or events used in this publication are not intended to correspond to any individual, group,
or association existing, living, or otherwise. Any similarity or likeness of the names, places, and/or events with the
names of any individual, living or otherwise, or that of any group or association is purely coincidental and
unintentional.
NO WARRANTIES OF ANY NATURE ARE EXTENDED BY ThIS DOCUMENT. Any product and related material
disclosed herein are only furnished pursuant and subject to the terms and conditions of a duly executed Program
Product License or Agreement to purchase or lease equipment. The only warranties made by Unisys, if any, with
respect to the products described in this document are set forth in such License or Agreement. Unisys cannot
accept financial, or other responsibility that may be the result of your use of the information in this document or
software material, including direct, indirect, special, or consequential damages.
You should be very careful to ensure that the use of this information and/or software material complies with the
laws, rules, and regulations of the jurisdictions with respect to which it is used.
The information contained herein is subject to change without notice. Revisions may be issued to advise of such
changes and/or additions.
Correspondence regarding this publication should be forwarded to Unisys Corporation either by using the User
Reply Form at the back of this manual or by addressing remarks directly to Unisys Corporation, SPG East Coast
Systems Documentation Development, Tredyifrin Plant, 2476 Swedesford Road, P.O. Box 203, Paoli, PA,
193010203, U.S.A.
PAGE STATUS SUMMARY
ISSuE: 7004 4623-000
Part/Section
Page
Update
Number
Level.
Part/Section
Page
Number
Update
Level
Part/Section
Page
Number
Update
Level
Cover
Title Page/Disclaimer
PSS
iii
About This Guide
v thru ix
Contents
xi thru xix
1
1 thru 13
2
lthrul3
3
lthru23
4
lthru5O
5
lthru4l
6
lthru78
7
lthrul7
8
lthrul5
9
1 thru 11
Appendix A
1 thru 9
Appendix B
1, 2
Appendix C
1 thru 20
Index
1 thru 13
User RepLy Form
Back Cover
Unisys uses an 11-digit document numbering system. The suffix of the document number (1234 5678-xyz) indicates the document level. The first digit of the suffix
(x)
designates a revision level; the second digit (y) designates an update level. For example, the first release of a document has a suffix of -000. A suffix of -130 designates the
third update to revision 1. The third digit (z) is used to indicate an errata for a particular level and is not reflected in the page status summary.
7004 4623-000
III
0
0
About This Guide
Purpose
This guide is one of a series of manuals designed to help the programmer use the
Unisys Operating Systeml3 (OS/3).
Scope
This guide specifically describes job control and explains how to use it.
Audience
The intended audience is the novice programmer with a basic knowledge of data
processing but with little programming experience and the programmer whose
experience is not on Unisys systems.
Prerequisites
Anyone using this guide should understand basic structured programming techniques.
How to Use This Guide
Read the entire guide to familiarize yourself with the basic concepts it presents; then
use it for reference as needed.
Organization
This guide is structured to proceed from the basic to the complex, addressing topics in
this sequence:
Job Control Overview
This topic is covered in Sections 1 and 2. It tells you what job control is and how it is
used by the operating system. You learn the basic concepts of a control stream and
the general program logic.
70044623-000
v
About This Guide
Basic Job Control Programming
This topic is covered in Sections 3, 4, and 5. In these sections, you become familiar
with the basic job control statements used to run your programs. You also learn about
job control procedure call statements (JPROCS) that can save you coding time and
reduce control stream coding errors.
Advanced Job Control Programming
This topic is covered in Sections 6, 7, 8, and 9, building on what you learned in
Sections 3, 4, and 5. You learn how you can get better performance and response from
the computer by using advanced job control statements that perform functions that
cannot be done with the basic set. You also learn how to write JPROC definitions that
you can store in the system and how you can call them when needed.
Appendixes
•
Appendix A discusses and illustrates the rules used in describing job control
statement formats. You also learn how you should code these job control
statements.
•
Appendix B contains supplementary information that increases your
understanding ofjob control.
•
Appendix C contains an alphabetical listing of all the job control statements and
JPROCS and their parameters. You can use this as a quick-reference chart.
Results
After reading this guide, programmers will be able to use job control statements and
JPROCS to specify to the operating system what specific work it must do.
Notation Conventions
The general conventions that apply to the coding formats presented in this guide are
described in Appendix A.
vi
70044623-000
About This Guide
Related Product Information
The following Unisys documents may be useful in understanding and implementing
job control.
Note:
Use the version that applies to the software level in use at your site.
Integrated Communications Access Method (ICAM) Utilities Programming
Guide, 7004 4565
This guide describes how programmers can use the utility routines provided by ICAM.
Job Control Programming Reference Manual, UP-9984
This manual is a quick-reference document for programmers familiar with OS/3. It
describes the job control statements and job control procedures used in a System 80
environment to communicate with job control. It also presents the procedure definition
statements that allow expansion and conditional modification of the job stream when
you start the job.
System Service Programs (SSP) Operating Guide, UP-8841
This guide describes the system service programs. They are utility programs that
support the operation and organization of the operating system. They include the
SAT and MIRAM librarians, the linkage editor, the disk, tape, and diskette prep
routines, and various copy routines.
Consolidated Data Management Macroinstructions Programming Guide,
70044607
This guide describes consolidated data management (CDM), a collection of program
modules that handles the movement of data between input and output devices on your
system. It also describes the consolidated data management macroinstructions, which
let you obtain information about the characteristics of your file or request that
consolidated data management process the files you defined for your program.
Models 8-20 Operations Guide, 70045208
This guide describes the hardware configuration of the System 80 models 8-20 and
presents procedures for initializing the system. It also covers all commands and
procedures used in the OS/3 environment.
Model 7E Operations Guide, 7002 3866
This guide describes the hardware configuration of the System 80 model 7E and
presents procedures for initializing the system. It also covers all commands and
procedures used in the OS/3 environment.
7004 4623-000
vii
About This Guide
Model 50 Detailed Operations Guide, 7004 1942
This guide describes the hardware configuration of the System 80 model 50 and
presents procedures for initializing the system. It also covers all commands and
procedures used in the OS/3 environment.
Supervisor Technical Overview, UP-8831
This manual presents an overview of the OS/3 supervisor and its functions for OS/3
high-level language programmers and site administrators.
Supervisor Macroinstructions Programming Reference Manual, UP-8832
This manual describes, for the assembler programmer, the OS/3 supervisor
macroinstructions used for program management, file space management, file access,
multitasking, and spooling. It also provides formats and coding conventions for the
macroinstructions, diagnostic and debugging information, and examples of
macroinstruction use.
Models 8-20 Installation Guide, 70045505
This guide provides the system administrator with the information and procedures
needed to install, tailor, and maintain 05/3 software in a System 80 models 8-20
environment.
Model 7E Installation Guide, 7002 3858
This guide provides the system administrator with the information and procedures
needed to install, tailor, and maintain OS/3 software in a System 80 model 7E
environment.
Model 50 Installation Guide, 70041892
This guide provides the system administrator with the information and procedures
needed to install, tailor, and maintain OS/3 software in a System 80 model 50
environment.
Interactive Services Operating Guide, UP-9972
This guide describes procedures used to communicate with the operating system
interactively through a local workstation or remote terminal. It also describes the
procedures for logging on and off the system and performing various interactive
commands.
File Cataloging Technical Overview, 70044615
This overview describes the 05/3 file cataloging facility in a System 80 environment
for the system administrator or programmers who are authorized to control use of the
system catalog file.
vi
7004 4623-000
About This Guide
Spooling and Job Accounting Operating Guide, 7004 4581
This guide describes basic spooling and job accounting concepts.
Screen Format Services Technical Overview, UP-9977
This overview describes how programmers can use screen format services to create
and maintain formatted screen displays to be used with their application programs.
Menu Services Technical Overview, UP-9317
This overview describes the procedures for creating and using menus. It also
describes how menus, displayed on the workstation screen, can be used with
assembler, COBOL, RPG II, and FORTRAN IVTM programs.
Dialog Processor Programming Guide, UP-8858
This guide provides the experienced programmer with information on the dialog
processor, which is the interface between the dialog (written in dialog specification
language) and the application program using the dialog.
Distributed Data Processing Programming Guide, 7004 4508
This guide describes OS/3 distributed data processing and the various distributed data
processing program products.
General Editor (EDT) Operating Guide, 7004 4599
This guide describes the commands and procedures needed to use the OS/3 general
editor to copy files, concatenate files, and create and modify library modules and data
files interactively from a workstation.
Consolidated Data Management Programming Guide, UP-9978
This guide describes consolidated data management and how it moves data between
peripheral devices and programs.
Assembler Programming Guide, 7004 4532
This guide describes the OS/3 basic assembly language (BAL) and its use. Included
are general language concepts, assembler instructions, and programming techniques.
Data Utilities Operating Guide, 7004 4516
This guide provides the information needed to use the data utilities. Included are
instructions on executing data utilities interactively and for batch jobs.
FORTRAN LV is a trademark of SuperSoft Associations.
7004 4623-000
ix
0
0
Contents
About This Guide
Section 1.
Overview
Why You Need Job Control
Job Control Statements and Job Control Streams
Job Steps
Job Control Procedures (JPROCS)
Job Control and the Operating System
Processing a Job Control Stream
Beginning Job Processing the Run Processor
Considering Jobs for Execution the Job Scheduler
Beginning Job Execution the Job Initializer
Initializing a Job Step the Job Step Processor
Ending the Job Step the Job Step Processor
Ending the Job the Job Terminator
Building and Storing Job Control Streams and JPROCS
Saving Translated, Expanded Job Control Streams
(Save/Restore Facility)
Running Job Control Streams
-
-
-
-
-
-
Section 2.
1-2
1-3
1-5
1-5
1-6
1-7
1-7
1-8
1-9
1-9
1-10
1-11
Basic Concepts
Assigning Devices and Files
Peripheral Devices and Logical Unit Numbers (DVC Statement)
Volume Serial Numbers for Disk, Diskette, and Tape (VOL Statement)
File Identifiers (LBL Statement)
Disk and Format-Label Diskette File Area (EXT Statement)
Data-Set-Label Diskette File Area (EXT Statement)
Logical File Names (LFD Statement)
Device Assignment Set Placement and Duration
Job Termination
Restarting a Job
Branching within a Control Stream
Jobs and Main Storage
Job Roll-Out/Roll-In
Minimum and Maximum Main Storage
Dynamic Expansion of Main Storage
7004 4623-000
1-1
1-1
2-1
2-2
2-3
24
2-5
2-6
2-7
2-9
2-10
2-11
2-11
2-12
2-12
2-12
2-13
xi
Contents
Section 3.
C
Minimum Control Stream Requirements
What Is a Minimum Control Stream’
Constructing the Minimum Control Stream
The Beginning of the Job
Identifying the Devices
Assigning a Logical File Name to the File
Executing the Program
Ending the Basic Control Stream
Ending the Card Reader Operation
The Control Stream So Far A Review
Adding Card Input
Card Input and Embedded Data
The Program Is Changed Another Device
What Is Needed to Use a Tape
7
The Logical Unit Number and File Name for the Tape
Supplying a Volume Serial Number for the Tape
Labeled Tapes for File Identification
Another Programming Change Another Device Assignment
The Device Assignment Set for a Disk or Format-Label Diskette
The Device Assignment Set for Data-Set-Label Diskette
The Device Assignment Set for a Workstation
The UID Job Control Statement
The USE Job Control Statement
Job Step Temporary and Job Temporary Files
Basic Job Control Statements
3-1
3-1
3-3
3-4
3-5
3-6
3-8
3-8
3-9
3-10
3-12
3-14
3-14
3-15
3-15
3-17
3-18
3-20
3-21
3-21
3-22
3-22
3-23
3-23
-
-
-
Section 4.
Getting the Most Out of the Basic Job Control Statements
Optional Parameters Can Improve Job Performance
Improving Your Control of the Job
A Selection Priority for the Job
Main Storage Needs
More Main Storage to Speed Up the Job
Multitasking Specification
The Processing Time for the Job
Debugging the Control Stream
Job Accounting and Spool Buffers
Printing the Job Log File and Page Headers
Identifying the Peripheral Devices a Little Further
Using Multiple Devices, SYSRES, or the Job’s $Y$RUN File
Specifying Multiple Workstations
More Control over Peripheral Devices
Assigning Devices by Physical Address and Assigning
Real Devices
Is This Device Needed for This Particular Run’
Different Volumes on the Same Device
4-1
44
4-2
4-2
4-3
4-4
4-5
4-6
4-7
4-8
4-9
4-9
4-10
4-11
4-11
4-12
4-12
C
xii
70044623-000
Contents
Multiple Volumes in a File? Use Alternate Devices to Decrease
Operator Setup Time
Ensuring that Workstations Are Connected to a Job
Specifying a Remote Disk File
Indicating Use of the DDP Program-to-Program Facility
More Information about the Characteristics of Your Volumes
More Than One Volume in a File
Special Characteristics of Tape Volumes
Extending Your Tape Volumes
Sharing Disk Volumes
Ignoring or Changing the Volume Serial Number
Multivolume Files Online Simultaneously
More Information on Disk and Format-Label Diskette File Allocation
The File Type
Formatting a File and Using Contiguous Space
Your Disk or Format-Label Diskette File Needs More Space
Terms of Allocation
Allocation Amounts
Changing the Specifications of a Previously Allocated File
Allocating Space in the Fixed-Head Area of Your 8417 Disk
No Terminate Option for Insufficient Extent Space
Information about Data-Set-Label Diskette File Allocation
Using Your File Identifier More Efficiently
Multivolume File? Assign Each Volume a File Serial Number
The Expiration and Creation Date of the File
Indicating the Position of the File when Several Are on a Tape Volume
Different Versions of a File
Changing the Label of a Disk File
Specifying Qualifiers for File Identifiers
More Information about the Logical File
Reserving an Extent Information Storage Area
Specifications for Existing Files
Indicating Where the Load Module Is Located
Task Switching Priority
Avoiding Abnormal Termination due to Program Errors
The Job Control Language So Far
Section 5.
Doing It the Easy Way with Procedure Calls
-
What Is a Procedure
7
Setting Up Temporary Work Files
Using Your Own Volume
Providing the Extent Specifications
Accessing Previously Allocated Files
Allocating a File with a JPROC Call
Too Many Devices for the Same Volume
Using the Linkage Editor
Generating LOADM and INCLUDE Linkage Editor Control Statements
Making the Linkage Editor Suit Your Needs
7004 4623-000
4-14
4.15
4-16
4-17
4-19
4-20
4-21
4-22
4-24
4-24
4-27
4-27
4-27
4-28
4-29
4-30
4-31
4-33
4-34
4-34
4-34
4.35
4-36
4-37
4-38
4-39
4-40
4.43
4-44
4-44
4-45
4-46
4-48
4-50
4-50
5-1
5-2
5-5
5-6
5-8
5-10
5-13
5-16
5-21
5-23
xiii
Contents
C
Personalizing Your Print Output
Controlling Spooled Output with a JPROC Call
Section 6.
5-33
5-38
Making Job Control Work for You
Advantages of Using Advanced Job Control Statements
Controlling Spooled Output with a Job Control Statement
Sending Spooled Output to Remote Batch Processing Terminals
Sending Spooled Output to DDP Sites and Auxiliary Workstation Printers
Spooling Input Card Data
Spooling Diskette Files
Equating Logical Unit Numbers to Device Type Codes
Specifying Unique Load Codes
Using Forms Control
Controlling Tape Units
Releasing (Freeing) a Device and Volume
Scratching Unwanted Files
File Cataloging
Selecting Optional Features
Using the SET Job Control Statement
Changing the Date
Setting the UPSI
The Communications Region
The User Local Data Area (LDA)
Restarting a Job
Restarting a Job from a Job Step
Restarting a Job from a Checkpoint Record
Issuing System Commands
Calling Control Streams
Using the RV/RUN Job Control Statements to Call Control Streams
Using CC SC/SI to Call Saved Translated Control Streams
Communicating with the System Operator or Workstations
Introducing Processing Options
Defining Software Facilities Needed by Your Job
Making Temporary Changes to a Load Module
Changing Your File Definition at Run Time
Adding Cards to a Stored Control Stream
Bypassing Job Control Statements
Bypassing Job Control Statements to Avoid Abnormal Termination
Dynamic Skip Function from a Workstation
Substituting Embedded Data
Replacing Embedded Data Sets in Expanded Control Streams
Job Control Considerations for Screen Format Services,
Menu Services, and Dialog Processing
The USE Statement for Screen Format Seriices
The USE Statement for Menu Services
The USE Statement for Dialog Processing
Source Module Access via the USE Statement
xiv
....
6-1
6-1
6-3
6-4
6-7
6-9
6-10
6-11
6-16
6-20
6-22
6-24
6-26
6-26
6-39
6-39
6-39
6-40
6-41
6-42
6-43
6-44
6-45
6-47
6-48
6-49
6-50
6-52
6-53
6-56
6-58
6-61
6-63
6-68
6-69
6-69
6-70
6-72
6-73
6-74
6-76
6-78
70044623-000
Contents
Section 7.
Run-Time Conditional and Set Symbol Job Control Statements
Run-Time Conditional Job Control Statements
Unconditional Branching
Conditional Branching
Providing Targets for Branching
Run-Time Set Symbols
Global Status Set Symbols
Local Status Set Symbols
Specifying Set Symbol Values in Quotes
Using Symbols to Examine Job and System Related Values and
Facilities
Priorities among Set Symbols, Keyword Parameters, and Positional
Parameters
Section 8.
8-i
8-1
8-3
8-3
8-4
8-5
8-6
8-7
8-9
8-10
9-1
9-3
9-8
9-8
9-9
Statement Conventions
Job Control Statement Format
How Job Control Statements Are Presented
Coding Conventions
Statement Continuation
Software Conventions
7004 4623-000
7-16
Using the Interactive Job Control Dialog
The Function of the Job Control Dialog
Building a Control Stream with the Job Control Dialog
Building a User JPROC with the Job Control Dialog
Entering Embedded Data
Changing Dialog Responses
Appendix A.
7-13
How to Write and Call a Job Control Procedure Definition
The Benefit of Procedure Definitions
Coding Rules
Parameter Types
The Start of the JPROC Definition
Naming the JPROC Definition
Ending the JPROC Definition
Calling JPROC Definitions
How JPROC Definitions Are Stored
Specifying an Alternate Library File to Be Searched for JPROCS
Parameter Referencing
Section 9.
7-i
7-i
7-2
74
7-5
7-5
7-10
7-12
A-i
A-2
A-7
A-8
A-9
xv
Contents
Appendix B.
Operation Considerations
System Libraries
Volume Table of Contents
Appendix C.
B-i
B-2
Job Control Statement Formats
Job Control Statements
Job Control Procedures
C-i
C-9
Index
User Reply Form
Xvi
70044623-000
Figures
1-1.
1-2.
Operating System/3 (OS/3)
Job Processing Flow
1-3
1-5
2-1.
Job Region in Main Storage
2-13
9-1.
9-2.
9-3.
Using the Job Control Dialog to Build a Control Stream or User JPROC
Audit Version of the Dialog Processor
Changing Your Dialog Responses
9-2
9-10
9-11
70044623-000
xvii
C
C
n
Tables
41.
Mode Characteristics
4-22
&1.
DD Supported Keywords
6-60
7-1.
Keywords and Symbol Values for 1/ INQ JOB and // INQ SYS
7-15
7004 4623-000
xix
0
0
0
Section 1
Overview
Why You Need Job Control
To process any program, the operating system must have some necessary instructions
and information. Should the system compile, link edit, or execute a program? Does it
know what files a program uses, which devices to reserve, and how much main storage
a program needs? Should it allocate space for a file? For the operating system to
know what specific work what job you want it to do and how, you must supply this
type of information to that part of OS/3 called job control.
-
-
To communicate with job control, you use OS/3 job control language (JCL), which
consists ofjob control statements and job control procedures (JPROCS). The
statements and JPROCS you code make up ajob control stream.
Job Control Statements and Job Control Streams
Each of the many job control statements has a different function but they are
combined in a control stream to do a singular job. OS/3 requires that every job have a
control stream. Using three statements, II JOB, II EXEC, and /&, we can show you the
following outline job control stream required for executing a program:
Job control
stream for
executing
a program
If JOB MYJOB
Identifies your job and indicates
the beginning of the control stream.
If EXEC PROG1
Specifies execution of the program PROG1.
/&
Indicates the end of the control stream.
These three statements illustrate the idea of a job control stream, but you’ll see in
later sections that you must also include statements identifying files and devices.
Additional statements are used, depending on the specific function needed to
accomplish your job. You can also include program data in the control stream.
In this guide, we’ll explain the function of each job control statement and its
parameters so you can build simple as well as complex job control streams.
70044623-000
1-1
Overview
Job Steps
Any job can have one or more steps. If, for example, you want to execute three
programs, one after the other, you can construct one job control stream with three (job)
steps like this:
1/ JOB MYJOB
Job step 1
/1 EXEC PROG1
Job named F4YJOB
.
Job step 2
II EXEC PROG2
Job step 3
II EXEC PROG3
/&
Ajob can have up to 254 job steps. The steps are processed serially and the EXEC job
control statement normally marks the end of each one.
Job Control Procedures (JPROCS)
Besides using individual job control statements in your control stream, you can use job
control procedures (JPROCS).
A JPROC is a series ofjob control statements that performs a certain function or
routine. JPROCS are supplied as part of the system End you can also write your own.
They are filed in a library and each JPROC has its own name. (See “Building and
Storing Job Control Streams and JPROCS” in this section.) When referenced by that
name in a job control stream, the statements that make up the JPROC are generated
and incorporated into the control stream.
You may frequently need some function that a specific group ofjob control statements
performs. Instead of coding the same group of statements in every job control stream
requiring that function, you can simply define the statements as a JPROC, then code
the JPROC name.
Compiling a source program, for example, is something that’s done often. If you
include a certain system supplied JPROC name in your job control stream, all the
statements necessary for the language processor to compile your source program are
generated. The following simplified control stream specifies the COBOL language
processor JPROC.
1-2
70044623-000
Overview
II JOB MYJOB
/1 COBOL
Causes the generation of job control
statements that identify files and
devices needed by the COBOL language
processor. Executes the language
processor so that a source program can
be compiLed.
/&
System-supplied and user-written JPROCS are explained in Sections 2 and 3.
Job Control and the Operating System
To better understand what job control does, it heips to know where job control fits into
the operating system.
Unisys Operating Systeml3 (OS/3) is divided into two parts: the executive and the
system support software components. Job control is part of the executive portion of
OS/3. Together, the supervisor and job control manage job processing for 05/3.
Figure 1-1 shows the executive and system support software components of OS/3.
EXECUTIVE
SUPERVI SOR
JOB CONTROL
SYSTEM SUPPORT SOFTWARE COMPONENTS
DATA
MANAGEMENT
DATA BASE
MANAGEMENT
SYSTEM
LANGUAGE
PROCESSORS
INTEGRATED
COMMUNICATIONS
ACCESS
METHOD
APPLICATIONS
SYSTEM
SERVICE
PROGRAMS
EMULATORS
I N FORMAT ION
MANAGEMENT
SYSTEM
DIAGNOSTIC
PROGRAMS
Figure 1-1. Operating System/3 (OS/3)
7004 4623-000
1-3
Overview
The supervisor controls the sequence and position of your programs and system
programs in main storage. For more information on supervisor facilities, see the
Supervisor Technical Overview, UP-8831.
Job control manages system facilities and prepares the system for job execution. In
general, it does the following:
•
Assigns ajob number to every active job and symbiont
•
Analyzes the job control stream
•
Checks the order and syntax of control statements
•
Expands job control procedures (JPROCS)
•
Schedules jobs and queues them according to priority
•
Allocates peripheral devices and main storage
These and some of the other functions that job control is responsible for are handled
by (system) programs called symbionts. Symbionts are normally executed in response
to a user request that may be in the form of a system console command, a workstation
command, or certain job control statements. Symbionts compete for main storage and
CPU time along with your jobs. The run processor, which begins processing your job
control streams, is a symbiont. We’ll be discussing the run processor in the next
section.
1-4
7004 4623-000
Overview
Processing a Job Control Stream
You can build job control streams on disk, data-set-label diskette, or on cards.
Looking at Figure 1-2, you can see that job processing involves several steps.
OR
OR
$V$RUN FILE
(CONTAINS TRANSLATED JOB CONTROL STREAM, INCLUDING EXPANDED JPROCSI
Figure 1-2. Job Processing Flow
A brief discussion of each step in the job processing flow should give you a general idea
of what happens after job control accepts a request to process a job.
Beginning Job Processing the Run Processor
-
The run processor begins job processing by scanning the control stream, translating
the job control statements into tables on disk, and expanding JPROCS. At this point,
it also checks the stream for order and syntax errors. If there are errors, no further
preparation of the job is made and job control error messages are generated.
7004 4623-000
1-5
Overview
Once the control stream is translated, the run processor places it in the $Y$RUN
system file (a $Y$RUN file is created for every job being processed). The name of the
job (obtained from the II JOB statement) is entered in a table called the job queue
table. The job queue table contains the names of all jobs waiting to be executed. The
jobs are ordered by a priority specified on the JOB statement (or, as you’ll see later, on
other jobcontrol statements or workstation/console commands). Within a particular
priority, the jobs are ordered on a first-in, first-out basis.
C
RUN PROCESSOR
•
Translates job control statements
•
Expands JPROCS
•
Checks order and syntax of control stream
•
Builds control blocks
•
Enters job name in job queue table
Creates $Y$RUN file
Considering Jobs for Execution the Job Scheduler
-
After the run processor prepares your job control stream, processing control passes to
the job scheduler, which checks the job queue table. If there are jobs in the queue
table, the scheduler determines which jobs will be executed next. The job priority and
the availability of system resources (peripheral devices and main storage) are the
basis for this determination.
Ajob can have one of four priorities: preemptive, high, normal, or low. At any one
time, the job queue table can contain the names of up to 15 preemptive priority jobs,
39 high priority jobs, 71 normal priority jobs, and 15 low priority jobs. The job
scheduler considers preemptive jobs for execution first, followed by high, normal
priority, and low jobs (in that order). Jobs are considered within each priority level on
a first-in, first-fit basis. Lower priority jobs are not considered until there are no other
higher priority jobs in the job queue table. Jobs in HOLD status are not considered at
all.
Before job execution can start, sufficient main storage and the necessary peripheral
devices must be available. The job scheduler checks for both; and if both are not
available, the job is left in the job queue table. A slightly different situation exists if
roll-out is configured with the system. (See “Job Roll-OutJRoll-In” in Section 2.)
C
1-6
70044623-000
Overview
In addition to checking priority and the availability of main storage and peripheral
devices, the job scheduler maintains the shared code directory, reserves volumes,
maintains a volume use table for all jobs, deletes your job name from the job queue
table, and displays your job name at the system console.
JOB SCBEDULER FUNCTION
•
Considers your job for execution by priority
•
Reserves devices and main storage for your job so that job execution can begin
•
Deletes the job name from the job queue table
•
Displays the job name on the system console
Beginning Job Execution the Job Initializer
-
Processing control passes to the job initializer when job execution is ready to begin.
Up to 47 jobs can be executed concurrently.
The job initializer also loads shared code modules, activates job accounting, and
updates job log status.
JOB INITIALIZER FUNCTION
•
Builds job preamble
•
Loads shared code modules
•
Activates job accounting
•
Updates job log status
Initializing a Job Step the Job Step Processor
-
The job step processor performs the functions necessary for initializing and completing
a job step. At this point in job processing, the program specified on the EXEC
statement is loaded and executed.
7004 4623-000
1.7
Overview
C
JOB STEP PROCESSOR FUNCTION (STEP INITIALIZATION)
•
Reviews volume requirements
•
Reviews device allocation
•
Updates system volume use table
•
Allocates devices and disk space
•
Locates and updates file control blocks
•
Locates and posts address of embedded data
•
Stores logging data
•
Performs utility functions (rewinding tapes, scratching files, etc.)
Ending the Job Step the Job Step Processor
-
The job step processor also performs the end-of-job-step housekeeping duties. If this
is the last step in the job, the job step processor passes processing control to the job
terminator; if not, it retains processing control for initialization of the next job step.
JOB STEP PROCESSOR FUNCTION (STEP TERMINATION)
•
Updates job preamble
•
Initiates burst mode printing of spool files
•
Records logging data
•
Scratches job step (temporary) work files
0
1-8
7004 4623.000
Overview
Ending the Job the Job Terminator
-
When the last step in the job has been processed, the job terminator receives control to
perform end-of-job housekeeping duties.
JOB TERMINATOR
•
Deletes job name from system console
•
Scratches job temporary files
•
Scratches job’s $Y$RUN file
•
Requests printing of log and spooi files
•
Displays job termination message
•
Frees memory and releases devices
•
Clears job entries from system volume use table
Building and Storing Job Control Streams and JPROCS
Here are some ways you can build and store job control streams:
•
If you have UDS-200 data entry equipment, you can use it offline to place job
control statements directly onto data-set-label diskettes.
•
You can use the general editor (EDT) to build control streams at a workstation.
Depending on the instructions you give the editor, the control stream can then be
placed on data-set-label diskette, in the spool file, in a permanent job control
stream library on disk or format-label diskette, or on cards. You can specifSr a
permanent SAT library of your own as the stream’s destination, or you can use
$Y$JCS, the system job control stream library. The General Editor (EDT)
Operating Guide, 7004 4599, explains the use of the general editor.
•
You can use the job control dialog to build control streams. The dialog stores the
completed stream in $Y$JCS. Section 9 explains the interactive job control
dialog.
•
If the control stream is already on data-set-label diskette, in the spool file, or on
cards, you can use a FILE system console command or the FILE workstation
command to place the stream in a permanent SAT library. The FILE system
console command is explained in your operations handbook and the FILE
workstation command is discussed in the Interactive Services Operating Guide,
UP-9972.
70044623-000
1-9
Overview
C
For JPROCS to function as intended, you must store them in $Y$JCS or your own
SAT library. So whether you use EDT or the job control dialog, the eventual
destination of the JPROC is a permanent library. See “How JPROC Definitions Are
Stored” in Section 8 for more information on storing JPROCS.
Saving Translated, Expanded Job Control Streams
(Save/Restore Facility)
Before a job can be executed, no matter how often its been executed already, it must
be translated and have any JPROCS expanded first. This is done by the run
processor, and for some jobs (especially those with many JPROCS) this takes a long
time. You can reduce this time by saving the control stream in its translated,
expanded state. Because the rim processor can skip the step of translating and
expanding this type of control stream when it is restored and job processing starts, the
job’s execution begins sooner.
To save a job control stream in its translated, expanded state, you simply include the
II OPTION SAVE or II OPTION NOSCHED statement in the control stream. (See
“Selecting Optional Features” in Section 6.) When job processing is initiated and the
run processor finishes expanding and translating the control stream, a copy of the
stream (as it appears in $Y$RUN) is placed in a permanent MIRAM library. You can
specify your own library or you can use the system library $Y$SAVE.
Depending on which OPTION statement you used, processing then proceeds through
execution (OPTION SAVE) or stops as soon as the expanded, translated stream is
placed in the specified library (OPTION NOSCHED). In either case, youll have a
copy of the expanded stream in a permanent library.
When a translated stream is processed, the OPTION SAVE/NOSCHED statement is
ignored. If you intend to process the untranslated stream, you should remove the
OPTION SAVE/NOSCHED statement. A command different from the one used to
initiate processing of the untranslated stream is used for the translated one. See
“Running Job Control Streams” later in this section for more information.
C
1-10
70044623-000
Overview
CONTROL STREAM CONIAINING
_1
OPTION SAVE. NOSCHED
OR
AN ALTERNATE
SAT
I
EXPANDED. TRANSLATED
CONTROL STREAM
I I
SYSSAVE
LIBRARY
oNrROL STREAM PROCESSING
I
I
YSJCS
OR
AN ALTERNATE
SAT LIBRARY
ALTERNATE
MIRAM
LIBRARY
ORIGINAL CONTROL STREAM
When deciding whether or not to save expanded, translated control streams, keep the
following in mind: these streams take up more disk space than untranslated ones, you
can’t use them to update a file catalog (see “File Cataloging” in Section 6), and you
can’t change parameters on any of the job control statements. Replacing embedded
data sets is the most extensive change you can make to these streams (see “Replacing
Embedded Data Sets in Expanded Control Streams” in Section 6). Remember, you
cannot use a hyphenated job name if you intend to save your translated control
stream; the save processor does not recognize the hyphen.
Running Job Control Streams
Running a job control stream is a term commonly used in place of processing a control
stream. In 05/3 there are several ways you can initiate the running of a control
stream. These include the RV/RUN system console and workstation commands, the
II RV/RUN job control statements, the SC/SI system console and workstation
commands, and the II CC SC/SI job control statements. The differences between these
commands and statements are summarized as follows:
RV system console or workstation command
This command initiates a stored control stream from a workstation or system
console that does not need an input device.
70044623-000
1-11
Overview
•
RUN system console or workstation command
This command initiates a job control stream from a workstation or system console
that needs an input device. This may mean the control stream to be run is on a
data-set-label diskette, in the spooi file, or on cards. It may mean the control
stream is stored in $Y$JCS or an alternate SAT library file but contains a CR job
control statement and, therefore, will need an input device to complete
processing. (See “Adding Cards to a Stored Control Stream” in Section 6.)
•
II RVjob control statement
This statement, when encountered in an executing job control stream, initiates
the running of another control stream.
•
II RUN job control statement
This statement, when encountered in an executing job control stream, initiates
the running of another control stream. You can use II RUN if the control stream
is on cards or is stored in a library but contains all CR statement because card
input is needed to complete job processing.
•
SC system console or workstation command
This command initiates an expanded, translated control stream (stored in
$Y$SAVE or an alternate MIRAM library) that does not require replacement of
embedded data and, therefore, does not need an input device.
•
SI system console or workstation command
This command initiates an expanded, translated control stream from $Y$SAVE
or an alternate MIRAM library that needs an input device for the replacement of
embedded data.
•
ll CC SC job control statement
This job control statement, when encountered in an executing control stream,
initiates an expanded, translated control stream from $Y$SAVE or an alternate
MIRAM library that does not require replacement of embedded data and,
therefore, does not need an input device.
•
II CC SI job control statement
This job control statement is used the same as II CC SC except that it initiates an
expanded, translated control stream from $Y$SAVE or an alternate MIRAM
library requiring an input device for the replacement of embedded data.
1-12
70044623-000
Overview
For information about system console commands, see the appropriate operations
guide. For information about workstation commands, see the Interactive Services
Operating Guide, UP-9972. For information about the II CC SC/SI and
II RV/RUN job control statements, see, respectively, the “Using the RV/RUN Job
Control Statements to Call Control Streams” and “Using CC SC/SI to Call Saved
Translated Control Streams” in Section 6.
70044623-000
1-13
C
Section 2
Basic Concepts
Assigning Devices and Files
An important part of writing ajob control stream is identifying devices and files and
establishing a logical connection between the files and the program using them. The
following job control statements help you do this:
DD
EXT
LED
UID
DST
LBL
ROUTE
USE
DVC
LCB
SPL
VFB
VOL
The DVC and LFD statements (in that order) are required for every type of file and
device you use. The other statements (when used) must appear between the DVC and
LFD statements. They’re necessary depending on the kind of file, or function you
want performed in relation to that file. As a group, these statements are called a
device assignment set.
If JOB MYJOB
Device
assignment
set for a
fiLe used by
PROG1
If DVC...
If LFD...
1/ EXEC PROG1
The CAT, DECAT, EQU, FREE, REN, and SCR job control statements are not coded
between the DVC and LFD statements; so, technically, they’re not part of a device
assignment set, but their function is related. We’ll talk about these in later sections.
In this section, we’ll talk about the DVC, VOL, LBL, EXT, and LFD job control
statements to help you become familiar with the overall function of a device
assignment set.
70044623-000
2-1
Basic Concepts
Peripheral Devices and Logical Unit Numbers (DVC Statement)
A peripheral device is any unit of equipment, distinct from the central processor and
main storage, that allows the system to send or receive data. Some devices, such as
card readers, only handle incoming data (input); some, such as printers and card
punches, can only handle outgoing data (ouput); while others, such as disks, formatlabel diskettes, tapes, and workstations, can handle both (input and output).
In OS/3, each type of peripheral device is assigned a specific number called a logical
unit number. You specify logical unit numbers in the DVC job control statement. This
tells job control (the job scheduler) which peripheral devices you need for your job.
Suppose you need a printer because your program produces printed output. The
following information (taken from Table A-i of the Job Control Programming
Reference Manual, UP-9984) shows some logical unit numbers for printers.
Logical
Unit No.
Device Type
Code
04040000
04010000
04020000
04FF0000
04400000
04100000
04200000
04800000
14,
16,
18,
20,
22,
24,
26,
28,
15
17
19
21
23
25
27
29
Device Type and Features
0791 correspondence quality printer
0798 printer, no optionaL features
0789 printer
Any printer, no features specified
9246 printer, no features specified
0776 printer, no optional features
AP9215 printer, no features specified
0770 printer, no optional features
If you need a Unisys 0776 printer, specify either 24 or 25 on the DVC statement. If
any printer will do, specify 20 or 21.
Device assignment
for the 0776
printer
f
II
If
1/
/1
JOB MYJOB
DVC 24 1
LFD....
EXEC PROd
Device
assignment
for any
avaiLabLe
printer
f
II
If
If
If
JOB MYJOB
DVC 20
LFD...
EXEC PROG1
Each logical unit number you use corresponds to a device requirement for your job.
So, if you specify logical unit number 20 in one job step and logical unit number 21 in a
following step, two printers must be available in order for your job’s execution to
begin, even if one is sufficient.
2-2
7004 4623.000
Basic Concepts
II JOB MYJOB
If DVC 20
1/ LFD..
•
1/ EXEC PROG1
•
Two printers must be
avaiLable for this
job to run.
II DVC21
If LFD.
-.
If EXEC PROG2
I&
Besides using logical unit numbers, disk devices can be assigned by specifying RES or
RUN. These and other functions of the DVC statement are further discussed in
Sections 3 and 4.
Note:
For 0776 printers, the CLASS=parameter should be used if a unique logical
unit number is required.
Volume Serial Numbers for Disk, Diskette, and Tape (VOL Statement)
Volume serial numbers are used to uniquely identify disk packs, diskettes (format and
data-set-label), and tape reels to the operating system. This number is written
externally (generally on a gummed label) and internally (on the actual recording
surface). Both numbers should match for identification purposes.
The assignment of volume serial numbers takes place when the prep routines
associated with disk, diskette, and tape are performed. See the System Service
Programs (SSP) Operating Guide, UP-8841, for information about prep routines.
When you specify a volume serial number in a VOL statement, job control checks to
make sure that a tape reel, diskette, or disk pack with the matching volume serial
number is mounted. If the wrong volume is mounted, the system notifies the operator.
7004 4623-000
2-3
Basic Concepts
In this example
II JOB MYJOB
Device
assigrnent
for a disk fiLe
II DVC 50
II VOL 12345A
II LFD...
Specifies any available
disk device
Specifies a disk pack with
the assigned volume serial
number of 12345A
/1 EXEC PROG1
the disk volume whose serial number is 12345A must be mounted for job processing to
continue.
We’ll discuss other functions of the VOL statement in Sections 3 and 4.
Notes:
1.
OS/3 assumes that all volume serial numbers are unique. The mounting of two
volumes with the same volume serial number at the same time yields unpredictable
results.
2.
OS/3 allows a maximum of 151 volumes to be in use by all active jobs. (The
maximum number of volumes allowed for a single job is also 151.)
File Identifiers (LBL Statement)
While a volume serial number identifies one tape, disk, or diskette volume, a file
identifier names (or identifies) a particular file on that volume. The file identifier is
an alphanumeric name physically written on the recording surface of the tape, disk, or
diskette (format and data-set-label). You specifSr a file identifier on the LBL job
control statement. If you’re creating the file, the identifier you specify is assigned. If
the file already exists, job control checks to see that the file identifier specified with
the LBL statement matches one already recorded for a file on a particular volume.
This ensures correct file use.
C,
2.4
70044623.000
Basic Concepts
If JOB MYJOB
Device
assignment set
for a disk file
/1
/1
/1
ii
DVC 50
VOL 12345A
LBL MYFILE
LFD...
If the fite,is being created,
I4YFILE is the identifier
assigned. If the file exists,
MYFILE is the identifier job
controL checks for.
II EXEC PROG1
A file identifier specified on an LBL statement is required for any file on disk,
diskette, or multifile tape volume. If a tape volume holds only one file, a file identif
ier
may be specified but isn’t required. As you’ll see in a later section on spoolin
g card
input, it is sometimes useful to specifSr an LBL statement (with a file identif
ier) in the
device assignment set for a card file that’s been spooled.
The LBL statement has other functions that are covered in Sections 3 and 4.
Note:
The prep routine for data-set-label diskette automatically assigns a file
identifier ofDATA unless you specify otherwise during the prep.
Disk and Format-Label Diskette File Area (EXT Statement)
Whenever you’re creating a disk or format-label diskette file, you allocat
e space for
that file in contiguous areas (on the recording surface) called extents. The
amount of
space as well as other characteristics of the file’s extent are specified using
the EXT
job control statement. The device assignment set for every disk or format
-label
diskette file you are creating must include an EXT statement. It is also
required if
you want to change certain extent specifications for a file that already
exists.
Using the EXT statement, space on disk or format-label diskette is
allocated in terms
of one of the following:
•
Number of cylinders
You specify the number of cylinders needed for the file.
•
Absolute cylinder address
You specify the number of cylinders needed for the file and you also
specify the
starting address of the file as an absolute cylinder address.
7004 4623.000
2-5
Basic Concepts
•
Number of tracks
You specify the number of tracks needed for the file.
•
Absolute track address
You specify the number of tracks needed for the file and you also specify the
starting address of the file as an absolute track address.
•
Number of blocks (by cylinder)
You specify the number and average length of the blocks needed for the file. Job
control converts this specification to the number of cylinders so the actual
allocation is by cylinder.
•
Number of blocks (by track)
You specify the number and average length of the blocks needed for the file. Job
control converts this specification to the number of tracks so the actual allocation
is by track.
You’ll learn more about file space allocation when we discuss the EXT statement in
Sections 3 and 4. For now, it is enough to know that an EXT statement must be
included in the device assignment set when you’re allocating space or making certain
allocation changes for a disk or format-label diskette file.
Device assignment
set for a
disk file.
II
II
II
/1
II
II
JOB MYJOB
DVC 50
VOL 12345A
LBL MYFILE
EXT MI,C,,CYL,4 —p
LFD....
This statement specifies four
cylinders of contiguous space
for a MIRAI4 (disk) file.
II EXEC PROG1
/&
Data-Set-Label Diskette File Area (EXT Statement)
The prep routine for a data-set-label diskette automatically allocates the entire
diskette for one file and assigns a file identifier of DATA unless you specify otherwise.
If the space was already allocated by the prep routine, there is no need for you to
include an EXT statement in your device assignment set. If, however, the space was
not previously allocated, you must use the EXT statement to allocate the space
yourself. Allocating the space yourself allows you to have control over how many files
the diskette can contain.
2-6
7004 4623-000
cD
Basic Concepts
Space on data-set-label diskette must always be allocated by block and it must be
contiguous. Data-set-label diskette files are always one-extent files. For information
about the EXT statement for data-set-label diskette, see “Information about Data-SetLabel Diskette File Allocation” in Section 4.
Logical File Names (LFD Statement)
We’ve already talked about how you specify a file identifier (a name that’s physically
recorded on the surface of a disk, tape, or diskette) on the LBL job control statement.
There is another name, however, that is required for every file (not just disk, tape, and
diskette) and must be included in every device assignment set. It is the logical file
name: the name your program references the file by.
You specifr it on the LFD (logical file definition) job control statement, which is
always the last statement in any device assignment set. The name you specify
logically (LFD) links the file (name) you reference in your program with the physical
file (LBL) defined in your job stream’s device assignment set. The names that you use
are:
InBAL
The name from the label field of the file definition macroinstruction.
If:
Then:
1
10
FILE1
CDIB
16
II
/1
II
II
II
DVC
VOL
LBL
EXT
LFD
50
12345A
MYFILE
MI,C,,CYL,4
FILE1
Device assignment
set for a newLy
allocated file
referenced by the
progran as FILE1
In COBOL
The LFD field of the implementor name from the SELECT clause.
If
Then:
12
SELECT CDS ASSIGN TO CARDREADER-INFIL-F
If DVC 30
II LFD INFIL
Device assignment set
for the card file
(In basic and extended COBOL, the LFD name corresponds to the first eight
characters of the file name from the SELECT clause. If an external name is specified,
however, then use the external name instead.)
7004 4623-000
2-7
Basic Concepts
•
In FORTRAN
The device number from the READ or WRITE statement, prefixed by FORT.
If:
Then:
7
10
/1
/1
Ii
If
READ(6, 10)
DVC
VOL
LBL
LFD
90
TAPEO1
PAYFIL
FORT6
Device assignment
set for a tape fite
InRPGII
The file name from the file description specification.
If:
FORM
TYPE
Then:
—
FILE TYPE
‘
If DVC 20
If LFD PRINT
FILE DESIGNATION
END OF FILE
Device assignment
set for a print fiLe
SEOUENCE
PAGE
NO.
FILE
NAME
LINE
NO
FILE FORMAT
o
123
567
QLI. JJL
PRII IT
1314151617181920
i
-
-
BLOCK
LENGTH
23
-
The file name used for a printer file in programs supplied by Unisys (such as the
compilers and the linkage editor) is a standard system file name, PRNTR. So, if you
want the printed output from a compilation, for example, the LFD statement for the
print file device assignment set is II LFD PRNTR. This logical file name applies only
to programs supplied by Unisys. In ajob or job step that executes a user program, you
must supply your own logical file names (for the printer plus any other files) on the
LFD job control statement.
When using any other Unisys routines (such as the data utility routines), specif the
standard system file names shown in the coding examples in the corresponding
publication.
These and other applications of the LFD statement are discussed in Sections 3 and 4.
C
2-8
7004 4623-000
Basic Concepts
Device Assignment Set Placement and Duration
There is no strict rule for the placement of a device assignment set in a job control
stream: simply place the device assignment set somewhere between the JOB
statement and the EXEC statement.
1/ JOB MYJOB
other job control statements
1/
1/
/1
II
DVC
VOL
LFD
LFD
50
12345
DSKFIL1
PAYROLL
other job control statements
II EXEC PROG1
I&
‘Where a multiple step job is concerned, just remember that a device assignment set
specified in one job step is normally effective for that step as well as any that follow.
Consider this example.
f
Job step 1
Ii JOB MYJOB
/1
Ii
1/
Ii
II
II
DVC
LFD
DVC
VOL
LBL
LFD
20
PRTFIL
90
T00001
TAPE1
PAYRATE
I
Device assignment sets for a print
file and a tape file. The assignments
are effective for job steps 1, 2,
and 3.
1/ EXEC PROG1
continued
7004 4623-000
2-9
Basic Concepts
Job step 2
If DVC 50
II VOL 1234A
/1 LBL DSKFIL1
II LFD PAYROL
Device assignment set for a disk
file. The assignment is effective for
job steps 2 and 3.
/1 EXEC PROG2
Job step 3
•
II EXEC PROG3
•
Any of the device assignments
specified in job steps 1 and
2 are effective for job step 3.
/&
In the preceding example, PROG1 can reference only PRTFIL and PAYRATE. It
cannot reference PAYROL. PROG2 and PROG3 can reference PRTFIL, PAYRATE,
and PAYROL.
Job Termination
There are two ways in which a job can terminate: normally or abnormally.
1.
Normal Termination
This is initiated by the control stream, the program, or the workstation or system
console operator. Generally, it occurs after the last job step, but it can also be
caused by the operator using the CANCEL or STOP operator command, or by the
program issuing a cancel instruction. If terminated by the CANCEL system
command or program instruction, the entire job terminates immediately. This
includes the currently executing job step plus all subsequent job steps (if any) in
the job. The STOP operator command terminates a job when the job step
currently executing is finished.
2.
Abnormal Termination
This is caused by program errors or by control stream errors (syntax order). If
caused by program errors, you can get a main storage printout (dump), which can
be used to debug your program, provided that you have placed an OPTION
DUMP statement in the control stream prior to the job step that caused
termination. The OPTION job control statement is covered in “Selecting Optional
Features” in Section 6. If caused by a control stream error, a message explaining
the error is displayed on the system console.
2-10
70044623-000
Basic Concepts
In anticipation of program errors, you may use the ABNORM=label parameter of the
EXEC statement. This parameter causes a skip forward in the job control stream so
that the job finishes executing and doesn’t terminate abnormally. If, however, the
operator issues a cancel instruction, the job terminates normally.
All terminations result in the deallocation of the system facilities (peripheral devices,
main storage, disk work areas, etc.) allocated to the job.
Restarting a Job
What if your job terminates abnormally specifically when your program is executing?
If the program only processes a few records, you can rerun the job from the beginning
without any great loss; but, if the program processes many records, rerunning the job
increases processing time and cost. To help avoid this, 05/3 provides a restart facility
that permits you to resume execution of your job from a particular job step or a
checkpoint record. See “Restarting a Job” in Section 6 for more information.
-
Branching within a Control Stream
When you write a program, you can set alternate paths for the program to take.
Normally, program statements execute consecutively in the order of their appearance.
However, it is often necessary to alter this normal sequence and skip forward to a
different point in the program this is called branching. Similarly, alternate paths
can be taken in job control streams. The SKIP and OPTION QUERY job control
statements allow you to skip forward in the job control stream during your program’s
execution to another job control statement. The ABNORM parameter of the EXEC job
control statement allows you to skip forward in the job control stream if your program
causes an abnormal termination. (See Section 6.)
-
You can also branch from one job control statement to another in a control stream by
using run-time conditional job control statements (they’re called run-time statements
because they are available and effective through the run symbiont). Run-time
conditional job control statements are interpreted and acted upon while the run
symbiont is scanning the control stream. They are not placed in the job’s $Y$RUN
file; their actions are completed when the run processor has acted upon them. Only
forward branches are allowed. The job control statements belonging to this category
are GO, IF, and NOP. They are explained in “Run-Time Conditional Job Control
Statements” in Section 7.
70044623-000
2-11
Basic Concepts
Jobs and Main Storage
After the supervisor is loaded into the system, the remaining main storage is available
to job control, symbionts (like the run processor and the job scheduler), your jobs,
shared code, and your programs. Naturally, the amount of available main storage
varies, depending on the jobs, symbionts, and programs executing at the time. Job
control assigns a portion of main storage to each job as the space becomes available.
The amount of main storage assigned is that which is needed to execute the largest job
step in the job. When ajob is completed, the space it occupied is returned to the
system.
Job Roll-Out/Roll-In
In “Considering Jobs for Execution the Job Scheduler” in Section 1, we mentioned
that the job scheduler considers jobs for execution by priority and the availability of
main storage and peripheral devices. In general, if the necessary main storage and
peripheral devices are not available, the job’s execution, regardless of its priority,
cannot begin. A different situation exists if roll-out (ROLLOUT=YES) is configured at
SYSGEN time.
-
With roll-out, high, normal, and low priority jobs are rolled out to disk to provide
enough main storage for preemptive jobs to be executed. When the preemptive
priority section of the job queue table is empty, the job scheduler rolls first the high,
then the normal, and last the low priority jobs back into main storage for execution.
Remember though, even if roll-out is configured, the peripheral devices needed for the
preemptive job must also be available; otherwise, roll-out does not occur.
Minimum and Maximum Main Storage
By minimum main storage size we mean the amount needed to successfully execute
the largest step of a job. The maximum size is the amount that can be used, if
available, to improve or speed up job step execution. As you’ll see in Section 4, you can
specifr the minimum and maximum main storage size on the JOB statement or on the
OPTION statement.
The total amount of main storage used by a job step also includes the size of the job
prologue. The prologue contains information (control tables) needed to regulate your
job. The size of the prologue, however, is automatically taken into consideration so
you don’t have to include it in any main storage size that you specify. Just keep in
mind that the job prologue is part of the true main storage requirement for a job. This
is illustrated in Figure 2-1.
C
2-12
70044623-000
-
Basic Concepts
JOB PREAMBLE
TASK CONTROL BLOCKS
JOB ACCOUNTING TABLE
—
JOB
PROLOGUE
LOCAL DATA AREA
SHARED CODE TABLE
DISK STORAGE
EXTENT INFORMATION
JOB REGION
LENGTH
SPOOL CONTROL TABLE AND BUFFERS
PHASE LOAD TABLES
SAT/CDM ACCELERATION
CODE
1— 1—
MINIMUM
LENGTH
JOB STEP LOAD MODULE
PROGRAM
AREA
MAXIMUM
LENGTH
Figure 2-1. Job Region in Main Storage
Dynamic Expansion of Main Storage
Your job may require dynamic expansion of its initial main storage allocation to load
software modules (data management modules, for example), or to accommodate other
program modules called by your job. This capacity for dynamic expansion of the
job
region is called the DLOAI) facility. For more information about this facility, see
“Defining Software Facilities Needed by Your Job” in Section 6.
