While I am new to using docbook, I'm quite sure this is not supposed to happen. My dtd is: <!DOCTYPE article PUBLIC "-//OASIS//DTD DocBook V4.1//EN"> sgmltools -b ps file.sgml generates the following error: [snip] Traceback (most recent call last): File "/usr/bin/sgmltools", line 93, in ? tool.processFile(curfile) File "/usr/share/sgml/misc/sgmltools/python/SGMLtools.py", line 230, in processFile dssslfile = utils.findStylesheet(stylesheet, self._aliases) File "/usr/share/sgml/misc/sgmltools/python/utils.py", line 307, in findStylesheet return findStylesheet(name, aliases) File "/usr/share/sgml/misc/sgmltools/python/utils.py", line 317, in findStylesheet raise IOError, "Couldn't resolve pubid [%s]" % id IOError: Couldn't resolve pubid ["-//SGMLtools//DOCUMENT Docbook Style Sheet for Print//EN"] [snip] Trying this without attempting a conversion to postscript but just html: [snip] Traceback (most recent call last): File "/usr/bin/sgmltools", line 93, in ? tool.processFile(curfile) File "/usr/share/sgml/misc/sgmltools/python/SGMLtools.py", line 230, in processFile dssslfile = utils.findStylesheet(stylesheet, self._aliases) File "/usr/share/sgml/misc/sgmltools/python/utils.py", line 307, in findStylesheet return findStylesheet(name, aliases) File "/usr/share/sgml/misc/sgmltools/python/utils.py", line 317, in findStylesheet raise IOError, "Couldn't resolve pubid [%s]" % id IOError: Couldn't resolve pubid ["-//SGMLtools//DOCUMENT Docbook Style Sheet for HTML//EN"] [snip] The SGML_CATALOG_FILES environment is: /etc/sgml/sgml-ent.cat:/etc/sgml/sgml-docbook.cat:/etc/sgml/openjade-1.3.1.cat:/etc/sgml/sgml-lite.cat:/etc/sgml/xhtml1.cat:/etc/sgml/html401.cat Is there something that can be done as a workaround so I can continue to hack around with docbook? Please let me know and I can provide any more info.
http://cvs.gentoo.org/~satai/sgmlfix.html
Seems like an odd way to fix things, but I'll bite. Is this simply a config file issue?
No, that has no effect. Is there any info I can send to help this along? The only thing that changed was the environment variable: [snip] paul cups # echo $SGML_CATALOG_FILES /etc/sgml/sgml-ent.cat:/etc/sgml/sgml-docbook.cat:/etc/sgml/dsssl-docbook-stylesheets.cat:/etc/sgml/openjade-1.3.1.cat:/etc/sgml/sgml-docbook-3.0.cat:/etc/sgml/sgml-docbook-3.1.cat:/etc/sgml/sgml-docbook-4.1.cat:/etc/sgml/sgml-docbook-4.0.cat:/etc/sgml/sgml-lite.cat:/etc/sgml/html401.cat:/etc/sgml/xhtml1.cat [snip]
A little more problem solving today. Executing the following command works with the dtd in my bug report: sgmltools -b pdf -s /usr/share/sgml/stylesheets/sgmltools/print.dsl test.sgml This seems to work for all backends so far. However, adding the catalog file to the SGML_CATALOG_FILES environment variable does not work.
Try putting quotation marks around the catalog files in /etc/sgml/catalog, see if that helps any. I'm working on a unified solution to this, but it might be a long time in coming, unfortunately.
*** This bug has been marked as a duplicate of 12792 ***
No, that no workie. It's also odd that if I create my own envitonment variable the program fails as well. Hmm. Well, I'll wait for the fix, in the meatime I'll keep specifying my dssl spec files.
Ok, then let's reopen. We'll see if we can nail this problem. I've got some ideas on how to do it, so let me work on them for a while and I'll get back to you.
Have you tried with the latest sgml-common? It's masked in package.mask.
I did some testing, and I got it to work with the new sgml-common, which has a change in the way catalogs are recorded. Let me know how that works for you. :)
OK. Will give it a shot.
Are you wanting me to test sgml-common r4? If so, that one works with docbook anyway. I havent tried it with any other catalogs though.
Well, I've been using it for most of the day and it appears to be working 100% now. In different formats: pdf, html, dvi, postscript and, rtf it is working well (the html did not work but I thought it was my fault). I have the ability to generate docbook 4.1/4.2 no problems without any extra command line. Are there any tests I can run for you?
Are you going to mark r4 stable and this closed?
Not stable; but it is arch masked, instead of package.mask'd. Sorry it took so long. As soon as I'm able I'll mark it stable.
Actually, closing this bug.
