John Maddock wrote:
A few issues:
1) I'm not sure if this is cygwin or quickbook issue, but it's a new
problem caused by the recent changes to mainline cvs. I'm seeing
spurious text appearing in the xml and html output. Mostly it's garbage
that appears at the end of paragraphs seemingly at random. Try building
the type traits docs and look at the "Type Traits by Category" page and
you'll see quite a few examples. I only found this, because sometimes
the extra garbage results in invalid xml so quickbook succeeds, but the
xml transform fails. The problem doesn't only occur at paragraph ends
either: I've also seen it at the start of some section titles. Looks
like a classic C array usage bug, but please tell me we're not using
those for string operations?
Hmmm... I couldn't reproduce that. I rebuilt quickdoc from the
CVS and buit the typetraits docs. I don't see any garbage.
And xml transform is not failing. Where is it actually? Which
paragraph in "Type Traits by Category"?
Did a nice elf fix the thing again while I was asleep? :)
Anyway... if not, my best guess is the CR/LF problem.
But then, IIRC, Joao already fixed that. Could you check
for CR/LF mismatch in your qbk?
2) I notice that [endsect] only works if there is a proceeding *blank
line*, otherwise a literal "[endsect]" gets put in the last paragraph in
the xml output. I'm not sure that this is a new issue, but if it's part
of the grammar then it really should be documented. It's also
asymmetrical, since [section] tags don't need a newline following them,
is there any reason why [endsect] shouldn't always be special, and
implicitly end a paragraph?
I'll check this out.
3) If there are mismatched [section] and [endsect] tags quickbook
doesn't care, but the xml parser obviously does. It would be nice if
this could be flagged up in quickbook, otherwise the error is very hard
to track down, especially in a big document.
Hmmmm.. It seems you have an older version of quickbook.
I did a fix for where mismatched section/endsect tags
are flaged as errors sometime last week. Do you have a sample
qbk file that exhibits the problem?
4) This is probably a wish for a new feature, but: if I specify a link,
but don't specify any text for the link, I assumed I would get an <xref>
xml element which would then fill in the title for you based on the
BoostBook setup. Instead I get a <link> with the title set to the id of
the link, which is pretty useless really :-) Yes, I realise this is
documented (now that I look!), but I honestly can't think of a use for
the current behaviour.
Ok... where can I learn about the "BoostBook setup" ?
I must've missed that.
Regards,
--
Joel de Guzman
http://www.boost-consulting.com
http://spirit.sf.net
-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
Boost-docs mailing list
[email protected]
Unsubscribe and other administrative requests:
https://lists.sourceforge.net/lists/listinfo/boost-docs