70044623-000
.
2.13
C)
C’
Section 3
Minimum Control Stream Requirements
What Is a Minimum Control Stream?
A minimum control stream consists of only those job control statements needed to
properly direct the execution of a job.
Let’s assume you want to execute a program that has been compiled, link edited, and
stored in a library. This particular program does not use any input (cards, tape, disk,
etc.) and the only output is directed to the printer. The purpose of the program is to
print constants on adhesive-backed mailing labels, like this:
NAME
ADDRESS
CITY
STATE
ZIP CODE
Granted, this isn’t a widely used application, but it illustrates a bare minimum control
stream.
Constructing the Minimum Control Stream
In order to run this program, we have to construct a control stream to tell the
operating system what to do with it. Since the needs of the program are simple, we
need very few job control statements.
70044623-000
3-1
Minimum Control Stream Requirements
First, a JOB statement is needed to indicate the beginning of the job to the operating
system. Every job entering the system must start with a JOB statement. Each job
step does not need a JOB statement, only one for the job as a whole. Next, since there
is a print output, a DVC statement is needed to assign a printer to the job. And
finally, every peripheral device we use has a file associated with it; every file needs a
file name. An LFD statement provides the file name.
The DVC and LFD job control statements make up a basic device assignment set.
Since the printer is the only peripheral device used by our program, no other device
assignment sets are required.
In fact, there are no other processing options needed for this program. We are now
ready to initiate the execution of the job step (our entire job consists of only one job
step). We need an EXEC statement for this.
Now our program has all the job control statements that it needs to function. But,
when it is finished, we have to tell the system that our control stream is finished. We
need a /& job control statement.
Briefly, we have indicated all the job control statements needed for this simple
prOgram. They are:
•
JOB
•
DVC
•
LFD
•
EXEC
•
/&
We will cover each of these job control statements in its proper sequence. We will
show all the parameters available for these job control statements, but, at first, only
those parameters that are required will be described, along with any parameters that
are generated by default. The optional parameters will be introduced into the
discussion ofjob control at the appropriate time.
But, before we start our control stream, you should read the statement conventions in
Appendix A. They explain how the job control statements are presented in text (how
you can tell which parameters are optional, which are required, how a default option
is shown, etc.) and how you code them.
C.,
3-2
70044623-000
Minimum Control Stream Requirements
The Beginning of the Job
The JOB job control statement is the first job control statement that you need. Its
format is:
Iltsyirbot) JOB jobname
,
P
t,min)t,max)[
tasks
max-time
L’
E,print-option-tist]E,acc-no][,nXm]
]L
,
ACT
LOG
NOACT
NOLOG
NONE
BOTH
[NOHDR
L
As you can see, it has quite a number of parameters. You can specifr the name of the
job, the priority, how much main storage is needed, the amount of tasks in ajob step,
how long the job should take, special information for display on the system console,
accounting information, spooling buffer size, and log information (where your
accounting record is kept).
The only parameter we are interested in right now is the jobname parameter, and any
default parameters (shown by shading) that are generated.
The jobname parameter does just what it implies: it names the job. It consists of one
to eight alphanumeric characters. Do not hyphenate the job name if you plan to save
the job. The save processor does not allow or recognize hyphens.
For example, we assign the name POCO to the job. It’s coded as:
II JOB POCO
By default, the job has a normal scheduling priority (N) and one task (1).
There is a special feature of the jobname parameter that helps you ensure that unique
job names are always assigned you can use trailing ampersands (&) in the job name.
You could, for example, code:
-
If JOB POCO&&&&
When the stream is processed, the system replaces the ampersands with unique
numbers.
7004 4623-000
3-3
Minimum Control Stream Requirements
When would you use this feature? If you have ajob control stream (POCO for
example) that is used frequently by different personnel perhaps even concurrently
from workstations all the users could use POCO&&&& and be assured of having
unique job names assigned. It is recommended that if you use this feature, you use at
least three trailing ampersands.
C
-
-
You can override the parameters specified on the JOB control statement through
selected features of the OPTION job control statement, which is explained in
“Selecting Optional Featured
t in Section 6.
Identifying the Devices
The next entry needed in the control stream is for the printer. The DVC job control
statement is used to request the assignment of peripheral devices to a job. Its format
is:
//Esymbot] DVC
fnnn[(n)]
RES
RUN
,
addr
OPT
IGNORE
ALT
0
REQ[(n)]
REAL
The DVC job control statement specifies the logical unit number associated with a
peripheral device type. It can also be used to assign alternate devices, or to specif5r
that the job should be executed even if the requested devices are unavailable.
Here, again, we are only interested in the required parameter specifring the logical
unit number. There are no default parameters.
The nnn is a decimal number indicating the logical unit number of the device. By
looking at the following information (taken from Table A-i of the Job Control
Programming Reference Manual, UP-9984) we see that the category for printers is in
the range of 14-29. If we are willing to use any printer that is available, we use logical
unit number 20 or 21. But, it just so happens that there also are a 0776 printer
subsystem and a 9246 printer subsystem available in the system. Our program uses a
special character that is only present on the 0776 printer, so we will use logical unit
number 24.
C.
3-4
7004 4623-000
Minimum Control Stream Requirements
Logical
Unit No.
20, 21
22, 23
24, 25
Device Type
Code
04FF0000
04400000
04100000
Device Type and Features
Any printer, no features specified
9246 printer, no features specified
0776 printer, no optional features
You’ll notice that there are two other choices for the DVC job control statement: RES
and RUN. They will be discussed and used in later examples.
We can now add the DVC job control statement to our control stream as follows:
1/ JOB POCO
//
Dvc.24
Notes:
1.
The (n)portion of the nnn parameter is used only when the logical unit number
indicates a workstation device.
2.
Logical unit numbers can be changed at system generation (SYBGEN) time to suit
the needs of a particular installation. You must be aware of any changes because
they could cause device assignment problems within your control stream, especially
ifyou’re using JPROCS supplied by Unisys.
Assigning a Logical File Name to the File
Every device assignment set in the control stream ends with the LFD job control
statement. This associates the file defined in the program with the file information in
the control stream. Its format is:
/ItsymboL] LFD
ffilename
l*fi1enI
r
r
1
i
11]
L
,
EXTEND
INIT
PREP
ID
IGNORE
The LFD job control statement specifies the file name of the file. It’s also used to
reserve main storage for information about disk file extents, write over the
information of the file, and add to the data already in the file.
70044623.000
3.5
Minimum Control Stream Requirements
The filename parameter specifies the name of the file you are going to use, and must
correspond to the name given to the file in the program. The file name for the LFD job
control statement is determined in the following manner:
•
The basic assembly language (BAL) programmer uses the name in the label field
of the file definition macroinstruction.
•
The COBOL programmer uses the external name from the SELECT entry in the
environment division. (If the external name is omitted in COBOL 68, use the file
name from the SELECT entry.)
•
The FORTRAN IV programmer uses the device number from the READ or
WRITE statement, prefixed by FORT.
•
The FORTRAN27 programmer uses FO followed by the unit number unless a
specific name was specified in the OPEN statement.
•
The RPG II programmer uses the file name from the file description specification.
The filename parameter is normally one to eight alphanumeric characters, but if you
are using a data management file, it has a maximum of seven characters. This is
because data management allows only one to seven characters in the label field of the
file defmition macroinstruction.
If an asterisk is placed in front of the file name on the LFD job control statement, it
means this is an input-only file; you cannot write on it. The operator should be
notified of this so he can take appropriate action.
For our control stream example, we’ll assume our program is a COBOL program. The
file name for the printer in the FD entry is WRITEOUT. We can now add the LFD job
control statement to our control stream.
II JOB POCO
II DVC 24
it.
Executing the Program
We have defined all the requirements of the program the operating system. Now
we have to provide a job control statement to call the sorted program from a library
and initiate execution. This is done with an EXEC job control statement. Before the
program is actually loaded, any outstanding tape and disk mounting requests are
completed.
3-6
70044623-000
C
Minimum Control Stream Requirements
The format of the EXEC job control statement is:
/I[symbot] EXEC programname[
I
[Library-name t,switch-priority]E,ABN0RM=tabeL]
$Y$RUN
L
The EXEC job control statement identifies the name of the load module. It is also
used to specify the library containing the load module, the task switching priority, and
any action to be taken if the program causes an abnormal termination.
Once more, we are only interested in the required parameter and any default
parameters generated.
The program-name parameter identifies the load module to be executed. Every
program that is successfully compiled and link edited creates a load module. Every
load module that is created and every routine supplied by Unisys must have a name.
The LOADM linkage editor control statement assigns a name to a load module; the
EXEC job control statement calls the load module by a program name. These names
must agree.
For example, you link edit your program with the module name TESTR on the
LOADM linkage editor control statement. The linkage editor creates the load module
with the name TESTR. When you want to execute this program, your EXEC job
control statement uses this same name: TESTR.
If, when you link edit your object module, you do not use a LOADM linkage editor
control statement, the load module name, by default, is LNKLOD.
Assume that this program is stored in a library from which it can be retrieved as many
times as needed. When the program was link edited, the linkage editor was
instructed to place the load module in a specific, permanent library; otherwise, it
automatically would have been placed in the job’s $Y$RUN file, which is only a
temporary file. Assume it is located in the system load library file ($Y$LOD), and the
load module name is LABELS. Since $Y$LOD is the default parameter generated for
the load library, we only need to specify the program name, which is the same as the
load module name: LABELS.
We can now add the EXEC job control statement to our control stream as follows:
II JOB POCO
1/ DVC 24
1/ LFD WRITEOUT
JLEX.EC;.iABEL
By default, the lowest available task switching priority established at system
generation time is used.
7004 4623-000
3.7
Minimum Control Stream Requirements
Ending the Basic Control Stream
So far, we have provided all the job control statements needed to construct a basic
control stream: JOB, DVC, LFD, and EXEC.
This control stream is all the system needs to execute our simple program. But, after
the program executes, the system returns to job control to obtain the next job control
statement. Because the job is finished, a /& job control statement is used to signal the
end of the job. Its format is:
/&
This statement has no parameters, but it can have comments. These comments have
no effect on the system; they only provide a means of annotation. The comments must
be separated from the /& job control statement by at least one blank column.
The statement conventions for coding more than one job control statement on a line
(multistatement coding) are presented in Appendix A. The I& job control statement,
however, must be the only job control statement on a line.
Adding the /& job control statement, along with some comments, our control stream
looks like this:
1/
If
1/
II
JOB POCO
DVC 24
LFD WRITEOUT
EXEC LABELS
Ending the Card Reader Operation
We have signaled the system we are finished processing. Now, we have to terminate
the card reader operation this informs the system that there are no more cards
associated with the job. We do this with a FIN job control statement. Its format is:
-
//Esymbot] FIN
There are no parameters.
We can now add a FIN job control statement to our control stream, as in the following
example:
II
II
II
II
JOB POCO
DVC 24
LFD WRITEOUT
EXEC LABELS
/&
END-OF-LABEL-JOB
/1 FIN
C
3-8
70044623-000
Minimum Control Stream Requirements
The FIN job control statement also signals the end of card input when merging job
control statements with stored control streams, submitting data cards as input for a
stored control stream, or storing a complete control stream.
Note:
Using the FINjob control statement is unnecessary when input is on data-setlabel diskette or in the input spool file.
The Control Stream So Far A Review
-
We have defined everything the system needs to know about the job. It has been
given a name, the system was instructed what load module to use, and the job has
been assigned the peripheral device it needs. The program is ready for execution.
This control stream represents only a minimum application. We have only scratched
the surface of the capabilities of the OS/3 job control. Throughout the rest of this
guide, we are going to build on this minimum control stream by adding and modifying
job control statements.
Let’s assume that the program with a load module name of LABELS was recompiled
and link edited after it was modified to accept input from the card reader. This new
input contains name and address information that will be printed on the adhesivebacked labels along with the constant information as shown in the following sample.
NAME
JOHN A. SMITH
ADDRESS
143 S. 52ND. ST.
CITY
ZIP CODE
7004 4623-000
HOMETOWN
STATE
PA.
18908
3-9
Minimum Control Stream Requirements
Adding Card Input
Since the job will now accept card input, we must provide a device assignment set for
the card reader. This means we have to insert a DVC and LFD job control statement
for the card reader into the control stream. Once again, their formats are:
//[syTnbot] DVC
[nnn[(n)]
RES
RUN
-
//[syiot] IFO
FfiLenne G
G*fitenameF
,
addr
[,HOSThost-id]
OPT
IGNORE
ALT
0
REQ[(n)]
REAL
fnl[
Ij 181]!
L
L
EXTEND
PREP
ID
IGNORE
The following section of Table A-i in the Job Control Programming Reference,
UP-9984, indicates that the category for card readers is 30-35.
LogicaL
Unit No.
Device Type
Code
30, 31
32, 33
34, 35
08FF0000
08200000
08800000
Device Type and Features
Any card reader subsystem, no features specified
0719 card reader, no features specified
0716 card reader, no features specified
For this example, we will assume the system you’re using has only one card reader, an
0719 card reader. For a logical unit number, there are four alternatives. We can use
32 or 33, which assigns a 0719 card reader specifically, or, since the 0719 card reader
is the only one we have, we can use 30 or 31, which allows us to use any available card
reader.
3-10
70044623-000
Minimum Control Stream Requirements
If the system had two card readers, both of a different type, and a particular card
reader is needed, you must be more specific in your assignment. If it’s immaterial
which card reader is used, you could assign the logical unit number for any card
reader (30 or 31).
A point to remember about logical unit numbers: if you don’t care about the specific
device type, use the logical unit number that assigns any device within the category
(20 and 21 for printer, 30 and 31 for card readers, etc.). In that way, if there is more
than one type of device, you get the first one available. For instance, suppose you
selected logical unit number 25 (Unisys 0776 Printer Subsystem) but there is also a
0770 printer connected to the system. The 0776 printer has 40,000 lines waiting to
print, while the 0770 printer has a backlog of only 500 lines. By specifying only the
0776 printer, you must wait for the other 40,000 lines to finish printing. By specif’ing
any printer, the output is sent to the first available printer. The logical unit number
we are going to use for the card reader is 30.
Note:
When requesting the assignment of more than one device of the same type (two
printers, for example), be sure you request the assignment of any specific
devices you need before you request the assignment ofgeneral ones. This
ensures that a specific device you may need (the 0770 printer, for example)
will not be allocated for use as a general printer when it’s needed as a specific
device.
Now that we have a DVC job control statement for the card reader, we need a
corresponding LFD job control statement. Since this program is written in COBOL,
we check the SELECT entry in the COBOL program and find that the file name is
CAROIN. This file name is coded in the LFD job control statement.
We can now add the device assignment set for the card reader to the control stream.
It can be placed anywhere in the control stream, with the following restrictions:
•
It must be before the EXEC job control statement.
•
It cannot be within embedded data.
•
It cannot be within the device assignment set (DVC through LFD sequence) for
another device.
7004 4623-000
3-11
Minimum Control Stream Requirements
Card Input and Embedded Data
To accept data input from a card reader, we must inform the card reader in some way
that it is data to be read. In many cases, this data is caused to be read at execution
time by data management. In this kind of application, the data cards follow the II FIN
card that caused the card reader to be turned off previously. All that is additionally
needed is a /* card after the data signifying end of data. There are no other
parameters required, and no comments are permitted in the comment area of the
card. This /* statement is always required for any type data. Thus, to our control
stream we can now add the data, followed by the 1* end-of-data statement, and run
our job, which consists of the LABELS program. Basically, we are saying to the
processor, run my job POCO which executes the program called LABELS my data is
a card file after the FIN statement when you are ready to execute. This will print the
name and address information, plus constants, as shown, on adhesive-backed labels
that the operator has previously placed in his printer. The following example
illustrates this control stream:
-
II JOB POCO
// DVC 24
If LFD WRITEO(JT
II DVC 30
II LFD CARDIN
II EXEC LABELS
/&
II FIN
data
END-oF-LABEL-JOB
cards
1*
Note:
You should be aware, however, that in the case of multiple files, if the first
program in the series does not read all of its data cards (along with the 1*
that signals end of data), the next program step will pick up where the
previous one left off. Additionally, ifyou are programming in higher level
languages, such as RPG, COBOL, or FORTRAN, you cannot read multiple
card files in a single program without closing and reopening the files.
Another way in which data cards may be accepted, and which informs the card reader
that data is being input, is the embedded data method. This means that the data is
embedded within the control stream itself. All it requires is a start-of-data (1$) job
control statement immediately after the EXEC statement, followed by the data and
the /* end-of-data. 1$ has no parameters, and may appear as the last job control
statement on a muitistatement line.
3-12
70044623-000
Minimum Control Stream Requirements
The advantage of this method is that the device assignment set is no longer required
for the reader, since the control stream is already being read. Additionally, the data
being read is instantly accessible, which is discussed later in Section 8. A
disadvantage is that embedded data in a prefiled job control stream is harder to
change than the data in a card file (which follows the II FIN job control statement).
This is because the embedded data is actually a part of your control stream rather
than a separate card file. Changing embedded data is discussed in “Substituting
Embedded Data” and “Replacing Embedded Data Sets in Expanded Control Streams”
in Section 6. An example of an embedded data control stream is:
II
1/
II
II
Is
JOB POCO
DVC 24
LFD WRITEOUT
EXEC LABELS
data-cards
1*
/&
END-OF-LABEL-JOB
If FIN
You can use this method when you become familiar with the programming techniques
needed by the language you’re using for example, a COBOL ACCEPT or FORTRAN
READ instruction. In fact, programs supplied by Unisys (such as the COBOL
compiler and the data utility routines) use this method. It entails the use of a
supervisor macroinstruction in the program (if it’s assembler language; if it’s one of
the other languages, there are similar instructions that are used). Again, if you decide
to use the embedded data method, the changes to your job control stream are:
-
1.
Remove the device assignment set for the card reader; it’s not needed.
2.
Place the data (1$, data cards, /*) after the EXEC job control statement. This is
what’s known as embedded data.
When you use the embedded data method, and you have an 0716 card reader
supporting the 96-column card feature, your data file can use the full 96 characters.
With data-set-label diskette, you can use up to 128 characters. But, even though your
control statements also can be on 96-column cards and data-set-label diskette, only
the first 72 columns (characters) can be used for job control statements.
In addition to embedded data, there is a dummy data set. A dummy data set consists
of only a 1$ and a /* This is used with some language JPROCS. More information
about dummy data sets can be found in the language manuals (COBOL, FORTRAN,
etc.).
You can replace embedded data sets in translated, saved job control streams by using
the DATA STEP job control statement. Refer to “Dynamic Skip Function from a
Workstation” in Section 6 for more information.
70044623-000
3-13
Minimum Control Stream Requirements
The Program Is Changed Another Device
-
So far, the program has been written to read name and address cards and print the
information, plus constants, on adhesive-backed labels. The program has been refined
once more. It is still going to print constants. However, the name and address file is
now on magnetic tape, in ZIP Code® sequence. This tape was created by someone
else’s job. We want to list only the name and addresses of certain ZIP Codes;
therefore, we modif5r the program to accept a table from the card reader. This table
contains only the ZIP Codes we want to print. The program instructs the system to
compare the ZIP Codes from the table with the file on the magnetic tape and print the
names and addresses that match the ZIP Code table.
We have already provided the device assignment sets for the printer and the card
reader. Even though the format of the card reader input is different (previously it was
the name and address file, now it is the ZIP Code table), no changes are needed to the
card reader device assignment set. It was a program change and does not affect the
job control stream. The logical unit number is still 30 (DVC job control statement),
and the file name in the program is still CARDIN (LFD job control statement). The
only new item we have to provide in the control stream is a device assignment set for
tape.
What Is Needed to Use a Tape?
We have already said that every peripheral device used needs the DVC and LFD job
control statements. For readers, printers, and punches, this is all that is needed to
complete the device assignment set. However, magnetic tapes have volume serial
numbers, and, optionally, file identifiers. So, the device assignment set for a tape file
could be either
/1 DVC
II VOL
II LFD
...
or
II
II
If
if
DVC
VOL
LBL
LFD...
The first step is to provide a logical unit number and file name.
ZIP Code is a registered trademark of the U. S. Postal Service.
3-14
70044623-000
Minimum Control Stream Requirements
The Logical Unit Number and File Name for the Tape
The range of logical unit number for magnetic tapes is 90-127. The name and address
tape is a 9-track, phase-encoded tape. We must be specific. The logical unit number
selected for the DVC job control statement is 100. This gives us any tape drive that
can read a 9-track, phase-encoded tape; the tape unit transfer rate is immaterial.
We can now add this partial device assignment set for tape to our control stream.
If
If
II
If
II
If
Ii
II
JOB
DVC
LFD
DVC
LFD
DVC
POCO
24
WRITEOIJT
30
CARDIN
1
D NANADD
EXEC LABELS
/&
END-OF-LABEL-JOB
II FIN
1*
These new DVC and LFD job control statements do not represent the entire device
assignment set needed for tape. If we tried to run the job now, it would abort.
Supplying a Volume Serial Number for the Tape
Every tape file used in ajob must have a VOL job control statement in the device
assignment set. This identifies the volume to be used. Its format is:
/I[synibol) VOL
F
voLsri-1I
Mcc
N
NMcc
I
voL sri-i
I
I
L
(NS)
(NOV)
L(PREP)
7004 4623-000
(NS)
(NOV)
(PREP)
VOLSfl-2r
L
SCRATCH
.c$>
SCRATCH
—
VOLSflr
2
ICS)
J(NS)
I
I
1(NOV)
L
[(PREP)
VOLSfl-3r
I
I
(NS)
(NOV)
(PREP)
L
(NS)
(NOV)
(PREP)
SCRATCH
3-15
Minimum Control Stream Requirements
The VOL job control statement supplies the volume serial number of the volume to be
accessed by the job. However, a tape volume does not necessarily need a volume serial
number, but it still must have a VOL job control statement.
You can also use the VOL job control statement to: count the number of blocks in the
file; specify the mode characteristics of the tape; request data management to write a
volume serial number; inhibit the checking of volume serial numbers if they are not
known; or, to indicate that the volume may also be used by someone else at the same
time that you are using it (this only applies to disk).
Again we are only interested in the required parameter. This parameter has several
different options, but for this job, only the volume serial number is needed.
The volsn-1 parameter is the 1- to 6-alphanumeric-character volume serial number of
the first volume of the file. A file may span more than one volume. Perhaps the
length of the file made it necessary to use three tapes (volumes) to hold the entire file.
Since this file is on only one volume, only one volume serial number is needed.
Assume it to be TAP111.
We can now add the VOL job control statement to our control stream as follows:
II JOB
// DVC
/1 LFD
If DVC
1/ LFD
Ii DVC
I!
II
II
1/
II
POCO
24
WRITEOUT
30
CARDIN
100
VOl. TAPIII
LFD NAMADD
EXEC LABELS
f&
END-OF-LABEL-JOB
FIN
data cards
1*
This control stream could now be run, provided that the tape is unlabeled (no file
identifier).
OS/3 data management supports a maximum of 151 explicit volume names per file for
disk, diskette, and tape files.
C
3-16
7004 4623-000
Minimum Control Stream Requirements
Labeled Tapes for File Identification
Just as there can be one or more volumes in a file, there can also be one or more files
in a volume. Suppose the tape volume contained five files. It would be necessary to
have file identifiers on each particular file to access the proper file. Single-file tape
volumes also can have file identifiers. This is done to ensure that the correct file is
used. Even though the volume serial number is checked to see if the proper tape is
mounted, it is possible that this tape does not have the proper file needed for the job.
For example, someone could have inadvertently written on the tape because it did not
have a file identifier to indicate that this tape already contains information to be
saved. By using a file identifier, you indicate this is a saved tape. Had there been a file
identifier on the tape, anyone trying to write on this tape would have been notified
that this is a saved tape.
The LBL job control statement is used to either check or create a file identifier. Its
format is:
//tsymbol] LBL.
[fiLe-identifier
1
l’fiLe-intifier ‘
ifiLe-seriaL-number1[,exptrat1on-date)
t 1VCHEcK
f]
L
e.number}
1
[Icreation.date][ {fi1e.seuenc
[
[
.[9eneration.number}1
ber}
1
{version.num
The LBL job control statement identifies the file. It also can be used to: ensure that
the correct members of a multivolume file are used; indicate the date the file can be
deleted (by a SCR job control statement); indicate the date the file was created;
indicate the position of the file in respect to the other files in a multifile tape vohime;
and, specify the generation and version number of a tape file, thus ensuring the most
current edition of the tape file is used.
We only want to ensure that the proper file is on the tape volume, so we need only the
required parameter.
The file-identifier parameter is 1 to 17 alphanumeric characters for tape, card, and
diskette files. It is 1 to 44 alphanumeric characters for a disk file unless that file is a
scratch (temporary) file; then the file-identifier is 1 to 39 alphanumeric characters. If
the file-identifier contains embedded blanks, it must be enclosed by single quotation
marks.
70044623-000
3-17
Minimum Control Stream Requirements
Assume that MASTERFILE is the file identifier assigned to this tape file when it was
created. We can now add the LBL job control statement to the control stream as
shown in the example.
1/ JOB POCO
// DVC 24
1/ LFD WRITEOUT
II DVC 3
/1 LFD CARDIN
// DVC 100
/1 VOL TAP111
/1 B1. ?4ASTERFIL
1/ LFD NAMADD
f/ EXEC LABELS
f&
END-OF-LABEL-JOB
II FIN
data cards
1*
The default parameters generated indicate this is the only file on the volume (1), and
it is the only edition of the file (1).
Note:
File identifiers prefixed by $SCR refer to job step temporary files; those
prefixed by $JOB refer to job temporary files.
Another Programming Change Another Device
Assignment
-
The site manager has determined the label program doesn’t fulfill all the
requirements for which it was intended. Once more, it must be changed.
The name and address file was copied from the tape volume to a disk volume by using
a Unisys data utility routine. Now, the input name and address file is on disk, the ZIP
Code table is still input from the card reader, and the selected names and addresses,
plus constants, are still printed on adhesive-backed labels. These selected names and
addresses are now going to be saved and output to a file on a tape volume for a later
processing application.
Although there may be many programming changes involved, the control stream
changes are minimal.
C
3-18
70044623-000
Minimum Control Stream Requirements
The device assignment set for the card reader, the printer, or the tape doesn’t need
changing. Even though the tape was used previously as an input file, converting it to
an output file is only going to involve changes in the program; it is not reflected in the
control stream. After the tape was copied to disk,, the information it contained was
deleted in another procedure. We can use this tape with a volume serial number of
TAP111 as the output tape. We can also use the same logical unit number in the DVC
job control statement. NAMADD is used as the file name for the output tape file in
the program. This allows us to continue using NAMADD as the file name in the LFD
job control statement. However, we are going to give this tape file a different file
identifier. In the previous device assignment set for the tape it was MASTERFILE.
We want to change it to reflect its purpose.
It is no longer a master file for input; it is an output tape let’s call it OUTPUTTAPE.
This requires a change to the file-identifier parameter of the LBL job control
statement for the tape device assignment set. We do not need to change it, but to
make the purpose and the name agree, we will. Changing the LBL job control
statement makes our control stream look like this:
-
II JOB POCO
II DVC 24
II LFD WRITEOIJT
II DVC 30
II LFD CARDIN
II DVC 100
II VOL TAP11l
ZL1.]E..LO1JTPUTTAPE
II LFD NAMADD
II EXEC LABELS
/&
END0F-LABEL-JOB
If FIN
data-cards
1*
We still must provide a device assignment set for the name and address file input
from disk.
70044623-000
3-19
Minimum Control Stream Requirements
The Device Assignment Set for a Disk or Format-Label Diskette
The following chart lists the necessary job control statements for the basic disk and
format-label diskette device assignment set.
SYSRES
or
$Y$RUN File
(Disk only*)
Allocation
Your
Disk or
FormatLabel
Diskette
Previously
Allocated
DVC
VOL
LBL
LFD
DVC
LBL
LFD
DVC
VOL
LBL
EXT
LFD
DVC
LBL
EXT
LFD
Not
Allocated
.
*A format-label diskette volume cannot be used as your SYSRES volume or the volume containing the $Y$RUN file.
In our case we have a disk file, the extent was allocated, and the file is not SYSRES or
the job’s $Y$RUN file. So the following job control statements are needed: DVC, VOL,
LBL, and LFD.
The disk pack used for the name and address file fits on an 8494 Disk Subsystem.
The logical unit number we are going to use for the DVC job control statement is 80.
Within the program, the file name from the FD entry is DKNAME. This is the file
name for our LFD job control statement.
We need a VOL job control statement to indicate the volume serial number of the disk
we are going to use. We need only the required parameter for the volume serial
number. Assume the site manager had the name and address file copied to the disk
with a volume serial number of DSKOO1.
Since most disk volumes contain many files, each file needs a file identffier. When the
site manager copied this file, he allocated it with a file identifier of DSKMASTFIL.
We must specify this in an LBL job control statement.
3-20
7004 4623.000
Minimum Control Stream Requirements
We now have all the information needed for the disk file. We can add the device
assignment set for the disk input file to our control stream and run the job.
JOB POCO
DVC 24
LFD WRITEOUT
DVC 30
// LFD CARDIN
/1 DVC 100
Ii VOL TAP111
If LBL OIJTPUTTAPE
Ii LFD NAMADD
ii DVC 80
II VOL OSKOOI
II LBL DSKMASTFIL
It1LFi..: PK1.AM
II EXEC LABELS
/&
II FIN
data-cards
/1
II
II
II
END-OF-LABEL-JOB
1*
The Device Assignment Set for Data-Set-Label Diskette
The prep routine for data-set-label diskette automatically allocates the entire diskette
for one file and assigns a file identifier of DATA (unless you specifSr otherwise). When
this file is used, you must include a device assignment set in your job control stream
that consists of the DVC, VOL, LBL, and LFD job control statements. For example:
II
II
II
II
DVC
VOL
LBL
LFD
130
DSLO1
DATA
FILE01
You only include an EXT statement in the device assignment set (and specifSr your
own identifier on the LBL statement) if the space wasn’t already allocated during the
diskette prep routine. See “Information about Data-Set-Label Diskette File
Allocation” in Section 4 for information about the EXT statement.
The Device Assignment Set for a Workstation
The DVC and LFD job control statements are required for a basic workstation device
assignment set. The UID, USE SFS, USE DP, and USE MENU statements are
included under certain circumstances.
70044623-000
3-21
Minimum Control Stream Requirements
The UID Job Control Statement
The UID job control statement may be used as part of the device assignment set for a
workstation when you want to ensure that specific workstations, identified by user-id
or device address, are automatically connected to a job. This is done before a job’s
execution begins (if the workstation has not already been connected via the
CONNECT command). Its format is:
//[synoL] UID
user-id-i
(addr-i)
user- Id- i(addr-1)
[...,
L
I
[user-id-255
(addr-255)
user- id-255(addr-255)
A maximum of 255 workstations may be specified. You can specify $Y$MAS as a
user-id to assign the job’s master workstation to a job. The user-id parameter is one to
six alphanumeric characters in length. A device assignment set that assigns the
workstation being used by user-id (JONES1) could look like this:
•
ii DVC 200
Ii UID JONES1
ii LFD WKSTN
Assigning workstations is discussed in more detail in “Specifying Multiple
Workstations” in Section 4.
The USE Job Control Statement
If you are preparing a control stream for a program that uses screen format services,
menu services, or the dialog processor, you must include a USE job control statement
as part of your workstation device assignment set. Three different forms of the USE
statement make it possible for you to specify which workstation service you want.
These are as follows:
/1 USE SFS...
Ii USE MENU...
If USE DP...
(for screen format services)
(for menu services)
(for dialog processing)
Each statement and its accompanying parameters is discussed further in Section 6 in
“The USE Statement for Screen Format Services”, “The USE Statement for Menu
Services”, and “The USE Statement for Dialog Processing”, respectively.
C
3-22
7004 4623-000
Minimum Control Stream Requirements
Job Step Temporary and Job Temporary Files
To satisfSr the needs of the software components for disk work areas, files lasting for a
job step and for the length of the job are provided. These files are deleted at the end of
the job step or the end of the job. While these files are primarily used by the software
components, the ability to allocate and use temporary files is also available to you.
Basically, you allocate job step temporary and job temporary files the same way you’d
allocate any disk file. The only difference is you must prefix your file identifier with
$SCR for a job step temporary file and $JOB for a job temporary file. For example, to
allocate ajob step temporary file, you could include the following device assignment
set in your job control stream:
II
II
If
Ii
/1
DVC
VOL
LBL
EXT
LFD
50
D12345
$SCRWORK1
MI,,,CYL,2
WORKFIL
When a temporary work file ($SCR, $JOB) is allocated, the file label is modified by job
control to allow concurrent jobs using the same file identifiers to access the proper
work file. Every job in the system is assigned a unique job number. The label $SCR1
in JCL is allocated as $SCRnnnnnl, where nnnnn is the job number.
Job step temporary files are automatically deleted at the end of the job step, while job
temporary files are automatically deleted at the end of the job. If the system is
reinitialized in the middle of your job, job control automatically scratches job
temporary files and job step temporary files when it reallocates them.
See “Setting Up Temporary Work Files” in Section 5 for information about using
JPROCS to allocate job step temporary and job temporary files.
Basic Job Control Statements
This section has covered the job control statements needed to run most jobs. In the
following section, we are goiPg to use the basic job control statements and add the
optional parameters, explaining how each parameter affects the performance of the
job.
70044623-000
3-23
0
0
Section 4
Getting the Most Out of the Basic Job
Control Statements
Optional Parameters Can Improve Job Performance
So far, in our discussions of basic job control statements, we’ve concentrated on the
required parameters. A great deal of work can be accomplished using just these
parameters. Sometimes, however, required parameters won’t provide enough
information. In other instances, the ability to provide more information to the system
will speed up job execution. Additional information about a job and its peripheral
devices is supplied via the optional parameters that are part of the basic job control
statements. This section describes these parameters and shows how they are used.
Improving Your Control of the Job
The JOB control statement was used to give a name to the job. It is used also to
specif5r the following: a selection priority; the main storage size for the job; how many
tasks are in any one job step; how long the job should take; a list of the control streams
on the operator’s system console for debugging purposes; and spooling buffer sizes.
Once again, its format is:
//[symbot) JOB jobname
,
P
[,min][,max][
max-time
tasks
JL
L’
[,print-option-List)E,acc-no][,nXm]
70044623-000
,
ACT
LOG
NOACT
NOLOG
NONE
BOTH
[NOHDR
L.
4-1
Getting the Most Out of the Basic Job Control Statements
As you can see, some optional parameters generate default values when they are
omitted. In the previous discussion of the JOB control statement, only the required
parameter -jobname was coded. By so doing, we indicated that, by default, the job is
to have normal priority (N) and there is only one task (1). This points up the fact that
when only the required parameters are specified, you are, in many cases, providing
more information about the job than is contained in the required parameters. The
default values were selected because they conform to the most frequently used
programming practices. This allows you to code as short a control statement as
possible. The less there is to code, the less chance there is of making a coding error.
-
Note:
The OPTIONjob control statement can be used to override individual
parameters of the JOB control statement. Refer to “Selecting Optional
Features” in Section 6 for more information.
A Selection Priority for the Job
Jobs are selected for execution on a priority basis. The second parameter on the JOB
control statement specifies the priority. There are four priorities: low (L), normal (N),
high (H), and preemptive (P). Remember our discussion on the use of priorities in
Section 1, where we outlined how the priority is used by the system for selecting jobs
and what each priority means?
Most jobs are normal priority, which is by default, the parameter generated. If you
need another priority, you have to specify it.
It so happens that the label job named POCO is needed in a hurry, so the system
administrator allowed you to assign high priority. Added to the existing JOB control
statement, it would be coded as:
/1 JOB POCO,H.
Main Storage Needs
When the load module named on the 1/EXECUTE statement is in a load library on a
mounted disk volume, you don’t have to indicate the minimum amount of main
storage to execute the load module. If the disk volume containing the load module is
not already mounted, you must indicate the minimum amount of main storage needed
to execute the module.
The mm parameter does this. The minimum main storage size is specified in decimal
or hexadecimal. The smallest amount that can be specified is 8K decimal bytes (2000
in hexadecimal). The area used by the job prologue is not included in this amount.
C
42
70044623-000
Getting the Most Out of the Basic Job Control Statements
Assume the label program needs approximately 12K (12,288) decimal bytes (3000 in
hexadecimal) and that it’s in a load library on your own volume. The JOB control
statement would now be:
If JOB POCO,H,3000
or
II JOB POCO,H,X’3000’
You can also specify the minimum main storage size in decimal. This is done by
coding D’number’ for the mm parameter, as illustrated in the following JOB control
statement:
If JOB POCO,H,D’12288’
For the sake of illustrating the omission of positional parameters, this JOB control
statement is coded as follows when the priority is omitted (it would be assigned the
normal priority, by default, by the system):
II JOB PoCo,,3ØØ
See “Coding Conventions” in Appendix A for information about coding numbers in job
control statements.
Note:
Ifa job consists of multiple job steps, specify only the minimum main storage
size needed by the largest load module.
Consider the possibility that you may be running a 3-job-step job, consisting of
perhaps a COBOL compile, followed by a link edit, and then the execution of the
generated load module. 05/3 knows how much main storage to allocate for both the
COBOL compiler and the linkage editor, but there is no way OS/3 can know how much
is required for the execution of your program, since it is not generated until after all
the job control has been interpreted. If yeur generated load module does not use more
main storage than the COBOL compiler (which is larger than the linkage editor, thus
the largest known job step), then your load module will have sufficient main storage
allocated. On the other hand, if your load module is larger than the COBOL compiler,
not enough main storage will be reserved.
More Main Storage to Speed Up the Job
In addition to specifying the minimum main storage, you can also request additional
main storage. This is an amount that can be used, but is not required, to speed up job
execution. However, the program must be structured to take advantage of the
additional main storage; for example, a segmented COBOL program. Some of the
routines supplied by Unisys that use extra main storage in this manner are
sortJmerge, linkage editor, and the language translators. Additional memory may also
be advisable when running large assembly programs using many tags. As the
minimum main storage size is specified in decimal or hexadecimal, so is the
maximum; it is the fourth parameter (max) shown in the format.
7004 4623-000
4.3
Getting the Most Out of the Basic Job Control Statements
We’ll asswne that the label program was structured to use 41K decimal bytes (A028
hexadecimal) of main storage, if it is available; also, that it uses the high scheduling
priority and needs at least 12K decimal bytes (3000 hexadecimal). Added to our JOB
control statement, it would be coded as follows:
II JOB POCO,H,3000,A028
You can also code X’A028’ to represent the maximum main storage size in
hexadecimal.
You can specif5r the maximum main storage size in decimal by coding D’number’ for
the max parameter (e.g., D’41 000’ instead of A028 or X’A028’).
If we omitted the scheduling priority (it would default to normal) and the minimum
main storage size, it would be coded as follows:
II JOB POCO,,,A028
Nate:
If either the mm or the max parameter is omitted, the value specified for one is
used for the other. If both are omitted, and the load module is not located in
$Y$LOD (on SYSRES) or in an alternate load library on either SYSRES or
the volume containing the job’s $Y$R UN file, job control automatically
allocates BK decimal bytes of main storage (2000 in hexadecimal). Ifyou have
requested ajob dump through the OPTION statement (JOBDUMP), and you
have not specified mm or max on the JOB statement, job control nearly
doubles the amount of main storage that is automatically allocated. Ifyou
specify mm or max and intend to request ajob dump, specify at least 14K
decimal bytes (3500 in hexadecimal).
Multitasking Specification
If a program is written in BAL, you can create multiple tasks within it by using the
task parameter. This is called multitasking.
So far, we have been saying that job POCO is written in COBOL. For this example,
assume that it is written in BAL, and that we are going to allow for 18 tasks to be
active. The job still needs 12K decimal bytes to execute, but it can use 41K decimal
bytes, and has a high scheduling priority. Adding the multitasking specification
would make our JOB control statement look like this:
1/ JOB POCO,H,3000,A028,18
44
7004 4623.000
Getting the Most Out of the Basic Job Control Statements
Each task specified requires 256 bytes in the job prologue. The maximum number of
tasks you can have within a job is limited by the maximum size of the prologue (65535
bytes). If we omit the task parameter, job control assumes 1 by default.
Note:
There are other tables which require prologue space and their size varies
depending, for example, on the number offiles and spool buffers declared
through job control. Ifyou exceed the prologue size (you receive an R289
message and the job is not scheduled), you can reduce the number of tasks,
files, or spool buffers specified.
The Processing Time for the Job
After the same job has run several times, you probably know how long it takes to
execute. Should it run longer, it may mean something is wrong perhaps there is a
“bug” that has never been encountered before. Rather than waste processing time,
you can set a processing time limit using the max-time parameter. If the job executes
beyond this time limit, a message is sent to the operator, who can either cancel the job
or extend the time limit by any increment. If you specifSr max-time, you should tell the
operator what action to take if the specified processing time is exceeded.
-
The max-time limit is specified in minutes. It refers to elapsed wall-clock time or to
elapsed CPU time, depending upon how your supervisor is configured. If you want to
suppress the max-time function completely for a particular job, you can specifr SUP in
the max-time parameter, rather than a number.
The system will adjust the max-time value to allow for the following conditions:
•
Checkpointfrestart
•
PAUSE job control statements
•
SET CLOCK commands
•
Roll-in/roll-out
If you omit max-time, the time limit set at system generation is used as the default
value. The max-time parameter is supported only on supervisors configured with
NORMAL or MAX timer services. If a timer service is not specified at system
generation, max-time specifications are ignored.
Suppose you know that the job POCO should take no more than 15 minutes to run.
Added to the other parameters of the JOB statement, the max-time parameter is
coded as follows:
II JOB POCO,H,3000,A028,18,15
70044623-000
4-5
Getting the Most Out of the Basic Job Control Statements
C
Debugging the Control Stream
With the print-option-list parameter, you can control the printing ofjob control
statements and JPROC listing by specifying one or more available options. In a
spooling system, statements are printed (without passwords) in the job log; otherwise,
they are displayed on the system console. This gives a graphic display or printout of
the control stream for debugging purposes. For example, if a particular control stream
is run for the very first time and there are syntax errors in the coding, the system will
generate an error message telling you so. If you have used one of the debugging list
options, you receive a listing of your control stream.
The options for this parameter are:
B
Lists job control statements with symbol substitution. This is the default in a
spooling system.
D
Lists job control statements (as they’re read in by the run processor) without any
symbol substitution
P
Lists completed job control statements, which are generated by a procedure call
statement in the control stream, showing the values assigned in the procedure
definition statements
E
Lists any data contained in the control stream
S
Lists all the job control statements skipped as a result of an IF or GO job control
statement
A
Combines all the options
W
Suppresses the display ofjob control warning errors on the console or workstation
but not on the job log
N
None of the options are in effect (the default in a nonspooling system).
C
You may specify more than one option on a JOB control statement. However, if more
than one option is specified, the parameter group must be enclosed in parentheses.
Each option must be separated by a comma and can be specified in any order. For
example, (S,P,E) or (P,E,S); when only one option is specified, no parentheses are
needed.
When the D, P, E, or S options are chosen (separately or in combination) you get a
listing of your basic job control statements with symbol substitution even if B is not
specified.
C
4-6
7004 4623-000
Getting the Most Out of the Basic Job Control Statements
Let’s assume this is the first time we are running job POCO, and we want to list the
basic job control statements with symbol substitution, the job control statements
generated by a procedure call, and the data. These are options B, P, and E, but since
the option B is in effect when either P or E is chosen, you don’t have to specif5r it.
Added to the other parameters of our JOB control statement, it would be coded as
either:
ii JOB POCO,H,3000,A028,18,15,.(P,E)
or
P)
1
1/ JOB POCO,H,3000,A028,18,15,(E
Job Accounting and Spool Buffers
Use the ace-no parameter to provide the account number that has been assigned to
you at your installation. This 1- to 4-alphanumeric-character parameter creates an
entry in the job preamble for this account number, containing the total elapsed wall
clock time. Wall clock time can be defined as the point in time when ajob is initiated
for execution, up to the time when the job terminates. Therefore, any time used by
spooi input and spool output is not included.
This parameter may or may not be required, depending on the accounting procedures
used at your installation.
Suppose the account number assigned to you is AOO1. Adding this information would
make the existing JOB control statement appear as:
ii JOB POCO,H,3øøøAø28,18,15,(E,B),AøØ1
The n.Xm parameter sets up buffers for the file. This buffer holds data from the time
it first becomes available until the time it’s needed for processing. Thus, the central
processor does not have to wait as long for data. The job log and any spooled files that
don’t have their own buffers can share these buffers.
When coded, the n is the number of buffers, Xis a constant, and the m is the number
of (256-byte) blocks. Whenever nXm is omitted, a single 256-byte buffer (1X1) is
reserved if only the job log is sharing the buffer with your spooi files. If other spool
files are also sharing the buffer, two buffers of 512 bytes each (2X2) are allocated for a
total of 1024 bytes.
For example, if you wanted to allocate two buffers of 2048 bytes total, you would code:
H,3000,A028,18,15,(E
POCO
B
),AGø1,2X4
II JOB 1
The only values accepted for m are 1, 2, 4, 8, 16, and 32. Numbers larger than 32
default to 32. Numbers not in the acceptable range are changed to the lower
acceptable constant (e.g., 6 is changed to 4).
70044623.000
4-7
Getting the Most Out of the Basic Job Control Statements
Printing the Job Log File and Page Headers
The job log file contains the job accounting records, dumps created as a result of an
OPTION job control statement with the DUMP parameter, and a log, or list, of
messages and job control statements that were displayed on the system console. You
can selectively print this job log file with your job, by using one of the following
parameter choices of the JOB control statement:
ACT
LOG
NOACT
NOLOG
NONE
BOTH
The ACT parameter forces the printing of accounting records, regardless of the system
options in effect. LOG forces the printing ofjob log records, regardless of the system
options in effect. The NOACT parameter, when used, suppresses the printing of
accounting records. The NOLOG parameter means do not print the log (which also
contains dumps generated by an OPTION DUMP job control statement). If you code
the NONE parameter, both the log and accounting records aren’t printed. The BOTH
parameter allows both the log and accounting records to print. If you don’t specify one
of these parameters, the system options in effect are used.
For example, if you want only the accounting information to print (no log records
NOLOG), you would code:
-
II JOB POCO,H,3øee,Aø28,18,15,(E,B),AØs1,2X4,NCG
Cancel and snapshot dumps are never suppressed. If you’re running in a nonspooling
environment, this parameter is ignored.
At the beginning of the job log and accounting record printout, a page header, which
consists of several lines of asterisks, is printed. This can be suppressed by coding the
NOHDR parameter on the job control statement; by default, HDR is generated.
Coded, it would be:
II JOB POCO,H,3ø0ø,Aø28,18,15,(E,B),Aøø1,2X4,NOLOG,NOjpR
This parameter is ignored if you’re not spooling.
Ajob log report program is also available that will provide you with a job accounting
report based on the contents of the log file. For more information about the job log
report program, refer to the System Service Programs (SSP) Operating Guide,
UP-8841.
4.8
7004 4623-000
Getting the Most Out of the Basic Job Control Statements
Identifying the Peripheral Devices a Little Further
The DVC job control statement associates a physical device type, specified by a logical
unit number, with your job. It can also be used to: assign multiple devices, in a serial
manner, during ajob step; provide the physical address of the unit for using a specific
device; or (in a DDP environment), indicate that a disk file is remotely located. Here,
again, is its format:
I/[symbot] DVC
nnn[(n)]
RES
RUN
,
addr
OPT
IGNORE
ALT
E,HOST=host-idJ
0
REQ[(n)]
REAL
Refer to this format when each new parameter is introduced.
Note:
A particularjob cannot mix RBP destinations with auxiliary printers or DDP
destinations.
Using Multiple Devices, SYSRES, or the Job’s $Y$RUN File
The first parameter has three choices: nnn, RES, or RUN. (Remember, the (n) portion
of nnn is only used when assigning workstations.)
We have already explained how to use nnn to specifr a logical unit number (see
“Identifying the Devices” in Section 3). However, if you want to use more than one
print, punch, or card file in a job, you should assign a different logical unit number to
each file because the run processor flags multiple occurrences of the same logical unit
number in the same job step. If your system contains only 0776 printers, for example,
you can use the logical unit numbers 20, 21, 24, and 25. Sometimes, in a spooling
environment, you may want to assign more than four virtual printers or punàhes. To
do this, you must use the EQU statement (see “Equating Logical Unit Numbers to
Device Type Codes” in Section 6) to equate additional logical unit numbers to your
devices. You can use any logical unit number that is not already in your system. The
EQU statement is placed just before the device assignment set. To get an 0776
printer when you have already used the logical unit numbers 20,21, 24, and 25, you
might use the logical unit number 10, as follows:
ii EQU 10,0410
II DVC 10
70044623-000
4-9
Getting the Most Out of the Basic Job Control Statements
The number used for the type parameter of the EQU statement, 0410, is listed in
Table A-i of the Job Control Programming Reference Manual, UP-9984, as the device
type code for the 0776 printer.
Note:
The maximum number of unique devices allowed in ajob is 255. The
maximum number of unit record devices (e.g., card readers, data-set-label
diskettes, printers) allowed in one job is 42.
You don’t have to supply a logical unit number for files in SYSRES or the volume
containing the job’s $Y$RUN file. Use RES to indicate that the file is on the SYSRES
volume, or RUN to indicate that the file is on the volume containing the job’s $Y$RUN
file. Whenever RES or RUN is used, you can omit the VOL job control statement in
the device assignment set. The system differentiates between which volume is the
SYSRES volume and which volume contains the job’s $Y$RUN file. RES or RUN can
only be used for disk files.
In our control stream, we used this device assignment set for the name and address
disk input file as follows:
1/
1/
1/
1/
DVC
VOL
LBL
LFD
60
DSKOO1
DSKMASTFTL
DKNAME
If, instead of using the disk with a volume serial number of DSKOO1, the site manager
puts the name and address file on the SYSRES volume, still using the file identifier of
DSKMASTFIL, and assuming the file name in the program is still DKNAME, then the
device assignment set is:
II DVC Ri::
1/ LBL DSKMASTFIL
1/ LFD DKNAME
The VOL job control statement is omitted because the file is on SYSRES.
Specifying Multiple Workstations
Suppose you want to access a workstation file from more than one workstation. The
(n) portion of the DVC statement’s nnn parameter allows you to associate up to 255
workstations of the type and characteristics specified by (nnn) with one file. Consider
the following example:
/1 DVC 200(4)
/1 LFDWKSTFILE
C
4-10
7004 4623-000
Getting the Most Out of the Basic Job Control Statements
When the DVC statement is specified like this, up to four workstations can be logged
on and then optionally connected (using the workstation CONNECT command) to the
same job. These workstations access WKSTFILE.
If all four workstations must be connected for the job to begin execution, use the REQ
parameter of II DVC, like this:
1/ DVC 200(4),REQ
The UID statement is used when you want specific required workstations
automatically connected to the job.
The REQ parameter and the UID job control statement are discussed further in
“Ensuring that Workstations are Connected to a Job” later in the section.
More Control over Peripheral Devices
The format shows there are eight possible choices for the second parameter of the
DVC job control statement: addr, OPT, IGNORE, ALT, I, 0, REQ, and REAL. They
are explained in the following paragraphs, except for I and 0, which are explained
when we discuss spooling diskette files. Refer to “Spooling Input Card Data” in
Section 6 for more information.
Assigning Devices by Physical Address and Assigning Real Devices
Every device has a physical address associated with it. This is a hexadecimal number
representing the channel number, control unit address, and device number. It is
assigned by a Unisys customer service engineer. You can specify it by using the addr
parameter of the DVC job control statement.
It is unlikely you will need to use the addr parameter because the system can best
assign devices, since it is aware of the requirements of all jobs being run. Your job
may have special needs, however. Suppose you are running in a spooling
environment. You have a large job where the format of the printed output is very
important. You want to bypass spooling so that you can check your printed output
immediately and stop the job, if necessary, to correct the format. Since it is a large
job, you do not want it to go first to a spool file and then print if there are formatting
errors. You would specify the physical address of a real (rather than a virtual) printer,
like this:
II DVC 20,160
70044623.000
4-11
Getting the Most Out of the Basic Job Control Statements
You may assign a real device and bypass spooling without specifring its physical
address if you use the REAL parameter. The following statement, for example, allows
you to request any real printer:
/1 DVC 20,REAL
If you use the addr parameter to request a specific tape or disk device, b sure the
volume you want is not mounted on another unit. The flUID job control statement
can be used to assign workstations by physical address. Refer to “Ensuring that
Workstations are Connected to a Job” later in this section for more information.
Is This Device Needed for This Particular Run?
