The following items are included in DITA Open Toolkit Release 2.1. Issue numbers
correspond to the tracking number in the GitHub issues
Maintenance Release 2.1.2
DITA Open Toolkit Release 2.1.2 includes the following bug fixes:
Index terms starting with “Ñ” are no longer dropped from Spanish PDF indexes. They are now handled with
the correct Spanish sort order and placed in a group between “N” and “O”, under the “Ñ” heading.
#2068 Add support for n-tilde in Spanish index
References to external code samples are now resolved properly when @coderef
target non-XML resources and the output directory is shifted by setting the
parameter to 3
#2044 Non-XML coderef fails on outer context
Warnings for deprecated templates are now displayed with the correct message prefix:
#2043 Warning for deprecated templates fails
Maintenance Release 2.1.1
DITA Open Toolkit Release 2.1.1 includes the following bug fixes:
Whitespace in <image>
elements is now ignored in PDF output to prevent errors when
#1985 XEP error: Element 'fo:external-graphic' must be empty
Any proportional table column widths expressed in the CALS table model as "*" are now normalized to the
equivalent "1*" value to prevent errors while generating XHTML output.
#1978 colwidth="*" causes Fatal Error! Cannot convert string "" to a double
In DITA Open Toolkit 2.0.1, the line ending characters in shell files were inadvertently changed to
Windows (CRLF) line endings. The correct UNIX (LF) line endings have now been restored.
#1954 bin/ant throws errors when executed in Ubuntu shell
Feature requests and changes
DITA Open Toolkit Release 2.1 includes the following new features and changes:
In PDF output, inline <codeph>
elements in topic titles now inherit the font size
from the surrounding title text.
Codephrase used in topic title has very small font
Image metadata processing has been refactored to improve the processing speed by determining width and
height without reading the entire image into memory.
Faster implementation for ImgUtils.getWidth/getHeight
Obsolete CHM configuration files have been removed from the PDF2 plugin.
DITA OT 2.1.dev chm properties folder in org.dita.pdf2
Milestone 1 included the following new features and changes:
The chapterBody XSL template mode has been refactored to create additional HTML5 groups
in the generated HTML <body> element.
DITA topics are now mapped to the HTML5 <article> element, DITA
<section> to <section>, and DITA <fig> to
accessibility attributes are also generated to associate each
article with its heading. This helps to improve the accessibility and interoperability of HTML output by
providing structural information to assistive technologies such as screen readers.
Added HTML generation extension for body content
Additional file formats referenced from <image>
elements are now copied to output.
Any references that are not DITA, DITA map, or image, are now treated as resources and also copied, so you
no longer need extra <copy>
operations in your project build scripts to include such
assets in output.
More recognized default resource extensions in the plugins base
DITA-OT now fails with a fatal error (DOTA013F) if a specified .ditaval filter file is not available.
This ensures that unexpected output is not generated if the filter file is missing or invalid.
Link to missing DITAVAL file should show a relevant error code in the console
options have been moved to runtime properties, so you can now pass these parameters at build time to
specify whether the @xtrf
debugging attributes should be
generated, and how the DITA-OT handles errors and error recovery. For more information, see common
Move configuration options to runtime properties
The deprecated Java command line tool has been removed in favor of the dita command
introduced in DITA-OT 2.0.
If you previously generated output via a Java command sequence such as
java -jar lib/dost.jar /i:input-file /transtype:transformation-type /parameter-name:value
you should now use the following syntax instead:
dita -f transformation-type -i input-file -o output-dir
Remove old Java command line tool
The custom FileUtils
code used to handle input and output in earlier versions of DITA-OT
has been replaced with the Apache Commons IO
Support for the args.odt.img.embed parameter has been removed from OpenDocument
The previous default behavior was to embed images as Base64-encoded text, but editors do not use this as
a default. Instead, office packages such as LibreOffice will convert embedded images into linked images on
opening and saving an ODT file.
Remove support for args.odt.img.embed
Keydef processing has been removed from the XHTML rendering code
Keys are now resolved in one preprocessing step, whereas in earlier versions of DITA-OT, the XHTML code
returned to the keydef.xml
file to look up targets for phrase elements and pull in
text when needed. This change affects non-linking elements that can't take @href
attributes, such as <ph>
, and <indexterm>
Remove keydef processing from XHTML rendering code
A new dita.parser extension point has been added to allow plug-ins to contribute a
custom parser for DITA files.
If a custom DITA parser is defined, the preprocessing routines will use it during the gen-list and
debug-filter stages to output DITA XML.
Extension-point for custom DITA parsers
DITA Open Toolkit Release 2.1 provides fixes for the following bugs:
Suspected memory leak in OT bug preprocess
Cannot convert string to double
integrator.xml should also load the necessary "commons-io.jar"
PDF2 output plugin fails rendering tables
Link to index term repeats itself
Searchtitle Incorrectly Used for HTML Title
build.xml help target documentation is incorrect ant interface
@class missing from choicetable XHTML output
properties table XHTML output
- 1894 Fix
Error reported during flagging
Index continued markers don't work if both primary and its secondary entries continue
- 1927 In
the key() function, the node supplied in the third argument (or the context node if absent) must be in a
tree whose root is a document node
when running ./dita, console shows "No such file or directory" when directory contains spaces
Milestone 1 provided fixes for the following bugs:
Filtering doesn't support default for rev flagging
Keyref processing doesn't use only first keyword or term
XSLT XPaths do not properly use the @class attribute to select elems
Maintenance Release 2.0.1
Note: DITA Open Toolkit Release 2.1 also included the following bug fixes that were
released earlier this year with maintenance version 2.0.1.
copy-to attribute not processed correctly (error DOTX008E) in XHTML transtype on a topicref included in a
Xref processing does not handle case of key with no remote resource or link text
are not preserved on <bodydiv> element
DITA-OT 2.0: PDF bookmarks (TOC, index) not in language indicated in source files
DITA-OT 2.0: typo in PDF-plugin, xsl/fo/toc.xsl ("boookmap")
Filtering doesn't support @deliveryTarget
Mappull step fails with nested concept
with links and nested task fails in XHTML
Convert String to double Error message when we try to generate pdf
2.0: Map with subject scheme and non-below map fails