Tim Williams wrote: > For the archives, it looks like this is the commit [ > http://svn.apache.org/viewvc?view=revision&revision=1239111 ]. If > this is a common need, perhaps a candidate to just get rolled into > Forrest itself, I dunno.
For the document-to-fo.xsl part. > FWIW, I'm pretty sure the output.xmap is unnecessary and you could > have used a project-specific locationmap to point to your custom > document-to-fo.xsl but maybe I'm missing something... Yes i reckon that is correct. -David > --tim > > On Wed, Feb 1, 2012 at 8:02 AM, Karl Wright <daddy...@gmail.com> wrote: > > FWIW, the $path has just what I hoped in it and works well. > > > > Thanks! > > Karl > > > > On Tue, Jan 31, 2012 at 9:23 PM, Karl Wright <daddy...@gmail.com> wrote: > >>> Then, inside your own document-to-fo.xsl you should have access > >>> to a "path" parameter. You may add an additional condition to the > >>> font of interest (e.g. rootFontFamily) that uses XSL string functions > >>> against your $path parameter. > >>> > >> > >> I've set things up so that my document path includes the language code > >> (e.g. 'en_US') as the first part of the path under > >> src/documentation/content/xdocs. For example, > >> src/documentation/content/xdocs/en_US/mydocument.xml would be the > >> starting point. The question is what the $path variable will contain > >> - if it's just "en_US/mydocument.xml" my job is easy. If it could be > >> "en_US\mydocument.xml" on Windows and "en_US/mydocument.xml" on Unix > >> it's a bit harder. But if it is the absolute path xsl expressions > >> aren't going to help me and I'd better find a better solution. > >> > >> What I'd do (and this might even make a decent general patch) is look > >> for the property with the language specifier first. For example, > >> instead of looking for "output.pdf.fontFamily.sansSerif" first, I'd > >> look for "output.pdf.fontFamily.sansSerif.en_US" first, and only look > >> for the other if that property is not found. > >> > >> So, what can I expect to see for the $path parameter? > >> > >> Thanks again for your help! > >> Karl