Sometimes, all the peripheral devices normally used by the job are not absolutely
needed. You may have a case where a job normally produces print and tape output.
Your system administrator needs the print output in a hurry, but is not worried about
the tape output at this time. If necessary, the job can be rescheduled to produce the
tape output.
Our control stream has device assignment sets for tape and print files. In the DVC job
control statement of the device assignment set for the tape file, we can use the OPT
parameter. This indicates that the peripheral device is optional; it is not essential to
the running of the job. If it is not available at the time the job is put into execution, all
references to this device are bypassed.
Added to our DVC job control statement for the tape output file, it would be coded as
follows:
/1 DVC 1øø,OP1
Different Volumes on the Same Device
Within a job step, job control normally allocates one device for each logical unit
number specified in the control stream. You might, however, have several different
volumes to be processed serially within the same job step. This could require several
different devices and your job would not be run until all the devices are free. You can
suppress job control’s check for one volume per logical unit number within a single job
step and reuse the same device serially by specifying IGNORE on the DVC statement.
Since IGNORE reduces the number of peripheral devices ajob needs, it increases the
chances of your job being run sooner.
If the first occurrence of a logical unit number does not specifr IGNORE in the DVC
statement, all subsequent references to that logical unit number must have IGNORE
specified in the DVC statements.
If you use the IGNORE parameter, processing for the first volume must be completed
before the second volume is needed, and so forth.
4-12
7004 4623000
Getting the Most Out of the Basic Job Control Statements
A typical application for the IGNORE parameter might be a program that takes
information from a tape file, updates it with information from a card file, and creates
a new tape. But ajob is scheduled that lasts most of the day, and it uses all but one of
the installations’s tape drives. Since you need two tape drives, you would have to wait
until that job was finished. However, you wrote the program so that it reads the input
tape file completely, updates the information, and then writes it out to a new tape.
Since the processing of the tape volume containing the input file is finished before the
program creates the new tape file, you can use the same device by using the IGNORE
parameter of the DVC job control statement in the device assignment set for the next
file to be processed (the output file, in this case).
The IGNORE parameter tells the system to disregard the fact that there already has
been a device assignment set for this logical number in this job step.
Suppose the input file is on a tape with a volume serial number of TAP111, a file
identifier of FIRST, and the file name for the input file is MASTIN. The output file
will be on a tape volume with a volume serial number of TAP222, have a file identifier
of SECOND, and a file name of MASTOUT. The logical unit number we are going to
use is 101.
The device assignment sets for the input and output files would be:
If
If
/1
If
Ii
If
1/
II
DVC
VOL
LBL
LFD
DVC
VOL
LBL
LFD
101
TAP111
FIRST
MASTIN
101 IGNORE
TAP222
SECOND
MASTOUT
When you use this feature ofjob control, make sure you inform the operator of the
tape mounting sequence.
Users of the Unisys sortJmerge routine will find the IGNORE parameter useful on
tape sort applications that use tape volumes as input, work areas, and output.
When a job consists of more than one job step, the system assumes that the first
device assignment set for a logical unit number will be used in subsequent job steps
until a new device assignment set for the same logical unit number occurs. For
instance, if you wanted to use the tape file with a volume serial number of TAP222 in
the next job step, you would have to specify the following device assignment set at the
beginning of the new job step:
/1
/1
If
II
70044623-000
DVC
VOL
LBL
LFD
101
TAP222
SECOND
xxxx (this depends on your program)
4-13
Getting the Most Out of the Basic Job Control Statements
Otherwise, the system assumes the tape with a volume serial number of TAP111 is to
be used.
Multiple Volumes in, a File? Use Alternate Devices to Decrease Operator Setup Time
The file is large in fact, so large it needs four tape volumes to hold it. When the
program uses four tape volumes, the operator can mount them, one at a time, on the
device associated with the logical unit number on the DVC job control statement.
When a volume is processed, the operator removes it from the device and mounts the
next volume on the device. Meanwhile, processing time is wasted while the system
waits for the new volume to be mounted. The operator must do this for every volume
of the file.
-
One way of avoiding this is to use the ALT parameter on the DVC statement. This
allows you to.alternate the same logical unit number between two devices, provided
that two devices of the same type are available. One device uses the logical unit
number while the first volume is being used, then the logical unit number switches to
the other device for the next volume. After the second volume is finished, and if there
are any more volumes in the file, the logical unit number is switched back to the first
device, and so on, until all volumes are used. In this way, the operator can mount two
tape volumes, on two different physical devices associated with a logical unit number,
in their proper sequence. When the first volume is finished, the system switches to
the device containing the second volume. Meanwhile, the operator can unload the
first volume and mount the third volume on the device. In this way, no time is wasted
because of setup time.. All alternate devices must be of the same type. This is
especially helpful when small tape reels are used. Note that alternating is restricted
to the boundaries of one job step, and that if only one device is available, a job will
execute with only one device (even though ALT is specified).
The ALT parameter of the DVC job control statement doesn’t work correctly if it is
used more than once in ajobstream. A separate drive is allocated for each ALT, and if
there are insufficient drives to accommodate all of the ALTs, only one drive is
allocated even if two drives are available. IftheALTfunction is needed more than
once in a jobstream, the following job control can be used:
II DVC 90 1/ VOL A
If DVC 91 If VOL B
Assume ajob has four tape volumes, using logical unit number 100. You can switch
between the two physical devices associated with logical unit number 100 by coding
the DVC job control statement as follows:
II DVC 100,J
444
7004 4623.000
Getting the Most Out o the Basic Job Control Statements
Of course, the VOL job control statement must be modified to indicate the volume
serial numbers of the four different tape volumes. We’ll discuss the use of optional
parameters for the VOL job control statement later. Briefly, the following example is
how multiple volume serial numbers are coded.
1/ DVC 100,ALT
/1 VOL T11111,T22222,T33333,T44444
To ensure that alternation occurs between devices, you may explicitly declare two
devices in your job control stream. This means you’ll have two DVC statements, each
specifying a different logical unit number. Consider the following example:
II
1/
1/
II
DVC
VOL
DVC
VOL
100
T11111,T33333
101
T22222,T44444
In this case, the operator can always alternate between the two devices specified by
the logical unit numbers 100 and 101, until all volumes are used.
Users of the sortlmerge routine will find it helpful to alternate when sorting many
tapes with the same label on a master tape.
Ensuring that Workstations Are Connected to a Job
You can use the REQ 1(n)] parameter of the DVC statement or the UID job control
statement when you want to ensure that workstations are connected to a job.
REQ tells the system that the workstations you’ve specified through the nnn[(n)]
parameter of the II DVC statement are required and must be connected (using the
workstation CONNECT command) for the job to begin execution. You can further
tailor the DVC statement by specifying that only a certain number of the workstations
must be connected before the job is executed. You do this with the (n) portion of the
REQ parameter. If you prepare your statement like this:
/1 DVC 200(8),REQ(1)
it tells the system that eight workstations can be connected to the job and that one of
the eight is required and must be connected for the job’s execution to begin.
Note8:
1.
7004 4623.OOb
The (n) portion of the nnn parameter and the REQ>(nXI parameter are used to
assign workstations only. Up to 255 workstations can be assigned to a single
workstation file.
4-15
Getting the Most Out of the Basic Job Control Statements
2.
The nnn parameter of / I DVC is used differently for workstations than for other
devices. Ifyou specify the logical unit number 200 (any workstations) and tailor
the specification by using the (n)portion of the nnn and REQ parameters, multiple
workstations (of any type) are assigned to the job.
Recall from “The UID Job Control Statement” in Section 3 that the UID statement is
used if you want specific workstations connected to ajob automatically. This is done
before the job’s execution begins (if the workstations specified have not already been
connected using a CONNECT command). You identify a particular workstation by its
user-id, device address, or both. For example:
II DVC 200
1/ UID WS1,(018),WS2(019)
II LFD WKSTFILE
The UID statement in this example indicates that the following three workstations
will automatically be connected: any workstation logged on with a user-id of WS1, the
workstation with the address 018 and logged on with any user-id, the workstation
with the address 019 and logged on with a user-id of WS2. If these three conditions
are not satisfied, the job remains in the scheduling queue. Remember that
workstations specified in the UID statement are required; therefore, the job will not
run until these devices are available (that is, logged on).
Although the (n) portion of the nnn parameter and the REQ 1(n)] parameter are
generally unnecessary in the DVC statement when the UID statement is used, you
may encounter a special situation. For example:
If DVC 200(4)
Ii UID WS1,WS2
/1 LFD WKSTFILE
The DVC statement indicates that the job can use up to four workstations. The two
identified in the UID statement are required and, provided they’re logged on, will
automatically be connected at execution time. Two more workstations (any two) can
optionally log on and then connect to the job with the CONNECT command.
Remember, you can specify $Y$MAS as a user-id to assign the job’s master
workstation to a job.
Specifying a Remote Disk File
To indicate that a disk file is located at a remote host in a DDP network, specify the
HOST=host-id keyword parameter on the II DVC statement. The host-id is one to four
alphanumeric characters long and identical to the label-id of the LOCAP
macroinstruction in your ICAM network. $HOST (in place of a host-id) indicates that
the file is located at the job’s remote originator (the remote host that initiated the job).
Cr
4-16
70044623-000
Getting the Most Out of the Basic Job Control Statements
Consider the following:
II JOB MYJOB
1/
II
II
II
DVC5Ø,HOST=A123
VOL 0000028
LBL FILE1
LFD REMOTE
II EXEC PROGA
I&
The DVC statement in the preceding device assignment set means that the disk file is
located at host A123.
Note:
The host you specify (using either a host-id or $HOST) must be a remote host.
Ifyou specify a local host, you’ll receive a data management error message
(DM21 INVALID DEVICE ASSIGNMENT).
For information about DDP facilities, see the Distributed Data Processing
Programming Guide, 7004 4508. For more information about the originator, see the
OPTION ORI statement in “Selecting Optional Features” in Section 6. See “How Job
Control Statements are Presented” in Appendix A for information about coding job
control statements containing positional as well as keyword parameters.
Indicating Use of the DDP Program-to-Program Facility
If your program is written in BAL and uses consolidated data management macros,
you can use DDP’s program-to-program facility. In its simplest form, this facility
allows a program at one host (the primary) to initiate communication with a program
at another host (the surrogate). The job control stream for each program participating
in this simple conversation must contain a DVC PROG job control statement. Used in
place of II DVC, II DVC PROG begins the device assignment set for the program-toprogram type file. The format is:
//Esymbot] DVC PROG t,job-name][,HOST=host-id)
You can specify one II DVC PROG statement in any single-step job control stream. (A
single-step job requests the execution of only one program.) The device assignment set
must contain all LFD statement and may contain all LBL statement for cataloging
purposes.
70044623-000
4-17
Getting the Most Out of the Basic Job Control Statements
The job-name parameter identifies the name of the other participant in the programto-program communication. For example, when specified in the II DVC PROG
statement for the primary, job-name identifies the surrogate. When specified in the
II DVC PROG statement for the surrogate, job-name identifies the primary. This
parameter is required in the II DVC PROG statement for the primary, but is optional
in the// DVC PROG statement for the surrogate.
The HOST=host-id parameter simply identifies a particular host in a DDP network.
The host-id is one to four alphanumeric characters long and identical to the label-id of
the LOCAP macroinstruction in your ICAM network. You use $HOST (in place of a
host-id) to indicate the originator (the host that initiated the job). Consider the
following control streams:
HOST AAAA
If JOB MYJOB
HOST BBBB
// JOB YOURJOB
1/ DVC PROG,YOURJOB, HOST=BBBB
/1 LFD THISFIL
If DVCPROG
// LFD THATFIL
If EXEC PROG1
If EXEC PROG2
I&
The II DVC PROG statement in MYJOB indicates that communication can only be
established with PROG2 the program identified in YOURJOB at host BBBB.
PROG1, in this case, must act as the primary. The II DVC PROG statement in
YOURJOB means that PROG2 is a surrogate in the program-to-program
communication with PROG1. PROG2 can also act as the surrogate when other job
control streams declare II DVC PROG,YOURJOB,HOST=BBBB. Now consider the
following:
-
HOST AAAA
HOST BBBB
If JOB MYJOB
II JOB YOURJOB
// DVCPROG,YOURJOB,HOST=BBBB
If LFD THISFIL
// DVC PROG,MYJOB, HOST=AAAA
1/ LFD THATFIL
1/ EXEC PROG1
I&
4-18
II EXEC PROG2
I&
7004 4623000
Getting the Most Out of the Basic Job Control Statements
These two job control streams indicate that only PROG2 at host BBBB and PROG1 at
host AAAA can communicate with each other. The first program to open the programto-program type file is considered the primary.
Although primarily intended for communication between programs executing on
different hosts, the program-to-program facility can be used between programs
executing on the same host. For more information about DDP’s program-to-program
facility, see the Distributed Data Processing Programming Guide, 7004 4508.
More Information about the Characteristics of
Your Volumes
We have used the VOL job control statement to specifr the volume serial number. It
also has additional parameters for further identifying each volume to the system.
Once again, its format is:
II [sythot) VOL
Mcc
N
NMcc
F-
votsn1I
I
(
L
(NS)
I(N0V)
I
votsn ir CS)
(NS)
(NOV)
(PREP)
(PREP)
Fvotsn 2j (5)
I (NS)
I (NOV)
(PREP)
L
SCRATCH
1
volsri 2r CS)
I
I
I
I
J
I
I
L
votsn
F3j CS)
(NS)
(NOV)
J
SCRATCH
()
(NOV)
(PREP)
LRE
SCRATCH
Refer to this format when each new parameter is introduced.
Notes:
1.
If all the volumes used to contain a multivolume file are going to be online
simultaneously (mounted on different devices during the course of a single job
step), the NOV and PREP options, if used, must be specified for each volume.
2.
The DVC specification in the device assignment set is used to determine if more
than one device is being used.
3.
In a multivolume file, if the individual volumes are mounted on separate devices,
the NOV and PREP options can be specified only for the individual volumes.
4.
If the PREP option is specified for any volume in a multivolume file sequentially
mounted on one device, it applies to all volumes in a multivolume file. NOV must
be specified for the last volume in the file for it to apply to all volumes in the file.
7004 4623-000
449
Getting the Most Out of the Basic Job Control Statements
More Than One Volume in a File
When we discussed the ALT parameter of the DVC job control statement, it was
stated that all volumes in the file must be specified on the VOL job control statement
of the device assignment set for the two devices sharing a logical unit number. (See
“Multiple Volumes in a File? Use Alternate Devices to Decrease Operator Setup
Time” earlier in this section). The example given was:
/1 DVC 1ØØ,ALT
II VOL T11111,T22222,T33333,T44444
Each group of numbers specified on the VOL job control statement (Tillil, T22222,
etc.) represents the volume serial number of the volumes in the sequence in which
they are mounted.
Remember, whenever there is more than one volume in a file, notify the operator of
the mounting sequence.
If more than eight volume serial numbers are listed, a nonbiank character must
appear in column 72 of the VOL job control statement and one or more continuation
cards (Appendix A) must follow. For example:
Column 72
(continuation)
II VOL T11111,T22222,T33333,T44444,T55555,T66666,T77777,T88888,
/11.
T99999,TAAAAA
1
X
Continuation CoLumn
Indicator
(Optional)
You can also specify multivolume files by using separate VOL control statements, like
this:
II VOL Tillil
II VOL T22222
1/ VOL T33333
This method has an advantage over the continuation method in that you can change
VOL specifications easier if they are coded separately.
C’
4-20
7004 4623-000
Getting the Most Out of the Basic Job Control Statements
The VOL statement’s (NOV) and SCRATCH parameters provide you with the option
of not listing each specific volume serial number in a multivolume file. For further
discussion of these parameters, see “Ignoring or Changing the Volume Serial Number”
later in this section.
Special Characteristics of Tape Volumes
Tape volumes have certain mode characteristics, such as bytes per inch, parity, and
the number of tracks (7 or 9). The mode characteristics of tape volumes are specified
using the Mcc parameter. The values force are given in Table 4-1.
Suppose you are using a UNISERVO® 12 magnetic tape subsystem, and the tape
volume is 7-track, 200 bytes per inch, even parity, with the translate and convert
features off. The mode setting is 20 and it would be coded as M20. The volumes being
used are coded as the remaining parameters.
II VOL M2,T11111,T22222
If the Mcc parameter is omitted, the mode settings specified at system generation
time are used.
If your supervisor supports block numbering and you have specified BKNO=YES in
your program’s file definition macroinstruction (or BC$CLNM for PIOCS), data
management will check block numbers on input tape volumes or write sequential
block numbers on output tape volumes. If you want to suppress block numbering or
checking during initialized processing, you use the N parameter on the VOL job
control statement. Initialized processing includes use of the TPREP utility routine or
the PREP option on the VOL statement as well as processing of input or output files
with nonstandard labels or no labels. When you specif5r N, block numbering is
suppressed for all volumes included on the VOL statement. For noninitialized
processing, the N parameter is ignored. That is, if your supervisor supports block
numbering and you have specified it in the file definition macroinstruction, you cannot
suppress checking or writing of block numbers by using the N parameter. For details
about block-numbered tapes, see the Consolidated Data Management
Macroinstructions Programming Guide, 7004 4607.
For example, to suppress block numbering on two tape output volumes with volume
serial numbers of T11111 and T22222, code as follows:
if VOL t1,T11111,T22222
When both the N and Mcc parameters are used, code them as one parameter. For
example:
1/ VOL Nt420,T11111,T22222
UNISERVO is a registered trademark of Unisys Corporation.
7004 4623.000
4-21
Getting the Most Out of the Basic Job Control Statements
C
Table 4-1. Mode Characteristics
Tape
cc
Bytes per Inch
Parity
Translate Feature
Convert Feature
UNISERVO 12/16 Magnetic Tape Volumes
7-track
10
20
28
30
38
50
60
68
70
78
90
A0
A8
B0
B8
200
200
200
200
200
556
556
556
556
556
800
800
800
800
800
9-track
C8
CO
800
1600
Odd
Even
Even
Odd
Odd
Odd
Even
Even
Odd
Odd
Odd
Even
Even
Odd
Odd
Off
Of f
On
Off
On
Of f
Of f
On
Of f
On
Off
Of f
On
Of f
On
On
Of f
Of f
Off
Off
On
Of f
Of f
Of f
Of f
On
Of f
Of f
Of f
Of f
Odd
Odd
Of f
Of f
Of f
Of f
UNISERVO 22/24 Magnetic Tape Volumes
9-track
C8
CO
800
1600
Odd
Odd__j
Off
Off
I
j
Off
Off
C,
UNISERVO 26/28 Magnetic Tape Volumes
9-track
CO
DO
1600
6250
Odd
Off
Of f
Off
Of f
Off
Off
Of f
Of f
Off
Of f
Of f
Of f
j__Odd
2145 and BT3200 Magnetic Tape Volumes
9-track
C8
CO
DO
800
1600
6250
Odd
Odd
Odd
5073 Magnetic Tape Volume
18-trackl
NA
38000
Odd
Extending Your Tape Volumes
If you recall, when we were assigning file names to files, we used the LFD job control
statement (see “Assigning a Logical File Name to the File” in Section 3). Well, now
we’ll use this same statement to extend our file. Once again, here is the format:
//Esymbo(] LFD
ffitename 1
[*filenrf
rExTEND
[ IN11[ 1IT
lafIL
L
C,,
4-22
7004 4623-010
Getting the Most Out of the Basic Job Control Statements
Looking at the format, we see the optional parameter EXTEND. The EXTEND
parameter lets us add information to the present end of a tape or disk file, provided
our program allows us to do so and the following job control conditions are met:
•
The PREP option is not specified on the VOL job control statement.
•
The file being extended is the only file on the volume.
•
The file uses standard labels.
•
The file specified is an output file.
The following example shows the use of the LFD statement to extend the file ADDR1:
/1 LFD ADDR1,,EXTEND
The following device assignment set, which includes this LFD statement, illustrates
how to extend a file (MAST) on volume Tilill.
II
II
ii
II
DVC
VOL
LBL
LFD
100
T11111,T22222,T33333
MASTER
MAST,,EXTEND
If you expect additional volumes will be needed to accommodate extension of the file,
you can add the volume serial numbers of any tapes to the VOL statement. The
following device assignment set indicates that the extension of MAST will result in a
multivolume file.
If
II
II
If
DVC
VOL
LBL
LFD
100
T11111,T22222,T33333
MASTER
MAST,,EXTEND
If you are extending a tape file that already has multiple volumes, your VOL
statement has to specify only the last volume containing the file plus any additional
volumes. You must include the serial number of the file’s first volume as the second
parameter (file-serial-number) of the LBL statement. See “Multivolume File? Assign
Each Volume a File Serial Number” later in this section for more information.
Suppose, for example, the file MAST is on volumes Tilill, T22222, and T33333. If
you expect the file’s extension to require an additional tape volume, you would code
the device assignment set as follows:
/1
II
II
If
DVC
VOL
LBL
LFD
100
,,,T33333,T44444
T11111
1
MASTER
MAST,,EXTEND
The volume serial number Tillil is required to identify T33333 and T44444 (the new
volume) as being part of the same file.
7004 4623-000
423
Getting the Most Out of the Basic Job Control Statements
Note:
When referencing multivolume files on the VOL statement, any undeclared
volume serial numbers must be represented with commas. Additionally, if
Mcc, N, or NMcc are not specified for the first positional parameter, you must
supply a comma. In the VOL statement in our previous example
(I / VOL ,,,T33333,T44444) the first comma represents the first positional
parameter. The second and third commas represent Tilill and T22222,
respectively.
The Consolidated Data Management Macroinstructions Programming Guide,
7004 4607, also contains information about extending tape files.
Sharing Disk Volumes
More than one job can share a disk volume. But suppose you are updating a file that
will be accessed by other user jobs. They should not access the file until the update is
completed, or else their output would not be the most current. You can indicate, on
the VOL job control statement, that the disk volume is nonsharable; thus the file
cannot be accessed. The system will not allow other jobs to begin execution until your
job has finished the update.
Assume the file being updated has a volume serial number of DSKO83 and it should
be nonsharable. You indicate this by using the (NS) parameter. The parentheses are
coded as part of the parameter, and there is no comma separating the volume serial
number and the (NS) parameter. This is coded as:
II VOL DSKO83(NS)
When there is more than one volume in the file (DSKO83, DSKO76, and DSKO93, for
instance) and they are all nonsharable, code it in this manner:
If VOL DSKe83(NS),DSKe76(NS),DsKg93(NS)
Sharable disk volumes are the default condition.
Ignoring or Changing the Volume Serial Number
Through the VOL job control statement, you have the option of ignoring volume serial
numbers. This allows the use of any available volume or one with an unknown
volume serial number.
For example, you want to create a tape file. The operator is told that you can mount
any unused tape with a volume serial number (it does not contain a permanent file,
and you do not want a scratch tape because you are creating this file for other jobs).
Since you don’t know what tape the operator will use, you don’t know the volume
serial number for your VOL job control statement. By using the (NOV) parameter and
a dummy volume serial number, you can use a volume without specifying the correct
volume serial number.
V
cV
4-24
7004 4623-000
Getting the Most Out of the Basic Job Control Statements
Code it this way:
II VOL DUMMY(NOV)
Notice that there is no comma separating the dummy volume serial number and the
(NOW parameter. The parentheses are part of the parameter.
After the job is processed, you should be informed, in some manner, of the volume
serial number of the created tape. This volume serial number must be used on the
VOL job control statement for any subsequent job using this tape volume.
Notes:
1.
The volume serial number DUMMY is used here just as an example. You can use
your own dummy volume serial number, but if it isn’t a unique one, keep the
following in mind: if two or more jobs use the same dummy volume serial number
for a disk volume, these jobs can run concurrently and share the same disk volume.
This may or may not be desired. Ifa job uses the (NOV) parameter with a dummy
volume serial number for one type of volume (e.g., a tape), and a second job uses
the (NOV) parameter with the same dummy volume serial number for another type
of volume (e.g., a disk) or, for another nonsharable volume (e.g., another tape), the
secondjob is not executed until the firstjob is finished.
2.
Ifyou specify a volume serial number and the volume with that serial number is
mounted on a device before the job goes into execution, that volume (and the device
on which it’s mounted) is used even ifyou’ve specified a different physical device
number on the DVC statement. If, however, you use I / VOL DUMMY (NOW the
physical request is not ignored.
With the VOL statement’s SCRATCH parameter you can specify a multivolume file
without listing each volume’s serial number. Consider this example:
/1 VOL VSN1,VSN2,SCRATCH
This statement declares a multivolume file and requests that the volume VSN1 with
the serial number be mounted first and volume VSN2 be mounted second. The
SCRATCH parameter indicates that after VSN2, any volumes can be mounted.
When you request scratch processing, a message to mount a scratch volume is
displayed (after any explicitly requested volumes have been taken care of) on the
system console. Any volume will then be accepted until the end of file. Remember,
because data management cannot check for the proper serial numbers at this point,
you should make sure that the operator knows exactly what volumes to mount and the
sequence to mount them in.
The SCRATCH parameter can also be used alone. For example:
/1 VOL SCRATCH
70044623-000
4-25
Getting the Most Out of the Basic Job Control Statements
This statement requests scratch processing for all volumes in the file.
You may want to use the SCRATCH parameter if you have a 20-volume diskette file
for example, and you don’t want to list 20 volume serial numbers in your job control
stream. When coding job control statements remember that the SCRATCH
parameter can only appear once in a VOL statement and it is always the last
parameter specified.
You can also suppress checking of volume serial numbers for all volumes of a
multivolume file by specifying NOV in the VOL statement for the last volume of the
file.
You can change a volume serial number by specifying the new volume serial number
followed by the (PREP) parameter. You can also use this to assign a volume that
currently does not have a volume serial number (scratch volume or a new volume).
Any information that is currently on the volume is scratched.
Your job creates an output tape that you want saved and to be assigned the volume
serial number of TAPO99. It would be coded as follows:
II VOL TAPØ99(PREP)
Once again, there is no comma separating the new volume serial number and the
(PREP) parameter. The parentheses are part of the parameter.
Notes:
426
1.
Be very careful when you use the PREP option on a file to be processed by the
librarian. When you specify the PREP parameter, the tape is prepped every time it
is opened as output. The librarian closes output tape files whenever they are to be
used as input and then reopens them as output. If a tape file is to be reused as an
output file within the same job, the librarian closes it as input and reopens it as
output. This reopening causes the tape to be reprepped (if PREP was specified),
thereby effectively erasing all the information previously produced. Therefore, use
this option only if the file will be output only, or output, then input. Otherwise, use
the TPREP utility to prep the file. The PREP option cannot be suppressed. You
must redefine the tape file without specifying the PREP option on the VOL
statement.
2.
For multivolume files, if PREP is specified for any of the volumes, all volumes in
the file are prepped.
3.
SCRATCH lets you mount additional tape volumes (unlimited processing);
however, these additional volumes are not prepped if PREP is specified. If they
must be initialized, use the TPREP utility routine.
7004 4623.000
Getting the Most Out of the Basic Job Control Statements
Multivolume Files Online Simultaneously
You may have an application, a data base system, for example, that requires a large
multivolume file, with the volumes online simultaneously (since they are accessed in a
random manner). Suppose you have a 3-volume file (volumes A B, and C). You would
code the device assignment set for the file like this:
1/
II
II
II
II
II
II
II
II
DVC 50
VOL A
DVC 51
VOL B
DVC 52
VOL C
LBL DATA
VMNTNO
LFD BASE
More Information on Disk and Format-Label Diskette
File Allocation
You use the EXT job control statement to allocate the space (extent) needed by a disk
or format-label diskette file. The format is:
//Esyrribo[] EXT
fMI
1STI
r
L
c
ii [ inch
FJJ
L
,
addr
Tccc:hh
fmi
L
l(bi,ai)
TBLK
CYL
TRK
OLD
fmj
L
l(bi,ai)I
,..1[,OLD][,FIX][ØNTERM)
]
All the parameters are optional.
The File Type
With the first parameter of the EXT statement, specify the type of file you’re
allocating the extent for.
7004 4623-000
4-27
Getting the Most Out of the Basic Job Control Statements
MIRAM files are discussed in the Consolidated Data Management Programming
Guide, UP-9978. System access technique files are described in the Supervisor
Technical Overview, UP-8831.
For the EXT job control statement, you can specify MIRAM (multiple indexed random
access method) files, indicated by coding MI, or SAT (system access technique) files,
indicated by coding ST.
If, for example, you wanted to use the multiple indexed random access method, you
would code:
II EXT MT,C,,CYL,1
Formatting a File and Using Contiguous Space
Files are formatted using the parameters F, BLK, and (bi,ai). These indicate that you
are going to format the file, F, in terms of blocks, ELK, to a certain length, (bi,ai). The
bi indicates the number of bytes in the block, and the ai indicates the number of
blocks in the file. Files can be formatted only in terms of blocks.
Suppose that you have a MIRAM file to allocate and it contains 5000 blocks, each 472
bytes long. Refer to the format of the EXT job control statement to see the correct
position of each of the parameters you are going to see: MI, F, BLK, and (bi[,ai]). It
would be coded as follows:
if EXT MI,F,,BLK,(472,5000)
You can set up your program to access a particular block (or blocks) within the file.
The EXT job control statement is also used to allocate space contiguously. When you
allocate a file, there may not always be a single extent (a single contiguous area)
available on the disk or format-label diskette. Suppose, for example, you need 10
cylinders for a file but there aren’t 10 contiguous cylinders anywhere on the volume.
Instead, there are 2 contiguous cylinders in one place, 3 in another, and 5 more in
another. If this is the case, OS/3 disk space management divides the file among 3
different areas resulting in a 3-extent file. The C parameter (shown as one of the
choices in the second parameter in the format) can prevent this from happening so
that if enough contiguous cylinders cannot be found, the file won’t be allocated.
Note:
A single file on disk or format-label diskette can have no more than 16
physical extents. If a file already occupies 16 extents but more are needed, you
must use another volume even if sufficient space is still available on the
original volume. (The file becomes a multivolume file.) A VTOC listing of the
volume will tell you in advance how many extents an existing file occupies.
Just remember there can be only 16 extents for a single volume file, 32 extents
if the file occupies two volumes, 48 for three volumes, and so on.
C>
4-28
7004 4623-000
Getting the Most Out of the Basic Job Control Statements
When you specify the absolute starting address (the addr parameter, explained in
“Terms of Allocation” later in this section), you must allocate contiguously. You must
also specifr addr in hexadecimal. The use of continguous space reduces file access
time, thus reducing job processing time.
To allocate a MIRAM file that contains 1000 blocks, each containing 1024 bytes, and
you want contiguous disk space, code as follows:
II EXT MI,C,,BLK,(1ø24,1øø)
The C and F parameters can be combined to form one parameter. Use this if you want
contiguous, formatted disk space. A comma is not needed to separate these
parameters.
For example, to allocate 300 blocks, each 256 bytes in a contiguous area, using the
multiple indexed random access method, code the following:
/1 EXT MI,CF,,BLK,(256,300)
Notice that we’ve been coding BLK in these examples. BLK, however, is the default
condition you could have coded the last example like this:
-
II EXT MI,CF,,,(256,300)
Your Disk or Format-Label Diskette File Needs More Space
When a disk or format-label diskette file is allocated, a certain area is reserved for a
file. It is possible, however, the amount that you estimate may not be enough. There
may be more information than you realized; an update of the file made it larger than
originally intended, or, you may be replacing existing information with new
information (this requires the use of the INIT parameter of the LFD job control
statement, which is explained in “The Expiration and Creation Date of the File” later
in this section). This new information may require more space than you had
previously allocated.
Job control can extend the requested area, if necessary. Let’s sayyou’re setting up a
file to contain 700 or 800 entries for an accounts payable procedure, and you estimated
the file would need 100 blocks, each 256 bytes in length. Since this is only an
estimate, you can use a parameter in the EXT job control statement to allocate more
space if it is needed. This is called dynamic extension. If it isn’t needed, it isn’t
allocated. In this way, you don’t waste space by allocating more than necessary.
The parameter used to provide this dynamic extension is the third parameter group in
the format. The inc parameter is the amount of additional space that you request.
This dynamic extension is in terms of cylinders.
70044623-000
4-29
Getting the Most Out of the Basic Job Control Statements
Specifying 0 indicates you do not want to allow for dynamic extension of the file. Use
this when you want to limit the amount of information placed in the file. If nothing is
specified, by default, one cylinder is generated.
Assume, for the accounts payable application, that we estimated 100 blocks, each 256
bytes long, on a formatted, MIRAM file. We want two additional cylinders if dynamic
extension is necessary. The coding would be:
II EXT MI,F,2,,(256,1øø)
Terms of Allocation
We’ve already covered some allocation terms in previous examples: BLK for allocating
in terms of blocks and CYL for allocating in terms of cylinders. With the addr
parameter you can also specify the absolute cylinder address in hexadecimal at which
the file is to begin. When you do this, allocation is in terms of cylinders.
Note:
The absolute address can be specified in decimal by coding D’number’, or
hexadecimal by coding X ‘number’ Any number not preceded by D or X and
enclosed in single quotes is considered hexadecimal.
Let’s say you need one MIRAM file, allocated contiguously, allowing 5 cylinders for
dynamic extension, and it must start at cylinder 78. Code it:
1/ EXT MI,C,5,4E
Do you recall specifying the amount of blocks needed for the file? One of the examples
looked like this:
if EXT IS,C,,BLK,(1024,1øøC)
Specifying (1024,1000) told job control how many blocks to allocate: 1000. When you
specify allocation in terms of cylinders or by absolute address, you must indicate how
many cylinders to allocate for the file by using the mi parameter.
If you wanted 10 cylinders, it would have been coded as:
II EXT MI,C,5,CYL,1[O
The TRK parameter allows you to allocate disk and format-label diskette files in
terms of tracks. The TBLK parameter allows you to allocate a file in blocks by track
rather than in blocks by cylinder (BLK parameter). The Tccc:hh parameter is similar
to the addr parameter because it lets you specify the absolute hexadecimal (X’number’
or number) or decimal (D’number’) starting address of the file. The address, however,
is a track address in cylinder/head format and the allocation is in terms of tracks, not
cylinders. For more information about file allocation by track, see the Consolidated
Data Management Macroinstructions Programming Guide, 7004 4607.
C
430
7004 4623-000
Getting the Most Out of the Basic Job Control Statements
Note:
You cannot allocate a disk file or format-label diskette file by track (using
TRK, TBLKor Tccc:hh) when creating MIRAM files with IRAM
characteristics.
Remember that when you specify CYL, addr, TRK, or Tccc:hhh, you must specify the
number of cylinders or tracks with the mi parameter.
Allocation Amounts
The parameters for indicating the amount of space wanted were shown, indirectly,
when we discussed formatting and terms of allocation. These were coded as the fifth
parameter, mi or (bi,ai).
The mi parameter is used with either the CYL, addr, TRK, or Tccc:hh parameter, and
indicates the amount of cylinders or tracks needed by the file. These were covered in
the last example of”Terms of Allocation” in this section.
The (bi,ai) parameter is used with the BLK or TBLK parameter for allocating in terms
of blocks (rounded up to cylinders or tracks, respectively). Remember BLK is the
default parameter so you don’t need to specify it. The bi indicates the amount of bytes
in the block, and the ai indicates the number of blocks in the file. For instance, this
example
II EXT t’1I,C,5,CYL,16
indicates an allocation of 10 cylinders, while either of these examples
II EXT MI,F,1ø,,{256,.1.øO)
100)
1
II EXT MI,F,1C,BLK,{256
indicates an allocation of 100 blocks, each 256 bytes in length.
You can specify any number of separate disk areas (extents) for an individual file. A
reason for using several different extents for a single file would be to decrease data
access time, thus reducing processing time. Assume the program is designed such
that the file can be divided into two different extents. The first extent contains data
used only by the first part of the program; the second extent contains data used only
by the second part of the program.
For instance, the first extent contains hourly pay rates for calculating gross pay, and
the second extent contains payroll deductions to subtract from the gross pay to get the
net pay. Once the gross pay is calculated, the first extent is no longer needed; the
• program will not need this information again. It only needs the deduction information
in the second extent to finish processing. In this way, one large extent is divided into
two smaller extents, reducing the amount of access arm movement for the disk unit.
70044623-000
4-31
Getting the Most Out of the Basic Job Control Statements
For example, you have a file divided into two different extents. The total size of the
file is 20 cylinders. The first part of your program uses 12 cylinders, and the second
part needs 8 cylinders. They can both be specified on the same EXT job control
statement. The information in the first four parameters applies to both extents in the
file.
Look at this portion of the format:
[jfmi
ilIJFmj 1
t,aif
L
JL li,ai)f
The mj parameter means the same as the mi parameter and the (bj,aj) parameter
means the same as the (bi,ai) parameter. The only difference is that mj and (bj,aj) are
used for additional extents in the file. So, we could code the two extent files (12
cylinders and 8 cylinders) as:
If EXT MI,C,1,CYL,12,8
1
23
Notes:
1 This applies to both extents.
2 This is the aNocation for the first extent.
3 This is the allocation for the second extent.
If you allocated in terms of blocks, with the first extent occupying 300 blocks, each 256
bytes in length, and the second extent occupying 700 blocks, each 256 bytes in length,
it would be coded as:
If EXT MI,C,1,BLK,(256,300),(256,700)
I
I
I
I
1
2
3
Notes:
1 This applies to both extents.
2 This is the allocation for the first extent.
3 This is the allocation for the second extent.
C’
4-32
70044623-000
Getting the Most Out of the Basic Job Control Statements
You can also specify separate extents for an individual file by coding separate EXT
statements, as we did when we coded separate VOL statements for a multivolume file.
Refer to “More Than One Volume in a File” earlier in this section for details. You
have coded separate extent specifications for our previous example, like this:
1/ EXT MI,C,1,BLK,(256,300)
LK,(256,700)
MI,C,1
B
/1 EXT 1
Changing the Specifications of a Previously Allocated File
Sometimes, you may want to change some of the information pertaining to a
previously allocated file. Use the OLD parameter to do this. The following portion of
the EXT job control statement format shows OLD as either the fourth or seventh
parameter:
addr
Tccc:hh
BLK
TBLK
CYL
TRK
OLD
,
1m
1
l(bi[,ai))I
rL
fmj
1 ,..1[,OLD][,FIX][NTERM)
bjE,ajDf
J
When coded as the fourth parameter, OLD means you want to change the automatic
allocation amount for dynamic extension (the third parameter) for a previously
allocated file. Suppose you specified one cylinder when a MIRAM file was originally
created. To change this specification to five, you code the EXT statement as follows:
II EXT ,,5,OLD
You can omit the first and second parameters, since they are ignored if specified.
When OLD is coded following the allocation amount (mi, mj, etc.), it increases the
original allocation amount for your extents.
70044623-000
4-33
Getting the Most Out of the Basic Job Control Statements
Let’s assume your file was originally a 30-cylinder, sequential file and you discover
you really need 50 cylinders. To obtain these extra 20 cylinders, you can change the
allocation amount for the file by using this EXT job control statement:
II EXT
,,,CYL,2@,OLD
When changing the allocation amount, you may omit the first, second, and third
parameters since they are ignored, if specified.
Allocating Space in the Fixed-Head Area of Your 8417 Disk
If you have an 8417 disk subsystem with a fixed-head feature, use the FIX parameter
with your EXT statement when you want to allocate the extent in the fixed-head area.
See the Consolidated Data Management Macroinstructions Programming Guide, 7004
4607, for information about the 8417 fixed-head disk.
No Terminate Option for Insufficient Extent Space
The NTERM option, when used, informs you if the extent cannot be allocated because
of insufficient disk space or because a specified absolute disk area is already in use
(error code 36). Rather than terminating the job, which is what happens without this
option, the system displays a JC48 message and waits for either a retry (R) or cancel
(C) reply. This allows your operator to evaluate the files currently on the disk and to
clear those that are not needed so your job can continue.
Information about Data-Set-Label Diskette File Allocation
To allocate space for a file on data-set-label diskette, include an EXT statement in the
device assignment set for the diskette.
A data-set-label diskette file is always a 1-extent, nonextendable, sequential file.
Therefore, several of the EXT statement parameters and options that we discussed in
the preceding section do not apply. To help you avoid confusion, refer to the following
EXT statement for data-set-label diskette:
Iltsymbot) EXT MI,C,ø,BLK,(bi,ai)[,NDI)
4-34
7004 4623000
Getting the Most Out of the Basic Job Control Statements
Just as for disk, the first parameter of the EXT statement indicates file type. The
extent for a data-set-label diskette file must be contiguous and cannot be dynamically
extended. So specify C for the second parameter and 0 for the third parameter. Space
on a data-set-label diskette is allocated by block, so BLK and (bi,ai) must be specified
for the fourth and fifth parameters respectively. Specify the last parameter, NDI
(non-data-interchange), for all System 80 data-set-label diskettes that are not basic
data exchange (BDE) diskettes. If you omit this parameter, it is assumed that you’re
allocating a BDE diskette (a single-sided, single-density diskette having 128-byte
sectors, 26 sectors per track, and 73 tracks.) For more information about the
characteristics of data-set-label diskettes, see the Consolidated Data Management
Macroinstructions Programming Guide, 7004 4607.
The following is an example of an extent statement for a data-set-label diskette file
having 100 blocks of 80 bytes each:
if EXT MI,C,ø,BLK,(80,1Oøø)
Using Your File Identifier More Efficiently
So far, the LBL job control statement was used to designate the individual files on a
volume by providing a file identifier (labeling a file).
We are now going to explain the optional parameters, and a special variation of the
file-identifier parameter, that improve file handling efficiency. Once again, the format
of the LBL job control statement is:
//[symboL] LBL
ffile-identifier 1
l’file-identifier’i
[,creation-date]
,
[
J [fi1e.seriat.number1[,expiration.date]
1_i
L 1VCHECK
1
{fite.seuence
.number}
[
1
{eneration.nu
nber}
version-number
As each individual parameter is introduced, refer to this format.
7004 4623.000
4-35
Getting the Most Out of the Basic Job Control Statements
But first, we’ll describe the special variation of the file-identifier parameter.
Sometimes, you may not want more than one job to access a particular file at the same
time, for example, when it is being updated. If it’s a disk file, you can make it lockable
by assigning a 6-byte lock ID as a prefix to your file identifier. Ninety-nine lock IDs
are available: $LOKO1 through $L0K99. The lock ID may be followed by up to 38
characters. The LBL statement for a lockable file might be coded this way:
II LBL $LOK15MASTERFILE
Once you have assigned a lock ID to the file, it is locked automatically each time it is
opened. The type of lock (read-only or write-only) is determined by the ACCESS
parameters in your file definition macroinstruction for the file. See the Consolidated
Data Management Macroinstructions Programming Guide, 7004 4607, for a complete
description of the file lock facility.
Multivolume File? Assign Each Volume a File Serial Number
When using a file consisting of multiple volumes, a file serial number can be assigned
to identifSr each volume as being a member of the file. In this way, a volume that is
not a member of the file cannot be used.
The file serial number is identical to the volume serial number of the first volume of
the file. For instance, there are four volumes in a file, in this sequence:
1.
XYZ
2.
P10
3.
A79
4.
TPL
The file serial number for all the volumes in this file would be XYZ.
You use the VCHECK parameter to either create a file serial number on output
volumes, or to check the file serial number on input volumes. This VCHECK
parameter instructs job control to use the first volume serial number specified on the
VOL statement as the file serial number.
Once again, we have the four volumes, XYZ, PlO, A79, and TPL, in that order, in a file.
We want to write a file serial number on them. Arbitrarily, the file identifier we are
going to use is OUTPUT. Your VOL and LBL statements would look like this:
II VOL XYZ,P10,A79,TPL
II LBL O1JTPUT,VCHECK
C
4-36
7004 4623.000
Getting the Most Out of the Basic Job Control Statements
If this file was already created with a file serial number (input rather than output), it
would be coded the same way. The VCHECK parameter writes on output and checks
on input.
The file-serial-number parameter is also used to write or check the file serial numbers
of volumes, but in a slightly different manner.
•
Again, we have these same four volumes (XYZ, PlO, A79, and TPL) in the file. But,
you only want to use the last two volumes, A79 and TPL, in that order, on this run.
This is a previously created file; when it was created, the VCHECK parameter was
used, giving a file serial number of XYZ to each volume. If we used the VCHECK
parameter now, while trying to read only these two volumes, A79 and TPL, job control
would use the volume serial number of the first volume specified on the VOL
statement, A79, as the file serial number value. Since these volumes were created
with a file serial number of XYZ, the job would not run. But, the file-serial-number
parameter allows you to specify the particular file serial number to use. This case
would be coded like this:
V
If VOL ,,,A79,TPL
II LBL OUTPUT,XYZ
Note:
/When referencing multivolume files on the VOL statement, any undeclared
volume serial numbers must be represented with commas. Additionally, if
Mcc, N, or NMcc are not specified for the first positional parameter, you must
supply a comma. In the VOL statement in our previous example
(II VOL ,,479,TPL) the first comma represents the first positional
parameter. The second and third commas represent XYZ and PlO
respectively.
V
If either VCHECK or the file-serial-number parameter is omitted when a multivolume
file is created, there is no file serial number for the file, or, if it’s a tape volume, there
is no VOL1 label.
The Expiration and Creation Date of the File
You can limit the life of files by writing an expiration date with the LBL statement.
This date indicates whether or not a file can be deleted by a scratch routine (by using
the SCR job control statement, explained in “Scratching Unwanted Files” in Section 6)
or by a function of data management. This is coded as the third parameter on the
LBLjob control statement, and can take either of two forms:
•
yyddd
In this form, yy is the year, and ddd is the day of the year. For example,
February 10th is the 41st day of the year (31 in January, plus 10).
70044623-000
V
4-37
Getting the Most Out of the Basic Job Control Statements
Rdddd
In this form, R is a constant, and indicates a retention cycle is being used based
on the creation date (either the next parameter, or the date set in the system).
The dddd indicates the amount of days (1 -9999).
For instance, you create an output tape with a file identifier of XRAY, and you want
it
to have an expiration date of the 98th day of 1991. This would be coded as:
II LBL XRAY,,91098
If you omit the expiration date when writing a file, the current date is inserted for you.
If you omit it when allocating a file, no date is specified and zeros are inserted. If
you
omit the date and allocate, then write to the file (in the same job step), the curren
t
system date is used.
The creation-date parameter indicates the date the file is generated. If omitte
d for a
tape file or a disk output file, the date stored in the job preamble is used. If omitte
d
for a disk input file, this field is ignored.
The creation date has only one form: yyddd, where yy is the year and ddd is the
day.
If you want a creation date of the file, identified by XRAY, to be the 100th day
of 1991,
code:
II LBL XRAY,,,91100
Indicating the Position of the File when Several Are on a Tape Volume
When you place more than one file on a single tape volume, you can indicate each
file’s
position on the tape by assigning sequence numbers. Later, if you want a particu
lar
file on that volume, you simply reference the file (in the II LBL statement) by its
identifier and sequence number. You can only assign sequence numbers to
standard
labeled tape flies.
When you create a tape file, you use the fifth positional parameter of the LBL
II
statement (file-sequence-number) to assign a sequence number. The following
statement, for example, assigns a sequence number of 3 to PRMAST the
third output
file on a volume to contain 5 files:
-
II LBL PRMAST,,,,3
Later, when you want to read (input) PRMAST, you can go directly to that file
by
including the same statement (II LBL PRMAST,,,,3) in your device assignm
ent set.
When you specifSr the file sequence number, data management search
es for the first
file with that number. If it’s found, data management then checks the
file identifier
for a match. (If the file sequence number you specifr is not found or if
the file
identifiers don’t match, a data management error results.)
4-38
7004 4623.000
Getting the Most Out of the Basic Job Control Statements
Remember, you must assign file sequence numbers when a tape file is created in order
to reference that file by sequence number later. If you don’t assign a sequence number
on output, data management assigns a number 1 to the file regardless of its position
on the tape volume. If you don’t provide a sequence number on input, data
management does not check for a sequence number but expects to use the first file
encountered. In either case, omitting file sequence numbers means using another
method to position the tape to the file you want (e.g., the II MTC statement or reading
and closing preceding files without rewinding until the desired file is reached).
Different Versions of a File
Ordinarily, only one generation of a file is used by a program. There are instances,
however, when more than one generation of the same file may be needed. For
example, one generation contains payroll deductions only used in January, March, and
May, and another generation has the payroll deductions used only in February, April,
and June. To indicate the different generations of a file, you can use the 1- to 4-digit
generation-number parameter of the LBL job control statement. This is used only
with tape files, and is the sixth parameter shown in the format. By using this
parameter, you can be sure the correct generation is used.
Suppose you did have two different generations of the payroll deduction file, with a
file identifier of CUSTMAST, and you want to use the second generation. This would
be coded as:
II LBL CUSTMAST,,,,,2
if you omit this parameter, data management assumes 0001.
Let’s go one step further. Each generation of a file can have several different versions.
Again, we have these two different generations of the CUSTMAST payroll deduction
file. Generation 1 is used in January, March, and May, and generation 2 is used in
February, April, and June. But, suppose each of these generations had two unique
sections. Version 1 is used in odd-numbered years, and version 2 is used in evennumbered years.
We could use the 1- to 2-digit version-number parameter to do this.
Suppose it is January, 1990. We need generation 1 (January) and version 2 (1990 is
even numbered). This would be coded as:
II LBL CUSTMAST,,,,,1,2
If the version-number parameter is omitted, data management assumes 01.
70044623-000
439
Getting the Most Out of the Basic Job Control Statements
C
Changing the Label of a Disk File
The REN statement is used to permanently change the label of a disk file through job
control a simpler procedure than the alternative methods for renaming disk files.
-
The format of REN is:
II[symbo[) REN tfdname, mew-labeL
l’new-Label
1
[SNTERM]
‘1
The lfdname parameter identifies the file to be renamed. It must match the lfdname
in the LFD statement for the file.
The file’s new label is specified in the new-label parameter. New-label replaces the
existing label identified in the device assignment set for the file. If new-label contains
embedded blanks, it must be enclosed by single quotation marks. It may be from 1 to
44 alphanumeric characters in length.
Specifring optional parameter NTERM causes any fatal errors encountered during the
renaming process to be ignored, but permits the job to continue. If this parameter is
present, the job continues running if a renaming error occurs, but the file is not
renamed. If NTERM is omitted, the job terminates at the point of error.
The REN statement is checked for syntax errors by the run processor during job
stream validation. If no errors are detected, the job is queued and becomes a
scheduling candidate. The run processor passes information from the REN statement
to the step processor, which performs the actual renaming during job execution.
The device assignment set for the file to be renamed must precede the REN
statement. It is a good idea to place the REN statement within the control stream as
close to the device assignment set for the file as possible, since II REN is only effective
against files on volumes mounted when the REN statement is encountered.
A file is renamed in the job step containing// REN, prior to execution of the program
for that step, or prior to job termination if no EXEC statement follows II REN.
Subsequent references to the renamed file must use new-label in the LBL statement
of the device assignment set for the renamed file.
Notes:
4.40
1.
The REN statement is used only to rename disk or format-label-diskette files; it
may not reference device assignment sets for data-set-label diskette or tape
volumes.
2.
REN statements are not permitted against files on SYSRES that begin with $Y$,
or against files on SYSRUN that begin with $Y$R.
3.
Don’t use /1 SKIP to bypass a device assignment set referenced by a REN
statement that is not also bypassed. Ifyou do, you’ll get an error during the
renaming process. (See “Adding Cards to a Stored Control Stream” in Section 6
for more information.)
.
7004 4623-000
Getting the Most Out of the Basic Job Control Statements
4.
Ifyou rename a cataloged file, you must recatalog the file under the new name.
Suppose you have a program that calculates the engineering department’s payroll and
outputs a disk file labeled EGRPAY. The control stream to rename the file EGRCOST
looks like this:
II
II
/1
II
DVC
VOL
LBL
LFD
50
DSKO1
EGRPAY
DSKO1JT
II REN DSKOUT,EGRCOST
The file’s label is now EGRCOST. Suppose that a subsequent job step uses EGRCOST
as input for calculating company-wide costs. Building on our first example, the
renamed file is referenced subsequently in the control stream like this:
II
/1
If
/1
DVC
VOL
LBL
LFD
50
DSKG1
EGRPAY
DSKOUT
/1 REN DSKOUT,EGRCOST
• II
II
II
II
7004 4623-000
DVC
VOL
LBL
LFD
50
DSKO1
EGRCOST
DSKIN
4-41
Getting the Most Out of the Basic Job Control Statements
A single REN statement applies only to the first volume in a multivolume file. To
rename a multivolume file, therefore, you must specify a unique REN statement for
each volume in the file.
