<?xml version="1.0" encoding="UTF-8" ?>
<rss version="2.0" xmlns:admin="http://webns.net/mvcb/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:creativeCommons="http://backend.userland.com/creativeCommonsRssModule" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:html="http://www.w3.org/1999/html" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:slash="http://purl.org/rss/1.0/modules/slash/"><channel><title>Ross Burton</title><link>http://www.burtonini.com/blog</link><description>A potted account of Ross' life</description><language>en</language><ttl>60</ttl><dc:creator>Ross Burton</dc:creator><admin:generatorAgent rdf:resource="http://pyblosxom.sourceforge.net/"/><admin:errorReportsTo rdf:resource="mailto:ross@burtonini.com"/><item><title>Give Me Pain</title><guid isPermaLink="false">computers/work-20040430</guid><link>http://www.burtonini.com/blog/computers/work-20040430</link><description>Work has been... interesting of late (as in &quot;may you live in interesting times&quot;). We have the usual vital deadline ...</description><content:encoded><![CDATA[  <p>
    Work has been... interesting of late (as in "may you live in interesting times").  We have the
    usual vital deadline in 3 weeks time, a crack-ridden compiler suite from a new embedded
    processor (a 16/32-bit smart card processor by Infineon), software which used to work and is now
    for some reason broken, and the office tension is slowly creeping up.
  </p>
  <p>
    Before this damn chip turned up I was assigned to personally interesting work: document our Java
    smartcard code, critique and write use cases for an Eclipse plugin we've written, and fix the
    documentation system.  This last one is so interesting to me that I've been hacking on it every
    day this week so far on the commute, and it is starting to really kick arse.
  </p>
  <p>
    Our tools parse the code and produce simple XML documents for each file with the
    (Javadoc-inspired) documentation comments in, which we used to process into HTML with XSLT.
    This was fine, but now I'm transforming the XML into a DocBook <tt>&lt;refentry&gt;</tt> and
    XIncluding it into a master DocBook <tt>&lt;book&gt;</tt>.  A couple of Python scripts parse the
    includes, generate the intermediate <tt>&lt;chapter&gt;</tt> files and Make dependencies do
    their magic and glue it all together.  Hopefully when I've battled with Make to my satisfaction,
    instead of having an API Reference in HTML, the programming guides in
    L<sup>a</sup>T<sub>e</sub>X, and all out of date, we'll have everything in DocBook.  I'm using
    the <tt>gtk-doc</tt> DocBook customisation layer as it looks <em>far</em> nicer and also
    generates a DevHelp file for me, so I must file patches with my changes (so far tooltips on the
    navigation buttons).
  </p>
  <p>
    Of course, in 45-minutes time I'll be sitting in front of <tt>brian</tt>, one of the Windows
    boxes, wondering why the Tasking post-locator says that the ELF file generated by the Tasking
    linker is corrupt...
  </p>
]]></content:encoded><dc:date>2004-04-30T07:14:27Z</dc:date></item></channel></rss>