XMLmind Word To XML Manual

Hussein Shafie
XMLmind Software
35 rue Louis Leblanc,
78120 Rambouillet,
France,
Phone: +33 (0)9 52 80 80 37,
Web: http://www.xmlmind.com/w2x/
Email: mailto:w2x-support@xmlmind.com (public mailing list)

Contents

1 Introduction 4

2 Installing w2x 5

2.1 Contents of the installation directory 6

3 Alternatives to using the w2x command-line utility 9

3.1 The w2x-app graphical application 9

3.2 The “Word To XML” add-on for XMLmind XML Editor 9

3.2.1 Installing the “Word To XML” add-on 10

3.3 The “Word To XML” servlet 10

3.3.1 Contents of the servlet software distribution 10

3.3.2 Installing the servlet 11

3.3.3 Configuring the servlet 11

3.3.4 Using the servlet to convert DOCX files 12

3.3.5 Non interactive requests 13

4 Getting started with w2x 15

4.1 How to generate useful multi-page HTML 17

5 Going further with w2x 19

5.1 Stock XED scripts 21

6 Customizing the output of w2x 24

6.1 Customizing the XHTML+CSS files generated by w2x 24

6.1.1 Using a XED script to modify the styles embedded in the XHTML+CSS file 24

6.1.2 Appending custom styles to the styles embedded in the XHTML+CSS file 24

6.1.3 Using an external CSS file rather than embedded CSS styles 25

6.1.4 Combining all the above methods 26

6.2 Customizing the semantic XML files generated by w2x 27

6.2.1 Converting custom character styles to semantic tags 27

6.2.2 Converting custom paragraph styles to semantic tags 28

6.2.3 The general case 30

6.3 Generating XML conforming to a custom schema 33

7 The w2x command-line utility 34

7.1 Variables substituted in the parameter values passed to the –p and –pu options 36

7.2 Default conversion steps 37

7.3 Automatic conversion step parameters 37

8 Conversion step reference 38

8.1 Convert step 38

8.2 Delete files step 40

8.3 Edit step 40

8.4 EPUB step 48

8.5 Load step 48

8.6 Save step 49

8.7 Split step 49

8.8 Transform step 51

8.9 Web Help step 55

9 Embedding w2x in a Java™ application 57

9.1 Extension points 58

9.1.1 Custom conversion step 58

9.1.2 Custom image converters 58

9.1.2.1 Specifying an external image converter 59

9.1.2.2 Controlling how image files found in the input DOCX file are converted to standard formats 60

10 Limitations and implementation specificities 62

10.1 About tab stops 64

Index 65

Introduction

Microsoft® Word is an amazing popular writing tool. However, its main drawback is that, once your document is complete, you cannot do much with it: print it, convert it to PDF or send it as is by email.

XMLmind Word To XML aims no less than to suppress Microsoft® Word main drawback. This 100% Java™ software component allows to automate the publishing —in its widest sense— of contents created using Microsoft® Word 2007+.

More precisely, XMLmind Word To XML (w2x for short) allows to automatically convert DOCX files to:

Clean, styled, valid XHTML+CSS, looking very much like the source DOCX files.

Because the generated XHTML+CSS file is clean and valid, you can easily restyle it, extract metadata or an abstract from it before publishing it.

Unstyled, valid, semantic XML (DITA, DocBook, XHTML, your custom schema, etc).

In this case, most styles are converted to semantic tags. For example, numbered paragraphs are converted to proper ordered lists.

Generating semantic XML out of DOCX files is useful for interchange reasons (e.g. implement open data) or because you want to port your existing documentation to a structured document format where form and content are completely separated (e.g. implement single source publishing).

Of course, deploying w2x does not require installing MS-Word on the machines hosting the software. Also note that w2x does not require the authors to change their habits while using MS-Word: no strict writing discipline, no specific styles, no specific document templates, no specific macros, etc.

This document explains:

how to install and use w2x;

how to customize the output of w2x;

because w2x has been designed to be easily embedded in any Java, desktop or server-side, application, how to embed a w2x processor in a Java application.

Installing w2x

Requirements

XMLmind Word To XML (w2x for short) requires the Oracle Java™ runtime 1.6 to 1.8 (new Java 9 not yet supported). However, w2x is officially supported by XMLmind only on Windows XP to 10, macOS 10.13 (High Sierra), 10.12 (Sierra) and and Linux.

On Linux, make sure that the Java bin/ directory is referenced in the $PATH and, at the same time, check that the Java runtime in the $PATH has the right version:

$ java -version

java version "1.8.0_131"

Java(TM) SE Runtime Environment (build 1.8.0_131-b11)

Java HotSpot(TM) 64-Bit Server VM (build 25.131-b11, mixed mode)

On Windows and on the Mac, this verification is in principle not needed as the java executable is automatically found in the $PATH when Java has been properly installed.

Install on Windows

Download the setup.exe distribution.

This setup.exe files includes a very recent, generally the most recent, Oracle® private Java™ runtime. Therefore, you don't need to install Java on your computer. Moreover, if you have Java already installed on your computer, then your public Java runtime will be ignored by w2x.

Double-click on the setup.exe file to launch the installer.

Follow the instructions of the installer.

Install on the Mac

Download the .dmg distribution.

Double-click the downloaded .dmg file to open it in the Finder.

Copy the WordToXML.app folder, an application bundle represented by icon ???, anywhere you want. For example, drag&drop this icon to the /Applications folder or to your desktop.

Start the w2x-app desktop application by double-clicking on the ??? icon (or use the Launchpad).

The first time w2x-app is started, your Mac will generally ask you to confirm that you actually want to open an application downloaded from the Internet. Click Open to confirm.

Don't worry, w2x-app has been digitally signed using a certificate issued by Apple itself. This confirmation is required for any digitally signed application not coming from the App Store.

Move the downloaded .dmg file to the Trash.

About Java on the Mac

The .dmg distribution includes a very recent, generally the most recent, Oracle® private Java™ runtime. However, this private runtime is not necessarily the one which will be used to run w2x-app:

- If you have an Oracle Java runtime at least as recent as the bundled one installed on your computer, this runtime will be used to run w2x-app.

- Otherwise, it's the bundled runtime which will be used to run w2x-app.

Manual install on any Java 1.6+ platform (Windows, Mac, Linux, etc)

Unzip the .zip distribution in any directory you want.

C:\> unzip w2x-1_3_0.zip

C:\> cd w2x-1_3_0

C:\w2x-1_3_0> dir

... <DIR> bin

... <DIR> doc

... <DIR> legal

...

XMLmind Word To XML is intended to be used directly from the w2x-1_3_0/ directory. That is, you can run the w2x command by simply executing (in a Command Prompt on windows, a terminal on Linux):

C:\w2x-1_3_0> bin\w2x

Usage: w2x [-version] [-v|-vv] [Options] [-liststeps] in_docx_file out_file

-version

Print version number and exit.

-v|-vv

Verbose.

-liststeps

List the conversion steps to be executed and exit.

Use '-?' to list options.

Contents of the installation directory

If the .dmg distribution has been used to install XMLmind Word To XML on the Mac, the following subdirectories are found in WordToXML.app/Contents/Resources/w2x/.

bin/w2x, w2x.bat

Scripts used to run XMLmind Word To XML (w2x for short). Use w2x on any Unix system. Use w2x.bat on Windows.

bin/w2x-app.exe, w2x-app.jstart

File w2x-app.exe is used to start w2x-app, a graphical application easier to use than the w2x command-line utility, on Windows. This .exe file is a home-made launcher parameterized by xxe.jstart, an UTF-8 encoded, plain text file.

bin/w2x-app, w2x-app-c.bat

Scripts used to run w2x-app, a graphical application easier to use than the w2x command-line utility. Use w2x-app on any Unix system. Use w2x-app-c.bat on Windows , but only when you need to start w2x-app with a console. On Windows, a console is needed to be able to see low-level error messages.

doc/index.html

Contains the documentation of w2x.

doc/manual/

Contains XMLmind Word To XML Manual. This document is available in source DOCX format, in PDF format and in all the output formats supported by w2x.

doc/manual/conv_manual.sh, conv_manual.bat

Scripts allowing to convert XMLmind Word To XML Manual to all the output formats supported by w2x. The files generated by these scripts are found in doc/manual/out/.

doc/xedscript/

Contains The XED scripting language.

doc/w2x_app_help/

Contains the online help of w2x-app, a graphical application which is easier to use than the w2x command-line utility.

doc/api/

Contains the reference manual of the Java™ API of w2x (generated using javadoc).

legal/, legal.txt

Contains legal information about w2x and about third-party components used in w2x.

lib/

All the (non-system) Java™ class libraries needed to run w2x:

jai_imageio.jar: Java Advanced Imaging Image I/O; needed to support the TIFF picture format.

resolver.jar: XML Entity and URI Resolvers.

saxon.jar: The Saxon 6.5.5 XSLT 1.0 engine.

w2x_all.jar: self-contained JAR containing everything needed to run w2x, that is, all the other JAR files and also all the scripts and the stylesheets found in subdirectories xed/ and xslt/.

w2x.jar: contains the w2x engine.

w2x_rt.jar: contains a runtime needed by the w2x engine. All these classes come from XMLmind XML Editor.

wmf2svg.jar: WMF to SVG Converting Tool & Library; needed to support the WMF picture format.

wmf_converter.jar: contains a picture format plug-in based on wmf2svg.jar.

whc.jar: contains the XMLmind Web Help Compiler engine.

snowball.jar: Snowball is used by XMLmind Web Help Compiler to implement stemming.

xed/

Contains the XED scripts used to convert styles to semantic XHTML tags.

xslt/

Contains the XSLT 1.0 stylesheets used to generate semantic XML.

Alternatives to using the w2x command-line utility

The w2x-app graphical application

Graphical application w2x-app should be easier to use than the w2x command-line utility. This application is found in w2x_install_dir/bin/. How to use it is explained in w2x-app - Online Help.

Figure 1 w2x-app window

???

The “Word To XML” add-on for XMLmind XML Editor

Graphical application w2x-app is also available as an add-on for XMLmind XML Editor. This add-on adds an "Import DOCX" item to the File menu. The "Import DOCX" menu item displays a non-modal dialog box almost identical to w2x-app. XML output files created using the "Import DOCX" dialog box are automatically opened in XMLmind XML Editor.

Installing the “Word To XML” add-on

This add-on is compatible with latest version of XMLmind XML Editor. In order to install it, please proceed as follows:

Start XMLmind XML Editor.

Select OptionsInstall Add-ons. This displays the “Install Add-ons” dialog box.

In the Install tab, click the checkbox found before the table row containing “Word To XML”.

???

Click OK to download and install the “Word To XML” add-on.

Restart XMLmind XML Editor as instructed.

Notice that the File menu has now an “Import DOCX” item.

???

The “Word To XML” servlet

The “Word To XML” servlet is a Java™ Servlet (server-side standard component) which has the same functions as the w2x-app desktop application.

Because it’s a server-side component and not a desktop application, please do not attempt to deploy the “Word To XML” servlet if you are an end-user of “Word To XML”. Please ask your IT personnel to do that for you.

Contents of the servlet software distribution

The “Word To XML” servlet comes in a software distribution of its own: w2x_servet-1_3_0.zip. This distribution contains a ready-to-deploy binary w2x.war, as well as the full Java™ source code of the servlet.

w2x.war

Ready-to-deploy Web application ARchive (WAR) containing the servlet.

src/

src/build.xml

The Java™ source code of the servlet. Run ant in src/ in order to use src/build.xml to rebuild w2x.war.

w2x/

Directory containing unpacked w2x.war. Needed to rebuild w2x.war.

lib/

Contains Java™ libraries needed to rebuild w2x.war.

Installing the servlet

File w2x.war may be easily installed in any servlet container implementing at least the Servlet 2.3 standard. Example of such servlet containers: Apache Tomcat, Jetty, Caucho Resin.

Though copying file w2x.war to the webapps/ folder of the servlet container and then restarting the servlet container is generally sufficient to deploy the “Word To XML” servlet, please refer to the documentation your servlet container to learn about the best deployment procedure.

On Windows, the .dll files contained in w2x_servlet_deployment_dir\WEB-INF\lib\ must be copied to a directory referenced by the PATH environment variable of the computer running the servlet.