If EGRPAY in our first example had been a multivolume file, we would have renamed
it this way:
1/
If
1/
II
DVC
VOL
LBL
LFD
50
DSKO1
EGRPAY
DSKOUT1
II
If
If
1/
DVC
VOL
LBL
LFD
51
DSKO2
EGRPAY
DSKOUT2
If REN DSKOUT1,EGRCOST
1/ REN DSKOtJT2,EGRCOST
Use the REN statement carefully to avoid renaming a file concurrently used by
another job. To help prevent this problem, establish nonsharable status (using the NS
option of the VOL statement) for endangered disk volumes, or use passwords known
only to selected personnel.
4-42
7004 4623-000
C
Getting the Most Out of the Basic Job Control Statements
Specifying Qualifiers for File Identifiers
The QUAL job control statement is used to prefix a qualifier to all subsequent file
identifiers in a job. The format of the QUAL statement is:
//[symboL] QUAL quatname
The qualname is a 1- to 8-character alphanumeric name. When specified, this name
followed by a slash becomes the qualifier, and is automatically prefixed to each
subsequent file identifier in your job control stream.
Consider the following example:
II
/1
II
1/
II
/1
II
QUAL SMITHCO
DVC 60 1/ VOL DISKO1
LBL PAYABLES.TAXES
LFD PAYFILE
DVC 60 1/ VOL DISKO1
LBL INCOME.INTEREST
LFD INFILE
In this example, SMITHCO is specified as the qualifier and will be prefixed, along
with a slash, to each subsequent LBL file identifier producing:
SMITHCO,PAYABLES.TAXES and SMITHCO/INCOME.INTEREST.
The qualifier remains in effect until the end of the job or until another QUAL
statement is encountered. If the next QUAL statement specifies another qualname,
that name becomes the qualifier for any subsequent file identifiers. If no name is
specified (e.g., II QUAL), use of the qualifier is terminated.
An LBL file identifier that is already prefixed with an alphanumeric name and a slash
overrides the QUAL statement qualifier. Consider this example:
II
1/
If
Ii
If
Ii
II
QUAL SMITHCO
DVC 60 II VOL DISKO1
LBL PAYASLE.TAXES
LFD PAYFILE
DVC 60 1/ VOL DISKO1
LBL INCOME/INTEREST
LFD INFILE
INCOME! in the second LBL statement is already considered a unique qualifier;
therefore, SMITHCO/ will be prefixed to PAYABLE.TAXES but not to
INCOME/INTEREST.
Because the QUAL statement is especially useful in identif’ing cataloged files (see
“File Cataloging” in Section 6), QUAL is also discussed in the File Cataloging
Technical Overview, 7004 4615.
7004 4623-000
4-43
Getting the Most Out of the Basic Job Control Statements
C
More Information about the Logical File
So far, you know the LFD job control statement is used to provide a file name that
associates the file defined in the program with the file information in the control
stream. Now, by introducing the optional parameters, you will see some of the other
functions it provides. Once again, its format is:
//Esyrnbot) LFD
ffitename
l*fllenamel
fn] [
[ 1]
L
L
EXTEND
INIT
PREP
ID
IGNORE
Refer to this format as each parameter is introduced.
We have already discussed the filename parameter. An asterisk (*) indicates that the
file label is lockable.
Reserving an Extent Information Storage Area
Files are defined on disk and format-label diskette volumes in terms of extents. An
extent is space on the volume made up of contiguous tracks. If you recall, we used the
EXT job control statement to split up a file into two extents. So, in the strict sense, an
extent is not always the entire disk area a file requires; at times it is, but at other
times it isn’t.
Information about the extents is placed in the job’s prologue along with other
information needed to regulate your job. (See “Minimum and Maximum Main
Storage” in Section 2.) On the JOB statement, you specify the minimum and
maximum amount of main storage needed to execute your largest job step. However,
in order for job control to reserve sufficient main storage for the extent information in
the prologue, you must specify the number of physical extents a file has in the second
parameter of the LFD statement. Assume, for example, that the file named DSKOUT
has 10 extents. To reserve space for information about these extents, you code the
LFD statement as follows:
II LFD DSKOUT,1G
The space acquired by using this parameter influences the total main storage
requirement for the job. If you specify a value of zero, job control does not reserve
main storage for extent information. If you omit this parameter, main storage
sufficient for information about eight extents is reserved. The maximum value you
can specify for this parameter is 20.
4-44
7004 4623-000
Getting the Most Out of the Basic Job Control Statements
Ifyou specify a greater number of extents on the LFD statement than a file
actually has, main storage is used unnecessarily for the extent information.
Although this should be taken into consideration, problems are more likely to
arise ifyou specify fewer extents on the LFD statement than the file actually
has.
Note:
Specifications for Existing Files
The third parameter of the LFD job control statement provides five different options:
INIT, EXTEND, PREP, IGNORE, and ID.
The INIT option is used to initialize an existing disk file; that is, INIT causes all
information except the allocated space and file identifier to be discarded when the
program using that file opens it. When you specifSr INIT for an output file, the output
will start at the beginning of the file. When INIT is specified for an input file, an endof-file will be indicated when your program reads the first record. You can specifSr
INIT for all disk and diskette files.
Suppose you already reserved for the old file rather than allocate a new one? In this
way, you are not leaving dead space on the disk volume.
We’ll assume that the file name is SORTOUT. The device assignment set to reuse
WORK2 on disk volume DPSO28 would look like this:
II
If
If
II
DVC
VOL
LBL
LFD
50
DSPO28
WORK2
SORTOUT,,INIT
Notice the logical unit number for the DVC job control is 50. This indicates any disk
device can be used. Also note the absence of an EXT job control statement. It wasn’t
needed; specifications for the new file are the same as the old one.
Note:
The INIT parameter must not be used for a file that contains checkpoint
records. The use of this parameter causes writing to begin at the start offile
every time you log a checkpoint record to the file, thus overwriting any
checkpoint records already existing on the file.
The EXTEND option allows you to add information to the present end of an existing
output file if the instructions in your program allow you to do so. EXTEND has no
effect on input files.
You can specify EXTEND for tape files; EXTEND logically does not affect disk and
diskette files.
Suppose you allocated four cylinders for your file, but filled only two cylinders with
information. Now, you have more information to add to this file, and your program
allows you to do so. You must also instruct job control that you intend to do this.
70044623-000
4-45
Getting the Most Out of the Basic Job Control Statements
If the file name were ADDON, you could extend the file like this:
II LFD ADDON,,EXTEND
Remember, whether or not you can actually extend a file depends on the instructions
given in your program. In COBOL, for example, an OPEN OUTPUT statement does
not permit file extension even if you specify the EXTEND parameter on the LFD
statement.
The IGNORE option lets you specify that a file is an optional file. This means you can
decide at job execution time, without having to change the program, whether accesses
to the file should be processed or ignored based on the file resources that are available.
This option should be used only with consolidated data management. It is ignored if
specified for a file that is accessed using basic data management (including SAT
files).
When the IGNORE option is specified, the following status is returned to the
executing program:
•
File initialization (OPEN) successful status
•
Input operation end of file (EOF) indication
•
All other operations successful status (request is ignored)
-
-
-
Use this option when a program reads or writes multiple files and it is not necessary
that all of the files be accessed. For example, a program updates some files and.
produces a report, but the report is not always needed. You would specify IGNORE
in
the II LFD statement of the printer device assignment set. Another example is
where
the program reads multiple input files, but one of them is not available at the time
and is not essential to the execution of the program.
When this option is specified for an input file, the program should be performing
sequential retrieval; otherwise, it may not be expecting an EOF indication.
Indicating Where the Load Module Is Located
An EXEC job control statement is required to call a load module and initiate
execution. Once again, the format is:
Il[symbol] EXEC program-name
(j
I
L
Ilibrary-name 1
1,[)switchpriority]1,ABI4ORM=tabet]
$Y$RUW
[.$Y$LOD
The second parameter indicates the name of the library (on disk) containing
the load
module. This can be either $Y$LOD, $Y$RUN, or the LFD name of the alterna
te load
library you have previously specified in the control stream.
C
4-46
70044623-000
Getting the Most Out of the Basic Job Control Statements
As you can see, the shaded default option is $Y$LOD. This is where you would store
most of your programs. When you omit the second parameter, $Y$LOD is searched.
If the load module is not found here, then the job’s $Y$RUN file is searched.
Another choice you have is the job’s $Y$RUN file, which is where the linkage editor
stores your load module if you have not indicated a specific library. You would code
$Y$RUN in the EXEC job control statement if you have a load module with the same
program name in $Y$LOD. Let’s assume that you have a load module named
PAYROL in $Y$LOD and that you want to make some changes to this program. Take
the source deck, make the necessary changes, and compile it with the same program
name: PAYROL. When it comes time to execute, the system is told that the load
module to fetch is PAYROL. Without specifying the library name on the EXEC job
control statement, the default ($Y$LOD) is assumed, so the system fetches the
PAYROL load module from $Y$LOD. But, you wanted the one from the job’s $Y$RUN
file. You are going to receive the wrong load module. So, in this case, you had better
indicate that you want to fetch from the job’s $Y$RUN file. Remember, the job’s
$Y$RUN file is only a temporary file. Any load module you store here is available
only during that job.
Let’s say your load module is named PAYROL and it is loaded in the job’s $Y$RUN
file. It would be coded as follows:
.11 EXEC PAYROL,$YSRUN
If the load module cannot be found in the job’s $Y$RUN file, $Y$LOD is searched to
see if it was stored there.
The remaining choice for this parameter, library-name, is used when the load module
is stored in a private load library of your own. if you do this, you must define this
library in a device assignment set, and the library-name must agree with the file
name on the LFD job control statement. Normally, if the module is not found in this
library, $Y$LOD and then $Y$RUN are searched. If, however, you specify NSRCH on
the OPTION job control statement, only the library named on the EXEC statement is
searched for the load module; $Y$RUN and $Y$LOD are not searched. (See “Selecting
Optional Features” in Section 6.)
Let’s say the load module is named PAYROL, and it is stored in a library with a file
identifier of PAYLIBRARYMAST, on disk volume DISKO1. You used PAYLIB as the
file name on the LFD job control statement, and, as the file identifier on the LBL job
control statement, you would, of course, have to use PAYLIBRARYMAST. The device
assignment set and the EXEC job control statement would be coded as:
II DVC 50
II VOL DISKO1
1/ LBL PAYLIBRARYMAST
// LFD PAYI..[B <
1iEXC:PAYRQL,PAX LIB
7004 4623-000
4-47
Getting the Most Out of the Basic Job Control Statements
If this library is not accessed by your program (if it is only accessed by the system to
obtain the load module named on the EXEC job control statement), the file name on
the LFD job control statement need not agree with any specification within your
program. It serves only to associate the library in the device assignment set with the
library on the EXEC job control statement. As the file name on our LFD job control
statement, we could have used any name as long as it agrees with the specification on
the EXEC job control statement.
If the load module is not located, $Y$LOD and then the job’s $Y$RUN file is searched.
Task Switching Priority
The EXEC job control statement is also used to specify task switching priorities. This
synchronization and rotation of central processor control from task to task is a
function of the supervisor, and is described in the Consolidated Data Management
Programming Guide, UP-9978.
The switch-priority is the third parameter of the EXEC job control statement. The
priority you specify can be an absolute value ranging from 1 to 60, with the lower
number representing the higher priority (1 is the highest priority). Assume, for
example, your job has one step and you want a switching priority of 10 assigned to the
specified program. (The load module name for the program-name is SWITCH and it
is stored by default in $Y$LOD.) You could code the EXEC statement as follows:
1/ JOB MYJOB
ii EXEC SWITCH,,1ø
You can also specify a relative value such as +3 or -3 to change priority for a program
specified in a particular job step with respect to the job’s overall priority (as set, for
example, by a SWITCH operator command or an OPTION PR! job control statement).
When specifring priorities this way, remember that a plus (+) value decrements the
overall assigned value. This results in a higher task switching priority. A minus C-)
value increments the overall assigned value. This results in a lower task switching
priority.
4-48
7004 4623-000
Getting the Most Out of the Basic Job Control Statements
Suppose you code the following:
II JOB MYJOB
/1 OPTION PRI=7
Assigns an overa(L task switching priority of 7
to each program.
// EXEC PROG1
•
if EXEC PROG2
The programs specified in these 2 job steps
have a task switching priority of 7.
1/ EXEC PROG3,,+2
The program specified in this job step has a
task switching priority of 5.
If EXEC ORIG4,,3
•
•
ii EXEC PROG5,,-4
The program specified in this job step has a
task switching priority of 3.
The program specified in this job step has a
task switching priority of 11.
I&
The OPTION PRI job control statement is discussed in “Selecting Optional Features”
in Section 6.
If you omit a task switching priority, the lowest available priority (the highest
number) is used.
You should understand that the task switching priority specified on the EXEC job
control statement is only the initial switching priority for that job step. There are two
ways it can be changed during the job step:
•
The operator can raise or lower the priority using the SWITCH console command
if ajob is getting too much or too little CPU time.
•
The program itself may raise or lower its priority using the CHAP (change
priority) macroinstruction. This function is described in the Consolidated Data
Management Programming Guide, UP-9978.
As you can see, the effect of the switching priority really depends on the task
switching priorities specified for other jobs running at the same time as your job. Your
job will not gain any advantage by specifying a task switching at priority of 1 (the
highest priority) if all other jobs also use priority 1. There is a case, however, where
the assigned switching priority is particularly significant. Recall from Section 1 that
the RUN syrnbiont is one portion ofjob control that reads and analyzes job control
streams. Tl)e RUN symbiont is only one of many OS/3 symbionts that perform system
functions, usually in response to operator console commands. Normally, all symbionts
run at priority 0, i.e., higher than any user job. A SYSGEN option, however, allows
70044623-000
449
Getting the Most Out of the Basic Job Control Statements
the supervisor to be configured so that all symbionts run at some lower (user) priority.
For example, suppose symbionts run at priority 2 under your supervisor. The only
jobs that should be run with task switching priority 1 would be those that are
extremely time-critical and cannot tolerate the loss of CPU time whenever a symbiont
is active. Other jobs should be run at priority 3 and lower.
Avoiding Abnormal Termination due to Program Errors
The ABNORM=label keyword parameter of the EXEC job control statement is used to
bypass job control statements if your program contains errors that may cause the job
to abnormally terminate. If the program has such errors, control of the job skips to
the statement whose label you specify in this parameter so that the job’s execution can
continue. Any subsequent action depends on the contents of the target job control
statement. A more specific example for using this parameter is given in “Bypassing
Job Control Statements to Avoid Abnormal Termination” in Section 6. For now, just
remember that ABNORM=label is a keyword parameter, not a positional parameter,
and, therefore, may be coded in any position. For example:
/1 EXEC MYPROG,ABNORM=ERROR
Also remember that the operator can still cancel (normally terminate) your job even
though you specify this parameter.
The Job Control Language So Far
We have now covered the job control statements you’ll probably use most frequently
for your jobs. Section 5 deals with system JPROCS provided in the basic OS/3
software package. Their use eliminates the need of repeatedly coding a series ofjob
control statements that perform a specific function.
7
C
4-50
7004 4623000
Section 5
Doing It the Easy Way with Procedure Calls
-
What Is a Procedure?
Have you ever heard someone say: “I’ve made that mistake before. There must be
some way, some procedure, to make sure it won’t happen again”? Common errors are:
keystroke errors, forgotten commas, statements out of sequence, ete. errors that
occur because of repetition rather than unfamiliarity. If we could reduce the number
ofjob control statements coded, the bulk of these errors would also be reduced. What
is needed is a procedure that allows you to write a series ofjob control statements,
store them for later use, and, by writing a single job control statement, call in these job
control statements whenever needed.
-
This procedure exists it allows you to write and call your own procedures, or to call
procedures supplied by Unisys. In Sections 6 through 9, you’ll learn how to write,
store, and call your own procedures. This section discusses how to use procedures
supplied by Unisys. These procedures are called by job control procedure call
statements (JPROC calls) in the control stream. Each JPROC call generates a readyto-use set ofjob control statements. Optional parameters in the JPROC call line
enable you to tailor the job control statements generated to suit your needs. (See “Job
Control Procedures” in Appendix C for a complete listing of JPROCS.)
-
When you use more than one JPROC call, keep this in mind: only one JPROC call can
appear on a single card. JPROC calls can be part of a multistatement line of coding,
but
•
It must be the only JPROC in the line.
•
It must be the last statement on the line.
You can code this:
II job control statement
If jproc call
but not this:
II jproc call.
II jproc call
and not this:
II jproc call
7004 4623-000
If job control statement
54
Doing It the Easy Way with Procedure Calls
-
Setting Up Temporary Work Files
Temporary work files are used extensively by programmers to store intermediate
processing results and data that will only be used in a particular job or job step.
Depending on file characteristics and the device used, from three to five job control
statements are needed in the device assignment set for each temporary work file. The
WORK and TEMP JPROC calls allow you to generate any device assignment set
needed for temporary work files.
The difference between the two JPROCS is that WORK sets up temporary files for one
job step and deletes them at the end of the job step. TEMP sets up temporary files for
the duration of the job, deleting them at the end of the job. WORK and TEMP also
generate different default file name values we’ll explain these in a moment.
-
The format for WORK and TEMP is:
//[Lfdname]
IWORKn1
lTEMI’If
DVC=rin,VOL=
VOL=
[EXTSP=
vol-ser-no
RES
RUN
voL-ser-no
RES
RUN
{}i
f
I
BLKnnnn
BLK=4000
CYL=nn
1 [TYPE= {fite.tYPe}
[SECALL= {n}
1
Suppose your assignment is to write a program that reports the grades for each
student in the local school district. The program must list each student’s name,
grouped by school, in descending grade order. The disk area that stores the data used
to calculate the order will never be used again once the job step terminates an ideal
candidate for a temporary file created by WORK
-
Ignoring all optional parameters, the basic WORK JPROC call is:
II WORKn
5-2
7004 4623.000
Doing It the Easy Way with Procedure Calls
-
Where n is a number in the range 1 through 10. Up to 10 temporary work files can be
set up for each job step (or job, if you’re using TEMP). If no specific device or volume
is requested, the file is allocated on either SYSRES or the job’s $Y$RUN file; oddnumbered files go to SYSRES and even-numbered files to $Y$RUN. So, if you want
one temporary file allocated on the job’s $Y$RUN file, for the duration of the job step,
you would code the following:
1/ WORK2
These job control statements would be generated:
II
/1
If
/1
DVC
EXT
LBL
LFD
RUN
ST,,1,BLK,(256,4000)
$SCR2,16
$SCR2
We’ll discuss the generated EXT statement in conjunction with the BLK, EXTSP, and
TYPE parameters. For now, it’s sufficient to know that 4000 blocks, each 256 bytes
long, are allocated by default.
The lfdname parameter of WORK and TEMP supplies a file name for the generated
job control statements. It is one to eight alphanumeric characters in length. The file
identifier on the LBL statement generated by WORK is always prefixed by $SCR,
which identifies job step temporary (scratch) files. The number after $SCR
corresponds to n in WORKn. If you omit the lfdname parameter of the WORK JPROC
call and code the following:
II WORK1
the generated statements are:
II LBL $SCR1
1/ LFD $SCR1
The file name in your program must also start with $SCR. In addition, you must use
the same WORK JPROC call each time the program is run. If the JPROC call is
changed toll WORK7, for example, the file name in your program must be changed to
$SCR7.
70044623-000
5-3
Doing It the Easy Way with Procedure Calls
-
For TEMP, unlike WORK, the generated file identifier is $JOB if you omit the
lfdname parameter. Therefore, if the file name in your program begins with $SCR,
you must use the lfdname parameter of the TEMP JPROC call, like this:
//$SCR1 TEMP 1
to generate:
/1 LBL $JOBI
/1 LFD $SCR1
If you had not used the lfdname parameter in this example, the generated file name
would have been $JOB1, which would not have matched the file name in your
program.
You can have the control statements generated by WORK and TEMP listed by
specifying the P option on the JOB statement. If you have spooling in your system,
the control statements will be printed in the job log. Otherwise, they will be displayed
on the system console.
When the job step terminates, all temporary files created by WORK are scratched.
Files created by TEMP are scratched at the end of the job.
The lfdname parameter can also indicate a file’s function when using the WORK
JPROC call. For example, if you code
//GRADEOUT WORK2
the generated job control statements are:
1/
/1
II
If
DVC
EXT
LBL
LFD
RUN
ST,,1,BLK,(256,4000)
$SCR2
GRADEOUT
It is easier to remember what GRADEOUT contains than it might be to remember
what $SCR2 contains.
The remaining optional parameters of the WORK and TEMP JPROC calls are
keyword parameters. if you are unsure of the rules for coding them, turn to Appendix
A to refresh your memory.
5-4
70044623-000
Doing It the Easy Way with Procedure Calls
-
Notes:
1.
Work file labels ($JOB and $SCR) are modified to be unique for the job by
inserting a job-id after the first 4 bytes of the label. This enables all residual work
files (for example, work files not scratched when an HPR occurs) to be cleared from
the disk during IPL.
2.
Within user JCL, work file labels (those starting with $JOB or $SCR) cannot
exceed 39 characters.
Using Your Own Volume
By default, temporary work files are allocated on SYSRES or $Y$RUN. But what if
you needed several work files and there isn’t enough available space on these
volumes? In this case, you would use your own volume by specif’ing the VOL
parameter. Building on our last example, if your own volume is DISKO1, you would
code:
//GRADEOUT WORK2 VOL=DISKØ1
This device assignment set is generated:
II: PVC .0
I/ VOL DLSKØI
if EXT ST,,1,BLK,(256,4000)
II LBL $SCR2
II LFD GRADEOUT
Note that the logical unit number generated for the DVC job control statement is 50.
The WORK and TEMP JPROCS automatically assign the first available logical unit
number in the range 50 through 59, but you can use the DVC parameter to assign
another logical unit number selected from Table A-i in the Job Control Programming
Reference Manual, UP-9984. In order to avoid a conflict, for example, you may want to
assign a different logical unit number to the temporary work file if youve already
assigned DVC 50 to a disk volume in your job control stream.
Suppose we select logical unit number 60 (indicating any 8419 disk) and add the DVC
parameter to our example, like this:
//GRADEOUT WORK2 VOL=DISKØ1,DVC=60
Since DVC and VOL are keyword parameters, they do not have to be in any specific
order. So, it could be coded:
J/GRADEOUT WORK2 DVC6O, VOLDI SKø1:
7004 4623-000
5-5
Doing It the Easy Way with Procedure Calls
-
Either of these two JPROC calls generate these job control statements:
iI:DVcø
ii VOL DISK1
II EXT ST,,1,BLK,(256,4@øø)
II LBL $SCR2
II LFD GRADEOUT
You can use the VOL parameter and omit the DVC parameter job control will assign
a logical unit number. The converse is not true; if you use the DVC parameter, you
must use the VOL parameter.
-
Providing the Extent Specifications
When the WORK or TEMP JPROC calls allocate temporary work files, they are, by
default, 4000 blocks each 256 bytes long. However, you can change this by using the
BLK parameter or the CYL parameter.
-
Possibly, your file doesn’t require 4000 blocks. Maybe you only need 1000 blocks.
Don’t tie up 3000 blocks that your program isn’t going to use. Use the BLK keyword
parameter to indicate that only 1000 blocks are needed:
//GRADEOIJT WORK2 DVC=60,VOL=DISKO1,BLK=1@OO
which would generate these job control statements:
//DVC6Ø
II VOL DISKØ1
II EXT ST,,i,BLK,(256,1øøø)
II LBL SSCR2
II LFD GRADEOUT
Suppose you want to allocate 3 cylinders for the file instead of 1000 blocks. In this
case, specify the CYL parameter in the JPROC as follows:
//GRADEOUT WORK2 DVC=60,VOL=DISKG1 ,CYL=3
This JPROC generates the following job control statements:
1/
/1
II
II
ii
DVC
VOL
EXT
LBL
LFD
60
DISKO1
ST,,1,CYL,3
$SCR2
GRADEOUT
C)
5-5
7004 4623-000
Doing It the Easy Way with Procedure Calls
-
In “Reserving an Extent Information Storage Area” in Section 4, we used the second
parameter of the LFD job control statement to tell the system how many extents
existed in the file. Job control used this to calculate the amount of main storage
needed to contain the information about the extents. For the WORK and TEMP
JPROC calls, you do this with the EXTSP keyword parameter.
When the number of extents is omitted, 16 is assumed. If you know your data will
take less than 16 extents, it’s a good practice to specify the EXTSP parameter. For
example, your data may only need one extent; it is foolish to let the system allocate 16.
Assuming only one extent, we would code:
//GRADEOUT WORK2 DVC=60,VOL=DJSKO1,BLK=1000,EXTSP=1
These statements would be generated:
ii
II
II
II
If
DVC
VOL
EXT
LBL
LFD
60
DISKO1
ST,,1,BLK,(256,1000)
$SCR2
GRADEOUT,1
In “Your Disk or Format-Label Diskette File Needs More Space” in Section 4, we
discussed the dynamic extension of a disk file. You can indicate how much additional
area to allocate on the WORK and TEMP JPROC calls, too. Use the SECALL
keyword parameter.
In the grading report, we estimated 1000 blocks were needed for 5000 students. If
this amount is exceeded, you will, by default, receive one additional cylinder. The
dynamic extension process takes a little time, which increases processing time.
Normally, one additional cylinder is enough extra space to contain any additional
information, but, at different times in the school year, you are called upon to do the
grading report for a neighboring school district. This district has 15,000 students.
This will no doubt exceed the 1000 blocks, and the overflow of data will take up more
than one cylinder; it will be closer to five cylinders. Job control will keep on
dynamically extending the file, in increments of one cylinder, until the needed space is
acquired. Since each dynamic extension takes time, why not request that the
extension be made all at once, by increasing the dynamic extension amount? This
additional space is only allocated when needed (and most times you run this job,
dynamic extension will not be needed). The relative cost of extra temporary space
acquired infrequently, by dynamic extension, is minimal compared with the processing
time cost required to allocate one cylinder five times. Since you know when the
special runs for the other school district will occur, they can be scheduled when the use
of these five additional cylinders will not hinder jobs being run.
70044623-000
5-7
Doing It the Easy Way with Procedure Calls
-
Let’s add a 5-cylinder dynamic extension to the example we’ve been using:
//GRADEO(JT WORK2 DVC=60,VOL=DISKO1,BLK=1000,EXTSP=11,SECALL=5
This generates these job control statements:
II
/1
II
II
II
DVC
VOL
EXT
LBL
LFD
60
DISKO1
(256,1000)
1
ST,,5,BLK
$SCR2
GRADEOUT,1
You should now be able to use the WORK and TEMP JPROC calls and tailor them to
your own needs.
By default, both the WORK and TEMP JPROCS set up temporary SAT files, but you
can also specify MIRAM files (MI) using the TYPE parameter:
For example, we can include the TYPE parameter in the previous example to indicate
a MIRAM file type. Code the JPROC as follows:
1/ GRADEOUT WORK2 DVC=60,VOL=DISKG1 ,BLK=2000,EXTSP=11 ,SECALL=5 , TYPE=MI
This generates the following job control statements:
II
/1
II
II
/1
DVC
VOL
EXT
LBL
LFD
60
DISKO1
MI,,5,BLK,(256,1000)
$SCR2
GRADEOUT,1
Accessing Previously Allocated Files
Ordinarily, to access a previously allocated disk file, you use the DVC, VOL, LBL, and
LFD job control statements. These statements aren’t needed, however, if you use the
ACCESS JPROC call. Its format is:
//tf&rnme ACCESS
lb(name
{(LbLname[
[
,
OVC=nn,VOL=
votsn
[{voL= {votsn} {RUN
{EXTEND}}
}
}1
This JPROC call can be used to access any tape or previously allocated disk file,
except a multivolume file. For instance, to access multivolume files, a file serial
number must be specified (otherwise, data management returns an error indication).
There is no parameter in the ACCESS JPROC call for this specification.
C
5.8
7004 4623000
Doing It the Easy Way with Procedure Calls
-
The ACCESS JPROC call uses both positional and keyword parameters; if you’re a
little hazy on the rules for coding, consult Appendix A.
Let’s digress a moment, and discuss the DVC and VOL parameters. The rules
governing their use are exactly the same as for the WORK and TEMP JPROC calls.
See “Using Your Own Volume” in Section 5. If you omit the VOL parameter, the file is
assumed to be on the volume containing the job’s $Y$RUN file.
Let’s set up a situation where the ACCESS JPROC call can be used to advantage.
Suppose we want to write an inventory control program for a metal fabricating plant.
This plant produces many different items: office furniture, aircraft parts, aluminum
siding, and such. Each item produced depletes a central metal inventory, and the
purchasing agent wants to know when to order new stocks of metals. After making
some further assumptions (DVC=60 and VOL=DKWORK) we have the information
needed to code a useful ACCESS JPROC call:
I/MM I F IL ACCESS METALMAST I NV, DVC=60,VOL=DKWORK
This ACCESS JPROC call generates this device assignment set:
If
II
II
II
DVC
VOL
LBL
LFD
60
DKWORK
METALMASTINV
MMIFIL
The ACCESS JPROC call has two optional positional parameters that allow you to
generate a complete LFD job control statement. In “Specifications for Existing Files”
in Section 4, we discussed how the optional parameters of the LFD job control
statement are used. Well, the optional positional parameters of the ACCESS JPROC
call correspond exactly to the parameters of the LFD job control statement.
Compare these formats:
f/Esymbol) LFD
//Lfdname ACCESS
ffilename 1
[*f[enamef
(bI.narne
{tbtn
F 11J
[nllE
L L
[{}1E
[EXTEND
uNIT
)
{EXTEND}1
}
,
DVCnn,VOL=
[{V0L
{ }}
voLsn
The two enclosed portions are identical, both in format and function.
7004 4623.000
5-9
Doing It the Easy Way with Procedure Calls
-
The n parameter specifies the number of extents reserved in main storage, and the
default value is 8.
The other optional positional parameter provides two different options: EXTEND and
INIT.
As a brief recap of “Specifications for Existing Files” in Section 4, we can say that
using the EXTEND option allows you to add information to the present end of the file.
With the INIT option, you can write over the existing information in the file (except
for the file identifier).
When you code any of these options, or specify the number of extents in the file, with
the ibiname parameter, you have to enclose them all within parentheses.
Since the metal fabricating plant buys and sells a lot of materials, the metal master
inventory file changes a great deal. You must update the metal master inventory file
to reflect any new materials purchased and sold, and perform the main processing
function.
All new material is purchased on the tenth of the month. On the eleventh, it’s time to
add the new material to the metal master inventory file. Our call line would look like
this:
//MMI F IL ACCESS METALMASTINV,DVC=60,VOL=DKWORK
By default, space is reserved for eight extents. The following device assignment set is
generated:
// DVC
/1 VOL
II LBL
/1 LFD
60
DKWORK
METALMASTINV
MMIFIL
While there are more minor limitations to the ACCESS JPROC call, there are many
instances where it’s very useful.
Allocating a File with a JPROC Call
You saw how we used the ACCESS JPROC call to access an existing disk file. This
replaced four job control statements, helping to reduce the possibility of coding errors.
You save more coding time by using the ALLOC JPROC call to allocate disk and
diskette files. It’s a combination of the ACCESS JPROC call and the EXT job control
statement.
5-10
7004 4623-000
Doing It the Easy Way with Procedure Calls
-
The format is:
//lfdname ALLOC
tb(name
(tb[name
Ij [nil 1
{F11
{11
votsn
RUN
*
7
IEXTEND1
T 11
Ll1JL1INI
EXT
DVC=nn,VOL
VOL= volsn
)
L
Tccc:hh
TBLK
CYL
TRK
OLD
*
[
{i,af)}1
IJ u,aj)I1
L
J
fmj
,..iE,0LD]E,FIX)E,ND1))
The EXT keyword parameter provides all the options available as positional
parameters on the EXT job control statement. The only difference is the equal sign
and the parentheses.
Note:
Section 4 describes the parameters and options available for data-set-label
and format-label diskette using the EXT statement.
Your site processes payrolls for 25 different companies. Each company has a file
containing each employee’s name and hourly wage. This file is accessed during the
processing of the company payroll (a use for the ACCESS JPROC call). Originally,
though, each company file was on punched cards, and each of them must be loaded
into its own disk area. (Here is one use for the data utility card-to-disk routines;. why
write your own program when one is already provided?) To do this, there must bea
device assignment set for each file being created. This means 25 device assignment
sets for the 25 files. Looking back at “Basic Job Control Statements” in Section 3, we
see that the site manager needed five job control statements to allocate his disk file:
DVC, VOL, EXT, LBL, and LFD. This means 125 job control statements would be
needed. The ALLOC JPROC reduces this to 25.
For our example, assume that the file requirements (such as access method area
needed, etc.) are identical for each of the 25 files, so most of the parameters for the job
control statements (and the ALLOC JPROC call) would be the same. Of course, each
file must have its own unique file identifier, but the information about the extents is
the same, all the files can be stored on the same disk volume, and, since you’re using
the same program to store them (the data utility routine, run 25 times), the file name
is the same.
70044623-000
5-11
Doing It the Easy Way with Procedure Calls
-
We’ll assume that disk volume DSPO28 will hold these files. It’s the only volume with
DSPO28 as the volume serial number, so a logical unit number in the range of 50
through 59 (any disk device) suffices. If we omit the DVC keyword parameter, job
control assigns the first available number in this range. Assume that the first one
available is 50. The data utility card-to-disk routine uses OUTPUTI as the file name
in the LFD job control statement; this is the value we must use as the ibiname
parameter. All the file names for the different data utility routines can be found in
the Data Utilities Operating Guide, 7004 4516.
We are going to take the default value for the number of extents (8), and we don’t
want to use any of the options for a previously allocated file.
After defining the extent information, we’ll have the parameters that are common to
all files. The only thing left will be to supply a unique file identifier for each file. AU
the files are MIRAM files (which is a default condition, MI), allocation is contiguous,
with one cylinder for dynamic allocation. The initial allocation is two cylinders. Now
we have what we can call our master ALLOC JPROC call for the 25 different files.
The only thing missing is the file identifier.
From the information we’ve gathered, our master ALLOC JPROC call for the file
identifier would look like this:
//O&JTPUT1 ALLOC xx...xx,VOL=DSPØ28,EXT=(,C,i,CYL,2)
Now, we need file identifiers for each file. Each of the 25 files must be given a unique
file identifier so the proper file can be accessed at the proper time. The names of two
of the companies are Target Manufacturing, Incorporated, and the Reality’s Dress
Company. Why not use TARGET and REALITYS as the file identifiers? It makes
them easier to remember and identify. The ALLOC JPROC call for Target
Manufacturing, Inc., would be:
//OUTPUT ALLOC TARGET,VOL=DSPØ28,EXT=(,C,1,CYL,2)
and the generated job control statements would be:
// DVC 50
// VOL DSPO28
II EXT ,C,1,CYL,2
If LBI TARGET
If LFD OUTPUT
The ALLOC JPROC call for the Reality’s Dress Company would be:
//OUTPUT1 ALLOC REAL[TYS,VOL=DSPO28,EXT=(,C,1,CYL,2)
The only difference in the generated job control statements is the file identifier of the
LBL job control statement: one is TARGET; the other is REALITYS.
5-12
70044623-000
Doing It the Easy Way with Procedure Calls
-
Note:
If the EXT keyword parameter is omitted, job control assumes one cylinder for
dynamic extension and, therefore, allocates one cylinder of extent space for a
MIRAM file.
Now, let’s see how to avoid the error of assigning one volume to different devices.
Too Many Devices for the Same Volume
Many applications use two files on the same volume. A common mistake is to assign
the files thus the volume to two different devices during the job. Using the
DVCVOL JPROC helps to avoid this. This JPROC assigns logical unit numbers for
the generated DVC job control statements. It also generates a VOL job control
statement with the volume serial number you specify in the JPROC call. The format
is:
-
-
//[symbol) DVCVOL [vo1ser-no
RES
IRUN
[,Lun][NOVOL= y
[
iN
The symbol in the label field is only used as a target for the job control statement that
causes a branch.
The DVCVOL JPROC assigns the logical unit numbers 50 through 59, in ascending
sequence, to the different volume sequence numbers in the order they are encountered
in the control stream. If you had three volumes, A, B, and C, in that sequence, A
would be 50, B would be 51, and C would be 52. It is possible, however, to override
these volumes and assign a specific logical unit number to a specific volume by using
the lun parameter.
The NOVOL parameter (NOVOL=Y) performs the same function as the NOV
parameter of the VOL job control statement. It suppresses the checking of volume
serial numbers.
Once a logical unit number is assigned by the DVCVOL JPROC call to a ‘‘olume, the
same logical unit number is assigned whenever this volume is encountered in the job.
If volume A was assigned 50 in one job step, and you tried to assign it to 51 in the next
job step, the system overrides the 51 and assigns 50.
If you tried to do this by using just the DVC and VOL job control statements,
assigning 50 in the first job step and 51 in the next job step, your job may run, but you
may have to demount the volume from DVC 50 and mount it on DVC 51.
When you use the DVCVOL JPROC call, the LBL and LFD job control statements for
the file must be present in the control stream after the DVCVOL JPROC call. If
you’re allocating a file on a disk volume, the EXT job control statement must, of
course, also be used.
7004 4623-000
5-13
Doing It the Easy Way with Procedure Calls
-
There is a limit to the number of volumes you can assign using the DVCVOL JPROC
call inajob: 10.
Another point worth remembering: the DVCVOL JPROC call can be a member of a
multistatement line of coding, but it must be the last statement on the line.
Let’s set up a control stream with some DVCVOL JPROC calls, and see what job
control statements are generated. The numbers refer to the explanation following the
example.
/1 DVCVOL
II LBL A
II LFD A
II LBL B
II DVCVOL
II LBL C
II DVCVOL
II LBL X
II LFD X
II DVCVOL
//LBLY
II LFD Y
1
2.
3.
4.
5-14
DISKØ1
If DVCVOL DSKØO2
If LFD B
DKØOO3,69
ii LFD C
DISKØ1
DKØOO3,67
1.
This is an example of a multistatement line of coding. Note that the
DVCVOL JPROC call is the last statement on the line. The next line and
the line after example 2 are also multistatement lines.
2.
This line assigns a specific logical unit number, 69, to the volume DK0003.
3.
This DVCVOL JPROC call is used again for the volume DISKO1. It was also
used in the first DVCVOL JPROC call on the first line. It will be assigned
the same logical unit number assigned to the first call for the volume
DISKO1. You’ll see this more clearly when we show the job control
statements generated by these DVCVOL JPROC calls.
4.
This is another example calling for the volume DK0003, which was already
assigned a logical unit by a DVCVOL JPROC call. Notice that it also
requests a specific logical unit number: 67. Since this volume already was
assigned to logical unit number 69 in example 2, the request for logical unit
number 67 is ignored, and it is assigned to logical unit number 69.
70044623-000
()
Doing It the Easy Way with Procedure Calls
-
Here are the generated job control statements. They should give you a clearer picture
of how each DVCVOL JPROC call functioned.
1.
II DVC 50
II VOL DISKO1
1/ LBL A
II LFD A
IIDVC51
2
II VOL DSKOO2
ii LBL B
II LFD B
3
//DVC69
VOL DK0003
LBL C
LFD C
DVC 50
VOL DISKO1
LBL X
LFD X
DVC 69
VOL DK0003
LBL Y
LFD Y
II
II
II
1/
II
II
II
II
II
II
II
4.
5.
1.
Volume DISKO1 was the volume encountered in the first DVCVOL JPROC
call it’s assigned to logical unit number 50. The LBL and LFD job control
statements are not generated by the JPROC call. Remember, these were
supplied in the control stream. If another DVCVOL JPROC call for volume
DISKO1 is encountered in this job, it is automatically assigned to logical unit
number 50.
-
70044623-000
2.
A DVCVOL JPROC call for volume DSKOO2 was the next one encountered.
It’s assigned the next available logical unit number. Since 50 was already
assigned to volume DISKO1, 51 is the next available logical unit number.
3.
The next DVCVOL JPROC call was for volume DK0003. Normally, it would
be assigned to logical unit number 52, which was the next one available.
But, the DVCVOL JPROC call for this volume requested a specific logical
unit number, 69, so that’s what is assigned.
4.
Another DVCVOL JPROC call for volume DISKO1 was encountered. Since
this volume was already requested and assigned earlier in the control
stream, this occurrence is assigned the same logical unit number: 50.
5-15
Doing It the Easy Way with Procedure Calls
-
5.
•
The volume DK0003 was requested by another DVCVOL JPROC call. Even
though a specific logical unit number, 67, was requested, it was assigned to
logical unit number 69, since this is the logical unit number assigned earlier
in the job. The first number encountered is used, and any other logical unit
numbers requested for the volume in the same job are ignored.
To assign multiple diskette volumes through a JPROC call, use the DVCDKT JPROC.
It functions the same as the DVCVOL JPROC except that it assigns the logical unit
numbers 130 through 132. Its format is:
//[syrnbol] DVCDKT voi.ser-no[,Lun)FNOVOL=
L
r’’
There is also a JPROC call for tape units: DVCVTP. Except for a few minor
differences, it functions the same as the DVCVOL and DVCDKT JPROCS. Its format
is:
Illsymbol) DVCVTP vol.ser.noEftun][PREP= {Y}
1 [NovOL=
ç}
The DVCVTP JPROC call assigns the logical unit numbers 90 through 99.
Additionally, DVCVTP has the keyword parameter PREP=Y. If specified, this
parameter functions the same as the PREP option of the VOL job control statement
(“Ignoring or Changing the Volume Serial Number” in Section 4); it causes any
information currently on the tape volume to be effectively erased.
Using the Linkage Editor
So far, we’ve discussed how to execute programs stored in a library. These programs
were not always located in this library. At one time they could have been on punched
cards in one of the programming languages, such as COBOL or RPG II.
These programs are compiled or assembled using a language translator, which
converts the program instructions into a form understandable to the computer (an
object module). Each language translator has a JPROC call you can use to generate
the job control statement needed to direct the operation of the language translator; in
other words, you get an object module from source input. The JPROC call for each
language translator can be found in the appropriate assembler, COBOL, FORTRAN,
and RPG II publications.
In this guide, we’ll explain the JPROC call for the linkage editor. But, before we do, a
word or two about the linkage editor.
5-16
70044623.000
Doing It the Easy Way with Procedure Calls
-
The linkage editor converts an object module into an executable load module. Only
load modules can be executed, and the only method of converting object modules to
load modules is by using the linkage editor. The function of the linkage editor is fully
covered in the System Service Programs (SSP) Operating Guide, UP-8841.
The format of the linkage editor JPROC call is:
JLINK
1LINKGI
rPRNTR=
I
L
[IN=
[RLIB=
L
N[,dest]
20[,dest]
(vol-ser-no, label )
(RES)
(RES, label)
(RUN, label)
(*, label)
(RUN $Y$RUN)
,OIJT=
(voL-ser-no,tabet)
(RES, Label)
(RUN, Label)
(*, I abet)
(N)
(vol-ser-no,Label) 1rALIB=
(RES,tabel)
(RUN, label)
(* (abet)
foil
rscRl= fvo( ser
J
(vol -ser-no,L abet)
(RES,L abet)
(RUN, [abet)
(*, Label)
rsTD= [YES1
1 rALTL0D=
L
1NO j U
I
L
E,OPT=Ioptionsl][JCLIB=
I
I
(vol ser no Label)
(RES,label)
(RUN,tabel)
(*,[abel)
(RUN,$Y$RUN)
(vot-ser-no,LabeL,modname)
(RES,Label,modname)
(RUN, label ,modname)
L
E,CMT=’coment’ i[,ENTER=expression]
7004 4623-000
5-17
Doing It the Easy Way with Procedure Calls
-
There are two choices in the operation field: LINK and LINKG. By specifying LINK,
you execute the linkage editor. By specifying LINKG, you execute the linkage editor
and the load module you just created (without using an EXEC job control statement).
As you can see, all the parameters are optional. But this JPROC call has default
values, which generate the job and linkage editor control statements sufficient to
accomplish a link edit, and assumes the following:
All the object code you specifically want included in the load module is in the job’s
$Y$RUN file.
•
Any object code that may have to be automatically included in the load module
(such as error processing routines) is in $Y$OBJ.
•
The load module produced is given the name LNKLOD, and it’s stored in the job’s
$Y$RUN file.
You can alter these default conditions using the optional parameters. There are also
parameters that allow you to choose special options (such as a specific printer, a
certain scratch work file, etc.).
Let’s see what job and linkage editor control statements are generated when you omit
all the parameters. We’ll use both the LINK and the LINKG operations. For these
examples, assume that the program was just compiled (or assembled) by a language
translator, and the object code was placed in the job’s $Y$RUN file. This occurred in
the last job step, but it is still the same job. The job’s $Y$RUN file is only a temporary
file, lasting for the length of the job. So, if you placed the object code in the job’s
$Y$RUN file in one job and tried to locate it in another job, you wouldn’t find it.
/*
(this is the end of the language translator job step)
/1 LINK
1/ FIN
Here’s what job control statements are generated:
/*
1
3.
4.
5.
F
1
/1
II
II
II
II
(this is the end of the language translator job step)
DVC 20
II LFD PRNTR
DVC RES
EXT ST,C,1,BLK,(256,10)
LBL $SCR1
/1 LFD $SCR1
EXEC LNKEDT
/s
i
LOADM LNKLOD
/&
II FIN
5-18
70044623.000
Doing It the Easy Way with Procedure Calls
-
1.
This is the device assignment set that’s generated for the printer. Notice
that we’ve used multistatement coding, showing the DVC and LFD job
control statements on the same line.
2.
The linkage editor always uses one scratch work area. The JPROC call
assigns it to the SYSRES device, and makes it ajob step temporary file.
(The file identifier begins with $.) This work area is scratched at the end of
the job step.
3.
This calls the linkage editor from $Y$LOD and initiates its execution.
4.
The generated load module must be assigned a name. The default is
LNKLOD. This is on the LOADM linkage editor control statement, which is
treated as data by job control, thus the 1$ and /* job control statements.
5.
As always, this indicates the end of the job.
This example is fine if you don’t want to execute the program, since default conditions
assign the load module to the job’s $Y$RUN file, which is only a temporary file. This
load module is not available to another job (but it is to another job step in the job).
This application is useful if you only want to see the output of the linkage editor; but it
isn’t much help if you want to execute. This does not mean that you can never access a
load module in ajob other than the one in which it was link edited. You can, but you
have to assign it to a library other than the job’s $Y$RUN file. You’ll see how later on,
when we discuss the optional parameters. But first, let’s see how to execute the load
module that was placed, by default, in the job’s $Y$RUN file.
There are two ways you can execute a load module placed in the job’s $Y$RUN file:
first, you can execute it in a subsequentjob step after link editing, using the default
LNKLOD load module name on the EXEC job control statement; or, second, you can
use the LINKG operation, which automatically executes the load module.
Here’s method 1 (LINK):
/* (end of Language translator job step)
II LINK
1/ EXEC LNKLOD,$Y$RUN
I&
II FIN
70044623.000
5-19
Doing It the Easy Way with Procedure Calls
-
The job control statements generated are:
/* (end of language translator job step)
II DVC 20
/1 LFD PRNTR
/1 DVC RES
II EXT.ST,C,1,BLK,(256,10)
II LFD $SCR1
If LBL $SCR1
If EXEC LNKEDT
Is
LOAPMIN.KLPP
1*
1
[1 EXEC LNKL0DSYSRUN
/&
f/FIN
The load module name on the LOADM linkage editor control statement and the
program name on the EXEC job control statement is the same: LNKLOD. Since we
know the linkage editor always assigns this as the default load module name, we use
it as the program name. Also note that $Y$RUN is the second parameter on the
EXEC job control statement. Remember, in “Specifring Qualifiers for File Identifiers”
in Section 4, we said this parameter indicates the name of the library containing the
load module. If omitted, $Y$LOD is searched and then the job’s $Y$RUN file. Since
the job’s $Y$RUN file is searched eventually, why specify it? Time. We know, it’s in
the job’s $Y$RUN file, so why search $Y$LOD needlessly? Go directly to the job’s
$Y$RUN file.
Now, here’s method 2 (LINKG):
/* (end of language translator job step)
II LINKG
I&
I/FIN
And here are the generated job control statements:
/* (end of Language translator job step-)
1.1 ..T[QN. GQ
1/ DVC 20
1/ LFD PRNTR
II DVC RES
If EXT ST,C,1,BLK,(256,10)
II LBL $SCR1
If LFD $SCR1
II EXEC LNKEDT
Is
LOADM LNKLOD
1*
/8
II FIN
5-20
70044623-000
Doing It the Easy Way with Procedure Calls
-
The only difference between this LINKG operation and the LINK operation is the
generated OPTION job control statement. The GO means the load module should be
automatically executed when linkage editing is completed. You don’t need an EXEC
job control statement.
The LINKG operation generates a load module name of LNKLOD and is loaded, by
default, in the job’s $Y$RUN file. This means it is not available after the job is
completed. The LINKG operation is useful when you’re testing programs or running
programs that are infrequently used.
So far, we’ve covered only the basic use of the linkage editor JPROC call. Now, let’s
add some optional parameters and make it do exactly what we want.
Generating LOADM and INCLUDE Linkage Editor Control Statements
Up until now, the module name for the generated LOADM linkage editor control
statement has been LNKLOD (the default name). You can override this using the
label field of the JPROC call, shown as symbol in this portion of the format:
IlEsymbot]
fLINK
tinput-moduLe-name-1,...,input-modute-riame-1ø)
The symbol parameter is a 1- to 6-alphanumeric-character name. If fewer than six
characters are specified, it’s padded on the right with zeros. If it’s omitted, the value
for the first input-module-name specified is used for the load module name. If the
input-module-name parameter is also omitted, LNKLOD is used.
Since we mentioned input-module-name, now is a good time to explain it. This
parameter allows you to specify up to 10 object modules to be included in the load
module you’re constructing. In other words, it specifies the module names for the
INCLUDE linkage editor control statements. Each input-module-name can be from
one to eight alphanumeric characters long. If this parameter is omitted, the value
specified as symbol is also omitted, all object modules residing in the job’s $Y$RUN
file are included in the load module. An explanation of how the linkage editor JPROC
searches for input modules to be included in the load module is given in the
description of the IN parameter (see “Making the Linkage Editor Suit Your Needs” in
Section 5).
If you are specif’ing more than one object module name, you may want to specify a
value in the symbol field that is representative of all the input-module-names to be
included. Also, if all eight positions are used for the first input-module-name and it is
also to be used as the symbol by default, the last two positions are truncated by the
linkage editor to obtain a 6-character symbol, and the linkage editor diagnostic
message K016 is issued.
Let’s look at examples showing different ways of assigning module names for the
generated LOADM and INCLUDE linkage editor control statements.
70044623-000
5-21
Doing It the Easy Way with Procedure Calls
-
Here’s the first example:
/*(end of language translator job step)
J/PROG LINK
II EXEC PRC.3,$Y$RUN
ii FIN
Here’s what is generated:
/* (end of language translator job step)
1/ DVC 20
1/ LFD PRNTR
II DVC RES
II EXT ST,C,1,BLK,(256,10)
/1 LBL $SCR1
If LFD $SCR1
II EXEC LNKEDT
/s
LOADM PROG
INCLUDE PROG
1*
11 EXEC PROG SYSRUN
/S
II FIN
By using PROG as the symbol, you get PROG as the module name on the LOADM
linkage editor control statement. By default, it’s also the module name for the
INCLUDE linkage editor control statement. (You’ll notice there’s no space between II
and PROG on the JPROC call.) You also use it as the program-name parameter on the
EXEC job control statement.
The same job and linkage editor control statements would have been generated if you
specified it like this:
/* (end of Language translator job step)
II LINK PROG
II EXEC PROd SYSRUN
/&
II FIN
Notice that PROG, in this case, is specified as the input-module-name, rather than the
symbol. Remember, one can substitute for the other if it’s omitted.
You could make this job a little easier by getting rid of the EXEC job control
statement, like this:
/* (end of Language translator job step)
II LINKG PROG
/&
/1 FIN
5-22
70044623-000
Doing It the Easy Way with Procedure Calls
-
The following subsections describe the rest of the parameters used with the linkage
editor JPROC call, and present examples showing what job control statements are
generated when you specify a particular parameter.
