Douglas Gregor writes:
> I think we should go for option #1 all the way, and use the hooks
> that the DocBook DTD gives us.
You spoke my mind; this is exactly what I think would be best!
> [XIncludes]
I can see the problem. From an aesthetic point of view, it would be
nice if all the documents were effectively one to the parser so that
we are guaranteed to have unique IDs throughout the documentation --
among other things. This would also allow for sophisticated
cross-indexing, etc. I personally tend to organize my documents so
that the "main" document just includes the bits from other files. This
way, I am free to refer to those parts from other documents as well.
It is clear, though, that we cannot expect all documentation authors
to adapt this technique. Particularly not, if they're working with
tools that do not directly support split documents the way Emacs/psgml
does.
My best guess would be that we -- at least for the moment -- stick to
XInclude and see what happens later. This is not an essential issue
for our task right now.
I added definitions for the XInclude elements to the DTD, but they're
not integrated into DocBook yet, meaning that the definitions
effectively do nothing. xmllint(1) will verify the documents fine if
you use "-postvalid" instead of "-valid", because then the validator
won't see the XInclude tags anymore.
-peter
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Boost-docs mailing list
[EMAIL PROTECTED]
Unsubscribe and other administrative requests:
https://lists.sourceforge.net/lists/listinfo/boost-docs