Configuring the servlet

The “Word To XML” servlet is configured by specifying a number of init-param parameters. These parameters are found in WEB-INF/web.xml, where folder WEB-INF/ is contained in w2x.war.

All these init-param parameters are documented in web.xml. Example, parameter workDir:

<!-- workDir =============================================================

Uploaded files and files generated during the conversion process

are stored in temporary subdirectories of this directory.

If specified directory does not exist, it will be created.

Value: this directory and its contents must be readable and writable

by the operating system account used to run the Word To XML servlet.

Default: dynamic; supplied by the Servlet Container.

====================================================================== -->

<init-param>

<param-name>workDir</param-name><param-value></param-value>

</init-param>

Using the servlet to convert DOCX files

Let’s suppose your servlet container runs on host localhost and uses 8080 as its port. In order to use the “Word To XML” servlet, please point your Web browser to http://localhost:8080/w2x/. This will cause the browser to display a page containing a simple DOCX convert form.

Figure 2 The Convert DOCX form (servlet container running on host 192.168.1.202 and using port 8080)

???

In order to convert a DOCX file to another format:

Click “Choose File” to select the DOCX file to be converted.

Select the desired output format using the “Output format” combobox.

Click Convert to download a .zip (or .epub) archive containing the result of the conversion. Generating this .zip (or .epub) file may take several seconds to several minutes depending on the size of the DOCX input file.

If the name of the DOCX input file contains non-ASCII characters (e.g. accented characters), please make sure to use Zip extractor software supporting .zip files having UTF-8 encoded filenames.

Note that most Zip extractor software do not support .zip files having UTF-8 encoded filenames1. Such extractors will succeed in unpacking the .zip file, but will generate files having incorrect names.

Non interactive requests

It’s also possible to use the conversion services of the “Word To XML” servlet by sending URL /w2x/convert an HTTP POST request having a multipart/form-data encoding.

cURL2 example:

curl -s -S -o manual_docbook5.zip \

-F "docx=@manual.docx;type=application/vnd.openxmlformats-officedocument.wordprocessingml.document" \

-F "conv=docbook5" \

http://localhost:8080/w2x/convert

Other example:

curl -s -S -o manual.epub \

-F "docx=@manual.docx;type=application/vnd.openxmlformats-officedocument.wordprocessingml.document" \

-F "conv=epub" \

-F "params=-p epub.identifier urn:x-mlmind:w2x:manual -p epub.split-before-level 8" \

http://localhost:8080/w2x/convert

The conversion request has three emulated form fields:

docx

Emulated <input type=”file”> field. Required. Contains the DOCX input file.

conv

Emulated <input type=”text”> field. Required. Contains the name of one of the conversionN.name init-param defined in WEB-INF/web.xml.

The stock WEB-INF/web.xml defines the following conversions to styled HTML:

xhtml_css (single page styled HTML), frameset (multi-page styled HTML, split on Heading 1), frameset2 (multi-page styled HTML, split on Heading 1, 2), frameset3 (multi-page styled HTML, split on Heading 1, 2, 3), webhelp (split on Heading 1), webhelp2 (split on Heading 1, 2), webhelp3 (split on Heading 1, 2, 3), epub (split on Heading 1), epub2 (split on Heading 1, 2), epub3 (split on Heading 1, 2, 3)

and also the following conversions to “semantic” XML:

docbook, docbook5, topic, map, bookmap, xhtml_strict, xhtml_loose, xhtml1_1, xhtml5.

params

Emulated <input type=”text”> field. Optional. Contains some w2x command-line options, generally -p parameters. These options are appended to the options of the conversion specified in the conv emulated form field.

The response to a successful conversion request is a .zip (or .epub) archive containing the result of the conversion.

Getting started with w2x

About Evaluation Edition

Note that Evaluation Edition is useless for any purpose other than evaluating XMLmind Word To XML. This edition generates output containing random words replaced by string "[XMLmind]". (Of course, this does not happen with Professional Edition!)

???

We’ll use this manual to explain the basic uses of the w2x command-line utility. This manual is found in DOCX format in w2x_install_dir/doc/manual/ and the w2x command-line utility is found in w2x_install_dir/bin/.

C:\w2x-1_3_0> cd doc\manual

C:\w2x-1_3_0\doc\manual> mkdir out

Convert manual.docx to out\manual.xhtml, containing clean, styled, valid XHTML+CSS, looking very much like manual.docx:

..\..\bin\w2x manual.docx out\manual.xhtml

If you want to generate XHTML which is treated by Web browsers as if it were HTML, simply use a .html file extension for the output file:

..\..\bin\w2x manual.docx out\manual.html

Doing this automatically turn on options3 which remove the XML declaration (<?xml version=”1.0” encoding=”UTF-8”?>) normally found at the top of an XHTML file and insert a <meta content=”text/html; charset=UTF-8” http-equiv=”Content-Type”/> into the html/head element of the output document.

Convert manual.docx to out\frameset\manual.xhtml, containing multi-page, clean, styled, valid XHTML+CSS, looking very much like manual.docx:

..\..\bin\w2x –o frameset manual.docx out\frameset\manual.xhtml

The above command generates multiple “.xhtml” files in the out\frameset directory which is automatically created4 if needed to.

Note that out\frameset\manual.xhtml contains a frameset. While an obsolete HTML feature, a frameset makes it easy browsing the generated XHTML+CSS pages. Moreover the table of contents used as the left frame, found in out\frameset\manual-TOC.xhtml, is a convenient way to programmatically list all the generated XHTML+CSS pages.

Convert manual.docx to out\webhelp\manual.html, containing a Web Help looking very much like manual.docx:

..\..\bin\w2x –o webhelp manual.docx out\webhelp\manual.html

The above command generates multiple “.html” files in the out\webhelp directory which is automatically created if needed to.

Convert manual.docx to out\manual.epub, containing a EPUB 2 book looking very much like manual.docx:

..\..\bin\w2x –o epub manual.docx out\manual.epub

Convert manual.docx to out\manual.xml, containing DocBook 4.5.

..\..\bin\w2x –o docbook manual.docx out\manual.xml

Convert manual.docx to out\manual.xml, containing DocBook 5.0.

..\..\bin\w2x –o docbook5 manual.docx out\manual.xml

By default, the generated DocBook files contain HTML tables. If you prefer DocBook to contain CALS tables, please use the following options:

..\..\bin\w2x –o docbook5¬

-p convert.set-column-number yes -p transform.cals-tables yes¬

manual.docx out\manual.xml

Convert manual.docx to out\manual.dita, containing a DITA topic.

..\..\bin\w2x –o topic manual.docx out\manual.dita

Generating a task having “MyTask” as its ID is equally simple:

..\..\bin\w2x –o topic¬

-p transform.topic-type task -p transform.root-topic-id MyTask¬

manual.docx out\manual.dita

Convert manual.docx to out\manual.ditamap, containing a DITA map.

..\..\bin\w2x –o map manual.docx out\manual.ditamap

Convert manual.docx to out\manual.ditamap, containing a DITA bookmap possibly having chapter topicrefs and nested topicrefs acting as sections and subsections (but no sub-subsections).

..\..\bin\w2x –o bookmap -p transform2.section-depth 3¬

manual.docx out\manual.ditamap

Convert manual.docx to out\manual.xhtml, containing “semantic”, unstyled XHTML5.

..\..\bin\w2x –o xhtml5 manual.docx out\manual.xhtml

Use the following options to generate other versions of semantic XHTML:

Option

XHTML Version

-o xhtml_strict

XHTML 1.0 Strict

-o xhtml_loose

XHTML 1.0 Transitional

-o xhtml_1

XHTML 1.1

-o xhtml5

XHTML 5.0

How to generate useful multi-page HTML

In order to generate multi-page HTML, that is, frameset, Web Help, EPUB, we need to automatically split the source DOCX document into parts.

A new part is created each time a paragraph having an outline level less than or equal to specified split-before-level parameter is found in the source. An outline level is an integer between 0 (e.g. style “Heading 1”) and 8 (e.g. style “Heading 9”). The default value of parameter split-before-level is 0, which means: for each “Heading 1”, create a new page starting with this “Heading 1”.

Frameset example: for each “Heading 1” and “Heading 2”, create a new page (out/frameset/manual-1.xhtml, out/frameset/manual-2.xhtml, ..., out/frameset/manual-N.xhtml) starting with this “Heading 1” or “Heading 2”:

..\..\bin\w2x -p split.split-before-level 1¬

–o frameset manual.docx out\frameset\manual.xhtml

EPUB example:

..\..\bin\w2x -p epub.split-before-level 1¬

–o epub manual.docx out\manual.epub

Web Help containing “semantic” XHTML 5 example:

..\..\bin\w2x -p webhelp.split-before-level 1¬

–o webhelp5 manual.docx out\webhelp\manual.html

Important tip

Generating any of the multi-page, styled HTML formats should work great if, for the DOCX document to be converted, you can use MS-Word's "References > Table of Contents" button to automatically create a table of contents.

Note that the source DOCX document is not required to have a table of contents, but MS-Word should allow to automatically create a good one.

In other words, automatically creating a table of contents using MS-Word is the best way to check that your outline levels are OK.

Going further with w2x

When you execute the following command:

..\..\bin\w2x –o docbook5 manual.docx out\manual.xml

you execute in fact a sequence of 3 conversion steps:

Convert the DOCX file to a styled, valid, XHTML 1.0 Transitional document, looking very much like the input DOCX file.

Apply a number of XED scripts to this document to convert CSS styles into semantic tags. For example, numbered paragraphs are converted to proper ordered lists .

The entry point of these “semantic” XED scripts is found in w2x_install_dir/xed/main.xed.

The XED scripts edit in place the input XHTML document. Therefore, the result of this step is the same XHTML document, still valid, but this time, containing no CSS styles whatsoever.

Apply an XSLT 1.0 stylesheet to the unstyled, valid, XHTML 1.0 Transitional document in order to generate the desired semantic XML format.

The XSLT stylesheets are all found in w2x_install_dir/xslt/. In the above case, we want to generate DocBook v5, therefore we use w2x_install_dir/xslt/docbook5.xslt.

This sequence of conversion steps can be made visible in every detail by specifying the –vv option (very verbose) :

..\..\bin\w2x –vv –o docbook5 manual.docx out\manual.xml

VERBOSE: Converting "manual.docx" to XHTML...

DEBUG: convert.xhtml-file=C:\w2x-1_3_0\doc\manual\out\manual.xhtml

VERBOSE: Editing XHTML document using "C:\w2x-1_3_0\xed\main.xed"...

DEBUG: edit.xed-url-or-file=file:/C:/w2x-1_3_0/xed/main.xed

DEBUG: Loading script "file:/C:/w2x-1_3_0/xed/main.xed"...

DEBUG: Loading script "file:/C:/w2x-1_3_0/xed/after-translate.xed"...

[...]

DEBUG: Loading script "file:/C:/w2x-1_3_0/xed/before-save.xed"...

VERBOSE: Transforming document using "C:\w2x-1_3_0\xslt\docbook5.xslt" then saving it to "C:\w2x-1_3_0\doc\manual\out\manual.xml"...

DEBUG: transform.out-file=C:\w2x-1_3_0\doc\manual\out\manual.xml transform.xslt-url-or-file=file:/C:/w2x-1_3_0/xslt/docbook5.xslt

[...]

In fact, option –o docbook5 is a shorthand for the following w2x command-line options:

-c

Execute a Convert step called “convert”.

-p convert.xhtml-file C:\w2x-1_3_0\doc\manual\out\manual.xhtml

Pass the above xhtml-file parameter to the conversion step called “convert”.

-e

Execute an Edit step called “edit”.

-p edit.xed-url-or-file file:/C:/w2x-1_3_0/xed/main.xed

Pass the above xed-url-or-file parameter to the conversion step called “edit”.

-t

Execute a Transform step called “transform”.

-p transform.xslt-url-or-file file:/C:/w2x-1_3_0/xslt/docbook5.xslt

-p transform.out-file C:\w2x-1_3_0\doc\manual\out\manual.xml

Pass the above xslt-url-or-file and out-file parameters to the conversion step called “transform”.

If you need to learn about the details of the conversion steps to be executed, the simplest is to use the –liststeps command-line option.
Example: w2x –o docbook5 –liststeps.

The order of the –c, -e and –t options is significant because it means: first convert, then edit and finally transform. The order of the –p (and –pu) options is not important, as a parameter name must be prefixed by the name of the step to which it applies.