Making the Linkage Editor Suit Your Needs
Once again, the format of the linkage editor JPROC call is:
//[symbol] fLINK
[input-module-name-i
1LINKGI
,IN=
input-moduIe-name-iØ]{PRNTR= tun[,dest)
N[,dest]
I
L
(vol-ser-no,labet)
(RES)
(RES, Label)
(RUN, Label)
Label)
,OUT=
(*, Label)
(N)
RtJN,$Y*RUN
,RUN,$Y$RUN
ERLIB=
I
I
(vot-ser-no,labet
(RES, label)
(RUN, labeL)
(vot-ser-no,label)
(RES,tabeL)
(RUN,Label)
1[ALIB
L
(voL-ser-rio,L abet)
(RES, labeL)
(RUN, Label)
(*,LabeL)
[SCR1= {vot.ser.no}
1 [STD= .[YE}
1 [ALTLOD=
(vol -ser-no, I abeL
(RES, label)
(RUN, label)
(*, label)
RtJN,$T$RUN
[,OPT=boptions9[CLIB=
(vol-ser-no,LabeL,modname)
(RES, Label ,modname)
(RUN, I abet ,rno&iame)
L
[,CMT=’conwnent’ )[,ENTER=expression]
7004 4623-000
5-23
Doing It the Easy Way with Procedure Calls
-
We’ve already covered symbol and input-module-name, and the difference between
LINK and LINKG. The remaining parameters are used to define particular input and
output files, to indicate libraries to be searched for modules to be automatically
included, to define scratch work areas, and to specify the alternate library that
contains.the linkage editor (normally it’s $Y$LOD). If you want to assign a specific
printer, there’s a parameter for that. And, if you are going to provide your own
linkage editor control statements (you might want to do multiple link edits in a single
job step), you must use a parameter to indicate this.
Let’s start with the PRNTR parameter. If PRNTR=N is specified, the LINK JPROC
does not generate a device assignment set for a printer. Also, it is assumed the
PRINT file is not to be sent to a terminal. Remember, since no device assignment set
is generated, you must supply your own. The lun subparameter is used if you want to
assign a logical unit number for a specific printer (20 is the default, indicating that
any printer can be used). The dest subparameter indicates the remote destination
identifier (one to six alphanumeric characters) for the print output file when dealing
with remote batch processing, which requires that every unit record device must have
a destination.
There may be times when you want to change the spooling environment, the standard
load code, or the vertical format buffer used by the linkage editor. (These buffers are
changed with the SPL, LCB, and VFB job control statements, described in
“Controlling Spooled Output with a Job Control Statement”, “Specifying Unique Load
Codes”, and “Using Forms Control” in Section 6.) This is accomplished by coding N as
the value of the PRNTR parameter. When you code N, the JPROC will not generate a
device assignment set for the printer; you must physically insert the printer’s device
assignment set into the control stream before the JPROC call. This device assignment
set consists of a DVC job control statement and an LFD job control statement (which
must have a value of PRNTR for the file name). The SPL, LCB, or VFB job control
statement you want to use is placed between the DVC and LFD job control
statements. For example:
II
II
II
/1
II
Note:
5-24
DVC 20
VFB LENGTH=78,OVF=75
SPL RETAIN
LFD PRNTR
LINK PRNTR=N
Other JPROCS allow you to use the PRNTR=Nparameter and supply your
own device assignment set for the printer. A.ll the language JPROCS and the
JPROCS for generating control streams for data utility routines allow you to
specify PRNTR=N. This parameter is used in these JPROS exactly as its
described for LINK/LINKG.
7004 4623.000
Doing It the Easy Way with Procedure Calls
-
Next, let’s look at the parameter for the input file definition:
,IN=
(vot-ser-no,tabeo)
(RES)
(RES, (abet)
(RUN, (abet)
( 1)
The linkage editor uses two processes to include modules specific and automatic
inclusion. Modules specified in the input-name parameter and modules specified on
embedded INCLUDE statements are specifically included. For each inp ut-modulename specified, the linkage editor performs a specific inclusion search in the following
manner: If the IN parameter is specified, only the file it identifies is searched; if the
IN parameter is not specified, first, $Y$RUN is searched for object modules to include.
Then the file defined in the RUB parameter is searched (if the RUB parameter was
specified) and, finally, $Y$OBJ is searched.
-
For automatic inclusion, the linkage editor performs a search in the following manner:
The file defined by the ALIB parameter is searched first (if the AUIB parameter was
specified), and then the file defined by the RUB parameter (or the default $Y$OBJ) is
searched. Modules are automatically included to satisfy the external requirements of
modules that have already been included by either automatic or specific inclusion.
Automatic inclusion may be suppressed by specifying the NOAUTO option.
Here are the options available to you through the IN parameter.
The first option is IN=(vol-ser-no,label). The vol-ser-no is the volume serial number of
the disk volume you’re using, and the label is the file identifier of the file used when
the file was created.
The next choice is IN=(RES). This means the file is on SYSRES in $Y$OBJ.
The following two choices are very similar: IN=(RES,label) and IN=(RUN,label). In
both, label stands for the file identifier. If you use RES, the file is on SYSRES; if you
use RUN, the file is on the volume containing the job’s $Y$RUN file. (Remember,
$Y$RUN can be on the SYSRES device.)
The next choice is IN=(*,label). This means the file is cataloged; therefore, its location
is obtained from the file catalog.
The default parameter CR UN,$Y$R UN) should not be coded when you want to use the
default; its use in coding can cause an invalid file name.
Whenever you use the iN parameter, with both subparameters (vol-ser-no,label , for
example), and STD=NO is omitted, an INCLUDE module-name/IN linkage editor
control statement is generated.
70044623.000
5-25
Doing It the Easy Way with Procedure Calls
-
The next parameter we’ll discuss defines the output file. Here’s what it looks like:
,OUT=
(vo(-ser-no,Iabel)
(RES, Label)
I
Label)
(* Label)
(N)
(PH.,
Quite frequently, you will not want to permanently save the generated load module,
particularly when you don’t have all the bugs out of your program. However, once the
program is working satisfactorily, you’ll probably want to save the load module, rather
than compiling (or assembling) and link editing it every time you run it (unless it’s
used only once a year, for example). This is done with the OUT parameter.
As we’ve said, most times you don’t want to save the generated load module for any
length of time (but you’ll probably want to execute it in the next job step to see how
close to the finished product you are). For this reason, the linkage editor JPROC
places the generated load module in the job’s $Y$RUN file by default.
But, once the module is ready to be saved, you override the default in one of these
ways.
You can specify OUT=(vol-ser-no,label). This is the volume serial number and the file
identifier of the file where you want to store the load module.
The following two choices are similar: OUT=(RES,label) and OUT=(RUN,label). This
is like the IN parameter we just discussed. label is the file identifier; .RES means the
file is on SYSRES; RUN means the file is on the volume containing the job’s $Y$RUN
file.
The next choice is OUT=(*,label). This means the file is cataloged; therefore, its
location is obtained from the file catalog.
The last choice is OUT=(N). This means you don’t want to save the load module at all,
not even for the next job step. When this option is used, all you get is a listing of the
load module, which you can use for debugging. The generated load module is not
placed in any file.
Just as with the IN parameter, the default (RUN,$Y$RUN) should not be coded.
Whenever the OUT parameter is coded, a PARAM OUT=OUT job control statement is
generated to specify the linkage editor option that an output file has been defined for
the load module. The PARAM job control statement is explained in Section 6.
5-26
7004 4623.000
Doing It the Easy Way with Procedure Calls
-
The linkage editor JPROC call assumes the output file is already allocated. If it isn’t,
you must allocate the file by placing a device assignment set in the control stream
before the linkage editor JPROC call. Let’s clarify this with an example. Suppose you
want to store the load module on disk volume DISKO1, and you want it placed in the
file identified by SAVEDPROGRAM. This file has never before been allocated. So,
what you have to do is allocate the file before you can link edit the module.
You’ve probably noticed that the logical unit number is not coded in the OUT
parameter (or any other except for the printer). This is because the linkage editor
JPROC call uses the DVCVOL JPROC call (a JPROC call within a JPROC call, which
is in turn converted to DVC and VOL job control statements). In “Too Many Devices
for the Same Volume” earlier in this section we explained how there can be conflicting
device assignments and how the DVCVOL JPROC call eliminates this conflict. So,
we’ll use the DVCVOL JPROC call in the device assignment set.
The OUT parameter generates a file name of OUT for the generated LFD job control
statement of the device assignment for the output file. So, we might as well use OUT
as the file name when we allocate the file. (We don’t have to, since the program does
not have to have a match for this file name; it’s only serving the purpose of completing
the device assignment to allocate a file. Think of it as a job step in itselfi Remember
that OUT is the file name used by the JPROC. In “Specifying Qualifiers for File
Identifiers” in Section 4 we said that, if the load module is stored in a user library (a
function of the OUT parameter), you have to use the file name of the device
assignment set for this library as a parameter in the EXEC job statement. This will
be a lot clearer in the example.
First we start to allocate the file
using the DVCVOL JPROC call.
II DVCVOL DISKØ1
Next, an EXT job control statement.
If EXT ST,C,3,CYL,1
Now the file identifier,
II LBL SAVEDPROGRAM
and then the file name
that allocated the file.
/1 LFD OUT
Now, the Linkage editor JPROC call
(Let’s call the load mobiLe XYZ),
II XYZ LINK OUT=(DISKØ1,SAVEDPROGRAM)
and execute the program.
/1 EXEC XYZ,OUT
—
If the file is already allocated, the load module created is appended to the present end
of the file. If a load module with the same name tlready exists in the file, it is
replaced by the new load module.
When you specify the LINKG operation, you can’t use the OUT parameter to define a
specific output file. You must use the job’s $Y$RUN file.
70044623-000
5-27
Doing It the Easy Way with Procedure Calls
-
Next, the parameters RLIB and ALIB name libraries that contain object modules,
such as your own (user-written) subroutines, for inclusion in the load module. To see
exactly how and why different object modules are included into your load module, see
the System Service Programs (SSP) Operating Guide, UP-8841.
By default, the linkage editor searches $Y$OBJ for the needed modules for automatic
inclusion processing. The ALIB parameter allows you to specifSr an additional file to
be searched. This file is searched first. If all the needed modules are not found here,
$Y$OBJ, or the file named by the RLIB parameter, is searched.
The RUB parameter names the file to be searched before $Y$OBJ during specific
inclusion processing, and in place of $Y$OBJ during automatic inclusion processing
when no ALIB parameter is specified.
Both the RLIB and AUIB parameters look very much alike:
LRLIB=
I
I
(voL-ser-no,Labet)
(RES,labet)
(RUN,Labe()
L
LALIB=
I
L
(vol-ser-no,tabet)
(RES,LabeL)
(RUN,tabel)
(*Lal,et)
In RLIB=(vol-ser-no,label) and ALIB=(vol -ser-no,label) , you provide the volume serial
number and the file identifier of the file containing the library you want.
RLIB=(RES,label) and ALIB=(RES,label) are similar, just as are RLIB=(RTJN,label)
and AUIB=(RUN,label). The label provides the file identifier; RES means the file is
on SYSRES; RUN means the file is on the volume containing the job’s $Y$RUN file;
the asterisk (*) means the volume is identified in the file catalog.
Whenever you use the RLIB orALIB parameters, PARAM job control statements are
generated to specify the linkage editor option for libraries for inclusion processing.
These PARAM job control statements are:
•
PARAMRLIB=RLIB
•
PARAM ALIB=ALIB
C
5-28
7004 4623-000
Doing. It the Easy Way with Procedure Calls
-
The linkage editor needs one scratch work file. Normally, SYSRES is used, but you
can use a different volume:
SCR1= [vot-ser-no
1RES
This parameter, whether specified directly or indirectly through default, generates all
the job control statements needed to allocate a job step temporary work file.
The linkage editor JPROC call often follows immediately after one of the language
translation JPROC calls. Each of the language translators also uses scratch work files
(the SCR1 parameter; some also use SCR2 and SCR3). The SCR1 parameter coded
for the linkage editor must agree L”ith the SCR1 parameter for the language
translator; you can’t contradict this assignment without getting errors. So, if you
specified SCR1tDSPO28 in the language translator JPROC call, you must do the same
in the linkage editor JPROC call.
You’ve already seen that the symbol field provides a name for the generated LOADM
linkage editor control statement, and the input-module-name parameters provide the
names for the generated INCLUDE linkage editor control statements. However, there
are times when you want to physically place these linkage editor control statements in
the control stream as data; you don’t want the JPROC call to generate them. You
indicate this by using the STD parameter.
For instance, you may want to include only certain parts of an object module to form a
load module. Since there’s no provision for doing this with the JPROC, you have no
choice but to physically place the linkage editor control statements you need in the
control stream. But, you have to use the STD parameter to tell the linkage editor
JPROC that you’re going to do this, or else it automatically looks at the inp ut-modulename parameters, and then the symbol field, for the name of an object module to
include. Since you didn’t specify the linkage editor control statements through the
JPROC call (they’re physically in the control stream), these fields would be blank.
Another case: you may want to use additional linkage editor control statements as
well. (OVERLAY, for example, there’s no parameter for this.) Whenever you place any
linkage editor control statement physically in the control stream, all the needed
linkage editor control statements must be physically placed in the control stream.
The STD parameter looks like this:
STD= iYE
1NO
Indicating NO means you’re goingto physically place the linkage editor control
statements in the control steam. The default value, YES, means you want them
generated automatically.
STD=NO tells the JPROC to ignore any specifications in the JPROC call for
automatically generating INCLUDE and LOADM linkage editor control statements.
70044623-000
5-29
Doing It the Easy Way with Procedure Calls
-
Next, let’s look at the parameter telling the JPROC where to find the linkage editor:
ALTLOD=
(voL-ser-no,Labet)
CRES, (abet)
(RIJN,( abet
(*, (abet)
Normally, the linkage editor resides in $Y$LOD. However, you may want to use a
copy of the linkage editor that is not in $Y$LOD. The ALTLOD parameter allows you
to identify the file that contains the linkage editor you want to use. You may specify a
volume serial number, RES, RUN, or an asterisk (*)• RES means the file is on
SYSRES; RUN means the file is on the volume containing the job’s $Y$RUN file; and
the asterisk means the volume is identified in the file catalog. In all cases, the label
provides the file identifier. If the ALTLOD parameter is omitted, the normal
procedure of searching $Y$LOD and the $Y$RUN is performed.
The next parameter we discuss is one making available certain linkage editor options.
The parameter looks like this:
OPTION=’options’
The options that may be specified here are all the keywords appearing in the linkage
editor /IPARAM and LINKOP control statements that do not need to be equated to
subparameters as, for example, SHARE, NOSHARE, AUTO, and NOAUTO. Refer to
the linkage editor section of the System Service Programs (SSP) Operating Guide
(UP-8841) for all the options.
The GLIB parameter looks like this:
CLIB=
(vot-ser-no,labe(,mothame)
(RES, (abet ,modne)
(RUN, I abet ,mocfriame)
(*, (abet ,modname)
You use this parameter to specify where the linkage editor control statements reside
that are to be processed for this link-edit job. As the parameter indicates, you must
supply the name of the source module and the file in which it resides. You must also
specify the disk volume on which the file resides.
The CMT parameter inserts a character string in the comment field of each phase
header record produced for the generated load module. Its format is:
CMT=’ comment’
The character string you choose may run up to a maximum of 30 characters and must
be enclosed in apostrophes. It may contain blanks, commas, and other special
symbols, excluding apostrophes.
C
5-30
7004 4623-000
Doing It the Easy Way with Procedure Calls
-
The ENTER parameter specifies the transfer address. The ENTER parameter looks
like this:
ENTER=expression
The expression is a decimal number from one to eight digits, a hexadecimal number
from one to six digits in the form X’nnnnnn’, a previously defined symbol, or a
previously defined symbol plus or minus a decimal or hexadecimal number, in the
form we’ve just discussed.
Now, let’s do some coding.
Cott.xm 72
1.
2.
II JOB LNKJPROC
//BELLPR LINK PAYROLL,IN=(DISKØ1,PRAREA),
OUT=(DISKØ1,BELLHRLPR)
I/i
I&
3.
X
II FIN
1.
This is the JOB control statement telling the operating system that the
name of the job is LNKJPROC.
2.
This is the JPROC call. (We’re only link editing, not automatically
executing, also. Thus, the operation is LINK, not LINKG. Besides, the OUT
parameter is used. When an output file is specified, the LINKG operation
can’t be used.) As you can see, we used the IN and OUT parameters. The
source deck was already compiled (let’s say yesterday), and the IN
parameter indicates it’s stored in the file identified by PRAREA, on disk
volume DISKO1. The OUT parameter indicates we also want the load
module to be stored on disk volume DI3KO1. This payroll is for the l3eil
Historical Library, so we chose a file identifier that closely represents the
name: BELLHRLPR. (Assume this file has already been allocated;
otherwise, we’d need a device assignment set to allocate the file.)
When the object module was created (compiled or assembled), it was given
the name PAYROLL. So, this is the name of the object module we want to
obtain from the file identified as PRAREA. This provides us with the inputmodule-name parameter, which generates an INCLUDE linkage editor
control statement for this module.
We’re providing a name for the load module by using the symbol field. We
also want to make this name readily identifiable with the company name.
Since the symbol field is limited to six characters maximum, we can’t use
BELLHRLPR, as we did for the output file identifier. (Also, two identical
names in the same JPROC call could cause confusion.) We’ll shorten it to
BELLPR. This is what will appear on the generated LOADM linkage editor
control statement. When you want to execute this load module, this is the
program-name you’d use on the EXEC job control statement.
70044623.000
5-31
Doing It the Easy Way with Procedure Calls
-
This ends the job and terminates the card reader operations.
3.
Now here’s what the JPROC call generated:
f
I
2.
II
II
/1
II
II
Is
3.
4.
5.
I/JOB
II DVC
/1 DVC
If LBL
if DVC
II LBL
INKJPROC
20
II LFD PRNTR
50
II VOL DISK01
PRAREA
II LFD IN
50
II VOL DISK01
BELLHRLPR
II LFD OUT
DVC RES
EXT ST,C,1BLK,(256,10)
LBL $SCR
1/ LFD $SCR1
EXEC LNKEDT
PARAN OUT=O(JT
LOADM BELLPR
INCLUDE PAYROLL,IN
f
1*
I&
II FIN
1.
This is generated by the IN parameter. The linkage editor uses the
DVCVOL JPROC (which we’re showing in its generated form: DVC and
VOL). DISKO1 is the first volume requested in the job, so it receives the first
logical unit number: 50. The IN parameter always generates a file name of
IN for the LFD job control statement.
2.
This is generated by the OUT parameter. Again, DISKO1 was requested in
the JPROC call, and since it was already assigned to logical unit number 50,
this number is assigned to this volume every time it’s encountered in the job.
The OUT parameter always generates a file name of OUT for the LFD job
control statement.
3.
This is the device assignment set for the scratch work area, which was
generated by default in this case.
4.
This is the PARAM job control statement generated by the OUT parameter.
5.
This is the object module name (PAYROLL) and the load module name
(BELLPR). These linkage editor control statements are generated by the
input-module-name parameter and the symbol field. The IN shown on the
INCLUDE linkage editor control statement is generated because both
subparameters on the IN keyword parameter are used.
We’ve now covered all the parameters of the linkage editor JPROC call and provided
examples of their use. You should now be able to use this JPROC call correctly.
5-32
7004 4623-000
Doing It the Easy Way with Procedure Calls
-
Personalizing Your Print Output
Unisys provides the WRTBIG and WRTSML JPROCS to produce block characters on
your printed output. Any type of information can be printed by WRTBIG and
WRTSML your name or a message, for example.
-
WRTBIG and WRTSML function in the same way; the only difference between the
two is the size of the block characters produced. Those created by WRTSML are
smaller than those created by WRTBIG.
WRTBIG and WRTSML produce block characters formed by the characters
themselves, arranged in the pattern of the characters being printed. You can print the
letters A through Z and the numbers 0 through 9. In addition, you can use WRTBIG
and WRTSML to print these special characters:
Left parenthesis
I
Left bracket
Right parenthesis
I
Right bracket
+
Plus
&
Ampersand
‘
Apostrophe
*
Asterisk
@
At
Hyphen
>
Greater than
/
Slash
<
Less than
?
Question mark
I
Vertical line
Colon
!
Exclamation Point
#
Number
;
Semicolon
=
Equal
-
$
Double quote
—
Underscore
Period
,
Comma
Dollar
%
Percent
Embedded blank
Note:
Some printers cannot print all of these characters check with your system
administrator.
-
Up to eight blocks, or lines, of print can be generated by WRTBIG and WRTSML.
7004 4623.000
5-33
Doing It the Easy Way with Procedure Calls
-
Each line produced by WRTBIG can contain up to 12 characters. A maximum of four
lines can be printed on each page. WRTBIG produces characters 10 characters high
and 8 characters wide. The letter P, for example, prints like this:
pppppp
ppppppp
pp
pp
pp
pp
ppppppp
pppppp
pp
pp
pp
pp
WRTSML produces characters seven characters high and five characters wide. Up to
20 characters can be printed on each line, and up to 6 lines can be printed on each
page. The number 7 produced by WRTSML looks like this:
77777
7
7
7
7
7
7
Note that the character produced by WRTSML is 7 characters high and the one
produced by WRTBIG is 10 characters high.
The format for WRTBIG and WRTSML is:
//[symboL)
fWRTBIG
lTsMIf
‘b[ock-1’[,’bLock-2’,...,’btock-8’]
f,IN=
I
I
I
L
[LUN
L
5-34
(vot-ser-no,tabet)
(RES,Labe[)
(RUN,Labet)
(*LaI()
(RE
C nnn
{2a
}
,
tfe
{PRNTR
}
[,dest])
7004 4623-000
Doing It the Easy Way -with Procedure Calls
The ‘block’ parameter is where you code the actual information you want printed on a
line. Notice there are eight ‘block’ parameters one for each line of print. Each
parameter is enclosed by apostrophes. You can use blanks anywhere in the field to
position the characters on the page.
-
For instance, if you coded this:
II WRTSML
‘
RETURN’,’
TO’,’
JOHN DOE’
you get:
RRRR
EEEEE
R
RE
R
RE
RRRR
EEE
RR
E
RR
E
R
R EEEEE
TTTTT
T
T
T
T
T
T
U
U
U
U
U
U
Ii RRRR
N
N
UR
RNNN
UR
RNNN
U RRRR
N NN
URR
N
N
URR
N
N
UUUU R
R N
N
TTTTT
T
T
T
T
T
T
JJJ 00000 H
H N
N
J
0
OH
HNNN
J
0
OH
HNNN
J
0
0 HHHHH N NN
JO
OH
HN
N
JJ
0
OH
HN
N
JJ
00000 H
H N
N
00000
00
00
00
00
00
00000
DDDD
00000 EEEEE
D
DO
OE
D
DO
OE
D
D 0
0 EEE
D
DO
OE
DO
D
OE
DDDD
00000 EEEEE
Notice that even though the field can be 12 characters, it does not have to be. You can
put the end apostrophe after the last character for the line. Also, note that if there are
over 12 characters for WRTBIG or over 20 characters for WRTSML the field is
truncated.
You can also use WRTBIG and WRTSML to print the date, the time the job started,
the system version number, and the job name from the JOB control statement. This is
done by coding the following as the first four characters in any ‘block’ parameter
(nothing else can appear in the parameter; the last eight positions must not be used):
o
TIM$
This prints the time of day in the form of hh:mm:ss (hours, minutes, seconds).
•
DAT$
This returns the date, in the form of yy/mm/dd (year, month, dày).
7004 4623-000
5.35
Doing It the Easy Way with Procedure Calls
-
•VER$
This gives you the version number of the operating system in use.
•
JOB$
This prints the job name from the JOB control statement.
Each option can be used alone, or combined with other options or information.
Look at this example:
Coti.ITrl 72
// JOB POCO
ii WRTBIG ************I,
‘JOBS’,
Ill
.‘DATS’,
//2
X
X
X
1/3
(Remainder of your
controL stream)
Note the use of statement continuation. The printout would look like this:
PPPPPP
PPPPPPP
pp
pp
pp
pp
PPPPPPP
PPPPPP
PP
PP
PP
PP
999
99999
99
99
99
99
99 999
99999
99
99
99
99999
999
5-36
000000
00000000
oo
00
oo
00
00
00
00
DC)
00
00
00
00
00000000
000000
0000
00 00
00
00
00
00
00
00
00
00
00
00
00
00
00 00
0000
CCCCCC
CCCCCCCC
CC
CC
cc
CC
CC
CC
CC
CC
CCCCCCCC
CCCCCC
/
II
III
II!
/1/
III
III
/1/
II
1
000000
00000000
00
00
oo
oo
00
00
00
00
00
00
00
00
00000000
000000
0000
00 00
00
00
00
00
00
00
00
00
00
00
00
00
00 00
0000
8888
888888
88 88
88 88
8888
888888
88
88
88
88
888888
8888
I
/1
///
III
/1/
III
III
/1/
//
/
0000
00 00
00
00
00
00
00
00
00
00
00
80
00
00
00 00
0000
8888
888888
88 88
88 88
8888
888888
88
88
88
88
888888
8888
70044623-000
Doing It the Easy Way with Procedure Calls
-
The IN parameter identifies the file containing either the load module WRTBIG or the
load module WRTSML. If you don’t specify this parameter, it is assumed that the
module you want is on SYSRES in the file $Y$LOD. If the load module is on SYSR.ES,
but in a file other than $Y$LOD, specify (RUN,label), where label is the file identifier.
To indicate that the load module is on the volume containing the job’s $Y$RUN file,
use (RUN,label). If the file containing the load module is identified in the file catalog,
use (*,label).
The LUN parameter provides the logical unit number of the printer to be used. By
default, 20 is used. But, if you want a specific printer, use the appropriate logical unit
number. (Make sure the rest of your print output goes to this printer.)
If the file name in the job is not PRNTR (which the programs supplied by Unisys use),
you indicate this through the lfdname of the LUN parameter (this is similar to the
LFD job control statement).
The dest subparameter indicates the remote destination identifier (one to six
alphanumeric characters) for the print output file when dealing with remote batch
processing, which requires that every unit record device must have a destination.
You can change the standard load code or vertical format buffer used for the job by
coding N as the value of the LUN parameter. This indicates that the JPROC is not to
generate a device assignment set for the printer; you must physically place the device
assignment set for the printer in the control stream before the JPROC call.
Suppose you wanted to use WRTSML to print the date at the beginning of the
printout, and the file name for the printer in the program is LISTER. You would code
it as:
/1 WRTSML ‘DAT$’,LUN=(,LISTER)
7004 4623-000
5-37
Doing It the Easy Way with Procedure Calls
-
Controlling Spooled Output with a JPROC Call
The maimer in which spooled output files (print, punch, or data-set-label diskette) are
handled is set at SYSGEN time, but you can alter the standard operation of individual
files with the SPOOL JPROC. To fully understand the function of this JPROC, you
should be familiar with spooling, which is discussed in the Spooling and Job
Accounting Operating Guide, 7004 4581.
When used, the SPOOL JPROC must be included in the device assignment set for the
spooled output file. The format of the JPROC is:
//[symbot] SPOOL, [EDIRECT= [DISK
11[,BUF=nXIn]LCOPIES
L
ITAPE
H
DISKETTEJJ
I
L
LSKIPCODE
L
LHDR= [NO
L
In11LREc0RDS= [n
lfJ[
[HOLD=
L
Note:
1[,FORMNAME=forms]
lø j
11IJTESTPAGE= INO 1iE,PAGEBRKn]
IYESIJ
1YESI]
LIJPDATE= [No
L
I
L
ii IJCOMPRESS=
1YESI]
fNP
1YESIJ
L
JSECURE
[NO
rio
ii [RETAIN= jYES
lsfJ L
INO
L
When using the SPOOL JPROC for a spooled data-set-label diskette output
file, only BUF, RETAIN, UPDATE, COMPRESS, and HOLD keyword
parameters are meaningful.
The REDIRECT keyword parameter redirects spooled output (output that’s already in
the spool file volume) to a disk, tape, or format-label diskette for temporary storage
-
the output is printed or punched later. A spooling component known as the output
writer assigns the tape, disk, or format-label diskette volumes to be used for the
redirected output so you don’t have to include a special device assignment set in your
job control stream for these volumes.
Notes:
1.
When you specify REDIRECT=TAPE, make sure that a DEV operator command,
directing all spooled output to a tape volume, is not in effect for this copy of the
output writer. A note to the operator should suffice.
2.
REDIRECT=DISKETTE means redirect the spooled output to a System 80 formatlabel diskette only.
CI
5-38
70044623-000
Doing It the Easy Way with Procedure Calls
-
The COPIES keyword parameter allows you to specify the number of times (up to 255)
you want a spooled file printed or punched (output). If you don’t specify this keyword,
the file is output only once and then deleted from the spooi file. If you specify 0, the
output is written to the spool file but is immediately deleted instead of being printed
or punched.
The BUF keyword parameter sets up buffers to be used by the spool subfile being
created. The n specifies the number of buffers, Xis a constant, and m specifies the
size of each buffer (in 256-byte increments). If you omit this parameter, the spooled
file shares the job log buffers along with other spooled files not having reserved
buffers.
You must specify SKIPCODE if you’re requesting a filed vertical format buffer (via the
II VFB statement) that has more than seven skip codes or if the system default vertical
format buffer has more than seven skip codes. Three skip codes are always included
in this count: home position for current page, overflow for next page, and home
position for next page. The four remaining are for user-specified skip codes. This
parameter, therefore, specifies the total count of lines on a form where a skip code is
allowed, plus three. Zero indicates no skip codes.
The RECORDS keyword specifies the number of records (lines, including spaces and
skipped lines for print files, cards for punch files) the spool file can contain before a
message asking if the job should be continued, breakpointed, or cancelled is sent to th
operator. The operator receives this message only when the specified number is
reached, and job processing stops until the operator replies. The specified number is
rounded to the next higher multiple of 1024. For example, if you specify 7000, it’s
rounded to 7168. The highest number you can specify is 262,144. Value 0 specifies no
limit is put on the number of records that can be entered into the spool file.
Note:
Ifyou’re executing a COBOL program that uses the WRITE verb with the
AFTER clause, the number you specify for RECORDS should be double that
of the actual number of records.
If your spooled file is to be output on a special printer form or on special cards, you
must identify the special form or card type in the FORMNAME parameter. The form
name you specify is a 1- to 8-alphanumeric-character name assigned by your
installation to each form. A message identifying the form or card type to be used is
issued to the operator. Remember, a form name specified in a VFB statement
overrides a form name specified in the SPOOL JPROC (see “Using Forms Control” in
Section 6).
The HDR parameter (HDR=NO) suppresses the printing of a page header in burst
mode at the beginning of the spooled file when it’s output. If omitted, a page header is
automatically printed.
If you specify the FORMNAME parameter, a query is directed to the operator asking
if a sample test pattern page should be printed. Specifying TESTPAGE=NO
suppresses this query. If the TESTPAGE parameter is omitted, the system default
(YES) is used. This query does not occur for STAND1 forms.
70044623-000
5-39
Doing It the Easy Way with Procedure Calls
-
You use the PAGEBRK parameter to specify the number of pages or cards to be
spooled out before the file is breakpointed and printed or punched. The highest value
you can enter is 32,000. When you omit this parameter, the file is printed or punched
according to the burst or nonburst operating mode in effect.
The UPDATE parameter (UPDATE=NO) specifies that the spool file subdirectory
entry is to be updated only when a file is closed. (In this case, if the system halts, you
lose any output the program generated prior to restarting the IPL with spool file
recovery.) If you omit this parameter, the spooler updates the subdirectory each time
it crosses a logical track in the program file. (In this case, if the system halts, you can
still print any output the program created prior to starting the IPL again.)
Using the COMPRESS keyword parameter (COMPRESS=N0) you can prevent the
system from attempting to compress data that’s directed to the output spool file.
Normally, you should not specify COMPRESS=NO if the data contains a large
number of embedded blanks or if the file has a block size larger than 120. Specifying
this parameter when the block size is 121 or greater results in an output spool file
containing only one line per sector and requires that n x m be at least 2 x 4.
,
If you specify RETAIN—_YES, the spooled output file is printed, punched, or placed on
data-set-label diskette, but it is also retained in the spool file to be printed, punched,
or output to data-set-label diskette again at a later time. If RETAIN is specified with
REDIRECT (the first keyword parameter), the output file is redirected to a tape, disk,
or format-label diskette and it is also retained in the spool file for printing, punching,
or outputting to data-set-label diskette at a later time.
The HOLD keyword parameter (HOLD=YES) simply holds the spooled print, punch,
or data-set-label diskette output file for later processing. (Files on hold are released
when the BEGIN SPL command is issued or when a CC job control statement
specifying the BEGIN SPL command is encountered in ajob stream.) This parameter
is useful if you have a large spooled file that will take a long time to output and you
don’t want to tie up the output device during peak processing time. Remember
though, since the file being held remains in the spool file, there is a possibility that the
spool file’s available disk space may be exhausted. Also, if you specify HOLD=YES in
conjunction with RETAIN, REDIRECT, or both, the or’tput file is put on hold and the
RETAIN or REDIRECT parameters are not acted upon until the file is released.
5-40
7004 4623-000
Doing It the Easy Way with Procedure Calls
-
The last keyword parameter (SECURE) determines whether print output that’s
destined for an auxiliary workstation printer is secured or not secured. (Spooled
output is directed to an auxiliary workstation printer via II ROUTE or II OPTION
OUT.) We say the print file is secured if the workstation to which the auxiliary printer
is physically connected must be logged on before the output file can be printed. If the
workstation isn’t logged on, the file will not be printed. If the file is not secure (this is
the default), the file will be printed at the specified auxiliary workstation printer
whether or not the workstation is logged on. Here is an example of a job using the
SPOOL JPROC to control output spooling.
/1
II
/1
II
II
ii
II
II
II
II
II
I&
If
JOB PAYROLL
DVCVOL DSPO28
LBL JONESPAYROLL
LFD JONESPAY
DVC 130
SPOOL BUF=4X32
LFD JONESYTD
DVC 20
SPOOL HOLD=YES
LFD JONESCHK
EXEC JONCKS
1
Device assignment set for
the input file on disk.
j
1
j
1
Device assignent set
for a spooled printer
output file.
Device assignment set for a
spooled data-set- label diskette
output file. (See “Spooling
Input Card Data” in Section 6
for information about the DVC
statement for spooled
data-set-label diskette files.)
FIN
We have now finished our discussion of what is known as basic job control. From this
point on, we enter the area of advanced job control programming. You’ll learn how to
use the advanced job control statements to perform functions that cannot be done with
the basic set. You’ll also learn how to write your own job control procedure definitions,
which you can store and call when needed.
By now, your grasp ofjob control should be such that you could construct control
streams for the majority ofjobs in your installation. When you finish Sections 6
through 9, you should be able to construct control streams for any job.
70044623.000
5-41
0
0
0
Section 6
Making Job Control Work for You
Advantages of Using Advanced Job Control Statements
As you have just seen, quite a lot can be done by using the job control procedure calls
(JPROCS) and the basic job control statements supplied by Unisys. Now we’ll see how
to increase performance by using the advanced job control statements and JPROCS
that you write yourself. Your basic objective is to run jobs in the most efficient, most
economical, and quickest way possible. This objective is achieved not only by how you
write a program, but also in the way you use job control.
Controlling Spooled Output with a Job Control Statement
In Section 5, we discussed how you can alter the standard operation (established at
SYSGEN time) of spooled output files with the SPOOL JPROC. The SPL job control
statement provides the same facilities and parameters as the SPOOL JPROC, so the
following brief description of the SPL job control statement is essentially a review of
“Controlling Spooled Output with a JPROC Call” in Section 5. When deciding
whether to use the SPLjob control statement instead of the SPOOL JPROC, keep the
following in mind: although the SPOOL JPROC is easier to code because of its
keyword (rather than positional) parameters, it takes more time for the run processor
to process the SPOOL JPROC.
The format of the SPL job control statement is:
//[syrnbol] SPL
HOLD
RETAIN
TAPE
DISK
DISKETTE
FI
L
[,nXm)
0
,
no-cop
J
1iiD1 E,NOCMP)E,RETAIN][,HOLD]E,SECURE)
1UPD
70044623-000
,
Ino-skpcodellE max-rec 7E,forms)j INOHDR11[ fNOTSTL
5120
I
I 1HDR ff1 1STL
f
L
E,brk-pge]
Note:
[{ }
f
When using the SPL statement for a spooled data-set-label diskette output file,
only the nXm, NOUPD, NOCMP, RETAIN, and HOLD parameters are
meaningful. The remaining parameters are ignored.
6-1
Making Job Control Work for You
The HOLD parameter holds the spooled output file (print, punch, or data-set-label
diskette) for later processing. Files on hold are released by a BEGIN SPL command
or by a CC job control statement specifying a BEGIN SPL command. You’ll notice
that HOLD is also the last parameter of the SPL statement. This is so you can specify
HOLD (as the first parameter) or choose one of the other options for the first
parameter and still specify HOLD (last).
With the RETAIN parameter, the spooled output file is processed (printed, punched,
or placed on data-set-label diskette), but it is also retained in the spool file for
processing at a later time. For the same reasons mentioned for HOLD, you can specify
RETAIN as the first or the twelfth parameter.
You use the TAPE, DISK, and DISKETTE parameters to redirect spooled output to
tape, disk, or format-label diskette for temporary storage. The output can be
processed (printed, punched, or placed on data-set-label diskette) at a later time.
The remaining parameters can be summarized as follows:
6-2
•
The nXm parameter establishes buffers for use only by the spool subfile being
created.
•
The no-cop parameter allows you to specify the number of times (from 0 to 255)
you want a spool file processed (printed or punched). Zero indicates no output.
•
The no-skpcode parameter must be specified if a filed vertical format buffer
(requested via II VFB) or the system default vertical format buffer has more than
seven skip codes. One skip code for forms overflow and two for home paper
position are always included in this count.
•
The max-rec parameter specifies the number of records the output file can
contain before a message is sent to the operator asking if the job should be
continued, breakpointed, or cancelled. Value 0 specifies that no limit is put on the
number of records that can be entered into the spool file.
•
The forms parameter identifies any special form or card type (other than the
standard paper or cards) needed when the spool file is output.
•
The NOHDR parameter suppresses the printing of a page header at the
beginning of a print file; the HDR parameter prints it.
•
The NOTSTL parameter suppresses the test line request message to the operator
when a forms change is required. This request message does not occur for
STAND1 forms. The STL parameter sends a test lines message to the operator
when a forms change is required. If both parameters are omitted, the system
default is used.
70044623-000
Making Job Control Work for You
•
The brk-pge parameter indicates a specific number of pages or cards to be spooled
out before the file is breakpointed and printed or punched.
•
The NOUPD parameter indicates that the spool file subdirectory entry be
updated only when a file is closed. Value UPD enables recovery of spooled data in
the event of a system crash.
•
The NOCMP parameter indicates the system should not attempt to compress
data that’s directed to the output spool file.
•
The RETAIN and HOLD parameters perform the same function they do when
specified as the first parameter. Remember, though, if you specify HOLD (as the
last parameter) with RETAIN; TAPE, DISK, or DISKETTE; or with RETAIN and
TAPE, DISK, or DISKETTE, the output file is first put on hold. The other
parameters are acted upon accordingly when the file is released. If you specify
RETAIN (the twelfth parameter) with TAPE, DISK, or DISKETTE, the output is
redirected to the appropriate device and a copy of the file is also retained (in the
spool file) for later use.
•
If specified, the SECURE keyword parameter indicates that the workstation to
which the auxiliary workstation printer is connected must be logged on before the
output file can be printed. If the workstation is not logged on and this keyword
parameter is specified, the file will not be printed.
Just as described in “Controlling Spooled Output with a JPROC Call” in Section 5 for
the SPOOL JPROC, the SPL job control statement must be placed in the device
assignment set for the spooled file.
Sending Spooled Output to Remote Batch Processing Terminals
The DSTjob control statement is used to send spooled output (print or punch) to RBP
(remote batch processing) terminals in your ICAM network. The format of the DST
statement is:
//Esymbot] DST dest-1[,dest-2,...,dest-16)
The dest parameter is one to six alphanumeric characters and defined by RBP. The
keywords OS3CTR or CENTRAL can be used to specify the local site’s printer.
•7004 4623-000
6-3
Making Job Control Work for You
The DST statement must appear within the device assignment set for the print or
punch file. When specif’ing multiple destinations, you can list several destinations in
one II DST statement or use several II DST statements each listing one or more
destinations. For example:
II JOB REMOTE
1/ DVC2B
1/ DST A,OS3CTR,c,D
If LFD PRINT
CN
/1 JOB REMOTE
or
1/ DVC2Ø
1/ DST A
1/ DST OS3CTR
II DST C,D
II EXEC PROG1
1/ EXEC PROG1
For more information on remote batch processing, see the Integrated Communications
Access Method (ICAM) Utilities Programming Guide, 7004 4565.
Note:
RBP output (specified by / / DST) and DDP and auxiliary printer output
(specified by I / ROUTE) cannot be mixed for any one job. For any job, all
output must be of one type or the other.
Sending Spooled Output to DDP Sites and Auxiliary Workstation Printers
C
The ROUTE job control statement routes print or punch output to printers and
punches at DDP sites and to auxiliary printers connected to a load or remote
workstation or terminal. You place the ROUTE statement in the device assignment
set for the file to be routed.
Notes:
1.
Output can be routed to the site central printer and up to seven auxiliary printers.
2.
The/I ROUTE and II DST statements cannot be mixed in the same job.
The ROUTE statement format is:
/ftsymbot)ROUTE destination-i,.. .,destination-8
You can specify up to eight destinations for non-DDP destinations, or one DDP site
destination. These destinations are the central printer or punch at a local or DDP
site, a workstation auxiliary printer at a DDP site, or an auxiliary printer that is
locally or remotely connected to your system.
You specifr the destination as follows:
[host-id: luser- Id-i,... ,user- id-8
C
6-4
70044623-000
Making Job Control Work for You
The host-id identifies a particular host in a DDP network. It is one to four
alphanumeric characters long and identical to the label-id of the LOCAP
macroinstruction in an ICAM network. You can also use $HOST to indicate the host
that initiated the job (the originator/master). The host in this case may be remote or
local. A host-id is optional but must be followed by a user-id if specified. Whenever
you omit a host-id, the local host (the processor on which the job is executing) is
assumed.
To identify an auxiliary workstation printer, specify a 1- to 6-alphanumeric character
workstation user-id. You can also use $Y$MAS to indicate an auxiliary printer at the
master workstation. The keyword CENTRAL in place of a user-id indicates a central
printer or punch. Any destinations that specify a user-id (other than CENTRAL) or
$Y$MAS denote auxiliary workstation printers and are valid only for print files.
Consider the following destinations:
•
CENTRAL or OS3CTR
The output goes to the central printer or punch at the local site.
•
host-id:CENTRAL
The output goes to the central printer or punch at a DDP site (identified by
host-id).
•
host-id:user-id
The output goes to an auxiliary workstation printer (identified by a user-id) at a
remote host (identified by a host-id) at a DDP site.. This destination is valid only
for print files.
•
user-id
The output goes to an auxiliary workstation printer (identified by a user-id) that
is locally or remotely connected to your system. This destination is valid only for
print files.
•
$Y$MAS
The output goes to the auxiliary printer if the master (the terminal or
workstation that has control of the job when the job is executed) at the local site
is a workstation. Otherwise, the destination is the central printer at the local
site. The $Y$MAS destination is valid only for print files.
70044623.000
6-5
Making Job Control Work for You
The following statements route output to the central printer:
/1 ROUTE OS3CTR
and
II ROUTE CENTRAL
The following statement routes output to up to eight auxiliary printers:
If ROUTE USERID1,....,USERID8
The following statement routes output to the central printer and up to seven auxiliary
printers:
1/ ROUTE OS3CTR,USERIO1,...,USERIDT
The following control stream contains a device assignment set for a print file which
includes the II ROUTE statement.
II JOB OUTPUT
II DVC 20
II ROUTE A123:CENTRAL
1/ LFD PRTFIL
1
J
Print file
device
assignment set
C.
/1 EXEC PROG1
The ROUTE statement in the preceding device assignment set routes the print output
to the central printer at a remote host whose host-id is A123.
C
6-6
70044623-000
Making Job Control Work for You
Notes:
1.
RBP output (specified by / / DST) and DDP or auxiliary printer output (specified
by / / ROUTE) cannot be mixed for any one job. For any job, all output must be of
one type or the other. Also, DDP destinations and local auxiliary printer
destinations cannot be used for the same print file.
2.
When a workstation or terminal starts ajob that directs printed output to an
auxiliary printer connected to a local or remote workstation or terminal (one that is
not the originator), the user at the other workstation or terminal must be logged on
with the same user-id as specified in the / I ROUTE job control statement and
must issue an RP command to starts printing. See the Interactive Services
Operating Guide, UP-9972, for more information on the RP command.
Spooling Input Card Data
Ajob that reads a large volume of data through the card reader ties up the operating
system by using a slow-speed device (card reader) as the means of supplying input to a
high-speed processor. You can avoid this by loading the card data into a spooi file
(high-speed disk device) for later retrieval. In this way, the card reader can be used to
transfer data to the spooi file while other jobs are being executed in the high-speed
processor. High-speed processing, therefore, goes on without interruption from a slowspeed card reader.
The system operator uses the IN command to initiate spooling. You must identify the
card file to be spooled to the system operator, precede these cards with a DATA
statement, and follow them with a FIN job control statement or another DATA
statement. DATA is a control statement that identifies (to the input reader) the card
data you want spooled. Its format is:
II DATA FILEID=fiLe-identifier[,RETAIN][,IGNORE]
When the operator places your cards in the card reader and issues the IN command at
the console, the card file is placed in the spool file along with the file-identifier from
the DATA statement. The FIN or the final DATA card terminates the card reader.
The spooled card file becomes a subfile.
Later, when your job stream is run, the subfile is read in (just as the cards are read in
at the card reader, only much faster). Spooled data cards may be read by ajob that’s
entered at a card reader or by a stored job control stream.
Note:
Input data doesn’t have to be spooled before yourjob’s processing begins, but it
must be spooled by the time your program attempts to open its files.
The job control stream must contain a device assignment set for a card file. If you’ve
included an LBL job control statement in the device assignment set, the file identifier
specified on the DATA card must match the LBL statement’s file identifier. If there
isn’t an LBL statement, the file identifier on the DATA card must be a concatenation
of the job’s name and the file name from the LFD job control statement. Either way,
an association is made between the file you defined in your job control stream and the
subfile.
7004 4623-000
6-7
Making Job Control Work for You
If this is the control stream,
II
1/
II
II
I&
/1
JOB
DVC
LBL
LFD
BALANCE
30
SPOOL1
READ
FIN
you code this DATA statement:
1/ DATA FILEID=SPOOL1
If this is the control stream,
1/ JOB BALANCE
If DVC 30
II LFD READ
‘a
II FIN
you code this DATA statement:
II DATA FILEID=BALANCEREAD
The RETAIN parameter is used to maintain the subfiie after it is processed. If you
specify RETAIN, only the DE SPL,RDR console command can delete the subfIle. The
following example shows the use of the RETAIN parameter:
II DATA FILEID=BALANCEREAD,RETAIN
data cards
II FIN
You can, if necessary, place all RUN/RV job control statement in the card deck. When
the deck is spooled, the run processor calls the specified job stream. Only one
RUN/RV statement may be placed within a DATA...DATA or DATA....FIN card
sequence. if more than one RUN/RV statement is present, only the last statement is
used.
The IGNORE parameter is used to permit RUN job control statements to be spooled
as data. It can be used, for example, for conversion jobs. Suppose you have a card
deck of control streams to be converted from another operating system to 05/3 and
you have several RUN statements in the deck. When you spool the card deck, you
don’t want the RUN statements to call stored control streams; you want them
converted to 0S13 RUN job control statements.
Let’s assume we are running a conversion job named CNVT with an input card deck to
be spooled named CARDIN. The DATA FILEID job control statement is coded like
this:
II DATA FILEID=CNVTCARDIN,,IGNORE
7004 4623-000
Making Job Control Work for You
Because we specified IGNORE in this example, RUN statements in the card deck are
spooled as data cards.
Spooling Diskette Files
Just as card input can be spooled, so can input from data-set-label diskette. The
operator uses the IN command to initiate the spooling and the data is placed in the
spool file. It remains there as a subifie and is retrieved by either a control stream
entered at a card reader, or by a prefiled job control stream. The data set label from
the diskette provides the label for your spool file, while the /* statement indicates the
end of the data file.
Whenever you’re using input that’s spooled from data-set-label diskette, specify the I
parameter of the DVC statement for the diskette. Remember, the format of the DVC
statement is:
//[symbo[] DVC
nnn[(n)]
RES
RUN
,
addr
OPT
IGNORE
ALT
0
REQ[(n))
REAL
The I parameter tells job control that your data is in the spool file. The data cannot be
retrieved from this file unless the I parameter is specified as follows:
II DVC 132,1
The 0 parameter is used when you want the spooled output to go to data-set-label
diskette.
70044623-000
6-9
Making Job Control Work for You
Equating Logical Unit Numbers to Device Type Codes
Since logical unit numbers can be changed at SYSGEN time, the possibility exists
that, when running your control stream on a system other than the one it was
designed for, one of your logical unit numbers may indicate a different device on the
other system. For example, the system your control stream was designed for might
have logical unit number 64 associated with an 8416 disk subsystem. But on the
system you are running under, logical unit number 64 may be an 8419 disk
subsystem wrong device. A way to get around this is to use the EQU job control
statement, which equates logical unit numbers to specific device type codes. (This
device type code is always associated with this device.)
-
The format of the EQU job control statement is:
//Esymbot) EQU Lun-1,type-i[,lun-2,type-2,...,Lun-n,type-n)
The lun parameter indicates the logical unit number you have on the DVC job control
statement in the control stream. The type parameter is the 4- to 8-hexadecimalcharacter device type code for the device you are using. See Table A-i of the Job
Control Programming Reference Manual, UP-9984, for the codes.
The EQU job control statement, which you must place before the device assignment
sets in the control stream, is effective for the entire job.
Let’s assume that a job is being run on a system other than the one it was written for
and that there’s a possibility the logical unit numbers in the second system were
changed at SYSGEN time. On your system, logical number 64 is the 8416 disk
subsystem. To ensure that we get an 8416 on the other system, we insert an EQU job
control statement coded as follows (the device type code 2002 was obtained from
Table A-i of the Job Control Programming Reference Manual, UP-9984).
-
Ii
/1
II
II
1/
EQU
DVC
VOL
LBL
LFD
-
64,2002
64
DISKO1
XYZ
TEST
You can also use the EQU statement to specify additional logical unit numbers for
virtual readers, printers, or punches. See “Using Multiple Devices, SYSRES, or the
Job’s $Y$RUN File” in Section 4.
C
6-10
70044623-000
Making Job Control Work for You
Specifying Unique Load Codes
A load code buffer controls what characters are printed by your printer. Codes
corresponding to the characters on your print band or cartridge are placed in the
buffer and whenever a particular code is encountered, the character equated (via the
load code buffer) with that code is printed. (To simplify this discussion, we’ll use the
term print cartridge from here on to mean print band or cartridge.)
For non-SDMA printers (0770 and 0776), the default contents of the load code buffer
are set at SYSGEN time; there is a unique buffer for each printer type. One of the
uses of the LCB job control statement is to override these specifications to equate
different codes with different characters so that you can change print cartridges. You
define a load code buffer by specifying an 8-bit code for each character on the
cartridge. Whenever that code is encountered, the corresponding character is printed.
-
For SDMA printers (class I, II, III), each print cartridge contains its own
corresponding load code buffer. Therefore, you don’t need to define a unique buffer in
an LCB statement when you change cartridges. if you do, it is ignored. As you’ll see,
though, the// LCB statement has other uses.
There are two formats for the LCB statement, one for non-SDMA printers and the
other for SDMA and AP9215-1 printers. The format for non-SDMA printers is:
//[symbot] LCB
1
fX’hex-strng-1’
rrx’hex-string.2’ 1
c’char-string-1’
Lld1tn9.2hi
i...,
fX’hex-stririg-n’
lC’char-string-n’
[,CARTNAME=symbot]
LNAME=
I
I
I
48-BUS
48-Sd
63-STD
OWNLC1
L
[JCARTID= IX’aa’
1C’c’
L
[,NUMBCHAR=n]
LTYPE= [0770
L
j0776
[JSPACE=[X! aa’
L
I.X’40’
continued
70044623-000
6-il
Making Job Control Work for You
JJMISM= ir.NQRE
1REPORT
L
[DUAL=
X’ xxyyxxyyxxyyxxyy’
I
C’abababab’
C’bbbb’
I
L
LMISMCHAR= fx’aa’
L.x
The only parameters that have practical use for SDMA printers are symbol,
CARTNAME, NAME, TYPE, and MISM.
The format of the LCB job control statement for SDMA and AP9215 printers is:
1/ [symbol] LCB Efl.4AME=symbol]
[JNAME=
L
48-BUS
48-Sd
63-STD
OWNLC1-OWNLC9
I
J
[,TYPE=SDMA]
LMISM=
L
f IGNORE
1REPORT
The symbol in the label field is a 1- to 8-alphanumeric character name and can have
one of the following uses:
•
To specify a default cartridge name when you omit the CARTNAME parameter
•
To specify the name of a filed load code buffer that you’re changing via the job
SG$PRB
Use of symbol will become more clear after we discuss the CARTNAME and NAME
parameters.
