Hello, On Fri, 7 Jan 2011 13:20:24 -0500 Mike Frysinger <[email protected]> wrote:
> i dont think this patch makes sense. it may have been necessary in
> the past where m-i-t didnt ship man pages as part of the distr, but
> now that that has been fixed, i think any remaining invocations are a
> distro problem. looking at buildroot, i dont see why it is necessary
> with m-i-t 3.12. docbook2man shouldnt be invoked for the normal
> unpack/configure/make cycle.
>
> if you dont want man pages *regenerated* because you're patching
> files, then simply touch them:
> touch *.[58]
I'm not patching any file in module-init-tools, and still it wants to
regenerate man pages with docbook2man:
if [ "docbook2man" = "docbook2man" ]; then \
docbook2man doc/modprobe.d.sgml > /dev/null 2>&1;
\
else \
docbook2man doc/modprobe.d.sgml 2>&1 > modprobe.d.5 | sed
's/^[^:]*://'; \
fi
make[1]: *** [modprobe.d.5] Error 127
make[1]: *** Waiting for unfinished jobs....
This isn't surprising, since in the freshly uncompressed tarball, one can see:
module-init-tools-3.12$ ls -l modprobe.d.5 doc/modprobe.d.sgml
-rw-r--r-- 1 test test 10460 May 4 2010 doc/modprobe.d.sgml
-rw-r--r-- 1 test test 20 Mar 10 2010 modprobe.d.5
The .sgml file is newer than the .5 file, so the following Makefile.am
rule gets triggered, unless I'm missing something:
%.5: doc/%.sgml
if [ "$(DOCBOOKTOMAN)" = "docbook2man" ]; then \
$(DOCBOOKTOMAN) $< > /dev/null 2>&1; \
else \
$(DOCBOOKTOMAN) $< 2>&1 > $@ | sed 's/^[^:]*://'; \
fi
So, I think that the patch is still necessary, or at least some
solution that does not require docbook2man to build module-init-tools.
Thanks!
Thomas
--
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com
signature.asc
Description: PGP signature