The Convert, Edit and Transform steps are the most important steps. There are other conversion steps though, which are all documented in chapter Conversion step reference. Moreover a Java™ programmer may implement its own custom conversion steps5 and instruct the w2x command-line to give them names (required to pass them parameters) and to execute them. See option –step.

A w2x processor executes a sequence of conversion steps whatever the output format. Simply the conversion steps, their order, number and parameters, depend on the desired output format. This is depicted in the figure below.

Figure 3 Anatomy of a w2x processor

???

The first sequence of in the above figure reads as follows: in order to convert a DOCX file to styled XHTML, first convert the DOCX file to a XHTML+CSS document, then “polish up” this document (e.g. process consecutive paragraphs having identical borders) using XED script w2x_install_dir/xed/main-styled.xed, and finally save the possibly modified XHTML+CSS document to disk.

Stock XED scripts

XMLmind Word to XML comes with two stock “main” XED scripts:

w2x_install_dir/xed/main-styled.xed

Invokes XED scripts used to “polish up” the styled XHTML 1.0 Transitional document created by the Convert step (e.g. process consecutive paragraphs having identical borders).

w2x_install_dir/xed/main.xed

Invokes XED scripts used to prepare the generation of semantic XML of all kinds: XHTML, DocBook, DITA. These scripts leverage the CSS styles and classes found in the styled XHTML 1.0 Transitional document created by the Convert step. They translate these CSS styles and classes (e.g. numbered paragraph) into semantic tags (e.g. ol/li).

Both the above “main” XED scripts are organized as sequences of simpler, short, XED scripts. Using –p or –pu options, these short scripts may be replaced or removed and may be passed parameters. It’s also possible to insert custom scripts before or after any of these short scripts.

Excerpts from w2x_install_dir/xed/main-styled.xed:

script(defined("before.init-styles", ""));

script(defined("do.init-styles", "init-styles.xed"));

script(defined("after.init-styles", ""));

script(defined("before.title-styled", ""));

script(defined("do.title-styled", "title-styled.xed"));

script(defined("after.title-styled", ""));

script(defined("before.remove-pis", ""));

script(defined("do.remove-pis", "remove-pis.xed"));

script(defined("after.remove-pis", ""));

script(defined("before.expand-tabs", ""));

script(defined("do.expand-tabs", "expand-tabs.xed"));

script(defined("after.expand-tabs", ""));

script(defined("before.borders", ""));

script(defined("do.borders", "borders.xed"));

script(defined("after.borders", ""));

script(defined("before.number-footnotes", ""));

script(defined("do.number-footnotes", "number-footnotes.xed"));

script(defined("after.number-footnotes", ""));

script(defined("before.finish-styles", ""));

script(defined("do.finish-styles", "finish-styles.xed"));

script(defined("after.finish-styles", ""));

Examples:

Remove script title-styled.xed:

-p do.title-styled “”

Replace script borders.xed by custom script “C:\Users\john\w2x tests\MyBorders.xed”:

-pu do.borders “C:\Users\john\w2 tests\MyBorders.xed”

Pass parameter finish-styles.css-uri to script finish-styles.xed:

-p edit.finish-styles.css-uri css/manual.css

By convention (this is not strictly required), the name of a parameter which applies to a given XED script is prefixed with the basename without any file extension of this script. Hence the full names of most parameters of Edit steps have the following syntax: step_name.script_name.parameter_name. Examples:

-p edit.prune.preserve “p-ProgramListing”

-p edit.inlines.convert “c-Code code”

Execute script customize\patch_manual.xed before script finish-styles.xed:

-pu edit.before.finish-styles customize\patch_manual.xed

Execute script customize\patch_manual.xed after script borders.xed:

-pu edit.after.borders customize\patch_manual.xed

Customizing the output of w2x

Customizing the XHTML+CSS files generated by w2x

Using a XED script to modify the styles embedded in the XHTML+CSS file

By default, w2x adds a number of CSS rules to the /html/head/style element of the generated XHTML+CSS file. Example: excerpts from w2x_install_dir/doc/manual/manual.html:

<style type="text/css">

body {

counter-reset: n-1-0 0 n-1-1 0 n-1-2 0 n-17-0 0 n-20-0 0;

font-family: Calibri;

font-size: 11pt;

}

...

</style>

A XED script allows to modify, not only the nodes of an XHTML document, but also its “CSS styles”. These “CSS styles” may be either style properties contained in the style attribute of an element or class names found in the class attribute of an element or the CSS rules of the document.

Therefore, when the desired customization is limited, suffice to execute a XED script in order to modify the XHTML+CSS document created by the Convert step. Example:

w2x -pu edit.before.finish-styles customize\patch_manual.xed¬

manual.docx out\manual.html

where w2x_install_dir/doc/manual/customize/patch_manual.xed contains:

set-rule(".p-ProgramListing", "white-space", "pre");

The above line adds CSS property “white-space: pre;” to the CSS rule having “.p-ProgramListing” as its selector. This CSS rule corresponds to custom paragraph6 style called “ProgramListing”.

Besides XED command set-rule, the following commands allow to edit the CSS styles contained in the XHTML+CSS document created by the Convert step: add-class, add-rule, remove-class, remove-rule, set-style.

Appending custom styles to the styles embedded in the XHTML+CSS file

XED script w2x_install_dir/xed/finish-styles.xed has a optional custom-styles-url-or-file parameter which makes it easy customizing the automatically generated CSS styles.

This parameter may be used to specify the location of a CSS file. The custom CSS styles found in specified file are simply appended to the automatically generated CSS styles. Example:

Example:

w2x -pu edit.finish-styles.custom-styles-url-or-file customize\custom.css¬

manual.docx out\manual_restyled.html

where customize\custom.css contains:

body {

font-family: sans-serif;

}

.p-Heading1,

.p-Heading2,

.p-Heading3,

.p-Heading4,

.p-Heading5,

.p-Heading6 {

font-family: serif;

color: #17365D;

padding: 1pt;

border-bottom: 1pt solid #4F81BD;

margin-bottom: 10pt;

margin-left: 0pt;

text-indent: 0pt;

}

.p-Heading1 {

border-bottom-width: 2pt;

}

...

.c-FootnoteReference,

.c-EndnoteReference {

font-size: smaller;

}

Using an external CSS file rather than embedded CSS styles

XED script w2x_install_dir/xed/finish-styles.xed has a optional css-uri parameter which allows to specify the CSS file where all CSS rules, whether automatically generated or custom, are to be saved.

Same example as above but using an external CSS file rather than embedded CSS styles:

w2x -p edit.finish-styles.css-uri manual_restyled_css/manual.css¬

-pu edit.finish-styles.custom-styles-url-or-file customize\custom.css¬

manual.docx out\manual_restyled.html

All the CSS styles, whether automatically generated or the custom ones found in customize\custom.css, end up in manual_restyled_css\manual.css. Moreover, out\manual_restyled.html contains a link to manual_restyled_css\manual.css.

<link href="manual_restyled_css/manual.css"

rel="stylesheet" type="text/css"/>

Combining all the above methods

It is of course possible to combine all the above methods. For example, the following w2x command is used to create w2x_install_dir/doc/manual/manual_restyled.html:

w2x -pu edit.before.finish-styles customize\patch_manual_restyled.xed¬

-p edit.finish-styles.css-uri manual_restyled_css/custom.css¬

-pu edit.finish-styles.custom-styles-url-or-file customize\custom.css¬

manual.docx out\manual_restyled.html

where w2x_install_dir/doc/manual/customize/patch_manual_restyled.xed contains:

for-each /html/body/p[get-class("^p-Heading\d$")] {

set-variable("class", get-class("^n-\d+-\d+$"));

if $class != '' {

set-variable("selector", concat(".", $class, ":after"));

if find-rule($selector) >= 0 {

remove-rule($selector);

set-variable("selector", concat(".", $class, ":before"));

set-rule($selector, "float");

set-rule($selector, "width");

set-rule($selector, "content",

concat(get-rule($selector, "content"), ' " "'));

set-rule($selector, "display", "inline");

}

}

}

The above XED script:

Delete CSS rules like this one:

.n-1-0:after {

clear: both;

content: "";

display: block;

}

Modify CSS rules like this one:

.n-1-0:before {

content: counter(n-1-0);

counter-increment: n-1-0;

float: left;

width: 21.6pt;

}

which becomes:

.n-1-0:before {

content: counter(n-1-0) " ";

counter-increment: n-1-0;

display: inline;

}

This script is useful because otherwise adding a bottom border to headings gives an ugly result. While the contents of the heading is “underlined”, the CSS float containing the numbering value of the heading is not.

Besides get-class, the following XPath extension functions may be used to access the CSS styles contained in the XHTML+CSS document created by the Convert step: find-rule, font-size, get-rule, get-style, lookup-length, lookup-style, style-count.

Why use XPath extension function get-class and not matches(@class,pattern)?

The answer is: because all class attributes have been removed by XED script w2x_install_dir/xed/init-styles.xed.

This script “interns” the CSS rules found in the html/head/style element of the XHTML+CSS document, the CSS styles directly set on some elements and the CSS classes set on some elements.

This operation is needed to allow an efficient implementation of the following XPath extension functions: find-rule, font-size, get-class, get-rule, get-style, lookup-length, lookup-style, style-count, and of the following editing commands: add-class, add-rule, remove-class, remove-rule, set-rule, set-style.

More information about “interned” CSS styles in command parse-styles (command invoked by w2x_install_dir/xed/init-styles.xed) and inverse command unparsed-styles (command invoked by w2x_install_dir/xed/finish-styles.xed).

Customizing the semantic XML files generated by w2x

Converting custom character styles to semantic tags

Converting a custom character style to an XHTML element (possibly having specific attributes) is simple and does not require writing a XED script. Suffice for that to pass parameter inlines.convert to the Edit step.

Example 1: convert text spans having a “Code” character style to XHTML element code:

-p edit.inlines.convert "c-Code code"

Notice that the name of character style in the generated XHTML+CSS file is always prefixed by “c-“.

The syntax for the value of parameter inlines.convert is:

valueconversion [ S ‘!’ S conversion ]*

conversionstyle_spec S XHTML_element_name [ S attribute ]*

style_specstyle_name | style_pattern

style_pattern → ‘/’ pattern ’/’ | ‘^’ pattern ‘$’

attributeattribute_name ‘=’ quoted_attribute_value

quoted_attribute_value → “’” value “’” | ‘”’ value ‘”’

Example 2: in addition to what’s done in above example 1, convert text spans having a “Abbrev” character style to XHTML element abbr having a title=”???” attribute:

-p edit.inlines.convert "c-Code code ! c-Abbrev abbr title='???'"

What if the semantic XHTML created by the Edit step is then converted to DITA or DocBook by the means of a Transform step?

In the case of XHTML elements code and abbr, there is nothing else to do because the stock XSLT stylesheets already support these elements:

w2x_install_dir/xslt/topic.xslt converts XHTML code to DITA codeph and XHTML abbr to DITA keyword,

w2x_install_dir/xslt/docbook.xslt converts XHTML code to DocBook code and XHTML abbr to DocBook abbrev.

The general case which also requires using custom XSLT stylesheets is explained in section The general case.

Converting custom paragraph styles to semantic tags

Converting a custom paragraph style to an XHTML element (possibly having specific attributes) is simple and does not require writing a XED script. Suffice for that to pass parameter blocks.convert to the Edit step.

Example 1.a: convert paragraphs having a “ProgramListing” paragraph style to XHTML element pre:

-p edit.blocks.convert "p-ProgramListing pre"

Notice that the name of paragraph style in the generated XHTML+CSS file is always prefixed by “p-“.

If you use the above blocks.convert specification, it will work fine, except that you’ll end up with several consecutive pre elements (one pre per line of program listing). This is clearly not what you want. You want consecutive pre elements to be merged into a single pre element. Fortunately implementing this too is quite simple.

Example 1.b: convert paragraphs having a “ProgramListing” paragraph style to XHTML element span (having grouping attributes; more about this below):

-p edit.blocks.convert "p-ProgramListing span g:id='pre' g:container='pre'"

When any of the target XHTML elements have grouping attributes (g:id='pre'7, g:container='pre', in the above example), then w2x_install_dir/xed/blocks.xed automatically invokes the group() command at the end of the conversions. This has the effect of grouping consecutive <span g:id='pre' g:container='pre'> into a common pre parent element.

