PD4ML – V4 Programmers Manual

The major difference with the previous PD4ML versions is that the new API performs the conversion in two phases.

First you need to invoke readHTML() method to read a source document. By default an HTML (styled with CSS) is

expected. However it can be arbitrary XML, where tag nature is specified with CSS display display property, e.g. <sideblock

style="display: block; float: right">Side content</sideblock> .

After the document is read and parsed you can render and write it as PDFPDF, PDF/APDF/A, RTFRTF or an raster image with

writePDF() or another corresponding method. If you need a conversion result in different formats – invoke

corresponding write*() methods multiple times, there is no need to reread the source HTML.

1. PD4ML pd4ml = new PD4ML();

3. String html = "TEST<pd4ml:page.break><b>Hello, World!</b>";

4. ByteArrayInputStream bais =

5. new ByteArrayInputStream(html.getBytes());

7. // read and parse HTML

14. pd4ml.writePDF(fos);

16. // alternatively or additionally:

17. // pd4ml.writeRTF(rtfos, false);

18. // BufferedImage[] images = pd4ml.renderAsImages();

20. // open the just-generated PDF with a default PDF viewer

21. Desktop.getDesktop().open(pdf);

Suggest edit

To use PD4ML in Maven environment you’d need to configure something like that:

1. Access to regular binaries

Project pom.xmlpom.xml

2. Access to the binaries with source code

Of course, it is recommended to store the password in encrypted form there. See Maven documentationMaven documentation how to do

Project pom.xmlpom.xml

choice and make sure there is a JRE v1.6+JRE v1.6+ installed on your workstation/server.

There are two tools available:

The command line tool is implemented as com.pd4ml.tools.Pd4Cmd.class . The class is specified as a main class of

the JAR, so it can be executed a simple way:

java -jar pd4ml*.jar

where java*.jar is an actual JAR file name, like pd4ml-4.0.1.jar .

There are four typical usage scenarios:

Customized HTML-to-PDF conversion

java -Djava.awt.headless=true -Xmx512m -jar pd4ml.jar "http://pd4ml.com" 1200

java -Djava.awt.headless=true -Xmx512m -jar pd4ml.jar "http://pd4ml.com" 1200 LETTER -

bookmarks HEADINGS -pdfforms -debug -out pd4ml.pdf

bookmarks HEADINGS ) and to convert HTML forms to interactive PDF forms ( -pdfforms ).

PDF page removal

The call prints to STDOUTSTDOUT basic PDF info: document permissions (as a hex number), document author, document title,

number of document pages (decimal number)

java -Djava.awt.headless=true -Xmx512m -jar pd4ml.jar -tools file:c:/docs/test.pdf -

pagerange 2-3,5+ -out c:/docs/newdoc.pdf

Also for runtime environment simplicity we’ll add the example PD4ML API calls to the application test suite. A typical test

case class looks like the example:

We are going to extend testApp() method with PD4ML usage example code.

Creating a PD4ML object

When writing an application with PD4ML API, the very first thing to do is to create the PD4ML object:

The constructor in the example has no parameters and it expects to find license pd4ml.lic file in the working dir of the

After you’ve got an instance of PD4ML you can read and parse a source HTML document with one of a variety of

readHTML() methods.

In the example it reads from a prepared input stream. But it can also read from an arbitrary URL.

If you want to point the converter to a font directory or to apply an additional style, headers, footers, watermarks etc – the

6. public BufferedInputStream getResourceAsStream(String resource, FileCache cache)

7. if (!resource.toLowerCase().startsWith(PROTOCOL)) {

8. return null;

9. }

11. // interpret the "resource" parameter according to your protocol (e.g. as a key to

A005AddCustomResourceLoader.java

Once PD4ML custom tags are deployed, its actions can be called from your HTML using an XML syntax.

First, reference a JAR file containing a tag library from your JSP: <%@ taglib uri="http://pd4ml.com/tlds/4.0"

prefix="pd4tl" %>

The uri attribute is a unique TLD identifier. It does not point to an existing Web document, but refers to pd4tl.tld

file, packaged to pd4ml*.jar and registered there under the same uri identifier.

1. <%@ taglib uri="http://pd4ml.com/tlds/4.0"

2. prefix="pd4tl"%><%@page

3. contentType="text/html; charset=ISO8859_1"%><pd4tl:transform

output to text mode and prefixes the PDF bytes with unexpected characters. As the last resort PD4ML taglib tries to reset

such undesired content (if any), but it is not possible in all environments and java.lang.IllegalStateException:

Prompt to save just-generated PDF

By default if you open a PD4ML-enabled JSP with a Web browser, it opens a generated PDF inline (within the browser

frame). If you’d like the browser to propose to save the PDF under particular name or open it with the system PDF viewer,

add two transform attributes inline and fileName

4. screenWidth="400"

5. pageFormat="A5"

12. body {

13. color: red;

14. font-family: Tahoma, "Sans-Serif";

15. font-size: 10pt;

16. }

17. </style>

18. </head>

19. <body>

20. <p>

21. Hello, World!

27. </tr>

28. </table>

29. </body>

30. </html>

31. </pd4tl:transform>

1. <%@ taglib uri="http://pd4ml.com/tlds/4.0"

2. prefix="pd4tl"%><%@page

3. contentType="text/html; charset=ISO8859_1"%><pd4tl:transform

4. screenWidth="800"

5. pageFormat="A4"

Identically you may define a watermark with <pd4ml:watermark> tag.

with a title value taken from HTML’s <title> > tag.

$[total] to be substituted with the current page number and total number of pages correspondingly.

How to add dynamic data (like current date) to PDF header or footer

It is JSP. Forget about PDF and do it you would normally do it in a regular JSP:

14. <body>

15. Hello, World!

16. </body>

17. </html>

18. </pd4tl:transform>

1. <%@ taglib uri="http://pd4ml.com/tlds/4.0"

2. prefix="pd4tl"%><%@page

3. contentType="text/html; charset=ISO8859_1"%><pd4tl:transform

4. screenWidth="800"

11. </head>

12. <body>

13. <pd4ml:page.header height=30 scope="1">$[title] Visible only on the first

14. <pd4ml:page.footer height=30>Page $[page] of $[total]</pd4ml:page.footer>

15. Page 1

16. <pd4ml:page.break>

17. Page 2

18. </body>

19. </html>

More IBM Notes URL syntax info…

If you run JSP infrastructure on IBM Domino server, another online method of PDF generation would be to use PD4ML

JSP custom tag library (TODO)(TODO).

If an online document generation method (involving HTTP, JSP etc) is undesired, there is a possibility to generate PDF

documents from their XML representations (so called DXL).

DXL is an adaptation of XML used to describe, in detail, the structure and data contained in a Domino database. It describes

data and design elements such as views, forms, and documents and provides a basis for importing or exporting XML

remote ssh/telnet sessions.

test.dxl 1200 are DXL locationDXL location and htmlWidthhtmlWidth (virtual “browser” frame width) parameters.

-xsl notesdefault applies XSL stylesheet to the input document. In the case it refers to the built-in default DXL-to-

HTML XSL, but it can be an URI of an arbitrary external stylesheet.

The default PDF document format: A4 / PORTRAITA4 / PORTRAIT

In the example 1200pxpx width of rendered document will be mapped to 595ptpt widths of A4 page format.

As long as an output file path omitted, the output is sent to STDOUTSTDOUTand can be piped to another application.

Suggest edit