On Mon, 03 Oct 2011 21:48:24 +0200
Andreas Hoenen <andr...@hoenen-terstappen.de> wrote:

> Jeffrey B. Green <j...@kikisoso.org> wrote:
> 
> <snip/>
> > It needs to be able to handle characters with umlauts and carets
> > expressed as a character entity references correctly (which it
> > does), and without the encoding setting it doesn't, i.e.
> >
> > <title> car&ecirc;t &ouml;laut</title>
> >
> > Hence I used it. It worked beautifully until the footnote problem.
> <snip/>
> > Your stylesheet indeed works in that the characters display
> > properly.
> <snip/>
> > Do you know offhand what aspects of the stylesheet you provided is
> > getting the character entity references to display correctly?
> <snip/>
> 
> I'm confused.  My stylesheet is optional, it shouldn't have any effect
> on entity references in the document's title.  Could you give a short
> example document that has problems without any user stylesheet?
> 

Hi Andreas,

It seems that an additional docbook.xsl file was imported via a very
innocent and slightly hidden line that was interacting with the
xsl:output statement. At this point I'm not sure why it was added
other than at some point some sort of encoding issue had arisen. It was
added several years ago without an explanatory note for that change. It
was mixed in with other modifications. Without it everything seems to
work okay, at least latex-wise. HTML output is also generated from this
file, so I'll need to check that. 

So at this point all seems to be working okay with the import and the
encoding lines removed.

Again, thanks for helping out on this. I'll go ahead and close this bug
after a check of the outputs from the various sources that have those
lines removed.

-jeff



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Reply via email to