For non-SDMA printers, you use the first parameter of the LCB statement to assign
the codes for each graphic symbol on the print cartridge by specifying either the
X’hex-string’ (hexadecimal) or C’char-string’. You need two hexadecimal characters or
one EBCDIC character for every symbol. The position of each in the string of
parameters must correspond to its position on the print cartridge. As many
6-12
70044623-000
Making Job Control Work for You
parameters as you need to specify the entire print cartridge may be used; you may
intermix the character and hexadecimal strings as required. Since the single quote
(apostrophe) and ampersand (&) symbols have special meanings to job control, they
must always be coded in hexadecimal. Statement continuation is only allowed
between parameters; individual character strings can’t be coded on one job control
statement and continued on another. When using hexadecimal character strings, the
number of digits must be even.
Note:
The character strings for your printer are shown in the appropriate printer
subsystem manual.
The CARTNAME parameter specifies the name of the print cartridge to be used. Your
installation is responsible for assigning a unique, 1- to 8-alphanumeric character
name to each cartridge. SCIENCE, for example, could be used for a scientific
character set.
When you provide a cartridge name in the II LCB statement, the operator. is requested
to mount the cartridge just before the file starts printing. The output is not printed
until the operator mounts the cartridge and replies to the message. Remember, if you
don’t specify a cartridge name, the cartridge that’s already on the printer is used. So,
to ensure use of the proper cartridge and to avoid printing of the wrong characters,
you should specify a cartridge name.
You can use symbol in the label field of the LCB job control statement (instead of
CARTNAME) to specify a cartridge name. If you use both symbol and CARTNAME to
specify a cartridge name, the CARTNAME parameter takes precedence.
You specify NAME when you want to use one of the filed load code buffers (48-BUS,
48-SCI, 63-STD, or OWNLCn) established at SYSGEN time or by use of the job
SG$PRB. There is aunique 48-BUS, 48-SCI, 63-STD, and OWNLCn foreach printer
type. (There is also a default load code buffer for each printer type when no II LCB
statement is specified.) NAME indicates that you want a filed load code buffer; you’re
not establishing your own. Therefore, CARTNAME, TYPE, and MISM are the only
other parameters you can specify when you use NAME.
The NMIIE parameter specifies the name of the filed load code buffer, which in turn
specifies a cartridge name. So, when NA?VIE is specified, CARTNAME is unnecessary.
As mentioned earlier, you can also use symbol for the name of a filed load code buffer.
This is done only when you are executing the job SG$PRB to change a filed load code
buffer (48-BUS, 48-SCI, 63-STD, or OWNLCn) via the job SG$PRB. If this is the case,
you use symbol to specify the name of the buffer to be changed. This is the only time
symbol indicates a load code buffer name. At all other times it indicates a default
cartridge name if you omit the CAl? TNAME parameter. See the appropriate
installation guide for more information about the job SG$PRB.
70044623-000
6-13
Making Job Control Work for You
The CARTID parameter specifies a cartridge or band identifier. It may be either two
hexadecimal digits (X’aa’) or one EBCDIC character (C’c’). This parameter is required
for non-SDMA printers (0776 and 0770) and must agree with the number found
physically on the cartridge.
The NUMBCHAR parameter applies only to non-SDMA printers and specifies the
total number of graphic symbols expected on the print cartridge. As a safety check to
make sure you specified all characters, this number should coincide with the number
of characters specified in the character strings. When you omit NUMBCHAR , the
number of characters specified by the character strings is assumed to be correct.
To identify the printer for which the LCB job control statement is constructed, you use
the TYPE parameter. From this, we can see that an LCB job control statement coded
for one type of printer cannot be used for another type.
The 0770 and non-SDMA 0776 printers are available only on the models 8/1 0/1 5/20/50
systems. For all other non-SDMA printers, enter * for the TYPE parameter and use
the logical unit number to specify the type of printer. Specify SDMA for the TYPE
parameter if you are a model 7E user with an AP9215-1 printer.
You specify the space, or nonprinting code, for non-SDMA printers through the
SPACE parameter. This code is not included in either the X’hex-string’ or
C’char-string’ parameters. It may be either two hexadecimal digits (X’aa’) or one
EBCDIC character (C’c’). The default value is X’40’.
A mismatch occurs when you try to print a character that is not in the load code buffer
or has not been specified as a dualed character. You can use the MISM parameter to
record character mismatch errors in the system error log by coding MISM=REPORT.
The default, MISM=IGNORE, means that mismatches aren’t recorded.
For the 0770 or 0776 printer, you have a choice as to the replacement symbol. If you
specify in EBCDIC, you would use the DUAL—C’abababab’ parameter, with a being a
character that is on the print cartridge and b being the character that a replaces. For
example, assume that the print cartridge contains the asterisk symbol (*) but not the
question mark symbol (?). You could substitute * for ? in the printout by specifying
DUAL=C?’. Every time the program outputs the EBCDIC code for a question mark,
an asterisk appears in the printed listing.
If you specify in hexadecimal, you would use the DUAL=X’xxyyxxyyxxyyxxyy’
parameter, where xx is the code for the character printed and yy is the code of the
character that xx replaces.
We’ve already said that when a character mismatch occurs, you can use the MISM
parameter to record it in the error log. For non-SDMA printers, you may also specify a
character that’s to appear on the printed output in case of a character mismatch;
otherwise, a blank will appear (the default value X’40’). This is done with the
MISMCHAR parameter. You can specify any character you want, in either
hexadecimal (X’aa’) or EBCDIC (C’c’), as long as the character also appears in either
the X’hex-string’ or C’char-string’ parameter.
C
6-14
70044623-000
Making Job Control Work for You
Here is an example of how the LCB job control statement is used for a non-SDMA
printer:
Co(umn 72
1.
2.
3.
4.
5.
70044623-000
I/DVC28
II LCB Ch/._* XJ7D! ,C’+,S’’)( Ø123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ’,
Ill
NUMBCHAR=48,CARTID=X’02’ ,TYPE=0770,DUAL=CI*? I
CARTNAME=SCIENCE
112
II LFD PRINTOUT
-
-
X
X
1.
The DVC job control statement has 28 for the logical unit number, indicating
that a 0770 printer must be used.
2.
This gives the actual character set for the load code buffer. Notice the
shaded area; this is where a switch is made from specifying in EBCDIC (C)
to hexadecimal (X). We did this because we want to specify a single
quotation mark (apostrophe) for the load code. Since a single quotation
mark begins and ends each character string, coding the single quotation
mark as an EBCDIC character would have terminated the character string,
and the remaining characters would be invalid. So, we ended the character
string after the last character before the single quotation mark (the
asterisk), specified the single quotation mark in hexadecimal (7D), and then
continued with the next character (a plus sign) in EBCDIC. The comma
character for the load code (after the plus sign) will not end the character
string because it’s enclosed within single quotation marks.
3.
The NUMBCHAR parameter indicates that there are 48 characters in the
print cartridge, if we missed specifying a character in the character string
parameter, this would cause an error, so we’d know that we forgot a
character. The CARTID parameter indicates a cartridge identifier of 02, and
we’re using a 0770 printer (TYPE parameter). The DUAL parameter
indicates that three nonprintable characters (?,>, and <) are going to be
appearing during the job, and gives the printable characters (*, “, and +) that
will replace them.
4.
When this print file is opened, the operator receives a message telling him to
mount the cartridge named SCIENCE.
5.
Provides the file name for the print output file and completes the device
assignment set.
&15
Making Job Control Work for You
Here is a similar example of how the LCB job control statement is used for an SDMA
printer:
II DVC 220
1/ LCB TYPE=SDMA,CARTNAME=SCIENCE
Some points to remember when coding the LCB job control statement are as follows:
•
You can always specify the CARTNAME and TYPE parameters.
•
If you specify NAME to indicate a filed load code buffer, you cannot specify any
other parameters except CARTNAME, TYPE, and MISM.
•
If you’re using the job SG$PRB to change a filed load code buffer, use symbol to
specify the name of the buffer rather than NAME.
Using Forms Control
A vertical format buffer controls a printer’s vertical form spacing. This applies to the
0770, 0776, and 0789 printers. Codes corresponding to specific lines on a printer form
are loaded into the vertical format buffer. You advance the form to a particular line
by issuing a skip command in your program and specifying the code. The default
vertical format buffer for each printer type is set at SYSGEN time. You can use the
VFB job control statement to specify a unique vertical format buffer for a print file.
You must place the VFB job control statement within the device assignment set for
the printer file to which it applies. The II VFB statement becomes effective when your
program opens the print file. The format of the VFB job control statement is:
//tsymbot] VFB
LENGTH=tines
[,FORMNAME=symbol]
[USE=
I
L
STAND1
OWNVF1
OWNVF2-OWNVF9
[DENSITY=
r
,TYPE=
L
(SDMA and 9215 printers only)
{}1
SDMA
0770
0776
[,OVF=(line-1,...,line-n)]
1
E,OVF
.
..,tine-n))[,CD1=(tine-1,...,l
2=(1in
e1
ine-n),...
£,CD15=(line-1,...,line-n)]]
[,HP=n]
6-16
70044623-000
Making Job Control Work for You
The symbol in the label field is a 1- to 8-alphanumeric character name and can have
one of the following uses:
•
To specify a default form name when you omit the FORMNAME parameter.
•
To specify the name of a filed vertical format buffer that you’re changing via the
job SG$PRB.
Use of symbol will become more clear after we discuss the FORMNAME and USE
parameters.
The LENGTH parameter indicates how many lines are on a form in the range of 1 to
192. You must use this parameter whenever you specify any of the VFB statement
parameters for forms overflow (OVF1 ,OVF2) or vertical line positioning
(CD1 ,...,CD15). LENGTH must also be specified whenever you specify DENSITY.
The FORMNAfi/IE parameter specifies the name of the printer form to be used. (This
is very useful when you want your output printed on a special form.) Your installation
is responsible for assigning a unique, 1- to 8-alphanumeric character name to each
form. PAYCHK, for example, could be the name used for payroll checks.
When you provide a form name in the II VFB statement, the operator is requested to
place that form in the printer before the file begins printing. The output is not printed
until the operator loads the form and replies to the message. Remember, if you don’t
specify a form name, the form that’s already in the printer is used. So, to ensure use
of the proper form and to avoid printing on any valuable special forms, you should
always specify a form name.
Remember, you can specify a form name using any of the following:
•
The SPOOL JPROC (See “Controlling Spooled Output with a JPROC Call” in
Section 5.)
•
The SPL job control statement (See “Controlling Spooled Output with a Job
Control Statement” in Section 6.)
•
The symbol in the label field of the VFB job control statement (A form name
specified this way takes precedence over a form name specified with
II SPOOL or II SPL.)
•
The FORMNAME parameter of the VFB job control statement (A form name
specified this way takes precedence over a form name specified with II SPL,
//SPOOL, or the II VFB statement’s symbol.)
70044623-000
6-17
Making Job Control Work for You
You specify the USE parameter when you want to use one of the filed vertical format
buffers (either STAND1 or OWNVFn) established at SYSGEN time or via the job
SG$PRB. There is a unique STAND1 and OWNVFn for each printer type. USE
indicates that you want a filed vertical format buffer you’re not establishing your
own. Therefore, FORMNAME and TYPE are the only other parameters you can
specify when you specify USE.
-
As mentioned earlier, you can use symbol for the name of a filed vertical format
buffer. This is done only if you are executing the job SG$PRB to change a filed buffer
(STAND1 or OWNVFn). If this is the case, you specify either STAND1 or OWNVFn in
the symbol field. You don’t specify the USE parameter. This is the only time symbol
indicates a vertical format buffer name. At all other times it indicates a default form
name if you omit the FORMNAME parameter. See the appropriate installation guide
for more information about SG$PRB.
DENSITY indicates the number of print lines per inch. (The default is 8.) An 11-inch
form, for example, printed at a density of 8 lines per inch has 88 lines; this same form
printed at a density of 6 lines per inch would have 66 lines.
Note:
Ifyou change the print density for a print file, the forms mount message is
displayed to your operator. This occurs even if the form name remains the
same, and the form is not to be changed. Once alerted, your operator must
reply to the message before any output printing can occur.
We refer to the remaining parameters of the VFB job control statement as skipcodes.
These codes indicate forms overflow and vertical line positioning. When you specify
any one of these, you must also specify the LENGTH parameter because line is a
decimal number in the range of 1 to whatever amount is specified by the LENGTH
parameter. When only one line is specified for a code, you may omit the enclosing
parameter. If you accidentally repeat a code for the same line, the first one is accepted
and the others are ignored. (In this case, a warning message is displayed.)
The OVF parameter specifies the forms overflow line indicator. When an overflow
code is placed in the vertical format buffer, any space operation (such as print and
space) that advances the form to or beyond the overflow position causes the hardware
to detect and indicate forms overflow. You can specify multiple overflow indicators.
For example, you might indicate a forms overflow routine through your program that
prints subtotals, and another overflow routine that goes to the top-of-forms (home
paper) position of the next page.
The OVF2 parameter specifies a secondary forms overflow position for use with the
0770 printer. You can specify multiple overflow indicators. For example, if you’re
going to print payroll checks and there are only 10 print lines for each check, setting
up a vertical format buffer at only 10 lines is impractical. Every time 10 lines are
printed, the vertical format buffer is rechecked to find the specifications for the next
paycheck (spacing, etc.), even though it’s the same form with the same spacing. This
takes time. But if you set up the vertical format buffer length for, say, 60 lines, you
could defme 6 paychecks in one buffer. In this way, the vertical format buffer is
checked after every sixth form instead of after every form.
C
6-18
70044623-000
Making Job Control Work for You
When you design the VFB, bear in mind that lines can be printed (and the form
advanced) beyond the overflow position. For printing of assemblies, librarian runs,
dumps, etc., you should provide at least four lines between the overflow position and
the bottom of the form.
The user should always specify an OVF parameter if the file is to be spooled or if you
specify the LENGTH parameter.
The CD1 through CD15 parameters are for the device independent control character
codes. These codes are used for vertical line positioning. For example, CD1=5 means,
every time this code is detected, each page of your report is skipped to the fifth line.
Not all codes may be used with all printers. The Consolidated Data Management
Macroinstructions Programming Guide, 7004 4607, lists the appropriate control
character codes for your printers in the section that explains the control printer forms
macroinstruction.
The HP parameter specifies the line number location of the home paper position.
Notes:
1.
In a spooling environment, space must be reserved for all lines with assigned skip
codes. Ifyou specify a / / VFB statement for a spooled file and provide a full
vertical format buffer specification (you do not specify a filed vertical format buffer
with USE or symbol), job control reserves enough space. If, however, you request a
filed vertical format buffer (STAND1 or OWNVFn) that has more than seven skip
codes, or ifyou use a system default vertical format buffer having more than seven
skip codes, you must specify the number of skip codes using the no-skpcode
parameter in the II SPL statement or the SKIPCODE parameter in the / /
SPOOL JPROC.
When you don’t use a / / SPL statement or / / SPOOL JPROC, the default is seven
skip codes. Three skip codes are automatically included in this count: home
position for current page, overflow for next page, and home position for next page.
The four remaining are user-specified skip codes. Therefore, the / / SPL statement
and the II SPOOL JPROC specify the total count of lines on your form where a
skip code is allowed, plus three.
2.
Repeat occurrences of the same skip code on more than one line are counted as
separate skip codes.
Consider the following. Suppose you want to produce a report on a special 11- by 14inch form that prints 12 lines of data at 6 lines per inch (lpi), then skips 3 lines; prints
another 12, skips 3; and so forth down the page a total of 4 times. Your VFB
statement might look like this:
If VFB FO=WORKSHT,DE=6,LE=66,OV=61
70044623-000
6-19
Making Job Control Work for You
You would have to identify your special printer form (WORKSHT) to the operator so
that it can be loaded on an available printer. Specify your desired printing density in
terms of lines per inch (lpi). Specify the overall length of the form as a function of the
number of lines that could be printed on the form; in this case, 66 (6 ipi xli inches).
And, finally, specify the line on the form at which you want the printer to advance the
paper to the top of the next page, which is called the home paper position. This
parameter is sometimes critical because the location of the home paper position
depends on where the operator physically aligns the form on the printer. If the home
paper position has been set by the operator to line 4 of the form and your program
prints before skipping any lines, the first print line will occur on line 4.
C
If we assume that the form we’re using is meant to be loaded at line 4 and that our
program prints before skipping, our OV specification would be 61 as shown in the
example. This would allow us to print 48 lines and skip 9, before advancing the paper
to the next top-of-forms, or home paper position (4 + 48 + 9 = 61). If, however, the
operator loads our form at print position 2, instead of 4, our OV specification would
have to be 59, instead of 61, to maintain our desired page format. The obvious lesson
in this example is that you must tell the operator how to load a special form when
your output format is critical. Most of the time, you’re not concerned with the exact
number of lines that are printed, but only that the printed output not continue beyond
a reasonable line on the form.
Some points to remember when coding the VFB job control statement are as follows:
•
You can always specify the FORMNAME and TYPE parameters.
•
If you specify USE to indicate a filed vertical format buffer (STAND1 or
OWNVFn), you cannot specify any other parameter except FORMNAME and
TYPE.
•
If you’re using SG$PRB to change STAND1 or OWNVFn, use symbol to specify
the name of the buffer instead of USE.
•
If you specify DENSITY, you must specify LENGTH.
•
If you specify any codes (OVF1, OVF2, CD1 through CD15), you must specify
LENGTH.
•
If you specify LENGTH, you should specify at least one overflow code (OVF).
Controlling Tape Units
You use the MTC job control statement to position a tape volume prior to or after job
step execution. With it, you can move the tape volume to a certain block within a file
or to a certain file within a multiple volume. A tape volume can also be rewound to a
load point, rewound and unloaded, or have tape marks written.
C,’
6.20
70044623-000
Making Job Control Work for You
You must insert the MTC job control statement at a point after the device assignment
set for that tape unit. The format of the MTC job control statement is:
/I[symbol) MTC lfdname,
BB,nn
BM,nn
FB,nn
FM,nn
WM, nn
RL
RU
The lfdname parameter specifies the same file name that was used in the device
assignment set for the tape volume.
The next parameter provides seven choices; they indicate the type of operation you
want done. They are:
BB
BM
FB
FM
WM
RL
RU
-
Backspace a specified number (nn) of blocks.
-
Backspace a specified number (nn) of tape marks.
-
Forward space a specified number (nn) of blocks.
-
Forward space a specified number Cnn) of tape marks.
-
Write a specified number (nn) of tape marks.
-
-
Rewind to load point.
Rewind and unload the tape volume.
The amount, nn, must be a decimal number.
The relationship between the number of tape marks to the number of files on a
volume is covered in the Consolidated Data Management Macroin.structioris
Programming Guide, 7004 4607.
The following example shows how the MTC job control statement is used:
II
II
II
II
If
II
11
I?
1/
I&
JOB TAPELIST
DVC 90
VOL T123
LFD TAPEIN
DVC 20
LFD PRNT
MTC TAPEIN FB,10
EXEC TPRNT
MTC TAP1N RU
II FIN
The first MTC job control statement spaces tape T123 forward 10 blocks prior to job
step execution. The second MTC job control statement rewinds and unloads the same
tape after the job step is finished. Note that the lfdname parameter of both MTC job
control statements agrees with the filename parameter of the LFD job control
statement.
70044623-000
62l
Making Job Control Work for You
C
Releasing (Freeing) a Device and Volume
Once a device and a volume are assigned to a job, they remain assigned until the job is
finished. This assignment applies to all job steps of the job. But, what if your job has
10 job steps, and a certain device or volume is only used in the first job step? In effect,
they can’t be used by any other job until this entire job is complete. You can use the
FREE job control statement to release the device and volume immediately after it is
no longer needed, even though the job is not completed. However, if a device and
volume are released during one job step and either one is requested in a later job step
in the same job, no release occurs. This protects you from not having a needed device
or volume available because it was released too soon. Remember, the entire control
stream is scanned before the job begins executing.
The format of the FREE job control statement is:
//[symbot] FREE tfdname-1[[(DEV)),...,lfdname-n[(DEV)]]
The lfdname parameter specifies the same file name used in the device assignment set
for the file.
The (DEW parameter indicates that the device and volume are to be released. There
is no comma between the lfdname and (DEV) parameters.
Note:
6-22
You should always specify (DEV) even though it’s shown as optional.
Additionally, you must specify (DEV) to free unit record devices such as card
readers, card punches, printers, and workstations.
70044623-000
Making Job Control Work for You
Here’s an example of a multiple-job-step job. The first job step needs the card reader.
After that, it’s not needed.
1/ JOB PAYROL
1/ DVC 50
// VOL DISKO1
II LBL DETAILS
II LFD PRDISC
1/ DVC 30
/1 1,FD READeI
II EXEC CARDTP
If FREE READ1(DEV)
1/ EXEC EDIT
II EXEC BALANC
1/ EXEC NEGBAL
If EXEC WORKP
Job steps that
do&t need the reader
If FIN
data file
1*
You can also use II FREE to allow a job to be scheduled that appears to use more
volumes or devices than are available.
Suppose, for example, that ajob (PAYROL) uses four cataloged tape volumes. Your
system has only two tape drives. The system assumes, for cataloged volumes, that
each unique volume requires a unique device. Your job is not scheduled because there
are not enough unique devices available for each unique cataloged volume. You can
get the job scheduled, however, if you use the FREE statement to release the tape
drives once the first two volumes have been accessed.
During execution of the job, you are still protected from a needed device or volume
being unavailable, because no actual release occurs if II FREE specifies a device or
volume needed in a later job step.
Your job stream might look like this:
II
II
1/
II
1/
If
II
II
II
1/
1/
II
II
JOB PAYROL
LBL FILA
LFD TAP1
LBL FILB
LFD TAP2
EXEC STEP 1
FREE TAP1(DEV)
FREE TAP2(DEV)
LBL FILC
LFD TAP3
LBL FILD
LFD TAP4
EXEC STEP2
Your job is scheduled, even though your system has only two tape devices, because the
drives are freed after the first job step. During execution of the job, volumes A and C
use one tape drive and volumes B and D use the other tape drive.
/1 FREE can be used to release a workstation when it’s no longer needed by a job. You
specify the workstation lfdname as it appears on the LFD statement in the
workstation’s device assignment set and code the FREE statement as follows:
7004 4623-000
6-23
Making Job Control Work for You
1/ FREE WRKSTN(DEV)
If this statement is specified, all workstations connected to the file are freed.
Scratching Unwanted Files
Once a disk or diskette file is no longer needed, it might as well be scratched, making
the space available for some other file. The SCR job control statement does this. Any
file or extent specified on this job control statement is scratched as soon as the SCR
job control statement is encountered by the job step processor. Therefore, the SCR
statement should only be specified after any job steps needing that particular file are
executed. Only files on volumes that are currently mounted when the SCR job control
statement is encountered are scratched. You can’t use the SCR job control statement
to delete a file on SYSRES that has $Y$ as the first three characters of the file label,
and you can’t use it to delete the $Y$RUN file from the RUN volume. Only one
volume serial number may be specified for any SCRjob control statement.
The format of the SCR job control statement is:
//[syinbot]SCR tfdname[
L
fDATE[,yyddd]
1PRE[,aaaa]
The lfdname parameter specifies the file name (of the file to be scratched) used in a
previous device assignment set in the control stream. Within that assignment set, you
must specifr the volume serial number and the file identifier. But, if you’re working
with a disk file and the next parameter is either DATE or PEE, you may omit the LBL
job control statement from the relevant device assignment set.
C
6-24
7004 4623-000
Making Job Control Work for You
The DATE and PRE parameters are used only for disk flies. When you use the DATE
parameter, all files on the disk volume that have an expiration date earlier than the
current system date are scratched. If you want to use a date other than the current
system date, include the yyddd parmeter as part of your specification (where yy is the
year and ddd is the day leading zeros must be specified). When you specifSr a date,
all the files on the disk volume dated earlier than the date specified are scratched.
-
The PRE parameter indicates that all files on a disk volume with a certain prefix are
to be scratched. You specify this prefix as the next 1-to-S-character parameter. The
first three characters of this prefix, however, cannot be $Y$ if the volume is SYSRES.
If you omit the aaaa parameter, the first four characters of the file identifier from the
associated LBL job control statement are used as the prefix.
If this parameter (DATE and PRE) is omitted, the entire file specified by the lfdname
parameter is scratched.
Here are three examples:
II
II
II
II
DVCVOL DSPØ28
LBL PAYROLLDETAILS
LFD PRDET
SCR PRDET
In this first example, the entire file identified as PAYROLLDETAILS is scratched.
The filename parameter of the LFD job control statement and the lfdname parameter
of the SCR job control statement must agree.
II DVCVOL DSPØ28
II LFD DELETES
II SCR DELETES,DATE,76002
In this example, all files on disk volume DSPO28 that have an expiration date earlier
than the second day of the year 76 are scratched. Notice the absence of an LBL job
control statement. When you use either the DATE or PRE parameter, an LBL job
control statement isn’t needed.
II
II
II
II
II
70044623-000
DVC
VOL
LBL
LFD
SCR
130
DKTOO1
ADDRFILE
ADDR1
ADDR1
6-25
Making Job Control Work for You
Our last example shows an entire file being scratched on our format-label diskette.
Remember, the filename parameter of the LFD job control statement and the lfdname
parameter of the SCR job control statement must agree.
Notes:
1.
The file to be scratched should not be in use by anotherjob.
2.
After an SCR job control statement is processed, the file is no longer available.
You can’t even refer to this file with another SCR job control statement or a FREE
statement.
File Cataloging
The file catalog ($Y$CAT) is a system resident file. It contains entries consisting of
file information about tape, disk, and diskette files in the system. The catalog enables
easy access to this file information for jobs and can also restrict files only to authorized
users.
The CAT, DECAT, and QUAL job control statements and a special form of the LBL
job control statement are used to create, access, and decatalog cataloged files. Their
use and a complete description of the OS/3 file cataloging facility are contained in the
File Cataloging Technical Overview, 7004 4615. The catalog manipulation routine
(JC$CAT) is also described there.
Selecting Optional Features
Unisys Operating Systeml3 provides optional features you can select whenever you
want. As you’ll see, some options (such as DUMP and GO) are only effective during
the job step in which they are specified, some (such as GABRDUMP, GDUMP, and
GSYSDUMP) are effective from the time the option is encountered until end ofjob,
while others (such as ACN, BUF, OFT, LOG, and SCAN) are in effect for the entire job
because they are acted upon when the run processor prepares the control stream for
execution.
This is the format of the OPTION job control statement:
IltsymboL] OPTION p-1[,...,p-n]
6-26
70044623-000
Making Job Control Work for You
As you can see, you can specify as many features as desired, as long as they’re
separated by commas (there can’t be any spaces). The features available are:
ABRDUMP
Provides a main storage dump in the immediate vicinity of the current TCB P5W
address. Displays current registers and buffers of all open DTFs.
•
ACN=account-number
Overrides the acct-no specified in the JOB control statement.
•
BOF
Your program is given control with binary overflow interrupt-enabled.
•
BUF=nxm
Overrides the nXm parameter specified in the JOB control statement.
•
DOF
Your program is given control with decimal overflow interrupt-enabled.
•
DUMP
Provides a job region dump at execution time in hexadecimal, ifjob step
termination is requested, or a snapshot dump in response to a SNAP
macroinstruction.
•
EOD=xx
Supplies substitute characters for the end-of-embedded-data (/*) job control
statement. Used when embedded data is DSL source code. The first character
specified must be a slash (I). The second character can be anything but a slash,
an asterisk, an ampersand, or a currency symbol (I, , &, $).
•
GABRDUMP
Specifies that OPTION ABRDUMP is in effect for every job step from the time
GABRDUMP is encountered to end ofjob.
•
ODUMP
Specifies that OPTION DUMP is in effect for every job step from the time
GDUMP is encountered to end ofjob.
70044623.000
6-27
Making Job Control Work for You
•
GJOBD{TMP
Specifies that OPTION JOBDUMP is in effect for every job step from the time
GJOBDUMP is encountered to end ofjob.
•
GO
Automatically executes a load module after link editing is completed. An
OPTION job control statement with the GO feature is generated automatically by
the ASMLG JPROC call statement, for example.
•
GSUB
Provides symbol substitution for all embedded data sets in the job stream. This
is a global SUB option.
•
GSYSDUMP
Specifies that OPTION SYSDUMP is in effect for every job step from the time
GSYSDUMP is encountered to end of job.
•
HDR= INOHDR
HDR
NOHDR suppresses the printing of page separators. HDR allows page separators
to be printed. OPTION HDR overrides the page separator specification in the
JOB control statement.
•
HOLD
Places a job containing it in hold status while the job is in the job queue table. A
job containing this option is not released until a BEGIN operator command is
issued, or until a CC job control statement with BE specified is encountered in a
subsequent control stream. CC BE cannot be used to release a HOLD within the
same job.
•
IMMOVE
Prevents memory consolidation movable shuffle in this job step.
•
JOBDUMP
Provides an edited version of a dump if a dump is requested and is in effect only
for the job step that contains this option.
6-28
70044623-000
Making Job Control Work for You
•
LDA
Directs job control to set up a 256-byte user local da area (LDA) in the job
prologue. This option is provided primarily for IBM Systeml34 compatibility
and is used if RPG programs or the assembler uses the LDA.
•
LINK
Automatically executes the linkage editor once the object module is created. This
allows you to compile and link edit without intervention from job control.
•
LOG= (logical-unit-number
Screen format file LFD name
II EXEC PRGRM2
When you run YOURJOB, PRGRM2 is executed. PRGRM2 contains an instruction to
open WORKSTN, which opens the screen format file FORMAT.
For more information about screen formats, see the Screen Format Services Technical
Overview, UP-9977.
The USE Statement for Menu Services
When your program needs to use menu services from a workstation, the USE
statement you specify takes this form:
//[symbot] USE MENU
menu-fite-LFD/$Y$FMT
I
$Y$FMT/menu-file-LFD
L
[linitial.menu)[ {flnn}
1
menu-format-1=aL las-if
0
[
6-74
menu-format-12=alias-12]]
70044623-000
Making Job Control Work for You
The USE MENU statement is similar to the USE SFS statement except that the
parameters refer to menu formats instead of screen formats.
The symbol parameter is used as the target of a branching statement. It is one to six
alphanumeric characters long, and the first character must be alphabetic.
The first positional parameter provides an LFD name for up to two menu format files.
Any name you use must match an LFD name specified in a previously defined device
assignment set for a menu format file. (Menu format files are always MIRAM files.)
The menu-file-LFD is one to eight alphanumeric characters long. When coding this
parameter, remember the following:
•
If you omit a forrnat-file-LFD name, it is assumed that all menus used reside in
the system file $Y$FMT.
•
When you code /menu-file-LFD, $Y$FMT is examined first, then the file
indicated by menu-file-LFD.
•
When you code menu-file-LFD/, the file indicated by menu-file-LFD is examined
first, then $Y$FMT.
The initial-menu parameter specifies the name of the first menu format to be used by
the application program. It is one to eight alphanumeric characters in length.
The nnn parameter specifies the number of menus to be resident in main storage at
one time, in the range 1 to 255. The default value is 1.
The menu-format=alias parameter equates a menu format name specified in an
application program (alias) to the actual menu format name (given when the menu
was created). A maximum of 12 alias name sets may be specified. The menu-format
name and alias names may each be from one to eight alphanumeric characters in
length.
7004 4623-000
6-75
Making Job Control Work for You
The control stream for a job that uses menu format services could include these job
control statements:
C
If JOB YOURJOB
II
II
II
II
DVC
VOL
LBL
LFD
50
ABC
MENUFILE
MENU1
If DVC 200
II USE MENU,MENU1
/1 LFD WORKSiIf
Device assignment set for the menu
format file
Device assignment set for the
workstation
Menu format file LFD name
II EXEC PRGRM1
/&
When you run YOURJOB, PRGRM1 is executed. PRGRM1 contains an instruction to
open WORKSTN, which opens the menu format file MENU1.
For more information about menu services, see the Menu Services Technical Overview,
UP-931 7.
The USE Statement for Dialog Processing
When your job needs the dialog processor to manage a dialog session at a workstation,
the USE statement you specify takes this form:
II[symbot] USE DP,dia[og-name[,printer-I.fd][,new-auditlfd][,old-audit-Lfd)
The files specified in the USE DP statement must have been previously identified
(through device assignment sets) in the control stream.
The symbol parameter is used as the target of a branching statement. It is one to six
alphanumeric characters in length, with the first character being alphabetic.
The dialog-name parameter specifies the name of the dialog you want to use; it must
match the LFD statement of the dialog file’s device assignment set. It is one to eight
alphanumeric characters in length.
C
6-76
7004 4623-000
Making Job Control Work for You
The printer-lfd parameter specifies the name of the printer file. It must match the
LFD statement of the printer’s device assignment set. It is one to eight alphanumeric
characters in length. This parameter is specified when you want to produce a printed
summary of the dialog session.
The new-audit-lfd parameter specifies the name of the new audit file output by the
audit version of the dialog processor. It must match the LFD statement of the new
audit file’s device assignment set. The new audit file contains a record of your
responses to a current dialog session. This parameter is one to eight alphanumeric
characters in length.
The old-audit-lfd parameter specifies the name of the old audit file used as input to
the audit version of the dialog processor. It must match the LFD statement of the old
audit file’s device assignment set. It is one to eight alphanumeric characters in length.
The old audit file contains a record of your responses to a previous dialog session.
The control stream for a job that calls the dialog processor could contain these job
control statements:
II JOB MYJOB
1
//DVC2O
/1 LFD PRNTR
II
II
/1
If
DVC5O
VOL DSKO1
LBL NEWAUDITFILE
LFD AUDIT1
If
II
II
II
DVC
VOL
LBL
LFD
f
Device assignment set for the printer
1
Device assignment set for the new
audit file
f
51
DSK02
DIALOGFILE
D!ALOG1
Device assignment set for the
dialog file
II DVC 200
II USE DP,DIALOG1,PRNTR,,AUDIT1 1
II LFD WKSTN
—
II EXEC PRURM1
/&
—
Device assignment set for the
workstation
New audit file tfd
Printer tfd
Dialog name
When you run MYJOB, PRGRM1 is executed. PRGRM1 contains an instruction to
open WKSTN, which, when processed, causes DIALOG1 to execute at the workstation.
Your responses to DIALOG1 are routed back to PRGRM1.
70044623.000
6-77
Making Job Control Work for You
c-)
For more information about dialog processing, see the Dialog Processor Programming
Guide, UP-8858.
Source Module Access via the USE Statement
Your programs can write (create) or read a source module that you identifSr in the USE
LIB job control statement. When included in the device assignment set for a library
file, II USE LIB indicates that the file contains source modules and the specified
module will be accessed by your program.
The format for II USE LIB is:
//[synibot] USE LIB,modutename ,TYPE= S
M
0
L
The module name you specifSr can be from one to eight alphanumeric characters long
and the first character must be alphabetic. The following job control stream indicates
that PROG1 will access a source module named MODULE1.
II JOB READMOD
/1
II
II
II
II
c
DVC5Ø
VOL 01234
LBL SRCLIB1
USE LIB,MODULE1
LFD SRCMOD
II EXEC PROG1
I&
Note:
Access of a source module by your program is limited to either a sequential
read or sequential write operation.
C
6-78
70044623-000
Section 7
Run-Time Conditional and Set Symbol Job
Control Statements
Run-Time Conditional Job Control Statements
GO, IF, and NOP are run-time conditional job control statements. They allow you to
branch to other job control statements in the control stream. Unlike SKIP job control
statements (effective during execution of your program), they are interpreted and
acted upon while the run processor is scanning the control stream, and then stripped
from the stream. Therefore, any devices and volumes specified on the bypassed job
control statements need not be available. Only forward branches are allowed for runtime conditional statements. Because GO, IF’, and NOP are processed only by the run
processor and their actions are completed when the run processor has acted upon
them, they are very useful when writing job control procedure (JPROC) definitions.
Unconditional Branching
The GO job control statement causes an unconditional branch to another job control
statement identified by a symbol. The destination can be a set symbol with a value
determined when the job stream is analyzed.
The format of the GO job control statement is:
//[symbol] GO destination
The symbol is only used when this job control statement is the target of another GO or
IF job control statement.
The destination parameter identifies the target job control statement and must agree
with the symbol in the label field of that statement.
Like the other run-time conditional statements, the GO job control statement is acted
upon by the run processor, before a job is scheduled, and then deleted from the control
stream. For this reason, the devices and volumes skipped by a GO statement need not
be available when the run symbiont is scanning the control stream.
Note:
70044623-000
Unlike GO, SKIP is effective during the execution of a program. Because ajob
is not executed until all the devices and volumes it uses are available to the
system, devices and volumes bypassed by SKIP must be available or the job
won’t be scheduled. However, devices and volumes bypassed by a SKIP
statement can’t be referenced in subsequentjob control statements in the
control stream because, even though they are available, they have not been
completely identified to the system.
7-1
Run-Time Conditional and Set Symbol Job Control Statements
The following is a stored control stream similar to the one shown with the SKIP job
control statement. (See “Bypassing Job Control Statements” in Section 6.)
JOB BALANCE
CR
DVC 90
VOL MASTO1
LBL DETAILS
LFD TAPEIN
GO DOlT
//CARD DVC 30
II LFD CARDIN
//DOlT EXEC EDIT
/&
1/ FIN
II
II
II
II
ii
II
II
If the input is on cards, you would place the following stream in the card reader:
Job control statements inserted in the stored stream
when II CR is encountered.
II GO CARD
If FIN
data cards
1*
-
Input card file
When the first CR job control statement from the stored control stream is encountered
by the run processor, it transfers control to the card reader, where the GO job control
statement causes the device assignment set for the tape to be skipped without any
processing. The tape volume and the device that would use it do not have to be
available. Therefore, they can be used by another job. If the input is on tape, a FIN
job control statement is all that’s needed in the card reader. The tape device
assignment set would be read, and the stored GO job control statement would cause
the device assignment set for the card reader to be bypassed.
Conditional Branching
The IF job control statement causes a conditional branch to another job control
statement, depending upon certain test conditions. This is similar to using the SKIP
job control statement conditionally, except that it’s interpreted and acted upon by the
run processor, just like the GO job control statement.
The format of the IF job control statement is:
//Esymbot] IF (a op b)destination
The symbol is only used when this job control statement is the target of another GO or
IF job control statement.
The test for a conditional branch is specified as (a op b), where a and b are the two
operands to be compared. You can compare two numeric operands (1 op 2) or two
alphabetic operands (a op b) but a run processor error results if you attempt a
comparison between one numeric and one alphabetic operand (1 op b).
C
7-2
7004 4623-000
Run-Time Conditional and Set Symbol Job Control Statements
The op in the expression is the relational operator that specifies the type of
comparison to be done. The values for op are:
EQ
NE
GT
LT
GE
LE
-
-
-
-
-
-
aisequaltob
a is not equal to b
a is greater than b
aislessthanb
a is greater than or equal to b
ais less than or equal tob
Remember, whenever you enclose an operand in quotes, the quotes are considered a
part of the operand. For example, (‘a’ EQ a) is an allowable comparison but the
operands are not equal because one value is ‘a’ and the other is a. (See “Specifiing Set
Symbol Values in Quotes” later in this section for more information.)
The operands are separated from the relational operator by spaces and the entire
parameter is enclosed within parentheses.
Note:
If a numeric comparison is made and neither a nor b is numeric, both the
greater than and less than conditions are set, resulting in all conditions except
equal being allowed to branch. If a character compare is being used and the
two operands are not of the same length, then the comparison is made on the
number of characters present in each, rather than on the contents of the
operands. Thus, a string offive characters will always be less than a string of
six characters, regardless of the character content of the comparand.s. Ifyou
have specified / / OPTION UNEQUAL, an error message is generated
whenever character strings of unequal length are compared. (See “Selecting
Optional Features” in Section 6.)
The destination parameter identifies the target job control statement that will receive
control if the transfer condition is true. This entry must agree with the symbol in the
label field of the target job control statement.
When scanning for the target job control statement, only the FIN job control
statement is acted upon. Therefore, you cannot branch out of the current job stream;
any procedure calls or CR job control statements that are skipped are not acted upon.
The comparand fields may be variable symbols, or dummy arguments, that can be set
in a JPROC definition. They’re called dummy arguments because the variable symbol
can be modified when called by the JPROC call.
7004 4623-000
7-3
Run-Time Conditional and Set Symbol Job Control Statements
Let’s look at an example. At first, this example will not be totally clear, but when
combined with the explanations of the remaining job control statement in this section
and the JPROC definitions in the next section, it will become clearer. The only
purpose here is to explain how the IF job control statement functions.
Consider this example:
II IF (‘&IN’ EQ ‘N’)EXIT
This job control statement is in a JPROC definition. When the PROC directi
ve was
written, it contained a parameter called IN. The ampersand of &IN identifies this
as
a variable symbol; this means, use the value of the IN parameter. EQ is the relatio
nal
operator. N is a value that can be supplied as a value for IN. Thus, if the value
specified by the IN parameter is equal to N, transfer control to the destination
supplied by the next parameter, which is EXIT. If IN is not equal to N, contro
l is
transferred to the job control statement immediately following the IF job contro
l
statement. Note in the example that spaces precede and follow both IF and
the
operator EQ. Note also the lack of spaces between the parentheses and the
‘&IN’ and
‘N’ terms and the lack of spaces or a comma before the word EXIT.
Providing Targets for Branching
The symbols in the label field of the job control statements provide the targets
for
branching job control statements. But the 1$, 1*, and I& job control statem
ents don’t
have a label field. You may also want to branch to the end of a JPROC
, which is an
END directive. This doesn’t have a label field that can be accessed
by a branching job
control statement. The NOP job control statement allows you to branch
to an
otherwise unaccessible position in the control stream.
The format of the NOP job control statement is:
/Isymbot NOP [QUERY]
This job control statement provides a target for a branching job contro
l statement.
The symbol must agree with the target defined in the sending job
control statement.
The optional QUERY parameter is used when you want to take
advantage of the
label-skipping facility of II OPTION QUERY. This facility is availab
le to workstation
users and console operators.
The following is an example based on the IF job control statement
example shown
earlier in “Conditional Branching” and using the END directive
as the target (it’s still
within a JPROC definition):
1/ EXEC LISTX
II IF (‘&IN’ EQ ‘N’)EXIT
II PARAM SPACE=TWO
//EXIT NOP
END
C
74
70044623-000
Run-Time Conditional and Set Symbol Job Control Statements
Notice that the IF job control statement was placed afer the EXEC job control
statement. This is allowable since it’s a run-time conditional job control statement,
which is acted upon by the run processor and then stripped from the control stream.
Note:
You can use the NOP statement to place comments in your control stream.
The comment is used in.place of the QUERYparameter, is separated from the
NOP statement by one or more blanks, and is enclosed in single quotes. When
used for this purpose, the NOP statement does not have to be the target of a
branching statement.
Run-Time Set Symbols
A set symbol is a type of variable that can be set to a value and used by the run
processor as a counter, switch, or value to control a job. Because the run processor is
responsible for making set symbols effective, they are called run-time set symbols.
There are two types of set symbols:
•
GLOBAL
A global set symbol, once declared, can be referenced anywhere in the basic
control stream as well as in any JPROC definition the control stream calls.
•
LOCAL
A local set symbol can only be declared and referenced within a JPROC
definition. (If a local and a global set symbol have the same name, the local
symbol is used within the JPROC.)
You use the following to declare run-time set symbols.
•
II GBL, II QGBL, RUN/RV (command), II RUN/RV
Declare global set symbols only.
•
IIJSET
Declares local set symbols and (if specified in a basic control stream after II GBL
or II QGBL) can be used to supply or change the value of a global set symbol
(without changing the symbol’s status to local).
Global Status Set Symbols
The GBL job control statement can be used to declare global set symbols. This
statement may appear anywhere in the control stream, and the symbols are global
from the point of declaration forward.
7004 4623-000
7-5
Run-Time Conditional and Set Symbol Job Control Statements
The format of the GBL job control statement is:
//[symbol]GBL set-id-1E=init-I][,set-i&2E=init-2
),
. .
.,set-id-nt=init-n))
The set-id parameter specifies the name of the set symbol. The mit parame
ter assigns
a value to the set symbol, provided a value has not already been assigned. For
example:
II JOB MYJOB
1/ GBLPRNTR=20
The set symbol defined in the preceding II GBL statement is PRNTR and the
value of
PRNTR (&PRNTR) is 20. The value 20 is substituted any time you referen
ce this
symbol by &PRNTR later in the control stream, or in any JPROC the contro
l stream
calls. For example:
If JOB MYJOB
1/ GBLPRNTR=20
II DVC &PRNTR=20
1/ LFD PRTFIL
Note:
1
j
The value 20 is substituted for &PRNTR when the
run processor encounters this statement. The
result is 1/ DVC 20.
The & used when referencing a set symbol is never used when defining the
set
symbol in the GBL job control statement.
The value assigned by the mit (value) parameter is used only if a value
is not assigned
by a preceding II GBL statement; is not assigned in the RUN/lW comma
nd (the
RUN/RV job control statement if you’re initiating one job from anothe
r); is not
changed later in the control stream by all JSET statement. Consider
the following:
1/ JOB MYJOB
II GBLPRNTR=20
1/ GBLPRNTR=26
C,
7-6
7004 4623.000
Run-Time Conditional and Set Symbol Job Control Statements
The value assigned by the first GBL job control statement applies for the entire
control stream any time &PRNTR is referenced. The second GBL job control
statement does not result in an error condition but has no effect on the value of
PRNTR. (You can use a JSET job control statement in place of the second GBL job
control statement to change the value of PRNTR. JSET is discussed in “Local Status
Set Symbols” later in this section.)
The effect of specifying a global set symbol and value in the RUN/RV command is as if
II GBL is inserted directly after the II JOB statement in the control stream. If, for
example, you use RV MYJOB,,PRNTR=28 to initiate a job, II GBL PRNTR=28 is
considered the first statement in the stream. You can reference &PRNTR any place in
the job stream and the run processor will substitute the value 28. Consider the
following:
/1 JOB MYJOB
•
•
ii DVC &PRNTR
II LFD PRTFIL
Note:
1
The global set symbol PRNTR was defined and given
a value of 28 in the RUN/RV comand. The run
processor substitutes 28 for &PRNTR resulting in
II DVC 28.
Remember to include a II OPTION SUB statement in your control stream if
you want values substituted for set symbols referenced in embedded data.
If you include all GBL statement for PRNTR in your control stream specifring one
value, and initiate that stream with a RUN/RV command specifying another value for
the same symbol, the value specified on the RUN/RV command is used. If, for
example, you use RV MYJOB,,PRNTR=28 to initiate the following stream:
II JOB MYJOB
II GBLPRNTR=20
II DVC&PRNTR
II LFD PRTFIL
the value 28 is substituted for &PRNTR. The value 20 is used only if you don’t supply
a value for PRNTR in the RUN/RV command.
7004 4623-000
7.7
Run-Time Conditional and Set Symbol Job Control Statements
Whenever you specify a set symbol in the II GBL statement without a value (for
example, II GBL PRNTR), you must use the RUN/RV command to supply the value, or
provide a value using the II JSET statement before the symbol is referenced.
Otherwise, the value of the symbol is considered null. This may or may not be desired.
Consider the following GBL job control statement:
II GBL PRNTR,TOKEN=DKIN
This statement declares global status for the set symbols PRNTR and TOKEN. The
value of TOKEN is DKIN. The value of PRNTR was previously defined in the
RUN/RV command, will be defined later in a JSET job control statement before
&PRNTR is referenced, or is a null value.
When coding the GBL job control statement, you cannot use the statement
continuation; specify separate II GBL statements.
With the QGBL job control statement, interactive users can declare global set symbols
in a job control stream and then specify values for those symbols through the
workstation at job run time. The format of the QGBL job control statement is:
II[symbol] QGBL setid-1E=init-1]E,set-id-2t=init-2],. ..,set-id-n[=init-n]]
The set-id parameter may be a maximum of eight characters and the mit (value)
parameter may be a maximum of 60 characters. When you run a control stream
containing all QGBL statement, the specified set symbol is displayed at the
workstation and you’re asked to provide a value for the set symbol. A null response
may indicate that a (default) value specified in the QGBL statement is valid. Suppose
you build a job control stream that includes these statements:
1/
If
If
If
JOB MYJOB
QGBL DVC=20
DVC &DVC
LFD PRNTR
I&
When you initiate the control stream (RV MYJOB) and the run processor encounters
the II QGBL statement, the following is displayed on the workstation screen:
03 ? JOB=MYJOB
SYMBOL=DVC
VALUE=20
*ENTER VALUE
If you don’t enter a value on the following line (for example, 03 22, indicating a specific
printer), the value specified in the II QGBL statement (20) is substituted for &DVC.
C
7-8
7004 4623-000
Run-Time Conditional and Set Symbol Job Control Statements
Suppose you write the following job control stream to prep a data-set-label diskette:
1/
II
/1
II
/1
II
1/
II
II
Is
JOB PREPDSL
QGBL ADDR,VSN,RCSZ,SPIRL,IPL
DVC 20
LFD PRNTR
DVC 130,&ADDR
VOL X(NOV)
LFD DISKIN
OPTION SCAN,SUB
EXEC DSKPRP
SERNR=&VSN,RECSZ=&RCSZ,SPIRL=&SPIRL, IPLDK=&IPL
VOL1
1*
Is
The// QGBL statement declares five global set symbols. One is referenced in the DVC
statement for the diskette device. The other four are referenced in the embedded
data. (The embedded data consists of keyword parameters whose values provide
necessary information for the diskette to be prepped.) When you initiate the job at
your workstation using RV PREPDSL, and the run processor begins job processing,
the following occurs:
•
A workstation screen display asks you to supply values for each of the set
symbols declared by the// QGBL statement. For example:
JOB=PREPDSL
SYMBOL=ADDR
VALUE IS NULL
*ENTER VALUE
(Assume that 320, DKOO1, 128, Y, and Y are the values you specify for ADDR,
VSN, RCSZ, SPIRL, and IPL, respectively.)
•
When DVC 130,&ADDR is encountered, the run processor substitutes the value
320 resulting in II DVC 130,320. (Had a null response been entered, then a
physical device would not be assigned.)
•
When the embedded data is encountered, the rim processor substitutes the
specified values (provided, of course, you included all OPTION SUB statement in
the job control stream) resulting in
SERNR=DKOO1,
RECSZ=128,
SPIRL=Y,
and
IPLDKY.
For information about prepping diskettes, see the Systems Service Programs (SSP)
Operating Guide, UP-8841.
If global symbols declared by II QGBL are given values through any other means (a
RUN/RV command, all GBL statement in the control stream, all JSET statement in
the control stream), you won’t be asked to submit a value at the workstation even
though the stream includes all QGBL statement.
70044623-000
7-9
Run-Time Conditional and Set Symbol Job Control Statements
Local Status Set Symbols
The JSET job control statement can be used to define a local set symbol or to change
the value of a global set symbol without changing its status to local.
The format of the JSET job control statement is:
I/symboL JSET value
The symbol specifies the name of the set symbol. The value of this set symbol is coded
as the value parameter. It may be a character string up to eight characters long
enclosed by apostrophes, if it contains blanks. For example:
1/ PRNTR JSET 20
(The symbol PRNTR is given a value of 20.)
Now, consider the following:
I/PRNTR JSET &DEVICE
In this statement, PRNTR will have whatever value is given to DEWCE. The value
for DEVICE can be supplied via the RUN/RV command, in a preceding II GBL or II
QGBL statement, in a JPROC call, or even in II JSET statement specified later in the
control stream (for example, I/DEVICE JSET 20).
The value can also be a simple 2-term expression such as &A+&B. The operations
allowed in a 2-term expression are:
Operator
Description
II
Covered quotient, A//B is equivalent to (A÷B-1)/B.
/
A/B means arithmetic quotient of A and B.
*
A*B means arithmetic product of A and B.
-
A-B means arithmetic difference of A and B.
+
A+B means arithmetic sum of A and B.
**
A**B means logical product AND of A and B.
++
A++B means logical sum OR of A and B.
A— —B means logical difference XOR of A and B.
7-10
7004 4623000
Run-Time Conditional and Set Symbol Job Control Statements
Whenever you’re performing an operation using a JSET statement, the operands upon
which the operation is to act must be numeric. Look at this example:
II GBL M=1,X2
//MX JSET &M+&X
The result of this operation is 1VIX=3.
If both the operands are not numeric, the operation is not performed and the result is
a concatenation of the values. If M had been set with the value of A in the preceding
example, the result would have been MX=A÷2. The operation would not have been
performed.
You can also use the JSET control statement to establish a null value. This can be
done by specifying either:
f/symbol JSET
or
I/symbol JSET
1
Leading zeros are not maintained for multiple-digit numeric values in a JSET control
statement. If a leading zero is required when the symbol is used, it must be created
via a second JSET control statement. For example, if you want the value of symbol P
to be 08, assign another symbol (K in this example) the value of 0, like this:
I/K JSET 0
Assign symbol P the value of 8, like this:
I/P JSET 8
When P is referenced, it must be prefixed by K Thus, the value of &K&P is 08.