Given the fact that XED command group() automatically removes grouping attributes when done and that w2x_install_dir/xed/finish.xed discards all useless span elements, this leaves us with clean pre elements containing text8.

The syntax for the value of parameter blocks.convert is:

valueconversion [ S ‘!’ S conversion ]*

conversionstyle_spec S XHTML_element_name [ S attribute ]*

style_specstyle_name | style_pattern

style_pattern → ‘/’ pattern ’/’ | ‘^’ pattern ‘$’

attributeattribute_name ‘=’ quoted_attribute_value

quoted_attribute_value → “’” value “’” | ‘”’ value ‘”’

Example 3: in addition to what’s done in above example 1.b, convert paragraphs having a “Term” paragraph style to XHTML element dt, convert paragraphs having a “Definition” paragraph style to XHTML element dl and group consecutive dt and dl elements into a common dl parent:

-p edit.blocks.convert "p-Term dt g:id='dl' g:container='dl' !¬

p-Definition dd g:id='dl' g:container='dl' !¬

p-ProgramListing span g:id='pre' g:container='pre'"

What if the semantic XHTML created by the Edit step is then converted to DITA or DocBook by the means of a Transform step?

In the case of XHTML elements pre, dt, dd and dl, there is nothing else to do because the stock XSLT stylesheets already support these elements.

The general case which also requires using custom XSLT stylesheets is explained in section The general case.

The general case

In the general case, customizing the semantic XML files generated by w2x requires writing both a XED script and an XSLT stylesheet.

For example, let’s suppose we want to group all the paragraphs having a “Note” paragraph style and to generate for such groups DocBook and DITA note elements.

The following blocks.convert parameter would allow to very easily create the desired groups:

-p edit.blocks.convert "p-Note p g:id='note_group_member'¬

g:container='div class=\”role-note\” ’"

However this would leave us with two unsolved problems:

A paragraph having a “Note” paragraph style often starts with bold text “Note:”. We want to eliminate this redundant label.

The stock XSLT stylesheets will not convert XHTML element <div class=”role-note”> to a DocBook or DITA note element.

A custom XED script

The first problem is solved by the following w2x_install_dir/doc/manual/customize/notes.xed script:

namespace "http://www.w3.org/1999/xhtml";

namespace html = "http://www.w3.org/1999/xhtml";

namespace g = "urn:x-mlmind:namespace:group";

for-each /html/body//p[get-class("p-Note")] {

delete-text("note:\s*", "i");

if content-type() <= 1 and not(@id) {

delete();

} else {

remove-class("p-Note");

set-attribute("g:id", "note_group_member");

set-attribute("g:container", "div class='role-note'");

}

}

group();

The “Note:” label, if any, is deleted using XED command delete-text. If doing this creates a useless empty (content-type() <= 1) paragraph, then delete this paragraph using XED command delete.

The above script is executed after stock script w2x_install_dir/xed/blocks.xed by the means of the following w2x command-line option:

-pu edit.after.blocks customize\notes.xed

A custom XSLT stylesheet

The second problem is solved by the following w2x_install_dir/doc/manual/customize/custom_topic.xslt XSLT 1.0 stylesheet:

<xsl:stylesheet version="1.0"

xmlns:xsl="http://www.w3.org/1999/XSL/Transform"

xmlns:h="http://www.w3.org/1999/xhtml"

exclude-result-prefixes="h">

<xsl:import href="w2x:xslt/topic.xslt"/>

<xsl:template match="h:div[@class = 'role-note']">

<note>

<xsl:call-template name="processCommonAttributes"/>

<xsl:apply-templates/>

</note>

</xsl:template>

...

</xsl:stylesheet>

This stylesheet, which imports stock w2x_install_dir/xslt/topic.xslt, is used for the topic, map and bookmap output formats (see –o option). Similar, very simple, stylesheets have been developed for the docbook and docbook5 output formats.

Note: Something like “w2x:xslt/topic.xslt” is an absolute URL supported by w2x. “w2x:” is an URL prefix (defined in the automatic XML catalog used by w2x) which specifies the location of the parent directory of both the xed/ and xslt/ subdirectories.

The above stylesheet replaces the stock one by the means of the following w2x command-line option:

-o topic -t customize\custom_topic.xslt

Do not forget to specify the –t option after the –o option, because it’s the –o option which implicitly invokes stock w2x_install_dir/xslt/topic.xslt (this has been explained in chapter Going further with w2x) and we want to use –t to override the use of the stock XSLT stylesheet.

Tip: You’ll find a template for custom XED scripts and several templates for custom XSLT stylesheets in w2x_install_dir/doc/manual/templates/.

For example, in order to create w2x_install_dir/doc/manual/customize/custom_topic.xslt, we started by copying template XSLT stylesheet w2x_install_dir/doc/manual/templates/template_topic.xslt.

Generating XML conforming to a custom schema

In order to use w2x to convert a DOCX input file to an XML output file conforming to your custom schema, all you have to do is write a custom XSLT 1.0 stylesheet converting the “semantic” XHTML 1.0 Transitional generated by the Edit step to your custom schema.

Let’s call your custom XSLT 1.0 stylesheet “C:\Users\John\foo\xsl\xhtml_to_foo.xsl”. Command-line tool w2x must then be passed the following options:

-c

Execute a Convert step called “convert”.

-e

Execute an Edit step called “edit”.

-pu edit.xed-url-or-file w2x:xed/main.xed

Pass this stock XED script (converting the styled XHTML  1.0 Transitional created by the Convert step to “semantic” XHTML) to the conversion step called “edit”.

-t

Execute a Transform step called “transform”.

-pu transform.xslt-url-or-file "C:\Users\John\foo\xsl\xhtml_to_foo.xsl"

Pass your custom XSLT 1.0 stylesheet to the conversion step called “transform”.

Stock XED script w2x:xed/main.xed creates a number of semantic XHTML elements having a class attribute starting with “role-“. Examples: <div class=”role-section1”>, <div class=”role-section2”>, <div class=”role-figure”>, <div class=”role-figcaption”>, <a class=”role-footnote-ref”>, <div class=”role-footnote”>, <a class=”role-xref”>, <span class=”role-index-term”>, etc. To learn how to process these elements, the simplest is to look how this is done in a stock XSLT stylesheet such as “w2x_install_dir/xslt/topic.xslt” or “w2x_install_dir/xslt/docbook.xslt”.

The w2x command-line utility

If the .dmg distribution has been used to install XMLmind Word To XML on the Mac, the w2x command-line utility is found in WordToXML.app/Contents/Resources/w2x/bin/.

Usage: w2x [-version] [-v|-vv] [Options] [-liststeps] in_docx_file out_file

Options are:

-o format

This option automatically adds all the steps needed to convert input DOCX file to an output file having specified format.

Possible formats: docbook, docbook5, topic, map, bookmap, xhtml_css (single-page styled HTML, that is, single-page XHTML+CSS), xhtml_strict, xhtml_loose, xhtml1_1, xhtml5, frameset (multi-page styled HTML), frameset_strict (multi-page XHTML 1.0 Strict), frameset_loose (multi-page XHTML 1.0 Transitional), frameset1_1 (multi-page XHTML 1.1), frameset5 (multi-page XHTML 5.0), webhelp (Web Help containing styled HTML), webhelp_strict (Web Help containing XHTML 1.0 Strict), webhelp_loose (Web Help containing XHTML 1.0 Transitional), webhelp1_1 (Web Help containing XHTML 1.1), webhelp5 (Web Help containing XHTML 5.0), epub (EPUB 2 containing styled XHTML 1.1), epub1_1 (EPUB 2 containing semantic XHTML 1.1).

The default output format is: xhtml_css (single-page styled HTML, that is, single-page XHTML+CSS).

-p name value

Set parameter name to value.

Use parameter step_name.param_name to parametrize the step called step_name.

Because they are used to parameterize named steps, the order of –p and –pu options relatively to options specifying conversions steps (-c, -e, -t, -step, etc) is not significant. For example: “-p convert.charset UTF-8 -c” is equivalent to “-c -p convert.charset UTF-8”.

-pu name URL_or_file

Same as -p, except that parameter value URL_or_file is first converted to an URL.

URL_or_file is an absolute or relative URL (relative to current -f options file if any, to current working directory otherwise) or the filename of an existing file or directory.

-c

Add or replace “convert” step. This step converts input DOCX file to an in-memory XHTML +CSS document.

-l

Add or replace “load” step. This step, mainly used to test XED scripts, loads input XML file.

-e xed_URL_or_file

Add or replace “edit” step. This step edits in place input XHTML document using XED script xed_URL_or_file.

-e2 xed_URL_or_file

Add or replace “edit2” step. This step edits in place input XHTML document using XED script xed_URL_or_file.

-t xslt_URL_or_file

Add or replace “transform” step. This step transforms input XML document or file using XSLT stylesheet xslt_URL_or_file.

The output file is specified by parameter transform.out-file.

-t2 xslt_URL_or_file

Add or replace “transform2” step. This step transforms input XML document or file using XSLT stylesheet xslt_URL_or_file.

The output file is specified by parameter transform2.out-file.

-s

Add or replace “save” step. This step saves to disk input XHTML document.

The output file is specified by parameter save.out-file.

-step:java_class_name:step_name

Add or replace step called step_name by an instance of Java™ class java_class_name deriving from com.xmlmind.w2x.processor.ProcessStep.

-f options_URL_or_file

Load one or more of the above options from options_URL_or_file, a plain UTF-8 text file

-v

-vv

-vvv

Verbose. More Vs means more verbose.

-version

Print version number and exit.

-liststeps

List the conversion steps to be executed and exit. This option is useful to determine how to customize the conversion steps. Example:

$ w2x -o bookmap -liststeps

-step:com.xmlmind.w2x.processor.ConvertStep:convert

-p convert.create-mathml-object no

-p convert.set-column-number yes

-step:com.xmlmind.w2x.processor.EditStep:edit

-p edit.xed-url-or-file file:/opt/w2x/xed/main.xed

-step:com.xmlmind.w2x.processor.TransformStep:transform

-p transform.out-file %{~pnO}.dita

-p transform.single-topic no

-p transform.xslt-url-or-file file:/opt/w2x/xslt/topic.xslt

-step:com.xmlmind.w2x.processor.TransformStep:transform2

-p transform2.xslt-url-or-file file:/opt/w2x/xslt/bookmap.xslt

-p transform2.topic-type %{transform.topic-type}

-p transform2.output-path %{~po}

-step:com.xmlmind.w2x.processor.DeleteFilesStep:cleanUp

-p cleanUp.files %{~pnO}.dita

Variables substituted in the parameter values passed to the –p and –pu options

The following variables are substituted in the parameter values passed to the –p and –pu options.

Variable

Description

Example

%{I}

Full path of the input DOCX file.

C:\My Docs\report.docx

%{O}

Full path of the output XML file.

C:\My Docs\out\report.xml

%{i}

Absolute URL of the input DOCX file.

file:/C:/My%20Docs/report.docx

%{o}

Absolute URL of the output XML file.

file:/C:/My%20Docs/out/report.xml

Variables %{I}, %{O}, %{i} and %{o} may all contain one or more of following modifiers. First modifier must be preceded by character “~”.

Modifier

Description

n

The name of the file or URL without any extension.

x

The extension of the file or URL. Starts with “.”.

p

The full path of the parent directory of the file or URL.

Note that combinations of modifiers other than “~nx”, “~pn”, “~pnx” do not make sense and that, for example, %{~pnxI} is equivalent to %{I}.

Examples: let’s suppose that command-line argument in_docx_file (see above) is “C:\My Docs\report.docx” and that argument out_file is “C:\My Docs\out\report.xml”.

%{~nI} is replaced by “report”.

%{~xI} is replaced by “.docx”.

%{~pI} is replaced by “C:\My Docs”.

%{~nxo} is replaced by “report.xml”.

%{~pno} is replaced by “file:/C:/My%20Docs/out/report”.

Default conversion steps

If none of the options creating a step (-l, -c, -e, -e2, -t, -t2, -s, -step) have been specified, w2x automatically adds the equivalent of –o xhtml_css, which consists in the following conversion steps:

-c

-e

-p edit.xed-url-or-file w2x:xed/main-styled.xed

-s

The above options convert the input DOCX file to clean, styled, valid XHTML. The resulting output file is not indented.

