Werner LEMBERG wrote:
> [forwarded to the groff list also]
>
>> One small problem is that it seems that either groff is no longer
>> using the number register called ":p" for storing the footnote
>> numbers, or else the instruction ".nr :p 0" isn't taking effect. So
>> we can't reset the footnote number to 0 at the start of each new
>> chapter.
>
> I'm no expert in mm; hopefully, someone on the list can help here...
>
> However:
>
>> Oh, I also get a segmentation fault in grohtml if I try to convert my
>> wife's thesis to html with -Thtml (in contrast to -Tps).
>
> This is something Gaius certainly want to investigate. Before further
> investigation I ask you to download the CVS version of groff, testing
> whether the crash is still triggered.
>
>> If you're interested, I can provide you with the data files.
>
> Please do so in case the CVS version doesn't improve the situation.
I fetched it, and updated texinfo so configure would work
>> it uses tbl in stressful ways, one footnote is so long that I had to
>> add a .vs -4 to avoid a bug in groff overprinting the following page
>> over the footnote body,
>
> This might be another bug in groff's mm implementation.
>
>> and it even uses pic to overdraw a line over a table on one page,
>> via the use of diversions).
>
> Is this a bug or a special feature of your wife's thesis?
It's a special feature of the thesis. On the original, we needed to
draw a smooth curve through a table of data; we couldn't work out an
easy way to do it at the time, so we just drew it by hand on the proof
copy before sending it to the book binder.
Since we're creating a PDF, that kludge won't work, so I figured out a
way to overdraw a pic figure over the top of the table. I should
improve the rough xfig curve I created, but it gives readers the rough
idea of what the original had.
One small oddity of the new source is that make clean now fails:
make[2]: Entering directory
`/home/stella/Archives/thesis/groff-src/groff-1.19.3-pre/src/include'
make[2]: /home/stella/Archives/thesis/groff-src/groff/Makefile.comm: No
such file or directory
make[2]: /home/stella/Archives/thesis/groff-src/groff/Makefile.man: No
such file or directory
make[2]: *** No rule to make target
`/home/stella/Archives/thesis/groff-src/groff/Makefile.man'. Stop.
make[2]: Leaving directory
`/home/stella/Archives/thesis/groff-src/groff-1.19.3-pre/src/include'
make[1]: *** [src/include] Error 2
make[1]: Leaving directory
`/home/stella/Archives/thesis/groff-src/groff-1.19.3-pre'
make: *** [clean] Error 2
Anyway, it still seems to crash.
: /home/stella/Archives/thesis/finalthesis.2007; groff --version
GNU groff version 1.19.3
Copyright (C) 2006 Free Software Foundation, Inc.
GNU groff comes with ABSOLUTELY NO WARRANTY.
You may redistribute copies of groff and its subprograms
under the terms of the GNU General Public License.
For more information about these matters, see the file named COPYING.
called subprograms:
GNU grops (groff) version 1.19.3
GNU troff (groff) version 1.19.3
: /home/stella/Archives/thesis/finalthesis.2007; make html
make: *** No rule to make target `html'. Stop.
: /home/stella/Archives/thesis/finalthesis.2007; vi Makefile
: /home/stella/Archives/thesis/finalthesis.2007; make html
cat .whole \
| sed 's/\\>/Q#Q#Q/g;s/>/ /g;s/Q#Q#Q/>/g' \
| groff -t -rO0.625i -rW5.5i -rL11.0i -mm -Thtml >
thesis.html 2> er
make: [html] Error 2 (ignored)
tbl:stdin:30822: multiple widths for column 5
tbl:stdin:30842: multiple widths for column 5
post-grohtml:<standard input> (stdin):30: bad font position `-1'
groff: post-grohtml: Signal 11 (core dumped)
I don't understand the tbl warning (since the other tables don't produce
that warning, nor do most of the rows in that specific table), and it
produces good Postscript.
>
> Werner
I hesitate to send all the data files to the list simply because they're
500k, so I'll send that just to you and Gaius separately, Werner, zipped
up. My wife is happy to provide the thesis as test data
Thanks for looking into this! She in particular can't help comparing
how helpful you all are, compared to how helpful Microsoft is with bugs
in their software. :-)
And I'm more than ever convinced of the importance of ascii data for
documents. It only took a day to get a 20 year old document reproduced
and looking better than the original!
Regards,
luke
