bug-gnulib
[Top][All Lists]
Advanced

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

Re: more m4 quoting


From: Jim Meyering
Subject: Re: more m4 quoting
Date: Tue, 13 Jan 2009 23:19:39 +0100

Bruno Haible <address@hidden> wrote:
>> I'm about to apply the changes induced by the above commands.
>
> What's the rationale of quoting purely numeric m4 arguments, like in
>    AC_DEFINE([FOO], [1], ...)  ?
> It looks a bit like clutter to me,

Initially I didn't do it, for just that reason,
but once Eric suggested it, and especially, once I
wrote the more general perl-based quote-adding transformer,
I decided I preferred the consistency.

> since quoting does not matter here. But
> if you and Eric and Simon are willing to do this purely for consistency,
> I'm going with it as well.
>
> Please exclude m4/intl.m4 - it is frozen until the next gettext release,
> see <http://lists.gnu.org/archive/html/bug-gnulib/2008-12/msg00298.html>.
> You are free to change the other gettext-borrowed .m4 files; I'll port
> back the changes to gettext.

Ok. thanks.

> Please don't do these 3 changes to m4/lib-link.m4:
> -dnl AC_LIB_HAVE_LINKFLAGS(name, dependencies, includes, testcode)
> +dnl AC_LIB_HAVE_LINKFLAGS([name], [dependencies], [includes], [testcode])
> -dnl AC_LIB_FROMPACKAGE(name, package)
> +dnl AC_LIB_FROMPACKAGE([name], [package])
> -dnl AC_LIB_APPENDTOVAR(VAR, CONTENTS) appends the elements of CONTENTS to 
> VAR,
> +dnl AC_LIB_APPENDTOVAR([VAR], [CONTENTS]) appends the elements of CONTENTS 
> to VAR,
> (Brackets in the comments usually denote optional parameters.)
>
> Likewise in m4/onceonly.m4 and m4/openmp.m4.

Done.

> Could you please bump the serial number of affected files? autopoint
> now considers the serial numbers of .m4 files. And the copyright year,
> of course.

Yep.  It was while preparing to update serial numbers that I noticed
all of the (gettext-...) notations.  It looks like I'll be automating
it with something like this:

  # normalize a few serial number lines
  g log -1 --name-only --pretty=oneline|tail -n+2|xargs grep -l serial \
    | perl -pi.bak -e 's,^#(serial \d),# $1,'

  # increment them
  g log -1 --name-only --pretty=oneline|tail -n+2|xargs grep -l serial \
   |xargs perl -pi.bak -e 's,^(\# (?:[\w-]+\.m4 )?serial +)(\d+),$1 . ($2+1),e'

  but that missed one, which I did manually:
  -# iconv.m4 serial AM6 (gettext-0.18)
  +# iconv.m4 serial AM7 (gettext-0.18)

> In m4/argp.m4, the second argument of two AC_DEFINE invocations is not
> quoted by your patch.

Sure.  FYI, they were missed due to different (nonstandard) spacing.

> In m4/signalblocking.m4, the fourth argument of AC_EGREP_HEADER is not
> quoted by your patch.

Done as well.

Thanks.

>> I've taken the unusual step of attaching the compressed diff below.
>
> This is welcome; how could I have reviewed the patch otherwise?

It was unusual in that it was compressed.
Since the patch was 100% generated using the perl code
in the log, you could have run those commands followed
by "git diff" to get a patch identical to the one I posted.

I'll push tomorrow after a little more testing.




reply via email to

[Prev in Thread] Current Thread [Next in Thread]