Note: Something like “w2x:xed/main-styled.xed” is an absolute URL supported by w2x. “w2x:” is an URL prefix (defined in the automatic XML catalog used by w2x) which specifies the location of the parent directory of both the xed/ and xslt/ subdirectories.

Automatic conversion step parameters

If the first conversion step is a Convert step, the following parameters are automatically added by w2x (unless, of course, they have already been specified by the user):

If out_file extension starts with “htm” or “shtm”,

-p step_name.charset UTF-8

The charset parameter allows to get Web browsers consider the generated document as being HTML, and not XHTML.

-pu step_name.xhtml-file out_file_with_an_xhtml_extension

If the last conversion step is a Save step, Transform step, Split step, Web Help step or EPUB step the following parameters are automatically added by w2x (unless, of course, they have already been specified by the user):

-pu step_name.out-file out_file

Conversion step reference

Convert step

Convert input DOCX file to a styled, valid, XHTML 1.0 Transitional document. The result of this step is this XHTML document.

For clarity, the “convert.” parameter name prefix is omitted here.

However when you’ll pass any of the following parameters to w2x, please do not forget this prefix. Example: -p convert.resource-directory images.

Parameters:

Name

Value

Description

charset

A valid character encoding (e.g. UTF-8, Windows-1252).

Default: no charset, add an XML declaration.

When a charset is specified, a meta element is added to the head element of the generated document:

<meta charset=”charset”/> if parameter version is “5.0”,

<meta content=”text/html; charset=charset” http-equiv=”Content-Type” /> otherwise.

If the specified charset is “UTF-8”, then the XML declaration (<?xml version=”1.0” encoding=”UTF-8”?>) is not to added to the generated document. This allows to get Web browsers consider the generated document as being HTML, and not XHTML.

converted-image-extensions

A list of image file extensions separated by space characters.

Default: “svg png jpeg”.

When the input DOCX file contains an image not having any of the file extensions specified in the converted-image-extensions list, attempt to convert this image to one of the formats of this list.

Each format is considered in turn, that’s why w2x will attempt to convert a WMF image to SVG first, before considering PNG and JPEG.

create-mathml-object

yes” | “no” | “auto

Default: “auto”.

When converting MS-Word math (that is, OpenXML math) to MathML:

yes

Generate an external file containing the converted MathML element and insert an object element pointing to the generated “.mml” file. Example: <object data="doc_files/math-010.mml" type="application/mathml+xml"/>.

no

Embed the converted MathML element in the XHTML document created by this step.

auto

Embed the converted MathML element in the XHTML document but only if parameter version is set to 5.09.

default-lang

A valid language code (e.g. en, fr-CA).

No default.

if parameter set-lang is not specified and if the main language of the document cannot determined by examining the contents of the input DOCX file, set the lang attribute of the html element to this value.

lower-case-resource-names

A boolean: true (same as: yes | on | 1) | false (same as: no | off | 0).

Default: false.

Not for general use. Specifying this parameter as true is needed to keep quiet epubcheck on platforms where filenames are case-sensitive (e.g. Linux).

resource-directory

A file path.

Default: if parameter xhtml-file is specified, basename of xhtml-file, without an extension, but followed by “_files”; otherwise the absolute path of an automatically created temporary directory.

Specifies the file path of the directory which is to contain copies of the images referenced in the input DOCX file.

A relative file path is relative to the value of parameter xhtml-file.

Note that, if it already exists, a resource directory specified this way is not automatically made empty by w2x before being used to store resources. Only the “automatic”, default, output_file_basename_files/ folder is automatically made empty by w2x (if this “automatic” folder already exists).

resource-prefix

A non-empty string not containing the file separator character (“/” or “\”).

Default: none, no prefix.

Specifies a prefix to be prepended to the names of resource files created by w2x.

This prefix is useful when used in conjunction with parameter resource-directory and when several files generated by w2x share the same resource directory.

set-column-number

A boolean: true (same as: yes | on | 1) | false (same as: no | off | 0).

Default: false.

If specified as true, insert in each table cell a column-number processing-instruction containing the column number of this cell. First column is column #1.

Example:

<?column-number 1?>

This processing-instruction greatly helps in generating CALS tables (DocBook, DITA) containing cells spanning several columns.

set-lang

A valid language code (e.g. en, fr-CA).

No default: set the lang attribute of the html element after examining the contents of the input DOCX file.

if specified, set the lang attribute of the html element to this value.

version

1.0_transitional (same as: 1.0_loose | 1) | 1.0_strict | 1.1 | 5.0 (same as: 5) | “”.

Default: 1.0_transitional.

Specifies which XHTML version to generate, hence which <!DOCTYPE> to add to generated XHTML document.

Note that XHTML 5.0 has no DTD, hence no <!DOCTYPE> for this version.

The empty string “” means: generate XHTML 1.0 Transitional , but do not add a <!DOCTYPE>.

xhtml-file

A file path.

No default .

If the generated XHTML document was saved to disk, this would be the path of its save file.

When specified (which is strongly recommended), this file path is used to give a base URL to the generated XHTML document.

Delete files step

Delete files or directories having specified path or matching specified glob pattern. The input of this step is ignored. The result of this step is thus equal to its input.

This step is used for example when generating a DITA map or bookmap. It is used to delete the intermediate topic file created by the first Transform step.

Parameters (for clarity, the “cleanUp.” parameter name prefix is omitted here):

Name

Value

Description

files

A file path or glob pattern.

No default (required).

Specifies which files or directories are to be deleted. A relative file path or glob pattern is relative to the current working directory.

Edit step

Edit in place input XHTML document using a XED script. The result of this step is the same XHTML document, but modified by the script.

For clarity, the “edit.” parameter name prefix is omitted here.

However when you’ll pass any of the following parameters to w2x, please do not forget this prefix. Example: -p edit.ids.generate-section-ids yes.

Parameters:

Name

Value

Description

xed-url-or-file

An absolute URL or the path of an existing file.

No default (required).

Specifies which XED script should be used to edit the input XHTML document. A relative file path is relative to the current working directory.

Any other parameter is passed to the XED script as a XED global variable.

XMLmind Word to XML (w2x for short) comes with two stock “main” XED scripts:

w2x:xed/main-styled.xed

Invokes XED scripts used to “polish up” the styled XHTML 1.0 Transitional document created by the Convert step (e.g. process consecutive paragraphs having identical borders).

w2x:xed/main.xed

Invokes XED scripts used to prepare the generation of semantic XML of all kinds: XHTML, DocBook, DITA. These scripts leverage the CSS styles and classes found in the styled XHTML 1.0 Transitional document created by the Convert step. They translate these CSS styles and classes (e.g. numbered paragraph) into semantic tags (e.g. ol/li).

Note: Something like “w2x:xed/main.xed” is an absolute URL supported by w2x. “w2x:” is an URL prefix (defined in the automatic XML catalog used by w2x) which specifies the location of the parent directory of both the xed/ and xslt/ subdirectories.

Table 1 Parameters common to w2x:xed/main-styled.xed and w2x:xed/main.xed

Name

Value

Description

finish-styles.css-uri

An absolute or relative “file:” URI.

Default: “”. “Interned” CSS styles, if any, are stored in a head/style element.

Global variable defined in w2x:xed/finish-styles.xed.

Store “interned” CSS styles, if any, in the CSS (UTF-8 encoded) file having this URI. A relative URI is relative to the URI specified by parameter xhtml-file.

More information about “interned” CSS styles in command parse-styles (command invoked by w2x:xed/init-styles.xed) and inverse command unparsed-styles (command invoked by w2x:xed/finish-styles.xed).

finish-styles. custom-styles-url-or-file

An absolute URL or a filename. A relative filename is relative to the current working directory.

Default: “” (no custom styles).

Global variable defined in w2x:xed/finish-styles.xed.

Specifies the location of a CSS file. The custom CSS styles found in specified file are simply appended to the automatically generated CSS styles.

Using this variable is the easiest way to customize the automatically generated CSS styles.

When generating multi-page styled or semantic XHTML of any kind (frameset, Web Help, EPUB)

Please use finish-styles. custom-styles-url-or-file to specify custom CSS styles.

No need to specify finish-styles.css-uri as all the CSS styles are anyway stored into an external “.css” file having the same basename as the main output file.

finish-styles.mathjax

yes” | “no” | “auto

Default: “no”.

Global variable defined in w2x:xed/finish-styles.xed.

Very few web browsers (Firefox) can natively render MathML. Fortunately, there is MathJax.

MathJax is a JavaScript display engine for mathematics that works in all browsers.

yes

Add a <script> element loading MathJax to the <html>/<head> element of the generated XHTML file.

auto

Same as “yes”, but add <script> only when the generated XHTML file contains MathML.

finish-styles.mathjax-url

String.

Default value: the URL pointing to the MathJax CDN, as recommended in the MathJax documentation.

Global variable defined in w2x:xed/finish-styles.xed.

The URL allowing to load the MathJax engine configured for rendering MathML.

Ignored unless parameter mathjax is set to “yes” or “auto”.

title.title-style-names

List of user-defined style names separated by space characters.

Default: “” (empty list).

Global variable defined in w2x:xed/title.xed.

Specifies which user-defined paragraph styles should be considered to be equivalent to standard style “p-Title”.

(Paragraph styles, whether user-defined or standard, are given a “p-“ prefix by the Convert step.)

title.subtitle-style-names

List of user-defined style names separated by space characters.

Default: “” (empty list).

Global variable defined in w2x:xed/title.xed.

Specifies which user-defined paragraph styles should be considered to be equivalent to standard style “p-Subtitle”.

(Paragraph styles, whether user-defined or standard, are given a “p-“ prefix by the Convert step.)

Table 2 Parameters which are specific to w2x:xed/main-styled.xed

Name

Value

Description

remove-pis.except

One or more processing-instructions targets separated by space characters.

Default: “” (remove all processing-instructions)

Global variable defined in w2x:xed/remove-pis.xed.

Specifies which processing-instructions should be kept in the styled HTML document.

By default, all processing-instructions are removed from the styled HTML document. Such processing-instructions are useful only when the styled HTML document created by the Convert step is used as an intermediate format in order to generate semantic XML.

Table 3 Parameters which are specific to w2x:xed/main.xed

Name

Value

Description

before-save.allow-flow

yes” | “no”.

Default: “no”.

Global variable defined in w2x:xed/before-save.xed.

If “yes”, allow flow elements (e.g. li) to directly contain text and inline elements.

If “no”, do not allow flow elements (e.g. li) to directly contain text and inline elements. Instead “wrap” these text and and inline elements in <p class=”role-inline-wrapper”> elements.

The “no” option greatly eases the generation of certain types of semantic XML (e.g. DocBook) during the Transform step.

biblio.style-names

List of user-defined style names separated by space characters.

Default: “” (empty list).

Global variable defined in w2x:xed/biblio.xed.

Specifies which user-defined paragraph styles should be considered to be equivalent to standard style “p-Bibliography”.

(Paragraph styles, whether user-defined or standard, are given a “p-“ prefix by the Convert step.)

blocks.convert

A conversion specification.

Default: “”. No conversions other than those performed by w2x:xed/blocks.xed.

Global variable defined in w2x:xed/blocks.xed.

Specified paragraph styles are converted to specified XHTML elements. See below.

blocks.convert-to-pre

A conversion specification.

Default: “”.

Global variable defined in w2x:xed/blocks.xed.

Specified paragraph styles are converted to specified XHTML elements. See below.

When using MS-Word, there two ways to represent code samples:

Use a sequence of paragraphs having the same style. Each paragraph contains one line of the code sample. Let’s call the style of these paragraphs Code1.

Use a single paragraph containing the whole code sample, which means that this single paragraph contains significant whitespace and line breaks. Let’s call the style of this paragraph Code2.

A sequence of Code1 paragraphs may be converted to an XHTML pre using:

–p edit.blocks.convert "p-Code1 span g:id='pre' g:container='pre'"

A Code2 paragraph may be converted to an XHTML pre using:

–p edit.blocks.convert-to-pre "p-Code2 pre"

captions.style-names

List of user-defined style names separated by space characters.

Default: “” (empty list).

Global variable defined in w2x:xed/captions.xed.

Specifies which user-defined paragraph styles should be considered to be equivalent to standard style “p-Caption”.

(Paragraph styles, whether user-defined or standard, are given a “p-“ prefix by the Convert step.)

convert-tabs.to-table

yes” | “no”.

