Thank you for your explanation, Bob. Do you have a DocBook test document with (nearly) all DocBook elements and combinations? During my work on the specification I'm also writing a lot of "test" code to proof the resulting PDF tag structure. Having such a document would speed up my work a bit ;-)
Another (for me) important topic I came across these days is the XSL-FO code generated by the DocBook-XSL files. They often contain deeply nested fo:block structures which will be tagged as paragraph (P) structure by FOP. This is the standard behavior that could not and will not be changed! This leads to an also deeply nested PDF tag structure. I couldn't find any solution for this problem so I wrote a Python script that optimizes the fo:block structures in side the document by merging fo:blocks as well as their attributes. You can find the script and an example (before/after) on github: https://github.com/hbast/dao Bye, Holger > Gesendet: Mittwoch, 26. April 2017 um 18:49 Uhr > Von: "Bob Stayton" <[email protected]> > An: "Holger Bast" <[email protected]>, [email protected] > Betreff: Re: [docbook-apps] tagged and accessible PDF document with DocBook > > Hi Holger, > There is a good reason simplelist and segmentedlist do not use > fo:list-block. That element can only format two columns, but both of > these kinds of lists can have any number of entries in a row, so a table > is required. In the case of simplelist, the flat list of members can be > laid out in any number of columns. In the case of segmentedlist, a > seglistitem can have any number of seg children. > > Bob Stayton > Sagehill Enterprises > [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
