[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[debian-devel:11584] Please test experimental groff 1.15-3.ja.1 (Re: NMU release of jgroff 1.15+ja-3.1 uploaded (Re: Re: Installed man-db 2.3.10-70 (source i386)))

At Fri, 11 Feb 2000 15:09:40 +0200,
Fabrizio Polacco <fab@xxxxxxxx> wrote:

> > Would you change man-db's depend line like this, please?
> > 
> >  Depends: groff (>> 1.15-2) | jgroff (>> 1.15+ja), ${shlibs:Depends}
> ok, I was just to upload a new man-db :-)

Thanks a lot.

> > Anyway, I don't think jgroff and groff are ready to merge, because
> > this ja.patch might have an effect on handling 8bit chars, especially
> > latin chars.  I'm not sure this ja.patch is not harmfull for latin texts.
> There's only one way: test it :-)

Yes, of course.
> Right now I'm uploading groff_1.15-3.ja.1_i386.deb into experimental.
> It's the 1.15-3 in frozen plus the patch, nothing else.

I tried it and see the diff.gz. I have some comments.

* 1.15-3.ja.1 has same bug as Bug#47800.
  In debian/rules build,

	cd xditview; $(MAKE) depend;  \

  So, DEFINES=".... -DNIPPON" will be overrided CCFLAGS/CFALGS supplied by
  this command line.  I think it should be, for example,

	cd xditview; $(MAKE) depend;  \

  or need not pass CCFLAGS and CFLAGS.

* jgroff patch will select dvi device at configure time.
  I think it's better that we have two type of dvi device, for example,
  dvi (modified for NTT-JTeX) and dvi-ascii (for ASCII pTeX).

* eqn/eqn.cc and eqn/eqn.tab.h are autogenerated from eqn/eqn.y 
  so you can remove it in debian/rules clean
* some .orig file still exist.  

> It should be quite the same as the other package, I suppose, but we can
> ask people in debian-i18l to use it instead of frozen groff, and report
> problems in their environment.

I agree, so I'll send this to debian-i18n.
Please test groff 1.15-3.ja.1 in experimental and report whether it works
for your environment.

Fumitoshi UKAI