Default: “no”.

Global variable defined in w2x:xed/convert-tabs.xed.

If set to “yes”, convert consecutive paragraphs containing text runs aligned on tab stops to a borderless table.

This option is turned off by default because, in the general case, it's not possible to emulate tab stops using tables.

convert-tabs.unwrap-paragraphs

yes” | “no”.

Default: “yes”.

Global variable defined in w2x:xed/convert-tabs.xed.

If set to “yes”, the cells contained in the borderless table used to emulate tab stops directly contain text runs rather than paragraphs.

headings.convert

A conversion specification.

Default: “”. No conversions other than those performed by w2x:xed/headings.xed.

Global variable defined in w2x:xed/headings.xed.

Specified paragraph styles are converted to specified XHTML heading elements (h1, h2, …, h6). See below.

Note that by default, script headings.xed automatically converts paragraphs having an outline level to h1, h2, …, h6 headings.

ids.generate-section-ids

yes” | “no”.

Default: “no”.

Ensure that all the sections found in the semantic XHTML resulting from the conversion of a DOCX file have a unique ID.

When this ID is missing, it is computed using the content of the h1, h2, ..., h6 heading which is the first child of the section. Example:

<div class="role-section2"

id="Title_of_this_section">

<h2>Title of this section</h2>

...

Setting ids.generate-section-ids to yes is especially useful when converting a DOCX file to a DITA map or bookmap. With this parameter, the filenames of the topics referenced by the generated map are guaranteed to have meaningful values (e.g. "Introduction.dita" rather than "d0e35.dita").

ids.section-id-max-length

An integer greater or equal to 1.

Default: 32.

Specifies the maximum length of the automatically computed ID when parameter ids.generate-section-ids is set to yes.

inlines.convert

A conversion specification.

Default: “”. No conversions other than those performed by w2x:xed/inlines.xed.

Global variable defined in w2x:xed/inlines.xed.

Specified character styles are converted to specified XHTML elements . See below.

inlines.generate-big-small

yes” | “no”.

Default: “yes”.

Global variable defined in w2x:xed/inlines.xed.

Specifies whether spans having a bigger (respectively smaller) font size than their parent elements should be converted to big (respectively small) elements.

prune.preserve

List of user-defined style names separated by space characters.

Default: “” (empty list).

Global variable defined in w2x:xed/prune.xed.

Empty paragraphs having a user-defined style found in this list will not be deleted by w2x:xed/prune.xed.

remove-styles.preserved-classes

List of user-defined style names separated by space characters.

Default: “” (empty list).

Global variable defined in w2x:xed/remove-styles.xed.

The CSS classes used to apply the user-defined styles specified in this list will not be removed by w2x:xed/removes-styles.xed.

Note that specifying both parameters prune.preserve and remove-styles.preserved-classes is currently the only way to keep in the generated semantic XHTML empty paragraphs having a given MS-Word style. For example, specifying -p prune.preserve p-PlaceHolder and -p remove-styles.preserved-classes p-PlaceHolder may be used to keep in the semantic XHTML output all empty paragraphs having the p-PlaceHolder style.

sections.max-level

An integer greater or equal to 1.

Default: -1. No maximum level.

Global variable defined in w2x:xed/sections.xed.

Wrap sequences of elements starting with a hN element (that is, h1, h2, h3, h4, h5, h6) into <div class=”role-sectionN> elements.

This parameter specifies the maximum level of nesting for such sections.

Simple conversion specifications

Above parameter blocks.convert (respectively inlines.convert) provides the user of w2x with a simple mean to convert p (respectively span) elements having certain paragraph (respectively character) styles to XHTML elements possibly having attributes.

The syntax of a simple conversion specification is:

specsimple_spec [ S ‘!’ S simple_spec ]*

simple_specstyle_spec S XHTML_element_qname [ S attribute_spec ]*

style_specstyle_name | style_pattern

style_pattern → ‘/’ pattern ’/’ | ‘^’ pattern ‘$’

attribute_specattribute_qname ‘=’ quoted_attribute_value

quoted_attribute_value → “’” value “’” | ‘”’ value ‘”’

Note that when specifying a XHTML_element_qname, you must restrict yourself to XHTML 1.0 Transitional elements. Specifying for example, XHTML 5.0 elements such as mark, aside, section, etc, will not give you the results you’ll expect.

Examples: stock styled span conversions used by w2x:xed/inlines.xed:

/Emphasis$/ em !

c-Strong strong !

c-BookTitle cite !

/((IntenseReference)|(SubtleReference)|(QuoteChar))$/ em !

/((itleChar)|(Heading\d+Char))$/ strong

Custom styled span conversions used to process this manual:

c-Code code

Stock styled paragraph conversions used by w2x:xed/blocks.xed:

/Quote$/ p g:id='blockquote' g:container='blockquote'

Custom styled paragraph conversions used to process this manual:

p-Term dt g:id="dl" g:container="dl" !

p-Definition dd g:id="dl" g:container="dl" !

p-ProgramListing span g:id="pre" g:container="pre"

Automatic grouping of the XHTML elements which are the results of the styled paragraph conversions

In the above examples, attributes having names prefixed with “g:” are in the “urn:x-mlmind:namespace:group” namespace. These attributes are called grouping attributes. Examples: g:id, g:container.

When parameter blocks.convert is used to create XHTML elements having grouping attributes, command group() is automatically invoked at the end of all the styled paragraph conversions. To make it simple, this command groups consecutive XHTML elements having the same g:id attribute into a common parent element. The parent element is specified by attribute g:container.

In the above examples,

Consecutive p elements having grouping attributes g:id='blockquote' and g:container='blockquote' are grouped into a common blockquote parent element.

Consecutive dt and dt elements having grouping attributes g:id="dl" and g:container="dl are grouped into a common dl parent element.

Consecutive span elements having grouping attributes g:id="pre" and g:container="pre" are grouped into a common pre parent element.

EPUB step

Splits input XHTML document, whether styled or semantic, into several pages and packages these pages as an EPUB 2 book. The result of the this step is the file containing the EPUB book.

No tab expansion for EPUB 2

By default, when generating styled HTML (that is, XHTML+CSS), some JavaScript™ code (w2x_install_dir/xed/expand-tabs.js) is added to the output file. This code computes and gives a width to all <span class=”role-tab> </span>. This allows to decently emulate tab stops in any modern Web browser. More information in About tab stops.

However, this cannot work in the case of the EPUB 2 output format because scripting is disabled in the styled HTML pages comprising an EPUB book.

Same parameters as the Split step, plus the following EPUB specific parameters (for clarity, the “epub.” parameter name prefix is omitted here):

Name

Value

Description

cover-image-url-or-file

An absolute URL or a filename. A relative filename is relative to the current working directory.

Default: none (no cover page).

Specifies an image file which is to be used as the cover page of the EPUB book. This image must be a PNG or JPEG image. Its size must not exceed 1000x1000 pixels.

default-lang

A language code conforming RFC 3066. Examples: de, fr-CA. Default value: en.

Main language of the EPUB book. This parameter is used only when this language cannot be determined by examining the input styled XHTML document.

identifier

String.

Default: dynamically generated UUID URN.

A globally unique identifier for the generated EPUB book (typically the permanent URL of the EPUB book).

out-file

A file path.

No default (required).

Specifies the path of the EPUB book. A relative file path is relative to the current working directory.

Load step

Loads an input XML file. The result of this step is loaded XML document.

This step is mainly useful to test XED scripts. Example:

w2x –l –e my_script.xed –s in.xhtml out.xhtml

Note that if loaded file starts with a <!DOCTYPE> pointing to a DTD, then a document loader created by this step will not attempt to load this DTD. The document loader will behave as if the <!DOCTYPE> was absent.

No parameters.

Save step

Saves input XHTML document to disk. The result of the this step is the save file.

Parameters (for clarity, the “save.” parameter name prefix is omitted here):

Name

Value

Description

encoding

A valid character encoding (e.g. UTF-8, Windows-1252).

Default: “UTF-8”.

Specifies the character encoding of the save file.

indent

A boolean: true (same as: yes | on | 1) | false (same as: no | off | 0).

Default: false.

Specifies whether the save file should be indented.

Note:

Do not specify indent=”true” in production.

The XML indentation created this way being very simple, this may add whitespace inside elements where space characters are significant.

out-file

A file path.

No default (required).

Specifies the path of the save file. A relative file path is relative to the current working directory.

Split step

Splits input XHTML document, whether styled or semantic, into several pages and saves these pages to disk.

This step also generates a frameset and a table of contents used as the left frame of the frameset. While an obsolete HTML feature, a frameset makes it easy browsing the generated pages. Moreover the table of contents used as the left frame is a convenient way to programmatically list all the generated pages.

The result of the this step is the file containing the frameset.

For clarity, the “split.” parameter name prefix is omitted here.

However when you’ll pass any of the following parameters to w2x, please do not forget this prefix. Example: -p split.split-before-level 8.

Parameters:

Name

Value

Description

allow-lonely-heading

A boolean: true (same as: yes | on | 1) | false (same as: no | off | 0).

Default: false.

If specified as true, allow a page to contain just a heading and nothing else.

indent

A boolean: true (same as: yes | on | 1) | false (same as: no | off | 0).

Default: false.

Specifies whether the save files should be indented.

Note:

Do not specify indent=”true” in production.

The XML indentation created this way being very simple, this may add whitespace inside elements where space characters are significant.

out-file

A file path.

No default (required).

Specifies the path of the file containing the frameset. A relative file path is relative to the current working directory.

This step always generates several files, all in the same directory as file out-file.

This output directory is created on the fly if needed too. However, the output directory, if it already exists, is not automatically made empty.

The file specified by out-file contains the frameset. Let’s suppose out-file is temp\foo.html.

The table of contents of the frameset, its left frame, is created in temp\foo-TOC.html.

Unless parameter use-id-as-filename has been specified as true, the styled HTML pages are created in temp\foo-0.html, temp\foo-1.html, temp\foo-2.html, …, temp\foo-N.html.

split-before-level

Outline level between 0 (e.g. style “Heading 1”) and 8 (e.g. style “Heading 9”).

Default: 0 (split at “Heading 1”).

In order to generate multi-page styled HTML, that is, frameset, Web Help, EPUB, we need to automatically split the input XHTML document into pages.

A new page is created each time a paragraph having an outline level less than or equal to specified split-before-level parameter is found in the source.

An outline level is an integer between 0 (e.g. style “Heading 1”) and 8 (e.g. style “Heading 9”).

The default value of parameter split-before-level is 0, which means: for each “Heading 1”, create a new page starting with this “Heading 1”.

See also Important tip.

use-id-as-filename

A boolean: true (same as: yes | on | 1) | false (same as: no | off | 0).

Default: false.

By default, the save files of the generated pages have the same basename as out-file, except that a number is appended to this basename. Example: out-file is temp\foo.html; the save files of the generated pages are thus: temp\foo-0.html, temp\foo-1.html, temp\foo-2.html, …, temp\foo-N.html.

In a MS-Word document, a heading is often given a bookmark. The Convert step translates this bookmark to an ID. When use-id-as-filename is specified as true, the save file of a page is given a basename corresponding to the ID of the heading used to start this page. When this heading ID is missing, the Split step fallbacks to the default behavior.

Transform step

Transforms input XML document or file using an XSLT 1.0 stylesheet. The result of the this step is the save file containing the transformed document.

Unlike the load step, if the input XML file starts with a <!DOCTYPE> pointing to a DTD, then the document loader created by a Transform step will silently skip this DTD.

For clarity, the “transform.” or “transform2.” parameter name prefix is omitted here.

However when you’ll pass any of the following parameters to w2x, please do not forget this prefix. Example: -p transform.cals-tables yes.

Parameters:

Name

Value

Description

xslt-url-or-file

An absolute URL or the path of an existing file.

No default (required).

Specifies which XSLT 1.0 stylesheet should be used to transform the input XML document. A relative file path is relative to the current working directory.

out-file

A file path.

No default (required).

Specifies the path of the save file. A relative file path is relative to the current working directory.

Any other parameter is passed to the XSLT stylesheet as an XSLT stylesheet parameter. Which XSLT stylesheet parameters are supported depend on the XSLT stylesheet being used.

