Geir's experiment was incomplete and he reported difficulty with some docs. Perhaps the following will help, or perhaps you already have a better solution.
There is a way to use Forrest itself to produce xml output ... ---------------------------- 0) Assuming that forrest is installed as per: http://incubator.apache.org/guides/website.html ---------------------------- 1) Temporarily add the "xml-link". This will cause forrest to also generate an xml version of each page. edit site-author/skinconf.xml at line 46 - <disable-xml-link>true</disable-xml-link> + <disable-xml-link>false</disable-xml-link> I gather that when Stefano built the orginal DTDs for Forrest that he might have started from the Anakia xml format. Anyway, whatever, the formats are close. ---------------------------- 2) Clean up the output xml * Remove the document type declaration. It seems that Anakia xdocs don't have one. * Transform some elements: document/header => document/properties link => a ---------------------------- 3) Copy all of the *.xml documents from site-publish/**.xml to incubator/site/xdocs/test2 In my experiment, i copied some test documents to infrastructure/site/ i.e. the www.a.o/dev/ etc. which also uses Anakia. They are processed reasonably well there. Geir's stylesheet seems to have different handling of section/title elements. ---------------------------- 4) Run anakia. cd incubator/site; ant ---------------------------- 5) Fix the links. This workaround is just saving the internal Forrest format to disk, so links that use Forrest's linking system are broken. ---------------------------- 6) Fix some other issues. In my experiment, the site-author/projects/index.html table is transformed well, but not with Geir's stylesheet (perhaps same issue as above). I also tested some status reports, some transform okay but some do not, perhaps there are problems with some of the source html. ---------------------------- It is interesting to note that Nicola Ken long ago moved the Incubator site to use html as the source format instead of the orginal xml - presumably people had problems editing xml. That has made it harder to convert the site back to xml again. -David --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
