se/SE/stuff/dita-ot-3.3.3/docsrc/reference/pdf-transformation-history.dita

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE concept PUBLIC "-//OASIS//DTD DITA Concept//EN" "concept.dtd">
<!--  This file is part of the DITA Open Toolkit project. See the accompanying LICENSE file for applicable license.  -->

<concept id="history-pdf-transformation">
  <title>History of the PDF transformation</title>
  <shortdesc>The DITA Open Toolkit PDF transformation was originally based on a third-party contribution by Idiom
    Technologies, and is commonly known as the “pdf2” plug-in.</shortdesc>
  <prolog>
    <metadata>
      <keywords>
        <indexterm><xmlelement>info</xmlelement></indexterm>
        <indexterm>legacypdf</indexterm>
        <indexterm>PDF2<index-see>PDF<indexterm>plug-in, history of</indexterm></index-see></indexterm>
        <indexterm>languages<indexterm>auto-generated strings</indexterm></indexterm>
        <indexterm>PDF<indexterm>plug-in, history of</indexterm><indexterm>HTML formatting
          differences</indexterm></indexterm>
        <indexterm>Idiom Technologies</indexterm>
        <indexterm>HTML<indexterm>PDF formatting differences</indexterm></indexterm>
        <indexterm>DITA 1.1</indexterm>
        <indexterm>index<indexterm>indexing domain</indexterm></indexterm>
        <indexterm>Customization directory</indexterm>
        <indexterm>org.dita.pdf2</indexterm>
      </keywords>
    </metadata>
  </prolog>
  <conbody>
    <p>When IBM developed the code that later became DITA-OT, it included only a proof-of-concept PDF
      transformation. IBM had their own processing chain for producing PDFs from SGML, which they had developed over
      several decades, so resources were focused primarily on XHTML output and preprocessing.</p>
    <p>Since the initial proof-of-concept transformation was not robust enough for production-grade output, companies
      began to develop their own PDF transformations. One company, Idiom Technologies, made their transformation (known
      as the “pdf2” transformation) available as open source on 23 February 2006. The Idiom plug-in was initially
      available as a separately-downloadable plug-in that could be installed into DITA-OT.</p>
    <p>Later the DITA-OT project formally incorporated the Idiom plug-in as a demonstration in the
        <filepath>demo/fo</filepath> directory. Beginning with DITA-OT version 1.5, released 18 December 2009, the
      “pdf2” code served as the main, supported PDF transformation. (The original PDF transformation was deprecated and
      renamed “legacypdf”.) In DITA-OT version 1.6, the “pdf2” plug-in was moved to
        <filepath>plugins/org.dita.pdf2</filepath>. </p>
    <p>The fact that the current PDF transformation was not originally developed in parallel with the other core DITA-OT
      transformations led to anomalies that often confuse users:</p>
    <ul>
      <li>Elements are often (by default) styled differently in the XHTML and PDF transformations. For example, consider
        the <xmlelement>info</xmlelement> element in a task topic. In HTML output, this is an inline element; in PDF
        output, it is a block-level element.</li>
      <li>The auto-generated strings used for localization are different, and so languages that are supported by DITA-OT
        differ based on whether the XHTML or PDF transformation is used.</li>
      <li>The Idiom plug-in used its own extension mechanism (the <filepath>Customization</filepath> folder) to provide
        overrides to the PDF transformation.</li>
      <li>Before the release of DITA 1.1 (which added support for the indexing domain), Idiom developed an index
        extension that used a FrameMaker-inspired syntax.</li>
    </ul>
  </conbody>
</concept>