Table 4 Parameters of w2x:xslt/docbook.xslt, docbook5.xslt, which are used to convert input XHTML document to DocBook v4 or v5

Name

Value

Description

docbook-version

DocBook version (e.g. “4.5” or “5.0”).

Default: “4.5” for docbook.xslt, “5.0” for docbook5.xslt.

Specifies the version of DocBook.

This number is used to specify which <!DOCTYPE> to add to the generate file or, in the case of DocBook 5, the value of the version attribute of the root element of the generated file.

Please remember that versions of DocBook older than “4.3” do not support HTML tables. (HTML tables, not CALS tables, are generated by default. See below.)

cals-tables

yes” | “no”.

Default: “no”.

If “yes”, generate CALS tables.

If “no”, generate HTML tables.

Note that cals-table=”yes” requires specifying Convert step parameter set-column-number=”yes”.

hierarchy-name

book” | “article” | “part” | “chapter” | “appendix” | “section” | “book-sect1” | “article-sect1” | “part-sect1” | “chapter-sect1” | “appendix-sect1” | “sect1” | “sect2” | “sect3” | “sect4” | “sect5” .

Default: “book”.

Specifies the root element name and type of sections of the DocBook document to be generated.

media-alt

yes” | “no”.

Default: “no”.

If “yes”, convert the alt attribute of XHTML element img to DocBook alt element.

If “no”, ignore the alt attribute of XHTML element img.

pre-element-name

An element local name. Default: “literallayout”.

Specifies to which DocBook element, an HTML pre element is to be converted.

Table 5 Parameters of w2x:xslt/topic.xslt, which is used to convert input XHTML document to a DITA topic

Name

Value

Description

root-topic-id

An XML ID.

Default: automatically generated ID.

Specifies the ID of the root topic.

single-topic

yes” | “no”.

Default: “no”.

If “yes”, convert input <div class=”role-sectionN”> to (non-nested) DITA section elements.

If “no”, convert input <div class=”role-sectionN”> to nested topics.

topic-type

topic” |“concept” | “generalTask” | “task” (same as: “strictTask” ) | “reference”.

Default: “topic”.

Specifies the type of topics to be created by the XSLT stylesheet.

pre-element-name

An element local name. Default: “pre”.

Specifies to which DITA element, an HTML pre element is to be converted.

shortdesc-class-name

A class name. Default: “”. Examples: p-Shortdesc, p-Abstract.

Specifies the class name of the XHTML <p> which acts as a short description of the section.

When this parameter is not specified (or is specified as the empty string which is its default value), the following style mapping, created by the w2x-app wizard:

-p edit.blocks.convert¬

"p-Shortdesc p class='p-Shortdesc'"

...

<xsl:template

match="h:p[@class='p-Shortdesc']">

<shortdesc>

<xsl:call-template

name="processCommonAttributes"/>

<xsl:apply-templates/>

</shortdesc>

</xsl:template>

causes DITA <shortdesc> elements to generated inside topic bodies, which is invalid.

After specifying

-p transform.shortdesc-class-name¬

p-Shortdesc

this issue is fixed and DITA <shortdesc> elements are generated before topic bodies.

Table 6 Parameters of w2x:xslt/xhtml_strict.xslt, xhtml_loose.xslt, xhtml1_1.xslt, xhtml5.xslt, which are used to convert input XHTML 1.0 Transitional document to XHTML having a different version

Name

Value

Description

discard-index-terms

yes” | “no”.

Default: “yes”.

If “yes”, discard <span class=”role-index-term”> elements.

If “no”, keep <span class=”role-index-term”> elements.

footnote-number-format

A valid XSLT number format (value of attribute format of element xsl:number).

Default: “[1]”.

When parameter number-footnotes is “yes”, specifies the format of the numeric label used for footnotes and footnote callouts.

generate-xref-text

yes” | “no”.

Default: “yes”.

If “yes”, add hyperlink text to a elements which are cross-references.

If “no”, keep empty a elements which are cross-references.

number-footnotes

yes” | “no”.

Default: “yes”.

If “yes”, add a numeric label to footnotes and footnote callouts.

If “no”, do not add a numeric label to footnotes and footnote callouts.

style-with-class

yes” | “no”.

Default: “no”.

If “yes”, add a class attribute to some elements to allow using a CSS stylesheet to style them. For example: convert <center> to <div class=”center”>.

If “no”, add a direct style to some elements to style them. For example: convert <center> to <div style=”text-align:center;”>.

Table 7 Parameters of w2x:xslt/map.xslt, bookmap.xslt, which are used to convert input DITA topic file to a map or bookmap

Name

Value

Description

add-index

yes” | “no”.

Default: “yes”.

bookmap.xslt only.

Ignored if the input topic document does not contain any index term.

If “yes”, add an indexlist element to the back matter of the bookmap .

If “no”, do not add an indexlist element to the back matter of the bookmap.

add-toc

yes” | “no”.

Default: “yes”.

bookmap.xslt only.

If “yes”, add a toc element to the front matter of the bookmap.

If “no”, do not add a toc element to the front matter of the bookmap.

output-path

An absolute or relative “file:” URI.

No default (required).

Specifies the URI of the directory which is to contain all generated files. A relative URI is relative to the current working directory.

section-depth

“1”, “2”, “3”, “4”, “5”, “6”, “7”, “8”, “9”.

Default: “1”.

Specifies the topicref structure of the DITA map to be generated.

Example 1: a bookmap generated using section-depth=”1” only contains chapter topicrefs.

Example 2: a bookmap generated using section-depth=”2” contains chapter topicrefs, themselves possibly containing plain topicrefs (acting as sections).

Example 3: a bookmap generated using section-depth=”3” contains chapter topicrefs, themselves possibly containing plain topicrefs (acting as sections), themselves possibly containing other plain topicrefs (acting as subsections).

topic-path

An absolute or relative “file:” URI.

No default: generate topic files in output-path.

Specifies the URI of the subdirectory directory which is to contain all generated topic files. A relative URI is relative to output-path.

topic-type

topic” | “concept” | “generalTask” | “task” (same as: “strictTask” ) | “reference”.

No default. See description.

Specifies the type of topics to be created by the topic.xslt XSLT stylesheet. See above.

This parameter is used to make a difference between a strict task and a general task. In all other cases, this parameter may be omitted.

Web Help step

Splits input XHTML document, whether styled or semantic, into several pages and compiles these pages into a Web Help. The Web Help compiler used to do this is free, open source, XMLmind Web Help Compiler.

This step always generates UTF-8 encoded, “.html” files, no matter the parameters specifying other values.

Same parameters as the Split step, plus the following Web Help specific parameters (for clarity, the “webhelp.” parameter name prefix is omitted here):

Name

Value

Description

add-index

yes” | “no”.

Default: “yes”.

If “yes”, automatically create an index.html file, if an index.html file does not already exist.

wh-* (examples: wh-local-jquery, wh-jquery-theme, wh-index-numbers)

String.

No default.

All parameters starting with “wh-“ are passed as is to XMLmind Web Help Compiler.

Example: -p webhelp.wh-collapse-toc yes.

These parameters are all documented in XMLmind Web Help Compiler, Parameters.

Embedding w2x in a Java™ application

Embedding w2x in a Java™ application is as simple as:

Create an instance of class Processor.

Configure it by passing an array of option strings identical to those of the w2x command line utility to method Processor.configure or (low-level) by directly adding conversion steps and parameters to Processor.stepList and Processor.parameterMap.

Invoke the configured processor to convert specified input file to specified output file. This is done invoking high-level method Processor.process or low-level method Processor.executeSteps.

About thread-safety

An instance of Processor cannot be shared by different threads.

It’s strongly recommend not to reuse an instance of Processor. That is, please create one instance of Processor per conversion.

The reference manual (generated using javadoc) of the Java API of w2x is found in XMLmind Word To XML Java™ API.

High-level example w2x_install_dir/doc/manual/embed/Embed1.java:

Processor processor = new Processor();

int l = processor.configure(args);

File inFile = null;

File outFile = null;

if (l+2 == args.length) {

inFile = new File(args[l]);

outFile = new File(args[l+1]);

} else {

System.exit(1);

}

processor.process(inFile, outFile, /*progress monitor*/ null);

Compile Embed1.java by executing “ant10 in w2x_install_dir/doc/manual/embed/.

Run “ant tembed1” in w2x_install_dir/doc/manual/embed/. This creates w2x_install_dir/doc/manual/embed/tembed1.dita.

Lower-level example w2x_install_dir/doc/manual/embed/Embed2.java:

Processor processor = new Processor();

ConvertStep convertStep = new ConvertStep("convert");

processor.stepList.add(convertStep);

EditStep editStep = new EditStep("edit");

processor.stepList.add(editStep);

processor.parameterMap.put("edit.xed-url-or-file",

"w2x:xed/main-styled.xed");

SaveStep saveStep = new SaveStep("save");

processor.stepList.add(saveStep);

processor.parameterMap.put("save.indent", "yes");

processor.process(inFile, outFile, /*progress monitor*/ null);

Compile Embed2.java by executing “ant” in w2x_install_dir/doc/manual/embed/.

Run “ant tembed2” in w2x_install_dir/doc/manual/embed/. This creates w2x_install_dir/doc/manual/embed/tembed2.xhtml.

Extension points

Custom conversion step

The stock conversion steps are: com.xmlmind.w2x.processor.ConvertStep, DeleteFilesStep, EditStep, LoadStep, SaveStep, TransformStep.

A custom conversion step may be implemented by deriving abstract class com.xmlmind.w2x.processor.ProcessStep. Such task poses no technical problems whatsoever. Suffice for that to implement a single method: ProcessStep.process.

See reference of class com.xmlmind.w2x.processor.Processor.

Custom image converters

Image converters are used to convert images having a format not supported by Web browsers (TIFF, WMF, EMF, etc) to a format supported by Web browsers (SVG, PNG, JPEG).

Image converters are specified by interface com.xmlmind.w2x.docx.image.ImageConverterFactory. XMLmind Word To XML ships with 4 classes implementing this interface:

com.xmlmind.w2x.docx.image.ImageConverterFactoryImpl

Image converter factory used to convert TIFF images to PNG or JPEG.

com.xmlmind.w2x_ext.wmf_converter.WMFConverterFactory

Image converter factory used to convert WMF graphics to SVG.

com.xmlmind.w2x_ext.emf2png.EMF2PNG

This image converter factory is available only on Windows. It leverages Windows own GDI+ to convert EMF (in fact, Windows metafiles of any kind, including WMF) to PNG.

This is not that great because, unlike above WMFConverterFactory which converts WMF (Windows vector graphics format) to SVG (standard vector graphics format), EMF2PNG converts a vector graphics format to a raster image format. However, having EMF2PNG is better than nothing at all.

EMF2PNG has one parameter called resolution. Its value is a real number expressed in Dot Per Inch (DPI). The default value of parameter resolution is 0.0 (see below).

The resolution parameter specifies the resolution of the output PNG file. 0 means: same resolution as the one found input EMF/WMF file; a positive number means: use this value to override the resolution found in the input EMF/WMF file; a negative number means: use specified absolute value but only if this absolute value is greater than the resolution found in the input EMF/WMF file.

com.xmlmind.w2x.docx.image.ExternalImageConverter

This image converter factory executes an external program to perform the conversion. See 9.1.2.1 below.

If you want w2x to support more image formats, you’ll have to create your own ImageConverterFactory and register it with w2x using method ImageConverterFactories.register.

About thread-safety

A single instance of a class implementing ImageConverterFactory is used by all instances of com.xmlmind.w2x.processor.Processor. This implies that an implementation of ImageConverterFactory must be thread-safe.

See reference of package com.xmlmind.w2x.docx.image.ImageConverterFactories.

Specifying an external image converter

Examples of W2X_IMAGE_CONVERSIONS specifications (see 9.1.2.2 below):

Convert EMF to SVG using OpenOffice/LibreOffice:

.emf.svg soffice --headless --convert-to svg -–outdir %~po %i

Convert EMF/WMF to PNG using ImageMagick:

.emf.png.wmf.png magick convert -density 288 "%I" -scale 25% "%O"

The command executed by an external image converter may contain the following variables:

Variable

Definition

%I

Absolute path of the input image file.

%O

Absolute path of the output image file.

%i

Same as %I but quoted, that is, equivalent to “%I”.

%o

Same as %O but quoted, that is, equivalent to “%O”.

%S

File separator: “\” on Windows, “/” on Mac/Linux.

