Support OT 2.x and 1.x

[drmacro]

Need to be able to run with OT 1.x or 2.x from same code distribution.

[drmacro]

Using this issue to fix some cosmetic issues in the EPUB transform (incorrect informational messages, debug handling, etc.).

[drmacro]

Partial support for OT 2.x implemented as of release 1.0.0RC18. However, full testing not yet done. XSLT uses now-deprecated templates for getting strings.

[raducoravu]

The generated EPUBs using DITA OT 2.x are invalid according to EPUB Checker because they contain HTML 5 elements like article, main and also attributes like "aria-labelledby" which do not point to a specific target.
Problem seems to be the fact that in the XSLT stylesheet:

 DITA-OT2.x\plugins\org.dita.xhtml\xsl\dita2xhtml-util.xsl

there is a template:

  <xsl:template match="/">
    <xsl:variable name="content" as="node()*">
      <xsl:apply-imports/>
    </xsl:variable>
    <xsl:apply-templates select="$content" mode="add-xhtml-ns"/>
  </xsl:template>

which no longer gets called for your output, template which removes various HTML 5 elements and attributes.
As a possible workaround in the XSLT stylesheet:

 DITA-OT2.x\plugins\org.dita4publishers.epub\xsl\html2xhtmlImpl.xsl

you could add your own equivalent filtering:

      <xsl:template match="nav | section | figure | article" mode="html2xhtml" priority="20">
        <div>
          <xsl:apply-templates select="@* except (@role, @aria-labelledby) | node()" mode="html2xhtml"/>
        </div>
      </xsl:template>

      <!-- Group for root document node does not need extra XHTML div -->
      <xsl:template match="main/article" mode="html2xhtml" priority="30">
        <xsl:apply-templates select="node()" mode="html2xhtml"/>
      </xsl:template>

      <xsl:template match="header | footer | main" mode="html2xhtml" priority="20">
        <xsl:apply-templates select="node()" mode="html2xhtml"/>
      </xsl:template>

      <xsl:template match="div/@role" mode="html2xhtml" priority="10"/>

[raducoravu]

Also the @scale attribute is not supported on images. There is a new DITA OT 2.x ANT target called "xhtml.image-metadata" which enriches the image elements in the DITA temporary topics with special attributes which list the original width and height detected from the image and it should get called before the main XSLT part of the EPUB plugin.