Mark J. Nelson wrote: > Mike Kupfer wrote: >> While updating the ON Developer Reference at the ON community page, I >> noticed that the web portal was displaying question marks in chapter >> headings. I tracked this down to the use of 0240 (0xa0) as a >> non-breaking space in the HTML that xsltproc was producing. >> >> I've made a small change to the framework for producing an HTML version >> of the DevRef and would like someone to review the changes before I >> commit them. The changes replace 0240 with " " in headings, and >> they replace 0240 with a regular space in anchor title attributes. >> (Those are the only places that 0240 appears.) >> >> Webrev is at http://cr.opensolaris.org/~kupfer/nbsp-fix/. It's a lot >> shorter than the last one I sent out. ;-)
In addition to these comments: > I was unable to observe the problem with the currently deployed ON > Developer Reference. > > In the webrev, I'm unable to discern the differences in the html > generated for the title attributes for the cross references, and it > seems like I would have expected more changes than I'm seeing. > > As an example, looking at these changes: > > http://cr.opensolaris.org/~kupfer/nbsp-fix/on-chunks/ch01.html.frames.html > http://cr.opensolaris.org/~kupfer/nbsp-fix/on-chunks/ch01.html.udiff.html > > - the change on line 1 is clear, and expected > - the change on 91 (and similar) is not clear to me, even from viewing > the page source from the udiffs > - I would have expected line 95 to show non-breaking spaces? Is this > just not getting the same treatment by xslt because it's a different > heading level? I add the following: - I looked at this a little more. The 0xa0 is coming from the definition for the "NO-BREAK SPACE" entity in usr/share/sgml/docbook/xml-dtd-4.4-1.0-30.1/ent/isonum.ent. (And, prior to version 4.4, iso-num.ent.) Surely we're not finding a fundamental error in the docbook dtd? Does my inability to reproduce the problem indicate a difference in browser/font/environment/locale/etc? - If we DO keep the changes you propose: -- the comments describing start_a() in fixup.py will need to change -- why are the results of fix_href() applied conditionally, and those from fix_title() unconditionally? (lines 163-168 in fixup.py) --Mark