As mentioned earlier, when you define a set symbol in a JPROC using II JSET, the
symbol is considered local and can only be referenced within the JPROC. JSET,
however, also allows you to change the value of a global set symbol without changing
its status to local. For example:
/1 JOB MYJOB
II GBL PRNTR=20
//PRNTR JSET 28
70044623-000
This statement changes &PRNTR to 28.
PRNTR is still a global set symbol.
7-11
Run-Time Conditional and Set Symbol Job Control Statements
If you define a global set symbol in the RUN/RV command or a GBL statement and
you don’t specify a value (for example, RV MYJOB,,PRNTR or II GBL PRNTR) you can
simply use II JSET to provide one or more values for PRNTR.
For example:
If JOB MYJOB
•
//PRNTR JSET 28
•
//PRNTR JSET 20
•
PRNTR was defined in the RUN/RV coninand or a previous
/1 JSET assigns 28 for the value of
PRNTR. Any time the run processor encounters &PRNTR,
28 is substituted until the next /1 JSET is encountered.
II GBL statement.
This statement changes &PRNTR to 20. Any time the
run processor encounters &PRNTR, 20 is substituted
until the end of job.
Specifying Set Symbol Values in Quotes
There are certain considerations you should take when assigning a value enclosed in
quotes to a set symbol.
Whenever you use II GBL or II QGBL to assign a quoted value to a set symbol, the
quotes are always considered part of the value. For example:
/1 GBL X=’ABC’,Y=XYZ
The value of X (&X) in this case is ‘ABC’ while the value of Y (&Y) is XYZ. This is
worth remembering especially if &X will be involved in a comparison using the IF job
control statement. (See “Conditional Branching” earlier in this section.) If, for
example, the value of X is set to ‘ABC’ as follows:
II GBL X=’ABC’
the following statement represents a character comparison match:
II IF (&X EQ ‘ABC’)LABEL
This statement results in a branch to LABEL because the value of X is ‘ABC’ and the
value you’re comparing X to is ‘ABC’. Consider the following statement:
II IF (&X EQ ABC)LABEL
This is not a character comparison match because the value of X is still ‘ABC’ while
the value you’re comparing X to is ABC.
7-12
7004 4623000
Run-Time Conditional and Set Symbol Job Control Statements
A different situation exists when you use II JSET to assign a quoted value because II
JSET always removes one level of quotes (if any). Consider the following:
lix JSET ‘ABC’
The value of X (&X) is ABC.
fiX JSET ‘‘ABC’’
The value of X (&X) is ‘ABC’.
IIX JSET ABC
The value of X (&X) is ABC.
//X JSET ‘‘ABC’’
The value of X (&X) is ‘ABC’.
lix JSET &X
X is now ABC.
This should also be considered when specif’ing a comparison with II IF that involved a
quoted value assigned by II JSET.
Using Symbols to Examine Job and System Related Values and Facilities
Through the use of symbols, the INQ job control statement allows you to examine job
and system related values (such as jobname, system time, and system date) or to
determine the availability of certain facilities (such as DDP and workstations).
The II INQ statement has two formats:
iisymbot INQ JOB,keyword
i/symboL INQ SYS,keyword
You use II INQ JOB to examine job related values and facilities and II INQ SYS to
examine system related values and facilities. In both formats, symbol defines the
variable symbol that is set to a value specified by keyword.
The keyword ORI, for example, sets the value of the symbol X in the following
statement to the user-id of the job’s originator (the workstation that initiated the job).
fiX INQ JOB,ORI
If you refer to the value of X (&X) elsewhere in the job control stream, the user-id of
the originator will be substituted for that value.
Consider the following:
ii JOB MYJOB
fix INQ JOB,ORI
ii OPR ‘DELIVER OUTPUT TO &X’,OPERATOR
i&
70044623.000
7-13
Run-Time Conditional and Set Symbol Job Control Statements
If USERO1 initiates the job, the run processor substitutes USERO1 for &X so that the
operator receives the message ‘DELIVER OUTPUT TO USERO1’. If USERO2 initiates
the job, the operator receives the message ‘DELIVER OUTPUT TO USERO2’.
Suppose you want to execute a program that can receive input either from a
workstation or diskette. If the job is initiated from a workstation, then workstation
input is preferred. The II INQ JOB statement, used with the keyword WKS, allows
you to determine whether the job is initiated from a workstation. This way you can
configure a job control stream that assigns a diskette device or a workstation,
depending on the situation.
The keyword WKS sets the value of the symbol X in the following statement to either
lorO:
ii X INQ JOB,WKS
If the value of X is 0, it means that a workstation is not initiated for this job. If the
value of X is 1, a workstation is initiated for the job. With this in mind, consider the
following job stream:
ii JOB MYJOB
lix INQ JOB,WKS
ii IF(&X EQ 0)DSKT
ii DVC 200
/I USE SFS
ii LFD INFO
ii GO NEXT
//DSKT
/1 VOL
ii LBL
ii LFD
i/NEXT
1
DVC 130
A123
FILE1
INFO
EXEC PROG1
This job stream is configured so that the device assignment set for the workstation is
skipped if the job is not initiated from a workstation and the device assignment set for
the diskette is skipped if the job is initiated from a workstation. Table 7-1 lists all of
the keywords that you can use with II INQ JOB and II INQ SYS.
7-14
7004 4623-000
‘I
Run-Time Conditional and Set Symbol Job Control Statements
Table 7-1. Keywords and Symbol Values for// INQ JOB and/I INQ SYS
Keyword
For// INQ JOB
For/I INQ SYS
Value of Symbol
NAME
The job name
ORI
The user-id of the originator
HOST
The host-id of the originator (null if none)
ORID
The device-id of the originator if a local workstation
WKS
0 if job is not initiated from a workstation
1 if job is initiated from a workstation
JBNO
A 4-byte job number
DDP
0 if remote DDP is not initiated
1 if remote DDP is initiated
RES
SYSRES volume serial number
RUN
SYSRUN volume serial number
DATE
The system date (yy/mnVdd)
DAY
The day of week (Sunday
MM
The month (01 through 12)
DD
The day of month (01 through 31)
VY
The year (specify last two digits)
TIME
The system time (hh.mrn.ss.)
HOST
The system’s own host-id
REL
The system release-id (w.r.rrr)
SUP
The supervisor’s name
DDP
0 if DDP is not available
1 if DDP is available
=
1, Monday
=
2, etc.)
continued
70044623-000
7-15
Run-Time Conditional and Set Symbol Job Control Statements
Table 7-1. Keywords and Symbol Values for// INQ JOB and/f INQ SYS (cont.)
Keyword
For/I INQ SYS
Value of Symbol
WKS
0 if workstation support is not configured
1 if workstation support is configured
S80
07 if running on model
08 if running on model
10 if running on model
15 if running on model
20 if running on model
50 if running on model
SPL
0 if spooling is not configured
1 if spooling is configured
JUL
Assigns Julian date YYDDD to specified symbol
7E
8
10
15
20
50
Priorities among Set Symbols, Keyword Parameters,
and Positional Parameters
External to a JPROC definition, the only possibility of substitution is for set symbols.
Inside of a JPROC definition, however, the possibility of a set symbol matching a
keyword parameter or positional parameter name does exist.
(Z:
The positional parameter name is maintained as a separate entity. Global set
symbols are maintained in a single list. Keyword parameter names and local set
symbols are maintained together, with a new definition replacing the old. The effect
of keyword parameter names and local set symbols being maintained together is to
force keyword parameter names to local status if they are mentioned in JSET job
control statements within the procedure.
When it’s determined that substitution should be performed, the following steps occur,
in the order given:
1.
A comparison is made with the positional parameter name. This test is done
first, since there is one name with many values, but it’s a relatively fast test.
Care must be taken to make the positional parameter name unique with respect
to all set symbols and keyword parameter names. A sublisted reference to a
keyword parameter cannot be distinguished from a reference to a positional
parameter.
2.
The list of local set symbols and keyword parameter names is scanned.
3.
The list of global set symbols is scanned.
C
7-16
7004 4623-000
Run-Time Conditional and Set Symbol Job Control Statements
The result is that if a keyword parameter name matches a local or global set symbol
within a procedure, the following occurs:
1.
A reference to the name obtains the keyword parameter value up until the
occurrence of a JSET job control statement for the name.
2.
From the point of occurrence of the JSET job control statement to the end of the
JPROC definition, the value of the most recent JSET job control statement is
used.
3.
At the end of the JPROC definition, the value reverts to the value of the global
set symbol at the time of entering the procedure.
Note:
70044623-000
Remember that set symbol substitution may increase the number of characters
in a value.
7-17
Section 8
How to Write and Call a Job Control
Procedure Definition
The Benefit of Procedure Definitions
Section 5 discussed the job control procedure (JPROC) call statements supplied by
Unisys. In this section, we’ll discuss how to write your own JPROC definitions and
how to call them.
A JPROC definition is similar to an assembler procedure definition, which is
explained in the Assembler Programming Guide, 7004 4532. However, the JPROC
definition is a series of job control statements and procedure directives, as opposed to
assembler instructions and directives. It consists of a PROC directive, one or more
NAME directives, a series of job control statements, and an END directive.
The PROC directive signals the beginning of the procedure, the NAME directive
declares a label by which the JPROC can be called, and the END directive signals the
end of the JPROC. Each time the series of job control statements is needed, a JPROC
call is used. Job control then inserts the necessary job control statements at the point
where the JPROC call was placed. The JPROC definition defines the coding and job
control statements needed for a particular operation, and the JPROC call specifies the
values for the variable parameters of the JPROC definition.
Coding Rules
The directives used in writing JPROC definitions take this form:
LABEL
AOPERATIONA
OPERAND
The label field extends from column 1 to column 8. At least one space must separate
the label field from the operation field, and also the operation field from the operand
field. Column 72 is used to indicate continuation, and columns 73 through 80 can
contain identification or sequence information.
Note:
For compatibility with job control statements, you can precede the label field
with two slashes (I I) in columns 1 and 2. In this case the label field extends
from column 3 to column 10.
The job control statements within a JPROC definition follow the same conventions as
regular job control statements. These are listed in Appendix A.
70044623-000
8-1
How to Write and Call a Job Control Procedure Definition
The characters that are allowable in directives and job control statements are as
follows:
Letters
Special letters
Digits
Special characters
A through Z
?$#@
0 through 9
+ * I, = ‘blank 0.
-
>
<&! :;
The terms you can use in the operand field of a directive may be symbols or character
strings, which are explained in the following paragraphs.
A symbol is a group of up to 240 alphanumeric characters used for parameter
identification and as labels. The first character must be alphabetic. Special
characters or blanks may not be contained within a symbol. The following are
examples of valid symbols:
V
CARDAREA
GS279
R$INTRN
DAVE
The
$ of R$INTRN is allowable, because it’s a special letter, not a special character.
For a symbol to be recognized by job control as a parameter identifier, it must be
immediately preceded by an ampersand.
(J
/
The following are not valid symbols:
READ ONE
SPEC’L
8AGN
-
-
-
embedded blank
special character
first character not alphabetic
The operand field in a NAME directive may be obtained by referencing the symbol
p(O), where p is the symbol used to reference any positional parameter in the
definition. The zero indicates the parameter of an operand field.
A character string can represent up to 252 valid characters, all of which must be
printable. Character strings containing embedded blanks or commas must be
enclosed in either quotation marks or parentheses. The enclosing quotation marks or
parentheses are considered part of the character string. Embedded quotation marks
are not allowed in the character string.
A null character string is represented by two consecutive quotation marks.
All parameter values are evaluated as character strings.
C:
8-2
70044623-000
How to Write and Call a Job Control Procedure Definition
Parameter Types
Parameters are used to pass information from the JPROC call to the JPROC
definition. These parameters can be equated to values, symbols, or character strings,
and may be used to specify file identifiers, file names, volume serial numbers, etc.
There are two types of parameters: positional and keyword. Positional parameters
are identified by their position within the operand field of the JPROC call; keyword
parameters are identified by the symbols assigned to them in the JPROC directive.
The rules for specifying positional and keyword parameters with respect to position,
order, omission, and format are covered in Appendix A.
Both positional and keyword parameters may be sublisted. Thus, each operand of the
JPROC call may represent one value or a series of values that may be referenced
independently. When a parameter is sublisted, the subparameters must be separated
by commas, and the entire list must be enclosed by parentheses.
For sublisted positional parameters, an operand would appear as:
(va(-1,vat-2,...,val-n)
For sublisted keyword parameters, an operand would appear as:
key=(vaL-1,vat-2,...,vaL-n)
An omitted positional parameter in a JPROC call takes the value of a null character
string. When a keyword parameter is given a value in the JPROC definition, it takes
that value if the keyword parameter is omitted in the JPROC call. When no value is
given to a keyword parameter in the JPROC definition, it takes the value of the null
character string when omitted.
Now, let’s explain the three JPROC directives.
The Start of the JPROC Definition
The PROC directive signals the start of a JPROC definition. It defines the number
and type of parameters that may be specified in the JPROC call.
The format of the PROC directive is:
LABEL
[[II]symbot)
AOPERATIONt
PROC
OPERAND
tpos,n][,k,. ..,k]
The symbol is a dummy label of one to eight alphanumeric characters. It’s used as an
entry point to the JPROC definition when it’s expanded and inserted into the control
stream. If the JPROC call also has a symbol, it replaces the symbol of the PROC
directive when the JPROC definition is called. If the JPROC call has no symbol, the
dummy label is replaced by a null character string. The characters & 0’, + / may
not be embedded in the symbol.
.
70044623-000
-
How to Write and Call a Job Control Procedure Definition
The pos parameter represents the symbol by which any positional parameter in the
body of the JPROC definition is referenced. If this parameter is omitted, no positional
parameters can be used in the JPROC call. The n is a decimal number that represents
the total number of positional parameters permitted in the JPROC call. If omitted,
zero is assumed. If you omit the pos and n parameters in this directive (thereby
indicating there are no positional parameters), you must still code two commas before
you can code any keyword name values.
The k parameter represents the name or names used in referencing keyword
parameters and their default values (if any).
To preset a keyword value, the k parameter takes the form:
E,k-1=va(ue,. ..,k-n=vatue)
In the following example, MOD1 is the symbol used as an entry point. One positional
parameter is allowed, and it’s referenced by the symbol P in the JPROC definition.
There are three keyword parameters allowed in the JPROC call; PRINTER, INPUT,
and OUTPUT. If the PRINTER keyword parameter is omitted, it defaults to 20.
MOD1 PROC P, 1,PRINTER=20, INPUT,OUTPUT
or
//MOD1 PROC P,1,PRINTER=20,INPUT,OUTPUT
Naming the JPROC Definition
The NAME directive supplies the name by which a JPROC definition is referenced. It
must immediately follow the PROC directive. More than one NAME directive can be
used, but all must be grouped at the beginning of the JPROC definition. Each such
NAME directive specifies a different name for the same JPROC definition. Multiple
NAME directives allow you to specify a different parameter in the operand field of
each directive.
Note:
You may not give a JPROC any valid job control statement names (DVC,
QGBL, etc.).
When you call the particular NAME directive on the JPROC call, you can reference
the parameter of the NAME directive with p(O), where p is the symbol used to
reference positional parameters. This will be shown in an example, which should
make this much clearer.
The format of the NAME directive is:
84
LABEL
tiOPERATION
OPERAND
[II)symbot
NAME
param
7004 4623-000
How to Write and Call a Job Control Procedure Definition
The symbol specifies the name of the JPROC definition. This is the name that’s used
on the JPROC call to obtain the JPROC definition. The param is a parameter or
parameter sublist that may be selected at job execution time.
Here’s an example of this procedure:
MOD1 PROC P,1
DUMPJOB NAME Y
DUMPSYS NAME X
II GO LABEL&P(S)
//LABELY OPTION JOBDUMP
ii GO NEXT
//LABELX OPTION SYSDUMP
I/NEXT NOP
or
//MOD1 PROC P,1
//DUMPJOB NAME Y
//DUMPSYS NAME X
1/ GO LABEL&P(Ø)
//LABELY OPTION JOBDUMP
II GO NEXT
//LABELX OPTION SYSDUMP
//NEXT NOP
This JPROC definition has two names: DUMPJOB and DUMPSYS. Positional
parameters are referenced by the symbol P, so the parameter of the NAME directive is
referenced as P(O). Assume that DUMPSYS is the name used on the JPROC call. The
parameter on this NAME directive is X. When the first GO job control statement is
interpreted, it would mean go to the job control statement with a symbol of
LABEL&P(O). This &P(O) references the parameter of the selected NAME directive.
In this case, it’s X. So X is added to LABEL, giving the symbol LABELX. This job
would have an OPTION SYSDUMP job control statement inserted at execution time.
The procedure would then go to the next job control statement, the NOP.
If DUMPJOB is the name used on the JPROC call, the parameter on the NAME
directive would be Y. When the GO job control statement is interpreted, it would
mean go to the job control statement with a symbol of LABELY (from the
LABEL&P(O)). This job would have an OPTION JOBDUMP job control statement
inserted at execution time; the GO job control statement means go to the job control
statement with a symbol of NEXT. This is the NOP job control statement; the
OPTION SYSDUMP job control statement is skipped.
Ending the JPROC Definition
The END directive indicates the end of the JPROC definition. Therefore, it’s the last
item in a JPROC definition. Everything between the PROC and END directives is
considered to be the body of the JPROC definition.
The format of the END directive is:
LABEL
[//]symbot
7004 4623-000
AOPERATIONA
END
OPERAND
unused
8-5
How to Write and Call a Job Control Procedure Definition
and, added to the PROC and NAME directives defined in “Naming the JPROC
Definition” earlier in this section, looks like the following example:
MOD PROC P,1
DUMPJOB NAME Y
DUMPSYS NAME X
any
job control
statements
needed
END
If you are submitting embedded data as part of a JPROC definition and the embedded
data contains the characters END, a special situation arises because the run processor
interprets the characters END as the END JPROC directive. To avoid this problem,
you must use all GBL job control statement to replace the END characters in the
embedded data. This is an example:
II OPTION SUB
II GBL X=END
II EXEC program-name
Is
1*
Calling JPROC Definitions
Once you’ve written and debugged a JPROC definition, use the file symbiont to store
it in the job control stream library file ($Y$JCS) or an alternate library file, and then
call it when you need it. Until that time, you can test it by placing the JPROC
definition within a control stream and issuing a JPROC call containing the name you
supplied on the NAME directive. (In this way, you can test a JPROC without having
to actually file it.) The JPROC definition is stored temporarily, in the job’s $Y$RUN
file. We’ll explain this in a little more detail in “How JPROC Definitions Are Stored”
later in this section.
To call the JPROC, you use a JPROC call in the control stream. When the run
symbiont encounters the JPROC call, it searches the job’s $Y$RUN file, then searches
the specified library file for the named JPROC definition, and then inserts the
selected job control statements from the JPROC definition into the control stream at
this point.
7004 4623-000
How to Write and Call a Job Control Procedure Definition
The format of the JPROC call statement is:
//[symbol] procname tpl,p2,...,pn,ki=vi,kj=vj,...,km=vm)
The symbol is a dummy label and is optional. When used, the symbol is substituted
for the symbol specified in the label field of the PROC directive.
The procname specifies the name of the JPROC definition. This must be the same as
that specified in the label field of a NAME directive in the JPROC definition being
called.
The p represents positional parameters, and the k=v represents keyword parameters
and their values.
Positional parameters specified in a JPROC call are associated with positional
parameters specified in job control statements in the body of the JPROC definition.
The PROC directive specifies the number of positional parameters allowed.
All parameters specified in the JPROC call must be separated by commas. Positional
parameters must precede any keyword parameters. When a positional parameter is
omitted, the comma must be retained to indicate the omission, except in the case of
omitted trailing positional parameters. When there are no positional parameters
preceding keyword parameters, two commas must precede the keyword parameters to
indicate the omission of the positional parameters.
Keyword parameters are identified by name, not by position, so an omitted keyword
parameter does not require a comma to indicate its omission. Keyword parameters
may be specified in any order.
No more than one JPROC call can be on a single line.
How JPROC Definitions Are Stored
The file symbiont stores JPROC definitions in $Y$JCS or an alternate SAT library
file. (See “Building and Storing Job Control Streams and JPROCS” in Section 1.) The
values to be used are not substituted for any preset values until the JPROC call is
issued. Substitution then takes place, and the JPROC definition is then considered to
be expanded.
A JPROC definition may be called as often an necessary, or until it’s deleted from the
library file.
7004 4623-000
8-7
How to Write and Call a Job Control Procedure Definition
The reading, verifying, and expanding of the entire control stream is a function of the
run symbiont.
Ajob input directly from a reader device may include JPROC definitions in its control
stream. The JPROC definition must appear in the control stream before any
reference to it is made. Therefore, if a JPROC definition pertained to assigning
devices to a job, it should be placed before any device assignment sets. Such JPROC
definitions apply only to that particular job; they aren’t stored in $Y$JCS or an
alternate library file, they’re stored temporarily in the job’s $Y$RUN file. They also
cannot be embedded within data.
Because the job’s $Y$RUN file is the first file searched for a JPROC definition, by
placing a JPROC definition in the job control stream you have the ability to test the
JPROC definition without storing it permanently. You can also use this facility to
temporarily override a JPROC definition that’s already stored. Whenever a JPROC
definition being called is found in the job’s $Y$RUN file, $Y$JCS or the alternate
library file is not searched.
A sample job using this facility would look like this:
If JOB TESTPROC
MOD PROC P,1
DUMPJOB NAME Y
DUMPSYS NAME X
any job control statements needed
END
II DUMPJOB
I&
II FIN
In this example, the JPROC named as either DUMPJOB or DUMPSYS would be
entered as a temporary JPROC definition, which is referenced later by the JPROC call
of II DUMPJOB. Upon encountering the PROC directive, the run processor will file
the statements up to the END directive into the job’s $Y$RUN file, which is scanned
when the JPROC call is encountered.
70044623-000
(-
How to Write and Call a Job Control Procedure Definition
Specifying an Alternate Library File to Be Searched
for JPROCS
The ALTJCS job control statement tells the run processor which alternate library file
is to be searched for JPROCS. An alternate library file is one other than $Y$JCS.
II ALTJCS specifies an alternate library file to be searched for JPROCS only, not job
control streams. An ALTJCS job control statement specification overrides an
alternate library specification on the RUN/RV command that initiated processing of
the job control stream.
You can specify multiple ALTJCS control statements in a control stream; the last
library file specified is searched for JPROCS until the next ALTJCS statement is
processed.
Note:
The run processor searches only the specified alternate library for JPROCS.
Specify ALTJCS ,,,FREE to revert to the prior order of search (alternate
library and then $Y$JCS). The FREE option must also be specified before
another ALTJCS statement can be specified.
The format of ALTJCS is:
/IEsymbot] ALTJCS [fite-Labet-id]
[I
I
[RES
[,rpw][
1
RUN
vot-ser-no]
1
1[
I
I
FREE 7[,LuNnnn]
ONLY
OFF
L°
The file-label-id is 1 to 44 alphanumeric characters long. It is optional if you’re not
searching a new library, but changing the last parameter (FREE, ONLY, OFF, or ON)
for an alternate library already defined in a previous ALTJCS statement. We’ll
discuss these options later. If you don’t specify a file-label-id, don’t specify
vol-ser-no or rpw.
The vol-ser-no parameter specifies the volume serial number of the disk where the
alternate library file resides. This parameter can also specify the volume serial
number of a format-label diskette. RES, RUN, or the actual volume serial number of
the disk or diskette may be specified. If no vol-ser-no is specified, the cataloged
vol-ser-no is used; if it is not cataloged, RES is used.
The rpw parameter specifies a read password associated with the alternate library
file. It must be specified if the file is cataloged with a read password. It is ignored if
no read password exists for the file or if the file is not cataloged.
The ONLY, OFF, and ON parameters specify order-of-search options. ONLY specifies
that only the identified alternate library file is to be searched. When OFF is specified,
the alternate library file remains open to the run processor and can be searched again
by the use of the ON or ONLY options. You specify this option if you no longer want
an alternate library file searched for JPROCS. ON specifies that the identified
alternate library file is to be searched first and then $Y$JCS. ON is the default
option. FREE is equivalent to OFF, except that it also frees the alternate device (from
the run processor).
70044623-000
8-9
How to Write and Call a Job Control Procedure Definition
Using the LUN keyword parameter, you supply a logical unit number to indicate the
device type and characteristics for the alternate library. LUN is never specified
unless a volume serial number is also specified. It is especially useful where either a
disk or format-label diskette can be the alternate library volume.
The volume serial numbers for disk and format-label diskette are syntactically the
same. As a result, the system cannot determine if a disk or format-label diskette is
required unless the volume is already mounted, or unless you use LUN parameter. If
you don’t specify a logical unit number and if the proper volume isn’t already
mounted, mount messages suggesting a disk drive, for example, could be directed to
the operator when a format-label diskette is actually required. The LUN parameter
helps avoid such confusion.
Notes:
1.
LUN is used only to determine the device type and characteristics. It has no
relationship to logical unit numbers used elsewhere in the job control stream.
2.
Confusion with mount messages is also avoided if the DVC-LFD sequence for the
file is cataloged. (See “File Cataloging” in Section 6.) By simply providing a filelabel-id in the ALTJCS statement, the correct volume serial number as well as
device type is extracted from the catalog (according to the label specified).
You can identify alternate libraries for control streams and JPROCS through the
FILE system console command and the RUNíRV workstation or console command.
Workstation commands are explained in the Interactive Services Operating Guide,
UP-9972. System console commands are explained in the operations handbook for
your system.
Parameter Referencing
The parameters ofjob control statements that require substitute values at execution
time must begin with an indicator of &.
For example, if, in the body of a JPROC definition, you have a DVC job control
statement in which you wanted to vary the logical unit number, it could be coded as
follows:
II DVC &P(2)
The P is an arbitrary symbol assigned by you in the PROC directive; the (2) indicates
that the logical unit number to be inserted is coded as the second positional parameter
on the JPROC call. The parentheses around the 2 are required.
8-10
70044623000
How to Write and Call a Job Control Procedure Definition
In this example,
II DVC &P(1,2)
the (1,2) indicates that the logical unit number to be inserted is coded as the second
subparameter under the sublist for the first positional parameter.
For each character string following the single ampersand, a substitution is made. If
the character string is invalid (not defined in the PROC directive), a null character
string is inserted.
Any job control statement may be continued between parameters, or between the
operation and the first parameter. No job control statement can exceed column 71.
This means that the total number of characters cannot exceed column 71, even after
substitution. The maximum of 71 characters includes embedded spaces. Column 72 is
used to indicate continuation.
The length of a single parameter is 242 characters. For positional parameters, this is
the value; for keyword parameters, it is the keyword and the value. If a parameter is
sublisted, the maximum length is decreased by 2 for each element of the sublist.
Note:
The maximum length of a single operand ofa job control statement is 252
characters. Thus, ifyou have a parameter of 242 characters, there are only 10
characters left for other parameters.
Here are some examples.
Example 1
In this portion of a JPROC definition, we’ll see how a value is given to the DVC
job control statement in the body of the JPROC definition.
PROC POS,1
ACTI NAME
II DVC
&POS(1)
END
Let’s say that the JPROC call is this:
II ACTI 10
the DVC job control statement that would be generated and inserted to the
control stream would be as follows:
II DVC 10
70044623-000
8-11
How to Write and Call a Job Control Procedure Definition
Example 2
If part of the JPROC definition looked like this:
PROC KEY 1=90
ACT2 .NAME
II DVC
&KEY1
END
and the following JPROC call was issued:
II ACT2 KEY1=20
this job control statement would be generated:
II DVC 20
If the JPROC call was issued without the KEY1 parameter, the value of 90 set in
the JPROC definition would be used.
8-12
70044623-000
How to Write and Call a Job Control Procedure Definition
Example 3
This JPROC definition has one positional and one keyword parameter, and two
NAME directives.
1.
2.
3.
LAB PROC POS,1,KEY1=10
MASTER NAME 20
DETAIL NAME 30
4.
5.
6.
//&LAB DVC &POS(0)
DVC &POS(1)
II
DVC &KEY1
II
END
When this JPROC call is issued:
I/Li MASTER 40,KEY1=50
these job control statements are generated:
I/Li DVC 20
DVC 40
II
DVC 50
1/
Line 4 in the JPROC definition means to take the value of the parameter in the
NAME directive that matches the name on the JPROC call MASTER. So the
first DVC job control statement has a logical unit number of 20. Line 5 means to
take the value of the first positional parameter in the JPROC call; the second
DVC job control statement has a logical unit number of 40. Line 6 means take
the value of the KEY1 keyword parameter; the third DVC job control statement
has a logical unit number of 50. LI is specified by the JPROC call as being the
substitute value for the symbol in the PROC directive. Line 4 will use this value.
So, the first DVC job control statement has a symbol of LI.
-
70044623.000
8-13
How to Write and Call a Job Control Procedure Definition
C
Example 4
A parameter sublist may be referenced. This is done with a secondary level of
indexing, which is shown in the following example:
PROC POS,1,KEY=(10,20)
EXAM NAME (30,40)
1.
2.
3.
4.
ii
II
II
/1
DVC&KEY(1)
DVC &KEY(2)
DVC &POS(0,1)
DVC &POS(0,2)
END
When the following JPROC call is used
II EXAM KEY=(50,60)
these job control statements are generated:
1/
1/
II
II
DVC
DVC
DVC
DVC
50
60
30
40
Line 1 of the JPROC definition means use the first subparameter of the KEY
keyword parameter. The JPROC call uses this keyword parameter, so its new
values (50 and 60) override the values assigned in the JPROC definition (10 and
20). Line 2 means use the second subparameter of the KEY keyword parameter.
Line 3 means use the first subparameter on the NAME directive (0,1), and line 4
means use the second subparameter on the NAME directive (0,2).
8-14
70044623-000
How to Write and Call a Job Control Procedure Definition
Example 5
A reference to a parameter may occur anywhere in the body of a procedure
definition. If the reference is the only field, and therefore naturally delimited,
there is not much likelihood of confusion. If the possibility of confusion exists, the
reference may be terminated with a period, which is a concatenation operation.
The period is dropped during the expansion of the control stream.
The following JPROC definition has two keyword parameters: KEY1 and
LABEL; neither has default conditions.
COM
PROC KEY1,LABEL
NAME
If OPR&KEY1.IS&LABEL.1990
END
If this JPROC call was used
II COM KEY1=TODAY ,LABEL=-OCTOBER-6-
this would be generated:
II OPR TODAY-IS-OCTOBER-6-1990
70044623-000
8-15
n
fl)
Section 9
Using the Interactive Job Control Dialog
The Function of the Job Control Dialog
The job control dialog is an interactive facility of 0S13 that guides you through the
process of building a job control stream or user JPROC from a workstation. To begin a
job control dialog session, key in SC JC$BLD. This activates the dialog processor and
opens the job control dialog file. Dialog text is displayed at the workstation screen
and your responses to the dialog are entered at the workstation keyboard. The dialog
processor passes your responses to the system program JC$BLD, which creates your
control stream or JPROC and stores it in the system file $Y$JCS. The functions of the
dialog processor, which manages a dialog session, are detailed in the appropriate
operations guide.
Note:
Ifyou encounter system errors when keying in SC JC$BLD, key in RV
JC$BLD and press XMIT. A short paragraph explaining RUN libraries is
then displayed followed by the question DO YOU WANT TO SAVE RUN
LIBRARIES? (Y OR N). Key in Y so that you’ll be able to enter the SC
JC$BLD command without encountering any errors in the future.
The job control dialog introduces the concept ofjob control and (if you’re building a
control stream) presents job control statements in the form of menu items from which
you choose the statements you want. If you need a dialog concept or particular
statement explained, you can ask for help by keying in HELP or a choice that
generates HELP screens. HELP screens explain the choice or statement parameters
to you. When you make a valid choice, the dialog resumes at the point where it was
interrupted. The HELP screen facility of the job control dialog can be used selectively
(statement by statement) so that you receive detailed explanations only when you
need them. More experienced users, then, can execute the dialog session quickly while
still being constrained to build syntactically correct statements. Figure 9-1 presents
an overview of the process of using the job control dialog to build a control stream or
user JPROC.
-
70044623-000
9-1
Using the Interactive Job Control Dialog
STEP 1
Key in the SC JC$BLD
command to initiate a
job control dialog session.
STEP 2
The dialog processor is
acth,ated and the job
control dialog file is opened
in response to the command
begin executing the dialog.
JOB
CONTROL
DIALOG
—
DIALOG
PROCESSOR
STEP 3
The dialog processor routes
your dialog responses to the
system program JCSBLD.
$
JOB
CONTROL
DIALOG
DIALOG
PROCESSOR
JCSBLD
STEP 4
JC$BLD uses your responses
to the dialog to build a job
control stream or user JPROC
and stores it in SY$JCS.
DIALOG
RESPONSES
f
JC$BLD
SYsJCS
Figure 9-1. Using the Job Control Dialog to Build a Control Stream or User .JPROC
9-2
7004 4623-000
Using the Interactive Job Control Dialog
Building a Control Stream with the Job Control Dialog
Let’s begin a sample job control dialog session. First, you perform the system LOGON
procedures described in the appropriate workstation user guide. Then, you key in SC
JC$BLD and its associated parameters. The first dialog screen looks like this:
DIALOG FOR JOB CONTROL
PROGRAM=
THIS DIALOG PREPARES A JOB CONTROL STREAM OR PROCEDURE
CJPROC). FOR AN EXPLANATION OF THE DIALOG PROCESS, ENTER
‘HELP’ IN THE SPACE PROVIDED. HELP
If you key in HELP, these screens are displayed:
THE DIALOG FOR JOB CONTROL IS A METHOD OF CONSTRUCTING
JOB CONTROL STREAMS AND PROCEDURES (JPROCS) USING COMPUTER
ASSISTANCE. PROMPTING FOR DATA ENTRY OR SELECTING FROM
AMONG AVAILABLE OPTIONS IS ALWAYS PROVIDED, AND YOU CAN
ASK FOR MORE DETAILED EXPLANATIONS OF STATEMENTS,
PARAMETERS, AND OPTIONS. AFTER A STATEMENT IS COMPLETED,
THE IMAGE BUILT BY THE COMPUTER AS A RESULT OF YOUR CHOICES
IS DISPLAYED ON THE WORKSTATION SCREEN. YOU MAY ACCEPT IT
FOR OUTPUT, CORRECT IT, OR REJECT IT ALTOGETHER.
Note:
To proceed from one screen to the next, you usually press the transmit key.
Whenever necessary, a note will appear at the bottom of the screen reminding
you to do this.
THE JOB CONTROL SETS ARE FORMED BY MAKING SELECTIONS FROM
MENUS OF AVAILABLE OPTIONS, AND ENTERING SOME TYPES OF
DATA DIRECTLY. THIS ALLOWS YOU AS MUCH FREEDOM IN YOUR JOB
CONTROL AS OTHER MEDIA, BUT AT THE SAME TIME PROVIDES
A STRUCTURE TO JOB CONTROL CREATION WHICH HELPS TO
PREVENT MANY COMMON ERRORS. REMEMBER, HOWEVER, THAT THE
DIALOG DOES NOT RECOGNIZE THE SAME JOB CONTROL ERRORS AS
THE RUN PROCESSOR. DIALOG ERROR CHECKING IS LIMITED TO
DIALOG OPERATION ERRORS, AND DATA TARGET MISMATCHES
(SUCH AS TRYING TO PUT ALPHABETIC DATA IN A STRICTLY
NUMBER FIELD).
70044623.000
9-3
Using the Interactive Job Control Dialog
C
The next screen asks what type of module you want to build:
JOB CONTROL MODULE TYPES
USE THIS MENU TO SELECT THE TYPE OF MODULE TO BE PREPARED:
1.
JOB CONTROL STREAM
2.
USER WRITTEN JOB CONTROL PROCEDURE (JPROC)
3.
HELP
SELECT ITEM BY ENTERING NUMBER.
If you ask for HELP, these screens are displayed:
IN ORDER TO EXECUTE ANY JOB, IT IS NECESSARY TO CONVEY TO
THE COMPUTER EXACTLY WHAT YOU WANT TO DO, AND WHAT RESOURCES
(PRINTER, READER, DISKS, ETC) ARE NEEDED. THIS IS ACCOMPLISHED
THROUGH THE USE OF JOB CONTROL. THERE ARE TWO TYPES OF JOB
CONTROL MODULES. THE COLLECTION OF JOB CONTROL STATEMENTS USED
TO RUN A JOB IS CALLED A JOB CONTROL STREAM, SOMETIMES REFERRED
TO AS THE JOB STREAM OR CONTROL STREAM. IN IT, THERE MAY BE JOB
CONTROL STATEMENTS, CALLS TO SYSTEM SUPPLIED PROCEDURES, AND THE
SECOND TYPE OF MODULE
USER-WRITTEN PROCEDURES (JPROCS).
-
JOB CONTROL PROCEDURES HAVE TWO PARTS
THE DEFINITION
AND THE CALL. THE DEFINITION IS THE JPROC MODULE CREATED
BY THE DIALOG. THE CALL IS A STATEMENT IN THE CONTROL
-
STREAM WHICH HAS THE JPROC NAME AS THE COMMAND, AND
PROVIDES ANY NECESSARY PARAMETERS. THE JPROC CALL IS USED
AS AN ABBREVIATION TO PREVENT CODING THE DEFINITION MANY
TIMES.
WHEN THE CONTROL STREAM IS PROCESSED, EACH CALL IS
REPLACED BY THE APPROPRIATE DEFINITION WHICH HAS BEEN PUT
AT THE BEGINNING OF THE STREAM OR STORED IN A SYSTEM FILE
($Y$JCS). THE RESULT IS THE SAME AS IF THE DEFINITION HAD
BEEN CODED INSTEAD OF THE CALL.
g.4
7004 4623-000
Using the Interactive Job Control Dialog
Once again, you’re asked what type of module you want to build.
JOB CONTROL MODULE TYPES
USE THIS MENU TO SELECT THE TYPE OF MODULE TO BE PREPARED:
1.
2.
3.
JOB CONTROL STREAM
USER WRITTEN JOB CONTROL PROCEDURE (JPROC)
HELP
SELECT ITEM BY ENTERING NUMBER.
You can ask that HELP screens explaining the choices be displayed again (by keying
in 3), but let’s assume you want to build a control stream. The next screen displayed
is the JOB control statement screen:
STATEMENT:
FORMAT:
FUNCTION:
JOB
I/SYMBOL JOB JOBNAME,PRI ,MINSTORE,MAXSTORE,TASKS,
TIME,OPTIONS,ACCT,BUFFERS,LOG,HDR
THIS STATEMENT IDENTIFIES A JOB AND INDICATES
THE BEGINNING OF CONTROL INFORMATION FOR THE
JOB. THE SAME NAME IS GIVEN TO THE JOB’S RUN
FILE ($Y$RUN).
IF YOU WILL NEED HELP WITH THIS STATEMENT, ENTER HELP.
What if you didn’t need HELP screens? The job control dialog screens vary according
to the responses you make to the dialog. The initial screen is the same:
DIALOG FOR JOB CONTROL
THIS DIALOG PREPARES A JOB CONTROL STREAM OR PROCEDURE
(JPROC). FOR AN EXPLANATION OF THE DIALOG PROCESS, ENTER
HELP IN THE SPACE PROVIDED.
70044623-000
9-5
Using the Interactive Job Control Dialog
Because you don’t need HELP screens to explain the dialog process, simply press the
transmit key to display the next screen. The next screen displayed is:
JOB CONTROL MODULE TYPES:
USE THIS MENU TO SELECT THE TYPE OF MODULE TO BE PREPARED:
1.
2.
JOB CONTROL STREAM
USER WRITTEN JOB CONTROL PROCEDURE (JPROC)
3.
HELP
You key in 1, indicating that ajob stream is being prepared. The next screen
displayed (since HELP screens weren’t requested) is the JOB control statement
screen:
STATEMENT:
FORMAT:
FUNCTION:
JOB
C,’
I/SYMBOL JOB JOBNAME ,PRI ,MI NSTORE,MAXSTORE ,TASKS,
TIME,OPTIOWS,ACCT,BUFFERS, LOG, HDR
THIS STATEMENT IDENTIFIES A JOB AND INDICATES
THE BEGINNING OF CONTROL INFORMATION FOR THE
JOB.
THE SAME NAME IS GIVEN TO THE JOB’S RUN
FILE ($Y$RUN).
IF YOU WILL NEED HELP WITH THIS STATEMENT, ENTER HELP.
As you can see, there is a big difference in the path the job control dialog takes,
depending on your responses to the dialog.
9-6
7004 4623.000
Using the Interactive Job Control Dialog
Let’s take the dialog one step further. If you key in HELP in response to the JOB
statement screen, each parameter of the JOB statement is explained. If HELP is not
requested, you are simply asked to key in the parametric values, without benefit of
prompting screens. When the JOB statement is built, it is displayed and you have a
final chance to change the parameters of the statement, with or without HELP
screens, or accept the statement as it appears. When the JOB statement is accepted,
the next screen presented is the job control statement master menu.
JOB CONTROL STATEMENT MASTER MENU
1.
2.
ALTER
ALTJCS
3.
CAT
4.
CC
CR
DATA
11. EXEC
12. EXT
13. FREE
21. LFD
22. MTC
23. NOP
31. ROUTE
32. RST
33. RUN/RV
24. OPR
25. OPTION
26. PARAM
35. SET
36. SFT
5.
6.
7.
DECAT
14. GBL
15. GO
16. IF
17. JNOTE
8.
DST
18. JSET
9.
DVC
19. LBL
20. LCB
10. EQU
27. PAUSE
28. QGBL
29. QUAL
30. REN
34. SCR
37. SKIP
38. SPL
39. UID
40. USE
41. VFB
42. VOL
43. 1$
44. /
45. /&
46. SYSTEM JPROCS
47. GENERAL ENTRY
48. END OF SESSION
49. HELP
ENTER SELECTION BY NUMBER
IF YOU WILL NEED HELP WITH THIS STATEMENT, ENTER HELP_
—
—
—
—
—
The rest of the job control dialog works in the same way as for the initial module-type
choice and the JOB statement screens. Each statement you choose from the master
menu is displayed and you are asked if you need help to build it. If you do, HELP
screens are displayed that explain the parameters of each statement.
Note:
The DD job control statement is not provided on the job control statement
master menu. To include this statement in yourjob control stream, make the
GENERAL ENTRY menu selection (47), then enter the statement and its
parameters in the space provided.
The control stream you create is stored in $Y$JCS. A printed summary of the dialog
session, organized by sequentially numbered paragraphs, is produced by the dialog
processor. The default logical unit number of the printer file (printed summary)
output is 20- any printer. You can accept this default or, during the dialog session,
provide a specific printer’s logical unit number. Table A-3 of the Job Control
Programming Reference Manual, UP-9984, lists the OS/3 logical unit numbers for
printers.
7004 4623-000
9-7
Using the Interactive Job Control Dialog
Building a User JPROC with the Job Control Dialog
The dialog for creating a JPROC guides you through the process of defining your
JPROC and building the job control statements and system JPROCS you want to
include in the body of the JPROC definition.
The procedure for initiating the dialog is the same as for building a job control stream:
perform the system LOGON procedures and key in SC JC$BLD.
When the job control dialog asks you whether you’re building a job control stream or
user JPROC, key in the choice for user JPROC. The dialog then presents menus for:
•
Beginning the JPROC (PROC, NAME)
•
Choosing job control statements
•
Choosing system JPROCS
•
Ending the JPROC (END)
As is the case when you’re building a job control stream, these menus generate other
menus based on your responses to the dialog.
You can request HELP screens at any point in the dialog where you need choices or
parameters explained. After the HELP screens are displayed and you make a valid
choice, the dialog returns to the point where it was interrupted.
JC$BLD uses your dialog responses to create a JPROC.
Note:
Ifyou store a JPROC in your own (alternate) library file instead of $Y$JCS,
you must include the ALTJCS job control statement in any subsequentjob
control stream that calls the JPROC. ALTJCS identifies the JPROC and
applies only to JPROCS.
Entering Embedded Data
To enter embedded data from a workstation, first choose the 1$ (start-of-data)
statement from the job control statement master menu. Then, when the master menu
is redisplayed, make the GENERAL ENTRY selection (47). Once this is done, you’ll
be able to enter your embedded data. When all embedded data is entered and the
master menu is presented again, choose the /* (end-of-data) job control statement.
If you plan to enter dialog specification language (DSL) source code as embedded data
from the workstation, a special situation arises because the characters that denote the
start of a DSL comment are the same as the end-of-data job control statement (/*)• It’s
necessary, then, to substitute another set of characters for the end-of-data job control
statement. You do this through the OPTION job control statement.
9-8
7004 4623-000
Using the Interactive Job Control Dialog
When the OPTION statement menu is displayed at the workstation screen, choose an
OPTION EOD statement. The format is OPTION EOD=xx. The first character you
select must be a slash (I). The second character can be anything but a slash (I), an
asterisk (*), an ampersand (&), or a currency symbol Cs). Let’s say you choose /Z.
Then, when the end-of-data statement is displayed as part of the job control dialog
menu, you choose GENERAL ENTRY and key in your substitute characters; /Z in this
case. The control stream you create, then, will include these job control statements:
II OPTION EOD=/Z
/S
(start of data)
(DSL source code)
IZ
(end of data)
You key in your DSL source code when the dialog requests it. By substituting
different characters for the end-of-data job control statement, you avoid any conflict
with the DSL start-of-comment delimiter.
Changing Dialog Responses
Once you build a control stream or JPROC from a workstation, you may be able to use
it for other jobs by making only a few changes to it or, you may discover that you need
to correct it. Rather than building a new control stream or JPROC from scratch to
incorporate the changes you want, you can use the audit version of the dialog
processor to change or edit the responses you made in a previous job control dialog
session. The audit version of the dialog processor outputs an audit file containing a
complete record of your dialog responses and accepts as input an existing audit file of
your responses to a previous dialog. An existing audit file used as input is considered
an old audit file. The audit file produced as output of the current dialog session is
considered a new audit file.
You begin a dialog session, which uses the audit version of the dialog processor, by
performing the system LOGON procedures and keying in RV JC$BLD. When you
identify a new and/or old audit file (by volume serial number and file label) during the
resulting dialog session, the system loads the audit version of the dialog processor.
Note:
Old and new audit file names cannot be the same when responding to
JC$BLD queries.
The audit version of the dialog processor (Figure 9-2) also outputs a printed summary
of a dialog session that is used as a guide to changing dialog responses in a subsequent
session. The summary is organized by sequentially numbered paragraphs. When you
70044623-000
9-9
Using the Interactive Job Control Dialog
use the audit file as input to the dialog processor in a subsequent session, the job
control dialog asks you to enter the numbers of the paragraphs you want to change.
The summary lists these paragraph numbers.
WORKSTATION
OLD
AUDIT
FILE
DIALOG
PROCESSOR
YOUR DIALOG
RESPONSES
JOB
CONTROL
DIALOG
NEW
AUDIT
FILE
SUMMARY
A18871
Figure 9-2. Audit Version of the Dialog Processor
Note:
Audit files must be previously allocated MIRAM files.
The audit version of the dialog processor allows you to present the job control dialog
quickly and create a
control stream or user JPROC by changing only the
responses that need to be changed. Unchanged responses are automatically routed
from the old audit file by the dialog processor to JC$BUILD without your
intervention. During the same session, you enter your new responses to the job
control dialog. You can also produce a new audit file (if you’ve specified it in the build
command) that contains a mix of responses from the old audit file and responses
entered during the current session. This audit file can then be used as input to the
dialog processor in a subsequent session.
“new”
-
Note:
Only control streams and user JPROCS created using the job control dialog
can be changed in a subsequent dialog session.
Suppose you build a control stream for a job that runs nearly every day with only a
few changes to the control stream. Perhaps you want disk and print output on some
days, and disk output only on other days. You first build the control stream on
Monday, specifying that a new audit file and a printed summary of the session be
produced. You use the audit file as input to Tuesday’s dialog session and use the
summary report as your guide to changing the appropriate dialog responses.
9-10
70044623.000
Using the Interactive Job Control Dialog
Figure 9-3 traces the process of changing your dialog responses in a subsequent
session.
ON MONDAY. you create a job control stream and
output a new audit tile (SESSION1) that contains
your responses to the job control dialog.
WORKSTATION
Jo.
DIALOG
DIALOG
PROCESSOR
RESPONSES
JCSBUILD
—CONTROL—ø’
STREAM
‘Ir
SESSION 1
(NEW AUDIT
FILE)
SUMMARY
PARA 1
PARA 2
SESSION1 IS SPECIFIED AS THE OLD
AUDIT FILE FOR TUESDAY’S SESSION
ON
SESSION I
(OLD AUDIT
FILE)
f
DIALOG
PROCESSOR
TUESDAY,
you create a
new control
stream, using
SESSION1 and the job control dialog as input to the dialog
processor. You change only those responses that need to be
changed, using Monday’s printed summary as a guide.
Unchanged responses are automatically routed from SESSION1
(old audit file) to JCSBUILD. In addition, you create a new audit
file (SESSION2) and a printed summary of Tuesday’s session,
which can be used as input to a subsequent dialog session
rewte
taken from SESSION1;
mae. in
JOB
..
JCS BUILD
ransi to the job control dialog
STREAM
‘Ir
SESSION2
(NEW AUDIT
FILE)
SUMMARY
PARA 1
PARA 2
Figure 9-3. Changing Your Dialog Responses
The appropriate operations guide has more information about using the audit version
of the dialog processor, including information about breaking off a session and
continuing it at a later time without losing your changed dialog responses.
-
7004 4623-000
9-11
0
0
Appendix A
Statement Conventions
Job Control Statement Format
Ajob control statement has a maximum of five fields, which must appear in the
following order:
1.
Indication Field
Distinguishes job control statements from data. It is required and begins with
either II, /&, 1$, or 1*.
2.
Label Field
Contains a 1- to 8-alphanumeric-character symbol; the first character must be
alphabetic. Unless this field is explained in a specified control statement, it is the
target address of a SKIP, GO, or IF control statements or the ABNORM=label
keyword parameter of the EXEC statement. This field is not separated from the
indication field by a space; it immediately follows the II.
3.
Operation Field
Contains the name of the function to be performed. It is required for all job
control statements having an indication field of II. At least one space must
separate the operation field from the label field.
4.
Operand Field
Contains the specific information concerning the items upon which a job control
function is to operate or the manner in which the function is to be performed. At
least one space must separate the operand field from the operation field.
5.
Comments Field
Contains any descriptive information desired but not processed. The field must
not contain a slash character. For those job control statements hi which an
operand is not permitted, such as the FIN control statement, all information
beyond the operation field is treated as the comments field.
70044623-000
A-i
Statement Conventions
Excluding the indication and label fields, consecutive fields must be separated by one
or more spaces. A space may not appear in a field except within apostrophes
(hexadecimal code 7D) or parentheses in an operand field.
Example
//MYTARGAD
L
J
1
I
1
2
LBL
‘MASTER CUST’
ILrJLrJL
J
1
I
I
636
4
NAME FILE
6
5
Notes:
1 Indication field
2 Label field
3 Operation field
4 Operand field: Note that spaces are allowable, because of the use
of apostrophes.
5 Comments field
6 Field separation spaces
How Job Control Statements Are Presented
The conventions used to delineate job control statements are:
•
Positional parameters must be written in the order specified in the operand field
and must be separated by commas. When a positional parameter is omitted, and
subsequent positional parameters are being specified, the commas associated
with positional parameters must be retained; otherwise, the specified parameters
will not be processed as required. If no subsequent parameters are being
specified, their associated commas should also be omitted.
For example, the ALTER job control statement has four optional positional
parameters. This is presented in text in the following format:
II[syrnbot] ALTER [phase-naine]E,address][,change)
[j
IRESET
Ll
C
A-2
7004 4623-000
Statement Conventions
Then, the statement may be written:
II
/1
II
II
II
II
II
ALTER
ALTER
ALTER
ALTER
ALTER
ALTER
ALTER
phase-narne,address,change,RESET
phase-narne,address,change
phase-name,address
phase-name
phase-name,,change
,,,RESET
phase-name,,,ORG
Note that three commas are required in both the last and next-to-last examples.
In the next-to-last example, the three commas are encountered before any
parameters and are thus used to imply that the first, as well as the second and
third parameters, were omitted. In the last example, a parameter is encountered
before any commas, and thus the first comma is used to separate the first
parameter from the omitted second and third parameters.
II ALTER ,,,ORG
If the last example used four commas, it would appear that ORG was the fifth
parameter. And, because job control only associates four parameters with the
ALTER job control statement, the ORG parameter specification would be invalid.
•
A keyword parameter consists of a word or a code immediately followed by an
equal sign, which is, in turn, followed by a specification. Keyword parameters can
be written in any order in the operand field. Commas are required only to
separate parameters.
