Integrated preprocessor patch - revised

[Zack Weinberg]

On Mon, Sep 11, 2000 at 05:21:17PM +0200, Philipp Thomas wrote:
> * Zack Weinberg (zack@wolery.cumb.org) [20000907 01:08]:
> 
> > 	* c-parse.gperf, c-gperf.h: Delete.
> 
> But you forgot to delete c-gperf.h from po/POTFILES.in :( This only shows
> when configuring with --enable-maintainer-mode, as only then gcc.pot will be
> built.

Bleah.  It is easy to forget, as you say:

>   - have a target in po/Makefile that checks the dependencies of
>     gcc.pot (listed in po/POTFILES.in) but doesn't try to build it.
> 
> That way bootstrapping gcc wouldn't fail only for me and thus such errors
> would be caught much earlier. If you agree, I'll try to come up with a patch
> that tackles the second point.

I like this idea.  Why don't you just change things so that this rule

$(srcdir)/$(PACKAGE).pot: $(POTFILES)
	$(XGETTEXT) --default-domain=$(PACKAGE) --directory=$(top_srcdir) \
	  --add-comments --keyword=_ --keyword=N_ \
	  --files-from=$(srcdir)/POTFILES.in --defines \
	&& test ! -f $(PACKAGE).po \
	   || ( rm -f $(srcdir)/$(PACKAGE).pot \
		&& mv $(PACKAGE).po $(srcdir)/$(PACKAGE).pot )

is always exposed, but without --enable-maintainer-mode the command is
'touch $(srcdir)/$(PACKAGE).pot' or similar?

Even simpler, have configure activate --enable-maintainer-mode if all
the commands it wants are found.  I think this is only xgettext at the
moment?

Unrelated question: are we ever going to go back to
--without-included-gettext default?  If you want to build a profiled
cc1 on a platform where gettext is in libc, you get link errors if you
don't use --without-included-gettext.

zw