Monday, January 10, 2011

Kernow 1.7 released


Kernow 1.7 is now available:
  • integrates LexEv (which now reports internal and parameter entities, along with CDATA sections, comments, numeric and entity refs, and the DOCTYPE)
  • On OSX and linux, it uses the "Nimbus" look and feel, on Windows it still uses native
  • includes the latest Saxon HE 9.3.0.2
  • directory transforms: input directory structure is now recreated in the output directory
  • files are correctly released, rather than held while Kernow is open
  • the caching entity resolver now correctly load all files from the cache dir first
  • updated to the latest Bounce version, which fixes some freezing in the sandboxes
  • includes Xerces 2.10 which partially supports XSD 1.1
  • fixed several small bugs
Kernow 1.7 is released as a trial version which allows you to hit the run button 100 times, after which you will need to enter a code to continue. The code can be purchased using the Help -> Purchase menu.

Monday, July 19, 2010

Kernow 1.7 beta available

Kernow 1.7 beta is now available:
  • integrates LexEv (which now reports internal and parameter entities, along with CDATA sections, comments, numeric and entity refs, and the DOCTYPE)
  • includes Saxon HE 9.2.1.1
  • directory transforms: input directory structure is now recreated in the output directory
  • files are correctly released, rather than held while Kernow is open
  • the caching entity resolver now correctly load all files from the cache dir first
  • updated to the latest Bounce version (enabling "folding" in the sandboxes)
  • includes Xerces 2.9.0 which partially supports XSD 1.1
  • fixed several small bugs

Tuesday, November 17, 2009

Mark Logic adds XSLT support

Mark Logic currently lacks any support for XSLT... but that's about to change: Norm Walsh has announced upcoming support for it in version 5.

Monday, September 14, 2009

XML Schema 1.1 tutorials

Roger Costello has created a couple of good powerpoint presentations on XSD 1.1
One tutorial is for developers, the other for managers which contains more general wordy descriptions of the benefits of 1.1

Friday, August 22, 2008

Some sample templates for use with LexEv

If your XML has been parsed using LexEv, here are some sample templates for handling the LexEv markup.

To output an entity reference:



<xsl:template match="lexev:entity">
<xsl:value-of disable-output-escaping="yes" select="concat('&amp;', @name, ';')"/>
</xsl:template>



To process a CDATA section as markup:


<xsl:template match="lexev:cdata">
<xsl:apply-templates/>
</xsl:template>


To output a DOCTYPE from the processing instructions:

In XSLT 1.0 the doctype-public and doctype-system attributes on xsl:output are static and need to be known at compile time, which means I'm afraid you have to do this:


<xsl:template match="/">
<xsl:value-of disable-output-escaping="yes"
select="concat('&lt;!DOCTYPE ', name(/*), '&#xa; PUBLIC &quot;',
processing-instruction('doctype-public'), '&quot; &quot;',
processing-instruction('doctype-system'), '&quot;&gt;')"/>
<xsl:apply-templates/>
</xsl:template>


In XSLT 2.0 you can use xsl:result-document where the doctype-public and doctype-system are AVTs which mean their values can be determined at runtime:


<xsl:template match="/">
<xsl:result-document
doctype-public="{processing-instruction('doctype-public')}"
doctype-system="{processing-instruction('doctype-system')}">
<xsl:apply-templates/>
</xsl:result-document>
</xsl:template>

Thursday, August 21, 2008

LexEv XMLReader - converts lexical events into markup

It's often a requirement to preserve entity references through to the output (which are usually lost during parsing) or to process the contents of CDATA sections as markup. The Lexical Event XMLReader wraps the standard XMLReader to convert lexical events into markup so that they can be processed. Typical uses are:

  • Converting cdata sections into markup:


    <![CDATA[ &lt;p&gt; a para &lt;p&gt; ]]>

    to:

    <lexev:cdata> <p> a para </p> </lexev:cdata>



  • Preserving entity references:


    hello&mdash;world

    is converted to:

    hello<lexev:entity name="mdash">—</lexev:entity>world


  • Preserving the doctype declaration:


    <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

    is converted to processing instructions:

    <?doctype-public -//W3C//DTD XHTML 1.0 Transitional//EN?>
    <?doctype-system http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd?>


  • Marking up comments:


    <!-- a comment -->

    is converted to:

    <lexev:comment> a comment </lexev:comment>


To use LexEvXMLReader with Saxon:


java -cp saxon9.jar;LexEvXMLReader.jar net.sf.saxon.Transform -x:com.andrewjwelch.lexev.LexEvXMLReader input.xml
stylesheet.xslt


Make sure LexEvXMLReader.jar is on the classpath, and then tell Saxon to use it with the -x switch (copy and paste this line -x:com.andrewjwelch.lexev.LexEvXMLReader)


To use LexEvXMLReader from Java:

XMLReader xmlReader = new LexEvXMLReader();


You can control the following features of LexEv:


  • enable/disable the marking up of entity references

  • enable/disable the marking up of CDATA sections

  • set the default namespace for the CDATA section markup

  • enable/disable the reporting of the DOCTYPE

  • enable/disable the marking up of comments


You can set these through the API (if you are including LexEv in an application), or from the command line using the following system properties:


  • com.andrewjwelch.lexev.inline-entities

  • com.andrewjwelch.lexev.cdata

  • com.andrewjwelch.doctype.cdataNamespace

  • com.andrewjwelch.lexev.doctype

  • com.andrewjwelch.lexev.comments


For example to set a system property from the command line you would use: -Dcom.andrewjwelch.lexev.comments=false


For support, suggestions and licensing, email lexev@andrewjwelch.com

Friday, July 18, 2008

Kernow 1.6.1

Kernow 1.6.1 (beta) is now availble both as a download and via web start.

Notable things in this release:

- Line numbers on the editor panes in the sandboxes (thanks to a new version of Bounce). You might not think so, but getting line numbers down the side of the editor pane is really involved. It's like block indenting (pressing tab or shift-tab when a block of text is selected) in that it's very low level and requires a lot of coding. Why it's not an intergral part of the editor pane I don't know...

- Improved the syntax-checking-as-you-type and highlighting, and added the ability to disable it.

- The output area is now also a JEditorPane using Bounce so it supports tag highlighting. This might slow things down because now it's an HTML document where every addition is inserted at the end of the document, instead of just appending to a JTextArea... if this proves to be A Bad Thing I'll revert it back to a plain old text area with plain text.

- You can now select which tabs are visible (in options -> tabs) so if you never use certain tabs (like Batch or Schematron) you can remove them.

- If you have Saxon SA you can use XML Schema 1.1 (options -> validation)

- Improved the parameters dialog to make it less fiddly to enter params

- Slight graphical tweaks and likely other things that I've forgotten...