The following modifiers may be applied to the %I, %O, %i, %o variables:

Modifier

Definition

~p

Absolute path of the parent directory of the file. For example, if %I is “C:\temp\doc_files\logo.wmf”, then %~pI is “C:\temp\doc_files”.

~n

Basename of the file. For example, if %I is “C:\temp\doc_files\logo.wmf”, then %~nI is “logo.wmf”.

~r

Basename of the file without any extension. For example, if %I is “C:\temp\doc_files\logo.wmf”, then %~rI is “logo”.

~e

Extension of the file. For example, if %I is “C:\temp\doc_files\logo.wmf”, then %~eI is “wmf”.

Also note that “%%” may be used to escape character “%”. More generally, just like in an URL, an %HH UTF-8 sequence may be used to escape any character. Example: “%3B” is “;” (semi colon), “%C3%A9” is “é” (“e” with acute accent).

Controlling how image files found in the input DOCX file are converted to standard formats

Conversion of images found in the DOCX file (TIFF, WMF, EMF, etc) to standard formats (SVG, PNG, JPEG) may be controlled using environment variable (or Java™ property) W2X_IMAGE_CONVERSIONS.

The default value of this variable is (all specifications on a single line):

.wmf.svg java:com.xmlmind.w2x_ext.wmf_converter.WMFConverterFactory;

.tiff.png java:com.xmlmind.w2x.docx.image.ImageConverterFactoryImpl

On Windows, the default value of W2X_IMAGE_CONVERSIONS is (all specifications on a single line):

.wmf.svg java:com.xmlmind.w2x_ext.wmf_converter.WMFConverterFactory;

.emf.png.wmf.png java:com.xmlmind.w2x_ext.emf2png.EMF2PNG resolution 0;

.tiff.png java:com.xmlmind.w2x.docx.image.ImageConverterFactoryImpl

The syntax of W2X_IMAGE_CONVERSIONS is:

specifications -> “-” | specification_list

specification_list -> specification [ “;” specification ]+

specification -> “+” | image_conversion

image_conversion -> extensions S ( java_image_conversion | external_image_conversion )

extensions -> [ “.” input_file_extension “.” output_file_extension ]+

java_image_conversion -> “java:” fully_qualified_java_class_name parameters

parameters -> [ S parameter_name S possibly_quoted_parameter_value ]*

external_image_conversion -> command_line

About this syntax:

-” means: no specifications; hence no image conversions at all.

+” means: insert default value of W2X_IMAGE_CONVERSIONS at this point. Example:

set W2X_IMAGE_CONVERSIONS=.emf.png magick convert %i %o;+

where default value of W2X_IMAGE_CONVERSIONS is (on Windows):

.wmf.svg java:com.xmlmind.w2x_ext.wmf_converter.WMFConverterFactory;

.emf.png.wmf.png java:com.xmlmind.w2x_ext.emf2png.EMF2PNG resolution 0;

.tiff.png java:com.xmlmind.w2x.docx.image.ImageConverterFactoryImpl

Note that the image conversion specifications are considered in the order of their declarations in variable W2X_IMAGE_CONVERSIONS. In the case of the above example, it’s custom “magick convert %i %o” which is used to convert EMF to PNG and not stock “java:com.xmlmind.w2x_ext.emf2png.EMF2PNG resolution 0”.

Limitations and implementation specificities

The Convert step does not support the following MS-Word features.

By “does not support”, we mean that w2x will not generate something useful corresponding to such features. We don’t mean that using such features in a DOCX file would cause w2x to fail or to generate invalid XML documents.

Right to left scripts.

Enclose characters.

Asian layout.

Cover Page. Blank Page.

Text wrapping of tables and pictures other than the simplest one.

Picture formats other than GIF, PNG, JPEG, BMP, TIFF and WMF are not supported. EMF pictures are supported only on Windows.

Clip Art. Shapes. SmartArt. Chart.

Header. Footer. Page Number.

Japanese Greetings. Text Box. WordArt. Drop Cap.

Object.

All features related to Page Layout except (to a minimal extent) page and column breaks and end of sections.

All features related to Mailings.

All features related to Spelling & Grammar, except of course the various languages used in the document (i.e. lang attribute).

Comments.

All features related to Change Tracking, (document) Compare, (document) Protect.

Macros.

Controls.

The Convert step generates XHTML+CSS documents having the following specificities:

Tab stops are converted to <span class=”role-tab> </span>. See About tab stops.

MS-Word document properties having no standard meta equivalent are given names starting with “ms-”. Example:

<meta content="Hussein Shafie" name="ms-cp-lastModifiedBy" />

MS-Word “styles” having no CSS equivalent are a given a “-ms-” prefix. Example:

.p-Heading3 {

-ms-outlineLvl: 2;

color: #4F81BD;

font-family: Cambria;

...

Page breaks are translated to <?break-page?>. Column breaks are translated to <?break-column?>. End of sections are signaled by <?end-of-section?>.

WMF pictures are converted to SVG.

OpenXML math, for example , is converted to MathML.

Conversion from OpenXML math to MathML is implemented by an XSLT 1.0 stylesheet called omml2mml.xsl coming from open source project XSL stylesheets for TEI XML. If you think you have access to a better XSLT stylesheet than open source omml2mml.xsl, then you may use it by specifying environment variable (or Java™ system property) W2X_MATH_CONVERTER_XSLT. Example:

set W2X_MATH_CONVERTER_XSLT=C:\Users\john\My better omml2mml.xsl

All simple and most complex fields are converted to a <?field code?> having a <span class="role-field"> parent. Example:

<span class="role-field">

<?field DATE \@ "MMMM d, yyyy" \* MERGEFORMAT ?>

August 27, 2014

</span>

Smart tags are enclosed between <?begin-smartTag tag?> and <?end-smartTag tag?>. Example:

<?begin-smartTag {urn:schemas-microsoft-com:office:smarttags}PersonName#0?>

<?begin-smartTag {urn:schemas:contacts}GivenName#1?>

Bill

<?end-smartTag {urn:schemas:contacts}GivenName#1?>

<?begin-smartTag {urn:schemas:contacts}Sn#2?>

Gates

<?end-smartTag {urn:schemas:contacts}Sn#2?>

<?end-smartTag {urn:schemas-microsoft-com:office:smarttags}PersonName#0?>

Controls are enclosed between <?begin-sdt control_id?> and <?end-sdt control_id?>. Example:

<?begin-sdt comboBox#6?>

<td class="tc-TableGrid--bb tc-TableGrid"

style="padding-bottom: 7.2pt; padding-left: 7.2pt;

padding-right: 7.2pt; padding-top: 7.2pt;">

<p class="tp-TableGrid p-Normal" lang="fr-FR">

<span class="c-PlaceholderText">Choose an item.</span>

</p>

</td>

<?end-sdt comboBox#6?>

About tab stops

Tab stops are converted to <span class=”role-tab> </span>. These span elements are processed as follows:

When generating styled HTML (that is, XHTML+CSS), some JavaScript™ code (w2x_install_dir/xed/expand-tabs.js) is added to the output file. This code computes and gives a width to all <span class=”role-tab> </span>. This allows to decently emulate tab stops in any modern Web browser.

If you don't want this code to be added to the output file, pass option -p edit.do.expand-tabs "" to w2x.

When generating semantic XHTML and all the other semantic XML formats (DocBook, DITA, etc), it's possible to convert consecutive paragraphs containing text runs aligned on tab stops to a borderless table.

However because, in the general case, it's not possible to emulate tab stops using tables, this XED script is disabled by default. If you really want to emulate tab stops using tables, pass option -p edit.convert-tabs.to-table yes to w2x.

Index

A

add-index, parameter 54

add-toc, parameter 55

allow-lonely-heading, parameter 50

B

before-save.allow-flow, parameter 43

biblio.style-names, parameter 44

blocks.convert, parameter 44

blocks.convert-to-pre, parameter 44

C

-c, option 19, 33, 34, 37

cals-tables, parameter 52

captions.style-names, parameter 44

charset, parameter 37, 38

Convert, step 38

converted-image-extensions, parameter 38

convert-tabs.to-table, parameter 44

convert-tabs.unwrap-paragraphs, parameter 45

cover-image-url-or-file, parameter 48

create-mathml-object, parameter 38

D

default-lang, parameter 39, 48

Delete files, step 40

discard-index-terms, parameter 54

DITA bookmap, output format 17

DITA map, output format 17

DITA topic, output format 16

DocBook 4, output format 16

DocBook 5, output format 16

docbook-version, parameter 52

E

-e, option 20, 33, 35, 37

-e2, option 35

Edit, step 40

encoding, parameter 49

EPUB, output format 16, 17, 39, 48, 51

F

-f, option 35

files, parameter 40

finish-styles.css-uri, parameter 41

finish-styles.mathjax, parameter 42

finish-styles.mathjax-url, parameter 42

footnote-number-format, parameter 54

frameset, output format 17, 49, 51

frameset, output format 16

G

generate-xref-text, parameter 54

H

headings.convert, parameter 45

hierarchy-name, parameter 52

I

identifier, parameter 48

ids.generate-section-ids, parameter 45

ids.section-id-max-length, parameter 45

indent, parameter 49, 50

inlines.convert, parameter 45

inlines.generate-big-small, parameter 46

L

-l, option 34

-liststeps, option 20, 35

Load, step 48

lower-case-resource-names, parameter 39

M

MathML 38, 42, 63

MathJax 42

media-alt, parameter 53

N

number-footnotes, parameter 54

O

-o, option 16, 17, 34

out-file, parameter 37, 48, 49, 50, 52

Outline level 17, 18, 51

output-path, parameter 55

P

-p, option 34, 37

-p, parameter 19, 20

pre-element-name, parameter 53

prune.preserve, parameter 46

-pu, option 34

-pu, parameter 20

R

remove-pis.except, parameter 43

remove-styles.preserved-classes, parameter 46

resource-directory, parameter 39

resource-prefix, parameter 39

root-topic-id, parameter 53

S

-s, option 35, 37

Save, step 49

section-depth, parameter 55

sections.max-level, parameter 46

servlet 10

curl 13

multipart/form-data 13

POST 13

set-column-number, parameter 39

set-lang, parameter 40

shortdesc-class-name, parameter 53

single-topic, parameter 53

split-before-level, parameter 17, 51

-step, option 20, 35

style-with-class, parameter 54

T

-t, option 20, 33, 35

-t2, option 35

tab stops 44, 45, 62, 64

title.subtitle-style-names, parameter 43

title.title-style-names, parameter 43

topic-path, parameter 55

topic-type, parameter 53, 55

Transform, step 51

U

us-id-as-filename, parameter 51

V

-v, option 35

-version, option 35

version, parameter 40

-vv, option 19, 35

-vvv, option 35

W

w2x-app 6, 7, 9

Web Help, output format 16, 17, 51, 55

webhelp.add-index, parameter 56

X

xed-url-or-file, parameter 41

XHTML 1.0 Strict, output format 17

XHTML 1.0 Transitional, output format 17

XHTML 1.1, output format 17

XHTML 5.0, output format 17

XHTML, output format 15, 17

xhtml-file, parameter 40

XMLmind Web Help Compiler 55, 56

XMLmind XML Editor add-on 9

xslt-url-or-file, parameter 52


1 However, “jar xvf converted.zip” works fine. jar is a command-line utility which comes with all Java Development Kits (JDK).

2 curl is an open source command line tool and library for transferring data with URL syntax.

3 This option is “-p convert.charset UTF-8”. See charset parameter.

4 But not automatically made empty if the output directory already exists.

5 A custom conversion step derives from abstract class com.xmlmind.w2x.processor.ProcessStep.

6 It’s a paragraph style because the CSS style name has a “p-“ prefix.

7 Any value would do (e.g. g:id=”foo” would have worked as well). Suffice for consecutive elements to be grouped to all have the same g:id attribute.

8 Unless you specify:

-p edit.prune.preserve "p-ProgramListing"

script w2x_install_dir/xed/prune.xed will cause open lines to be stripped from the generated pre element.

9 Because only XHTML 5 documents may embed MathML. With any other version of XHTML, this would cause the document to become invalid.

10 Apache Ant is a command-line utility for automating software build processes. By default, ant uses an XML file, called build.xml to describe the build process and its dependencies. In the case of the two above code samples, this file is w2x_install_dir/doc/manual/embed/build.xml.