The VFB job control statement has the following format:
//[symbot] VFB LENGTH=1ines[DENSITY= f611[,FORMNAME=symboL)
11J
L
1
[USE= [STAND1
OWNVF1
I
IL
L04NVF2-OWNVF9
[TYPE=
I
I
SDMA 1
E,OVF=(1ine-1,...,tine-n)]
0770
0776
*
t,OVF2=(line-1
line-n))(,CD1=(Une-1,...,Une-n),...]
[,CD15=(tine-1,...,tine-n)j
[,HP=n)
However, for the purpose of explaining the use of keyword parameters, we’ll use
only the first four parameters. Thus, we arrive at the following format:
//[symbol] VFB LENGTH=l.Ine5rDENSITY= 1611[,F0RMNAME=symbol][,USE=STAND1)
L
70044623-000
181]
A-3
Statement Conventions
C
Then, this job control statement may be written as:
II
II
II
If
II
•
VFB
VFB
VFB
VFB
VFB
LENGTH=Lines,DENSITY=6,FORMNAME=symboL,USE=STAND1
USE=stand,FORMNAME=symbol,DENSITY=6,LENGTH=Lines
DENSITY=6,LENGTH=Unes
LENGTH=Unes
FORMNAt4E=symbo,USE=STAND1
Ajob control statement may consist of a group of positional parameters followed
by a keyword parameter (as the last parameter).
For example:
I/[symboL] EXEC program-name[ [Ubrary-name
I
L
$Y$RUN
[$t
1
[,E±]switch-priority][,ABNORM=Labe[]
Since the last parameter is a keyword (not the last positional) parameter, this
statement may be written as follows:
II EXEC program-name,ABNORM=labet
II EXEC program-name, I ibrary-name,ABNORM=Iabet
Commas for the omitted positional parameters may be retained if desired. For
example:
II EXEC program-name,, ,ABNORM=(abel
II EXEC program-name,t ibrary-name, ,ABNORM=Iabet
The conventions for coding commas when a positional parameter is omitted and
subsequent positional parameters are being specified still apply. When the
second positional parameter is omitted, for example, the EXEC statement must
be coded as follows:
II EXEC program-name,,switch-priority,ABNORM=Iabel.
•
A positional or keyword parameter may contain a sublist of parameters called
subparameters, which are separated by commas and enclosed in parentheses.
The parentheses must be coded as part of the list. The subparameters within the
parentheses may be positional, in which case the associated commas must be
retained if a parameter is omitted, except for the case of trailing parameters, or
they may be nonpositional. The description of the subparameters will indicate
whether or not they are positional or nonpositional.
For example:
[,OVF=(Une-1,.,J.ine-n)][,OVF2=(tine-1,...,Iine-n))
A4
70044623-000
Statement Conventions
Capital letters, commas, equal signs, and parentheses must be coded exactly as
shown.
CMcc
X ‘aa’
NUMBCHAR=n
(NOV)
•
Lowercase letters and words are generic terms representing information that
must be supplied by the user. Such lowercase terms may contain hyphens and
acronyms (for readability). For example:
phase-name
max-time
destination
filename
•
Information contained within braces represents mandatory entries of which one
must be chosen, such as:
BB,nn
BM,nri
FB ,nn
FM,nn
WM,nn
RL
RU
•
Information contained within brackets represents optional entries that
(depending upon program requirements) are included or omitted. Braces within
brackets signify that one of the specified entries must be chosen if that parameter
is to be included. For example:
Esched-priority]
r
I
L
70044623-000
ALT
OPT
IGNORE
A-5
Statement Conventions
•
An optional parameter having a list of optional entries may have a default
specification that is supplied by the operating system when the parameter is not
specified by the user. Although the default may be specified with no adverse
effect, it is considered inefficient to do so. For easy reference, when a default
specification occurs in the format delineation it is printed on a shaded
background. If, by parameter omission, the operating system performs some
complex processing other than parameter insertion, it is explained in the
parameter description.
r
$Y$RUN
L
FYPE=
L
1SDMA
•
An ellipsis (series of three periods) indicates the presence of a variable number of
entries.
•
When a portion of a parameter is underlined, only that portion need be specified.
For example:
FORMNAME=symbol
can be coded as:
FO=symbot
C
A-6
70044623-000
Statement Conventions
Coding Conventions
All the job control statement information starts in position 1 and is not permitted to
extend for more than 71 positions. Job control statements begin with either one or
two slashes. In those with only one slash, no space is permitted between the slash and
the next character. However, one space must appear between this character and the
operand field. In job control statements beginning with two slashes, at least one space
must appear between the last slash and the operation field (except when using the
continuation statement (fin) or the label field).
More than one job control statement of the type beginning with two slashes may be
written on a card, but must not extend beyond position 71. At least one space must
precede the slashes denoting the beginning of the second job control statement; this is
referred to as multistatement coding.
Numbers required for particular parameters can be expressed in decimal or
hexadecimal. Numbers preceded by D’ are considered decimal. Numbers preceded by
X are considered hexadecimal. (A trailing quote may optionally be specified.) All of
the following represent the same value:
X’FF
X’FF’
D • 255
D’255’
Numbers not preceded by X’ or D’ are automatically considered decimal except in the
following cases when they default to hexadecimal:
•
Main storage sizes specified on the JOB statement (mm and max parameters)
•
Memory sizes specified on the OPTION MIN and OPTION MAX job control
statements
•
Absolute disk addresses specified on the EXT statement (addr or Tccc:hh
parameters)
•
Address on the ALTER statement (address parameter)
•
Expansion limit on the SFT statement’s DLOAD option (expansion-limit
parameter)
Character strings on the ALTER, LCB, and SET job control statements must be
specified as shown in their formats.
7004 4623.000
A-7
Statement Conventions
Statement Continuation
A continuation line is not considered to be a job control statement in itself It is a line
that contains the continuation of a job control statement in a preceding line. A
nonbiank character must appear in position 72 of the line containing the statement to
be continued. Continuation maybe used with any job control statement that contains
at least the first two fields.
A continuation statement must begin with either the 3-character sequence lln, or just
a simple II, which then must be separated by one or more blanks from the continued
portion of the job control statement. The continued statement takes the form:
f/En) param-1. .param-n
.
The n is a decimal number in the range of 1 through 9. The numbers do not need to be
consecutive; however, each number must be greater than or equal to the preceding
number used in the control stream. This is an optional field and may be left blank, or
numbers can be used so you can keep a visual record of the amount of continuation
statements used.
For example, you could code the continuation as either
Column 72
II parameters
I/i parameters
//2 parameters
X
X
II parameters
//i parameters
I/i parameters
X
X
II parameters
/1 parameters
// parameters
X
X
<
or
or
The param-1...param-n are the parameters required to continue the immediately
preceding job control statement.
Continuation can only occur at the blanks following the operation or operand fields, or
after the comma following a parameter in the operand field. When you continue job
control statements, the positions between the last used position and position 72 must
be blank. Any information you intended as a comment for this line would be treated
as data.
0
A-8
7004 4623-000
Statement Conventions
An error message occurs if:
•
Column 72 contains a nonbiank character and the card is not a valid continuation
•
Comments extend past column 71
•
A parameter list is not delimited by a comma
An example of the continuation of a multistatement line of coding is as follows:
If DVC 56
I/i CYL,1
1/ VOL ABC123,T12345,T57341
1/ LBL MASTER II LFD FILEX
If EXT ST,C,3,
X
Software Conventions
The following rules and conventions apply to the processing ofjob control statements
and directives:
•
Data cannot be contained on ajob control statement.
•
Embedded data is normally assumed to be 80 characters long; when input from
diskette, data can be 80 or 128 characters long.
•
Comments cannot contain a slash.
•
Job control does not scan past position 72; however, embedded data of up to 128
bytes is passed through.
•
The CR job control statement, and a JPROC call when used, must be the last
statement on the card.
•
The following job control statements and JPROC directives cannot be part of a
multistatement line:
i/JOB
II FIN
II PROC
If NAME
I/END
The II need not start in column 1, but must be first on the card. The II is optional
for PROC, NAME, and END.
•
The following job control statements cannot be part of a multistatement line.
They need not start in column 1, but must be first on the card.
1*
/&
/s
70044623-000
A-9
C
Appendix B
Operation Considerations
System Libraries
There are five primary system program libraries stored on the system resident device
(SYSRES). The format of these libraries conforms to the standards established by the
librarian. For a description of these standards, see the System Service Programs
(SSP) Operating Guide, UP-8841. As in all disk files, an entry for each library file is
maintained in the volume table of contents (VTOC) on SYSRES. These files may be
accessed by your program without specifying a DVC-LFD sequence, provided the file
name you use in your program is the same as the file identifier. For example:
$Y$LOD.
The five library files are:
•
System Load Library File
This file contains the load modules that are generated as output from the linkage
editor or the librarian. This includes system software load modules. This file is
used as the default input file to the system loader.
The file identifier for this file is $Y$LOD.
•
System Object Library File
This file contains the object modules generated as output from the language
translators. This includes system software object modules. This file is the
default input file to the linkage editor.
The file identifier for this file is $Y$OBJ.
•
System Macro Library File
This file contains the standard system macro definitions, and is used as the
default input file for these definitions by the assembler.
The file identifier is $Y$MAC.
•
System Source Library File
This file provides permanent storage for source modules consisting of source
coding processed by the language translators. This file is used only when
specifically referenced in the control stream. It’s never used as a default input or
output file.
70044623-000
B-i
Operation Considerations
The file identifier is $Y$SRC.
System Job Control Stream (JCS) Library File
This file provides for the permanent storage of control streams and JPROCS. It’s
used as the default output file by the file symbiont and as the default input file by
the run symbiont.
The file identifier is $Y$JCS.
Volume Table of Contents
For each file on a direct access volume, there exists a set of control blocks in the VTOC
area of the volume. Each set indicates the attributes and extents of the file, and may
contain up to two control blocks. The information contained in these blocks is used by
data management to control access to files. In case of multivolume files, there is a set
of control blocks for the file in the VTOC of each volume.
For a complete description of these control blocks, see the Consolidated Data
Management Macroinstructions Programming Guide, 7004 4607.
B-2
70044623-000
Appendix C
Job Control Statement Formats
Job Control Statements
II[syrnbot] ALTER Ephase-name)[,address]E,change)
fRESET
L
//[symbot] ALTJCS Ef I Le- [abet-id)
I
L
RES
RUN
vot-ser-no
10RG
1I [,rpw] I
FREE
ONLY
OFF
1 [,LUN=nnn]
//[symbot] CAT LfdnameE,catpw]E,SCR)E,GEN=nn]
//[symbot] CC
fcomand
l’comand and parameters’
//[symbol]
CR
If DATA FILEID=fite-identifier[,RETAIN)[,IGNORE]
If DATA STEPnnn
70044623-000
C-i
Job Control Statement Formats
f/Esymbol] DD[CFM=
I
I
I
FIXBLK 1[,BKSZ=n][,RCSZ=n][,SIZE=AUTO][,SIZEn=n]
FIXUNB
UNDEF
VARBLK
L
fKLEN
L
1
lKLE1
.1
‘-°“
fKLOC
IL
,ACCESS=. EXC
EXCR
SRDF
SRDO
SRD
SADD
UcP
=][,INDS=n]
J
[REWIND= INORWD
1UNLOAD
L
I
NORWD 1
[FILABL=
RWD
FREE
ASSIGN JL—
[,OPRW=NORWD]LCLRW=
I
L
E,TPMARK=NO) ,RECV=
ALL
[VSEC=
LOAD
L
NO
NSTD
STD
TYES1
iVMNT= IONEIl RCB= JNO
Vb0
1YES
l 1J L
‘I L
FCE
OFF
[,OFFSET=1] [RESTORE=
L
In
ii IJCACHE=
1VES!]
L
[,KEYREF=n]
1N0
ii EMSGSUPP=
1YESTJ
DM36
L
I/[symbol] DECAT (fdnameE,catpw]E,SCR][ fGEN
1R0L
L
//[symbol] DST dest-1[,dest-2,...,dest-16]
IlEsymbol) DVC
,
RES
RUN
J
addr
ALT
IGNORE
OPT
0
REQ[(n)]
REAL
C
C-2
70044623-000
Job Control Statement Formats
//[symbot] DVC PROGE,job-namell,HOST=host-id)
//[s’rnbot] EQU tun-1,type-1E,lun-2,type-2,...,lun-n,type-n]
//Esymbot] EXEC program-namer [ubrary-name 1[,[+]switch-priority][,ABNORMtabet]
I $Y$RUN
L Isys
EXT (For disk and format-label diskette)
//[symboL) EXT
fMI1
1STI
t
L JL
fmj
L
11j
[C
jCF
1
inc
1
,
addr
c:hh
[ml
L
l(bi,ai)
TBLK
CYL
TRK
OLD
,.i[,OLD][,FIX][,NTERM]
l(bi,ai)i
J
EXT (For data-set-label diskette)
//[symbot] EXT MI,C,ø,BLK,(bi,ai)E,NDI]
//[symboL] FIN
//[symbot] FREE Lfdname-1 EE(DEV)],. ..,Lfdname-n[(DEV)]]
//[symboU GBL set-id-1E=init-1],set-id-2[=init-2),...,set-id-n[=init-n]]
//[symbot] GO destination
//[symboL) IF (a op b) destination
Ii [symbol.) INQ JOB, keyword
II [symbol] INQ SYS, keyword
//[symbot] JNOTE coment-(ine[,destination-1,...,destination-n]
7004 4623-000
C-3
Job Control Statement Formats
/Itsythol.) JOB Jobname[
{P}
E
1
iminH;max[
[
1
{tasks}
{rnax.time}1
E,pririt-option- List]
E,acc-no]E,nXm]
,
ACT
LOG
NOACT
NOLOG
NONE
BOTH
fNOHDR
L.
IlEsymbot] ISET value
//[sythot] LBL
ffite-identifier 1
l’fiLe-identifier’J
[ ffite-serial-numberll[,expiration-date]
JJ
L
1VCHECK
[,creation-date]
[
[
//[symbot] LBL
IJ
L
[
1
{fiteseuence.
number}
[
{9enerationnL11ber
1
{version.num
ber}
[quaL/]tevel-id-1 [Level-id-2...[iLevet-id-n]
[{nn}
]
1
[rPw/wPw
‘[quat/]1eveL-id-11eveL-id-2...t,level-id-n]
[{nn}
]
1
t(rPw/wPw))1
ffiLe-s
[
1
,expiration-d
eriat-n
ate]E,creatio
n-date]
umber
l’CHECK
JJ
1
{fi1e.seuenc
enumber}
[
1
{9eneration.n
umber}
[
1
{versionnum
ber}
7004 4623-000
Job Control Statement Formats
LCB (For Non-SDMA printers)
fX’hex-string-l’ 1
C’char-string-1’f
I/[symbot] LCB
L
fX’hex-string-2’ 1
c’char-string-2’f
,...,
fX’hex-string-n
1C’char-string-n’
[ CARTNAME=symbo I. I
,
NAME=
48-BUS
48-Sd
63- STD
OWNLC1
OWNLC2
[X’aa’
•IC’c’
IJCARTID
L
E,NUMBCHAR=n]
LTYPE=
0770
0776
*
IJSPACE=
I
Xaa’
C’c’
L
LMISM= IIGN0RE
L
[DUAL=
1REPORT
X
‘C’abababab
C’bbbb’
L
IJMISMCHAR= fx’aa’
L
LCB (For SDMA and AP9215-1 Printers)
//[symboll LCB [CARTNAME=symbotl
IJNAME=
48-BUS
48-Sd
63-STD
L
[,TYPE=SDMA]
LMISM= rIGN0RE
‘[REPORT
L
7004 4623-000
C-5
Job Control Statement Formats
//[syffbot] LFD
ffiLename
l*f11en11
F;- 181]
fllF
L
1
L
//[syTrbot) MTC Lfdname,
C
EXTEND
PREP
ID
IGNORE
BB,nn
BM,nn
FB,nn
FM,nn
WM,nn
RL
RU
//[syrr&L) NOP [QUERY)
//[syirboL) OPR coment-Une[,destination-1,...,destination-n)
I/[syithoL] OPTION p-1[,...,p-n)
I/[symbol.) PARAM operand-1[,...,operand-n)
//[syrrbol) PAUSE coniiient-tine[,destination-i ,...,destination-n)
//[syrrbol] QGBL set-id-1[=init-1][,set-id-2[=init-2],...,set-id-n[=init-n))
//[symbot] QUAL [quatname)
II[syrrbo[) REN tfdname, mew-label 1 [,NTERM]
mnew-tabel’f
1
//[symboL] ROUTE destination-i
destination-8
1f[syrr&t) RST fitename,checkpoint-id,step-number[,jobname[(rename)))[,pri]
[,key-1=val-i,...,key-n=val-n]
Jf[syr*ot)
RUNrrjobnne[(new.name)]
Llem
jobname[(new-name)]
RV
:att-fi tename
:(att-fitename,
fRES )
RUN
lvsn
rIREs
1
LJ
I
L
C-6
ERE
HIGH
NOR
LOW
lrftime 11[,key-i=vat-1,...,key-n=vat-n)
HL ltime+nI
7004 4623-000
Job Control Statement Formats
//[symbol] SCR tfdname[ fDATE[,yyddd]
1PREE,aaaa]
L
//[symboL] SET COMREG,char-string
//[sythol] SET DATE,yy/m/dd[,t-date)[,d-date]
//[symbol] SET UPSI,switch-setting
//[symbol] SFT
1
modute-1[i...imodule-n)][DLOAD=[caLls].[{exPansion-Umit})
DLOAD=[([catls]I
[
exPansiontimit}l)1
//Esyrnbot] SKIP target-label ,mask[j
L
ALL
ANY
NONE
II [symbol] SET LDA,n,m, fstririg
‘strng
//[symbol] SPL
HOLD
TAPE
DISKETTE
EJ
LlH
1
[,nXm]
[
{nocoP}1
[,ISTL)[,brk-pge]
1
1NOKDR
//[symbol]UID
[
{maxrecf] E ,forms]
roskPcode}1
,NOUPD
f]
[user-id-i
1(addr-1)
luser-id-i(addr-i)
user-id-255
L
(addr-255)
user-id-255(addr-255)
//[symbot]USE DP,diatog-name[,printer-tfd][,new-audit-Lfcf][,old-audit-lfd]
[ {} I
//[symboL] USE LIB,module-name ,TYPE= L
10044623-000
C-7
Job Control Statement Formats
[j
//[symbol] USE MENU
fmenu-fiLe-LFD/YSfMT 1
[,initia[-menu][
/menu-1eLF
L
finn
L
]
menu-format
.
1
m
enu-format-1
..
-1=aUas-1[
2=aUas-12j
/f[syTTbol]USE SFS[
[[format-fite-LFD-1/[format-fiLe-LFD-2]
format-fite-LFD
I
L
[uinitial-screen][ {nnn}
1
[,screen-format-1=atias-1[,...,screen-format-12=atias-12]
II [symbol] VFB
LENGTH=lines
[,FORMNAME=symbo(]
[USE
L
STAND1
OWNVF1
OWNVF2-OWNVF9
rDENSITY=
L
r
,TYPE=
(SDMA printers only)
1611
181]
6770
0776
C
[,OVF=(line-1,....,tine-n)]
L
[,OVF2=(tine-1,..,tine-n)][,CD1=(line-1,...,tine-n),...]
[,CD15=(Line-1,...,line-n)]
[,HP=n]
//[sythot]VOL
Mcc
N
NMcc
r
L
J
NS
(NOV)
(PREP)
I
1
L
I
]
CS)
(NS)
(NOV)
(PREP)
1
L
I
L
SCRATCH
1
VOLSfl-3
volsn.2r
SCRATCH
S)
(NS)
(NOV)
(PREP)
(NOV)
(PREP)
I
]
L
NS)
(NOV)
(PREP)
SCRATCH
Is
1*
C
C-8
7004 4623-000
Job Control Statement Formats
Job Control Procedures
//[symbot] procname [pl,p2,...,pn,k1vi,kj=vj,. .,km=vm)
//Lfdname ACCESS
DVCnn,VOL
lbtname
{(LbLname[
votsn
RUN
*
{fl}1[
{EXTEND}1)
VOL=
votsri
RUN
*
//Lfdname ALLOC
Ibiname
{(tblname[
{}1E
{EXTEND}1
I
,EXT=
(L{sT}1
{CF}1[
{inc}1
[
{VolsflJ
Tccc:hb
TBLI(
[
{i.ai)}1
TRK
OLD
—
7004 4623-000
L
fmj
1 ,..1L,OLD][,FIX][,NDI])
l(bj,aj)f
]
c-g
Job Control Statement Formats
C
/JEsymbot]
ASM 1 [JPRNTR= [lurit,dest] 1IJIN=
ASML
N[,dest]
I
II
ASMLGJ I
2et,destj
L.
—H
L
,OtJT=
(vot-ser-no,tabet)
CRES)
(RES,Label)
(RUN,tabel)
C*,tabet)
(vot-ser-no,label)
(RES, Label)
(RUN,1 abet)
(* Label)
(N)
RUNY$R]UNY
vot-ser-no-1,tabet-1
RES,labet-1
RUN,tabet-1
*,Label.1
N
,LIN=(
,
vol-ser-no-2,tabel-2
RES,Iabet-2
RUN,tabeL-2
*,label..2
N
)
vot-ser-no-2,Label-2
RES,labeL-2
RUN,Labet-2
*,label..2
N
)
RES,$$MAc
,COPY=(
vot-ser-no-1,Label-1
RES,Labet-1
RUN,label-1
*,tabet
N
ii [SCR1=
JLST= loption
L
loPtli...oPtnh]L
[SCR2= fvot-ser-noll ,ALTLOD=
J I
IL.
,
fvol-ser-no
1RE.s
(vot-ser-no,Labet)
(RES,Labe()
(RUN,Labet)
(*
Label)
(RES,$YSLQD)
(R*,$Y:$RUN)
[SYSM64= IYES
L
C-b
1NQ
70044623-010
Job Control Statement Formats
I/[symboL)
AUTO
AUTRPG
AUTRPGL
AUTRPGLG
[RNTR= 1
[Lun[,dest]
[
IN=
N[,dest)
2ØE desti
‘—
J H
L
,OUT=
(vot-ser-no,(abel)
(RES)
(RES,(abet)
(RUN,tabet)
(*LaL)
(vol-ser-no,labet)
CRESS (abet)
(RUN, Label)
(*, label)
(N)
(ji.N,$y$RtjN)
rOUTSRC= f(vol-ser-no,Label,Lfd-name,moduLe-name)1 ,LST=
ci[
L
LSCR1=
L
Ivol-ser-noll [SCR2=
1R
,ALTLOD=
1] L
{}1
K
Ivol-ser-no
l1m
(vol-ser-no,labet)
CRESS label)
(RUN, Label)
(*, label)
(RE,:$Y$LOp
[EMB= fNO llrMoD= [3
1ESfJ
L
[copY n=
I
1E,5K1PC)
L J
(vol-ser-no,(abeL,[fd-name)
(RES,label,lfd-name)
L
E,ERRFIL=(vol-ser-no, label ,module-name)]
70044623-000
C-li
Job Control Statement Formats
f/syi±ol {CCOMP
CCOMPL
CCOMPLG}IN=(tabet,vot-ser-no)
[PRNTR=((NILunIZØ})) [,LIB1=(tabet,vot-ser-no)]
[,LIB2=(tabeL,vot-ser-no)) [,LIBC=(Labet ,vot-ser-no)]
(,OUT=({Labet,vo(-ser-nolN))] (,DEBUGC$1140)YIYES)]
E,DEFINE=(define_symbo(E=initiaL_vatue){,
[,INTERACT={NINO}] [,DUMP=CNINO}] [,LIsT=CNINOIYIYEs)]
...
E,SOURCE=CNINOIYIYES)] [,NRENT=CNINQIYIYES)]
,ALTLOD=(tabet,vot-ser-no)] t,MAINCNINQIYIYES})
[,PP={NIJIc:IYIYES}J
f/Esymbot)
COBL74
COBL74L
[RNTR=
COBL74LG
L
rIN=
I
L
(vot-ser-no,tabet)
(RES)
(RES,Labet)
(RUN,t abet)
(* (abet)
[LIN=(
(vot-ser-no,tabel) )
(RES,Labet)
(RUN,tabet)
ç*,Label )
(RESI$Y$SRC)
I
L
,OBJ=
tun[,dest]
N[,dest)
JJLINn=( vot-ser-no,tabet )
RES,Labet
RUN,Labe(
I
I
L
(vo(-ser-no,tabet) i[SCR1= fvot-ser-noii
(RES,Labet)
L.
(RUN,labe()
(RUN ,SY$RUN)
II
.1
(*, (abet)
[JSCR2= [vot-ser-noll [SCR3= [vot-ser-no
L
‘
,ALTLOD=
]L
1JN
(vot-ser-no,tabet)
(RES, (abet)
(RUN, label)
(*, label)
t,option=specification]
(RES $Y$L,OD)
(RUN,$Y$RUN)
[,ERRFIL=(vol-ser-no,tabet ,module-name)]
C’
-
C-12
70044623-000
Job Control Statement Formats
/IEsymboL]
[
COBL85
COBL85L
COBL85LG
I
,IN=
1
J
Lun[,vot-ser-no]
N
(voL-ser-no,label)
(RES)
(RES, label)
(RUN, Label)
(* Label)
,LINn=
,LIN=
,ALTLOD=
1
,SCR2=f vot-ser-no
I
L_
COBOL
COBOL I
COBOLLG
[OBJ=
I
I
I
L
I
(voL -ser-no,tabet)
(RES,Labet)
(RUN,tabet)
(*,Labet)
I
,SCR3=f vol-ser-no
J
RES
[RNTR=
COBOLBLG
(vol-ser-no,L abet)
(RES. label)
(RUN, Label)
(*, label)
.OBJ={
(vol-ser-no,tabet)
(RES, labeL)
(RUN, [abet)
(*1 LabeL)
(RES,$Y$LOD)
(RES,$Y$RUN)
COBOLB
COBOLBL
I
(RE.$Y$SR
(vol-ser-no,Labet)
(RES,tabet)
(RUN,tabeL)
(*,[abe[)
,SCR1=f vot-ser-no
1RES
/I[syrrt.ol]
PRNTR=
E,optton=specification)
LunE,destJll [1N=
N[,dest)
(vol -ser-no, Label)
(RES)
(RES, Label)
(RUN, labeL)
Label)
20[,dest]
(vo[-ser-no,[abel)
(RES,Labet)
(RUN,Labet)
(*[)
1[LIN=
C.RUN:g.S?$RUN..
]L
II
(voL-ser-no,tabet)
(RES,tabet)
(RUN,[abet)
(*[[)
[,OUT=(p-1[,...,p-n)][,LST=(p-1,..,p-n)]
rSCR1= fvoL-ser-no1 jSCR2= JvoL-ser-no
L
1RES
JL
IJSCR3= Ivol-ser-noll
I
1RUN
I I
—
lRs
,ALTLOD=
(volser-no,labe[)
(RES,Labe[)
(RUN,labeL)
(*, [abet)
CRESS $YSQD
(RUN,$Y*RU
704 4623-000
C-13
Job Control Statement Formats
/l[syutoL] DVCDKT vol -ser-no[, [un] [NOVOL= {Y}J
//[syr±oL] DVCVOL [vot-ser-nol [,[un][NOVOL= fy
RES
RUN
l.
I
J
L
/I[syrriiot] DVCVTP voL -ser-no[,[un] [PREP= {Y}
1 [NOVOL= [Y}
1
//[symbo[]
FORT
FORTL
FORTLG
FOR
FORL
FORLG
FOR4
FOR4L
FOR4LG
[RNTR=
I
IL_
[Iun[,dest]
N[,dest]
2ø[,cIest1
1[1N=
—
L
,OUT=
(vot-ser-no,Label)
(RES)
(RES,LabeL)
(RUN,Label)
(* Label)
(voL-ser-no,tabet)
(RES,LabeL)
(RUN, label)
(*, Label)
NO
(RUN,$Y$RuN)
Lscgi=
IL.
Ivol-ser-noll IJALTLOD=
I I
—I
I
(vol-ser-no,Iabe[)
(RES Label)
(RUN,(abel)
(*,[abel)
L
RuN,$Y$RUN
[,OPT(D,N,X)][,NDEI][,STX=opt ions]
[,CNL=k][LIN=
L
ff iLename1
lL1.bi
IJLST= fk
‘J L
optton
M
[
L
A
1
)][SIZE=
AP=(S .[L}
1
[,ERRFIL(vol-ser-no,labe[,module-name)]
C
C-14
70044623-000
Job Control Statement Formats
//[symbot] [FOR4
N=
1
71j
FRNTR= ILunE,dest)
N[,dest)
I
I
FOR4L
IFOR4L.G1
(votser-no,tabet)
(RES)
(RES,Labet)
(RIJN,Labet)
(* Label)
j
IZØ[,destl
H
—U
L
,OUT=
(vot-ser-no,tabeL)
(RES,Labet)
(RUN,Labet)
[SCR1= fvot-ser-no
IL.
[abet)
N
RuN;$Y$RN
,ALTLOD=
(vot-ser-no,Labet) 7[,OPT=(S,N,X))[,LIN=fitename]
(RES, (abet)
(RUN, (abet)
(*,tabet)
(RL:,$Y$RU.N
tLST=oPtion)EiMAP(SiAiL)][S1ZE= {L}
1
E,ERRFIL(vot -ser-no, (abet ,modute-name)i
I/Esyrrioti [FOR?7
[RNTR=
IFOR77L
IFOR77LG
N
(1
I
L
(Un
7IJIN=
[,vot-ser-no)})
N
‘
‘
rouT=
I
L
II
II
(vot-ser-no,(abe()
(RES,tabet)
JL
(vot-ser-no,tabet)
(RES,tabe()
(RUN, (abet)
(* label)
RUN,$Y$RUN
rSCR1=c vot-ser-no j1JSCR2=f voL-ser-no
[RE$
1RES
L
flL
FSCR3=r vot-ser-no
L
lRS
j1IJSCR4=f
L
FALTLOD=
(vot-ser-no,tabet)
I
I
(RES,Labet)
(RUN,tabe()
vot-ser-no
jRES
7
[,LIN=fitename]
L
[LIN=f (vo(-ser-no,labe()
(RES,(abet)
I (RUN,tabet)
I
L
70044623000
I
j]
H
EMIN= ,MAX= 1, [UtiP=f Y
N
I
1
C45
Job Control Statement Formats
1
lLdJ
//[sythot) [LINK
Einput-modute-name-1,...,input-mo&ite-name-1ø]
FPRNTR= [[un[,dest]
INL,dest]
I
L
,IN=
1øe
(vot-ser-no,tabe[)
(RES)
,OUT=
(RES, (abet)
(RES, (abet)
(RUN, [abet)
(*, (abet)
(RU N,$X$RU.N)
rRLIB=
I
I
L
[ALIB
I
I
L
[ALT
[CLIB=
L
(vo[-ser-no,Labet)
(RUN, labe[)
(*,[abel)
(N)
(RUN.;$i$RUN):
(vol-ser-no,label)
(RES,labet)
(RUN,Label)
(*,tabe()
(vot-ser-no,labet)
(RES,tabet)
(RUN,[abel)
(*,(abel)
lrscRl=
fvot
-ser-no1
‘1 L
[STD= IYES
V0
[,OPT=’optons]
1
(vol-ser-no, label )}
(RES, Label)
(RUN, Label)
(*, label)
(RUN ,$Y$RUN)
(vol-ser-no,label,modnarne)
(RES, Label ,modname)
(RUN, label ,modname)
(*, label,modname)
C;
1E,CMT=’coment’]
[,ENTER=express ion]
C
C-16
7004 4623-000
Job Control Statement Formats
II [symboL] PASCAL [PRNTR=N]
,IN=
(vsn,tabet)
(RES)
(RES,tabet)
(RUN,tabet)
(vsn,tabet)
IjLIN
I
(RUN,tabet)
1
[ALTLOD={ (vsn, [abet)
(RES, [abet)
(RUN,[ abet)
(RES,SYSLOD)
roBJ=
I
I
(RES,tabet)
(RUN,tabet)
L
(RUN,$Y$RUN)
(REIYSRUN)
LOBJLST[ YES ][,MEM=vatueK]LDUMPf YES
1NOJ
iN.
L
//[symbot]
RPG
RPGL
RPGLG
E
L
[RNTR= [tun[,dest]
N[,dest]
[20[,des.t]
1[IN=
L
[OUT=
(vot-ser-no,[abel)
(RES,[ abet)
(RUN,tabet)
(*, [abet)
(N)
(vol-ser-no,t abet)
(RES)
(RES, [abet)
(RUN, [abet)
(*, [abet)
,LST=
{K}J
(RUN,$Y$RUN)
JSCR1= fvot-ser-no
1 IJSCR2= Ivot-ser-no
L
[ALT
1]L
iR
iR
(vot-ser-no,t abet)
(RES, [abet)
(RUN, [abet)
(*, [abet)
C RUN , $.Y RUN )
[,CONSOLE=tfd-narne]
[EMB=
,MOD= {3
I RAM
}
[,COL=7]
[,ERRFIL=(voL-ser-no,tabet,modute-name)]
[,MI RAM=ALL) [JUNPKDS=IYES11
L
7004 4623-000
i1_]
[FNANE8= [YES
L
1N.Ô
C-17
Job Control Statement Formats
C)
If Esy±ot) SPOOL [EDIRECT=
I
DISK
11[,BUF=nXm)EJCOPIES= fn
TAPE
DISKETTEJJ
LSKIPCODE
[nil
l!1J
H
L
L
I
n ii [,FORMNAME=forms]
[RECORDS
L
[HDR= [NO ll[JTESTPAGE
L .vcj L
[No il[,PAGEBRKn]
lEs1J
11 LCOMPRESS= [No ii ERETAIN=
1s1JL
l’1JL
LUPDATE= [NO
[YES
L
iNc
[HOLD=
L
IYES11
[JSECURE
[YES
i1JL
I/ignored UDD IN=( vol-ser-no
RES
RUN
1No
,tabet[
[noextil)
118
1
L
[noextlllj 1T
[EXTEND11)
1] L
f]
,OUT= C vot-ser-nol ,tabe1[
L
.1
[PRNTR=
I
L
LPUNCII=
L
lun[,dest]
N[,dest]
de.st I
1
20.
[YES11 IJCOMPARE=
lf] L
,EXT=([MI]
I
IL.
Jmi
L
C-18
18
[
c j
CF
FJ
—a---
I
II
[YES
inc
0
I
j
,
—i
il,rrmj. i
Ll(tuiai)f
a11j
1
l(b1
addr
Tccc:hh
BLK
TBLK
CYL
TRK
OLD
,...i[,OLD][,FIX])
]
70044623-000
Job Control Statement Formats
I/ignored UDT IN=(
vot-ser-no ,tabetIj Inoextll)
l
RES
RUN
I
,OUT=(vot-ser-no,LabeU [PRNTR= fun[,dest]
N[,dest]
I
LZcit:I
L
LPUNCH=
L
IN ii
1YESI]
[COMPARE=
L
1YES
I/ignored UTD IN=(vol-ser-no,(abet),
i7[J FEXTEND17)
1
J
L
OUT= (Ivot-ser-nol 1
LabetIj
7
IJPRNTR= [tun[,dest]
181...] L
[PUNCH= INO
JL
L
ii [COMPARE
1YESI]
1 1[
[No
L
[1ncJ
[EXT=w4IJi ICF}
fmi
L
//[Lfdname] IWORKn1
JL
fmj
1
l(bi,ai)1
,..7E,OLD]E,FIX])
]
DVC=nn,VOL= [RES
RUN
1TEMPnI
Ivot -ser-no
VOL=
BLK=
{4000
70044623-000
•
l(bi,ai)i
Tccc:hhJ
RES
RUN
vot-ser-no
J
[EXTSP= {nn}
[SECALL= {nn}
[TYPE= {f lie
tYPe}1
C-19
Job Control Statement Formats
//[symboL] IWRTBIG1 ‘btock-11,’bLock-2’,...,’btock-8’]
1LWRTSMLI
(vot-ser-no,t abet)
(RES,[abet)
(RUN,Labet)
I
I
I
(*,tabet)
cRL$oD.
L
[LUN=
{
({nnn}
{tfd.name}1[Idest])
C’
C-20
7004 4623-000
Index
A
B
ABNORM parameter, EXEC statement,
4-50, 6-68
Abnormal termination, 2-10, 4-50, 6-68
ABROUMP option, 6-27
Absolute address, 2-5
ACCESS JPROC call, 5-8
Access method, specifying, 4-26
Account numbers, 4-7
Account records, suppressing printing, 6-34
ACN=account-number option, 6-26
ALIB parameter, LINK JPROC call, 5-28
ALLOC JPROC call, 5-10
Allocation, file, 4-30, 4-31, 4-33, 5-10
ALTER job control statement description,
6-56
Alternate devices, 4-14
Alternate library files
for job control streams, 1-10
for JPROCS, 1-10
for saved, translated control streams,
1-10, 6-36
searching for JPROCS, 8-9
storing control streams in, 1-10
storing saved, translated control
streams in, 1-11, 6-36
ALTJCS job control statement, 8-9
ALTLOD parameter, LINK JPROC call,
5-30
Audit version, dialog processor, 9-9
Automatic inclusion, 5-25
Backspacing, 6-20
BAL, 2-7
Basic assembler language (BAL), 2-7
Binary overflow interrupt, 6-27
BLK parameter, changing extent
specifications, 5-6
Block characters, printing, 5-33
Block numbering, tape volumes, 4-21
Blocks
allocation amounts, 4-31
changing extents, temporary work files,
5-6
file allocation, 2-7
BOF option, 6-27
Branching
conditional, 7-2
directing program control, 2-11
providing targets, 7-4
unconditional, 7-1
Buffers
load code (See Load code)
spool, 4-7
vertical format (See Vertical format
buffers)
BUF=nXm option, 6-27
Building job control streams
description, 1-9
using the job control dialog, 9-1
7004 4623-000
Index-i
Index
C
CACHE parameter, DD statement, 6-59
Card data, input spooling, 6-7
Card input, adding, 3-10
Card reader
device assignment set, 3-10
ending operation, 3-8
start of data and end of data, 3-12
Cards, adding, 6-61
CARTID parameter, LCB statement, 6-14
CARTNAME parameter, LCB statement,
6-13
Cartridge (See Print cartridge)
CAT job control statement, 6-26
Catalog, file (See File cataloging)
CC job control statement
calling saved translated control
streams, 6-49
description, 6-45
CD1 through CD15 parameters, VFB
statement, 6-17
Changing dialog responses, 9-9
Character strings
LCB statement, 6-11
phase header record comment field, load
modules, 5-30
Characters, block, 5-33
Checkpoints
INIT parameter, 4-45
restart facility, 2-11
RST statement, 6-44
CUB parameter, linkage editor JPROC
call, 5-30
CMT parameter, linkage editor JPROC call,
5-30
COBOL, naming your files, 2-7
Coding conventions, A-7
Commands, issuing (CC statement), 6-45
Comments field, A-i
Communications region, SET statement
(SET COMREG), 6-40
Conditional branching, 7-2
Continuation lines, A-8
Control fields, modifying, 6-39
Control streams (See Job control streams)
CR job control statement, 6-61
Creation date, file, 4-37
Cylinders, file allocation, 2-5, 4-30
Index-2
D
Data
compressing, 6-3
definition, 6-58
embedded (See Embedded data)
start of data and end of data, 3-12
DATA job control statement, 6-7
Data management
assigning a file name, 3-5
modules not in $Y$LOD or $Y$RUN,
6-53
Data-set label diskette (See Diskette files)
DATA STEP job control statement, 6-70
Date
block characters, 5-35
changing, 6-39
file expiration and creation, 4-37
DATE parameter, SCR statement, 6-25
DD job control statement
description and format, 6-58
keyword parameters, (table) 6-1, 6-59
DDP program-to-program facility, DVC
PROG statement, 4-17
DECAT job control statement, 6-26
Decimal overflow interrupt, 6-27
DENSITY parameter, VFB statement, 6-17
Destination, specifying
DST statement, 6-3
JNOTE statement, 6-52
OPR statement, 6-50
OPTION LOG statement, 6-29
OPTION MAS statement, 6-30
OPTION ORI statement, 6-32
OPTION OUT statement, 6-33
PAUSE statement, 6-52
ROUTE statement, 6-4
DEV parameter, FREE statement, 6-22
Device assignment sets
card reader, 3-10
different volumes on same device, 4-14
disk, 3-20
diskette, 3-20
DVCVOL JPROC call, 5-13
file name assignment, 3-5
job control statements, 2-1
minimum control stream, 3-i
renamed file, 4-40
tape, 3-14
7004 4623-000
C,
Index
temporary work files, 5-2
workstation, 3-21, 3-22
(See also Devices)
Device independent control character codes,
6-19
Device type codes, equating logical unit
numbers, 6-10
Devices
adding, 3-14
alternate, 4-14
assigning by physical address, 4-11
assigning multiple workstations to a file,
4-10
different volumes on same device, 4-12
identifying, 3-4, 4-9
logical unit numbers (See Logical unit
numbers)
multiple volumes in a file, 4-14
optional device assignment, 4-12
releasing (freeing), 6-22
too many on same volume, 5-13
using, 2-2
using multiple, SYSRES, or $Y$RUN
file, 4-9
Dialog processor
audit version, 9-10
device assignment set for workstation,
3-21
job control considerations, 6-27, 6-76
Dialog responses, changing, 9-9
Dialog session, control stream Section 9
Disk device assignment set, 3-20
Disk file area allocation
amounts, 4-31
changing specifications, 4-33
contiguous space, 4-28
cylinders, 4-30
description, 2-5
EXT statement, 4-27
formatting the file, 4-28
more disk space needed, 4-29
Disk files
changing label, 4-40
reinitializing, 4-45
scratching, 6-24
Disk volumes
file allocation, 2-5
reserving extent storage area, 4-44
sharing, 4-24
temporary work files, 5-5
7004 4623-000
(See also Volumes)
Diskette device assignment set, 3-20, 3-21
Diskette files
area allocation, 2-5, 4-34
data-set-label, 2-6
EXT statement, 4-34
format-label, 2-5
scratching, 6-24
spooling, 6-1, 6-9
Diskette volumes extent information
storage area, 4-44
multifile, 5-13
DLOAD facility, 2-13, 6-54
DLOAD parameter, SPF statement, 6-54
DOF option, 6-27
DST job control statement, 6-3
Dummy data set, 3-13
Dump, edited, 6-28
DUMP option, 6-27
DVC job control statement
adding card input, 3-10
assigning device by physical address,
4-11
assigning multiple workstations to a file,
4-10
assigning optional devices, 4-11
description, 4-8
device assignment sets, 3-4
different volumes on same device, 4-11
disk, 3-20
diskette, 3-20, 3-21
JPROC calls (DVCVOL), 5-13
minimum control stream, 3-1
multiple volumes in a file, 4-13
specifying a remote file, 4-16, 6-5
tape, 3-15
using multiple devices, 4-9
workstation, 3-21
DVC parameter, temporary work files, 5-5
DVCDKT JPROC call, 5-16
DVC PROG statement description, 4-17
DVCVOLJPROC call 5-13
DVCVTP JPROC call
description, 5-16
linkage editor JPROC call, 5-23
Dynamic expansion
main storage, 2-13
overriding SYSGEN limits, 6-56
user job region, externally referenced
program modules, 6-56
Index-3
Index
Dynamic extension, disk file
description, 4-29
JPROC calls, 5-6
Dynamic skip function, 6-69
E
Embedded data
entering from a workstation, 9-8
EOD option, 6-27
JPROC definitions, 8-6
sets, replacing in expanded control
streams, 6-70
start of data and end of data, 3-12
substituting, 6-69
END directive
ending JPROC definition, 8-5
target for branching, 7-4
End-of-data (/*)job control statement, 3-12
End-of-job (/&)job control statement, 3-7
End-of-job process, 1-9
End-of-job-step process, 1-8
ENTER parameter, linkage editor JPROC
call, 5-31
EOD=xx option, 6-27
EQU job control statement
description, 6-10
multiple devices, 4-9
Error messages
undefined set symbol, 6-37
unequal length character strings, 6-38
Errors
abnormal termination, 2-10
renaming disk files, 4-40
testing UPSI byte, 6-64
EXEC job control statement
abnormal program termination, 4-50,
6-68
format and description, 3-6
job step delimiter, 1-1
locating load module, 4-46
minimum control stream, 3-2
specWying alternate library file for
JPROCs, 8-9
task switching priority, 4-48
using the linkage editor, 5-20
Executive, 1-3
Expanded control streams, replacing
embedded data, 6-70
Index4
Expiration date, file, 4-37
Exponent underflow exception interrupt,
6-38
EXT job control statement for disk
allocating disk area for new files, 3-20
allocation amounts, 4-31
changing specifications of previously
allocated file, 4-33
cylinder allocation, 4-30
description, 2-5, 4-27
device assignment set for diskette, 3-20,
3-21
dynamic extension, 4-29
formatting a file and using contiguous
space, 4-28
specifying file access method, 4-27
EXT job control statement for diskette, 2-5,
2-6, 4-34
EXT parameter, ALLOC JPROC call, 5-11
EXTEND option, access JPROC call, 5-10
Extending files, 4-45
Extent information storage area, 4-44
Extents
allocating disk area for new files, 3-20
allocating file with JPROC call, 5-11
allocation amounts, 4-31
changing specifications, 5-6
data-set-label diskette EXT statement,
4-34
description, 2-5, 2-7
disk EXT statement, 4-27
format-label diskette EXT statement,
4-27
LFD statement, 3-5
reserving, 4-44
EXTSP parameter, 5-6
-
F
FD entry, changing, 6-60
File access methods, 4-27
File allocation
amounts, 4-31
changing specffications, 4-33
data-set-label diskette, 2-7, 4-34
disk, 2-5
format-label diskette, 2-5
JPROC call, 5-10
terms, 4-30
7004 4623-000
C
Index
File cataloging
description, 6-26
SKIP statement, 6-63
File definition
changing at run time, 6-59
linkage editor JPROC call, 5-21
File identifiers
description, 2-4
job step temporary files, 3-23
JPROC calls, 5-2, 5-12
labeled tapes, 3-17
qualifiers, 4-43
using efficiently, 4-35
File names
assigning, 3-5
description, 2-7
JPROC calls, 5-2
tape, 3-15
File serial numbers, multivolume files,
4-36
File symbiont, storing JPROC definitions,
8-7
FILE system console command, 1-9
FILE workstation command, 1-9
FILEID parameter, DATA statement, 6-7
Files
accessing previously allocated, 5-8
allocating with a JPROC call, 5-10
cataloging (See File cataloging)
different versions, 4-39
existing specifications, 4-45
formatting and using contiguous space,
4-28
identifiers, 2-4
job step temporary, 3-23, 5-2
job temporary, 3-23, 5-2
logical, 4-44
logical names, 2-7
multivolume (See multivolume files)
naming, 2-7, 3-5
renaming, 4-40
scratching, 6-24
spooling, 6-1
(See also Disk files)
FIN job control statement, 3-8
Floating-point significant exception
interrupt, 6-37
Format-label diskette (See Diskette files)
FORMNAME parameter, VFB statement,
6-17
7004 4623-000
Forms control, 6-16
Forms, special (SPL statement), 6-2
FORTRAN, naming your files, 2-8
Forward spacing, 6-20
FREE job control statement, 6-22
G
GABRDUMP option, 6-27
GBL job control statement, 7-5
GDUMP option, 6-27
Generation number, file, 4-39
GJOBDUMP option, 6-28
Global set symbols
calling control streams, 6-47
description, 7-5
GO job control statement
branching, 2-11, 7-1
description, 7-1
GO option, 6-28
Graphic symbols, print cartridge, 6-11
GSYSDUMP option, 6-28
H
HDR option, 6-28
HELP screens, job control dialog, 9-3
Hexadecimal characters
LCB statement, 6-11
SET COMREG statement, 6-41
HOLD option, 6-28
HOLD parameter, SPL statement, 6-2
Host-id, user-id pair (See Destination)
HOST parameter
DVC statement, 4-16
DVC PROG statement, 4-18
HP parameter, 6-19
IF job control statement
branching, 2-11
description, 7-2
IGNORE parameter
DATA statement, 6-8
LFD statement, 4-44
IN parameter, linkage editor JPROC call,
5-24
Index-5
Index
Temporary work files
changing extent specifications, 5-6
job step, 3-23, 5-2
setting up, 5-2
using your own volume, 5-5
Termination, job, 2-10
TEST option, 6-37
Test pattern page, 6-3
Time of day, block characters, 5-35
TRACE option, 6-37
Tracks, 2-6
Transfer address, ENTER parameter
(LINK JPROC call), 5-31
Translated control streams, calling, 6-49
TSK option, 6-38
TYPE parameter
LCB statement, 6-14
VFB statement, 6-18
U
UID job control statement, 3-22, 4-11, 4-16
Unconditional branching, 7-1
UNDEFINED option, 6-38
UNEQUAL option, 6-38
USE job control statement
dialog processor, 3-22, 6-76
library services, 6-78
menu services, 3-22, 6-74
screen format services, 3-22, 6-73
source module access, 6-78
USE parameter, VFB statement, 6-18
User-id, host-id pair (See Destination)
User local data area, 6-41
User program switch indicator (UPSI)
setting, 6-39
testing (SKIP statement), 6-64
V
Version number
block characters, 5-36
file, 4-39
Vertical format buffers
changing, 5-24
skip codes, 6-2
VFB statement, 6-16
Vertical line positioning, 6-16
VFB job control statement
description, 6-15
linkage editor JPROC call, 5-24
SPL statement, 6-3
VOL job control statement
description, 4-19
device assignment set for disk, 3-20
device assignment set for diskette, 3-20,
3-21
device assignment set for tape, 3-15
DVCVOL JPROC call, 5-13
extending tape volumes, 4-23
ignoring or changing volume serial
number, 4-24
multivolume files, 4-19, 4-27
sharing disk volumes, 4-24
tape volumes, special characteristics,
4-21
VOL parameter, temporary work files, 5-5
Volume serial numbers
alternate library file, 8-9
description, 2-3
ignoring or changing, 4-23
multivolume files, 4-19, 4-36
tape, 3-15
temporary work files, 5-5
VOL statement, 4-19
Volume table of contents (VTOC), B-2
Volumes
data-set-label diskette file allocation,
2-7
different on same device, 4-12
disk file area allocation, 2-5
disk, sharing, 4-24
format-label diskette file allocation, 2-5
identifying files, 2-3
multiple, assigning file serial numbers,
4-36
multiple, online simultaneously, 4-27
releasing (freeing), 6-22
tape, extending, 4-22
tape labels, 3-17
tape, special characteristics, 4-21
temporary work files, 5-2, 5-5
too many devices, 5-13
VOL statement, 3-14, 4-19
C)
C)
Index-12
7004 4623-000
Index
w
SYMBOLS
Work files
linkage editor, 5-29
temporary, setting up, 5-2
(See also Scratch files)
WORK JPROC call
changing extent specifications, 5-6
description, 5-2
using your own volume, 5-5
Workstation
assigning more than one to a file, 4-10,
4-15
building a job control stream,
(figure) 9-2
changing control stream execution, 6-35
communicating with operator, 6-50
device assignment set, 3-21
dynamic skip function, 6-69
master, reassigning, 6-30
originator, reassigning, 6-32
releasing (FREE statement), 6-23
WRTBIG JPROC call, 5-33
WRTSML JPROC call, 5-33
V
$Y$CAT, 6-26
$Y$JcS
adding cards to control streams, 6-61
calling control streams, 6-48
description, B-2
restarting a job, 6-42
storing job control streams, 1-9
$Y$LOD
description, B-i
locating load module, 4-46
$Y$MAC, B-i
$Y$OBJ
ALIB and RLIB parameters (LINK
JPROC call), 5-25
description, B-I
$Y$RUN
locating load module, 4-46
preparing ajob for execution, 1-6
temporary work files, 5-3
using the linkage editor, 5-18
$Y$SAVE, running a job control stream,
1-12
$Y$SRd, B-i
x
XIJF option, 6-38
7004 4623-000
Index-13
C
C
Source Exif Data:
File Type : PDF File Type Extension : pdf MIME Type : application/pdf PDF Version : 1.6 Linearized : Yes XMP Toolkit : 3.1-701 Create Date : 2012:06:29 17:04:41+01:00 Creator Tool : Xerox WorkCentre 7535 Modify Date : 2012:07:01 23:12:09+02:00 Metadata Date : 2012:07:01 23:12:09+02:00 Producer : Xerox WorkCentre 7535 Format : application/pdf Document ID : uuid:dce53d31-1601-432c-8f40-9f1f0f182932 Instance ID : uuid:8e8f961f-6e08-48c8-a250-d75a2fb3e551 Page Count : 334 Creator : Xerox WorkCentre 7535EXIF Metadata provided by EXIF.tools