[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: gnulib-izing sharutils
From: |
Bruce Korb |
Subject: |
Re: gnulib-izing sharutils |
Date: |
Sun, 30 Aug 2009 15:13:09 -0700 |
User-agent: |
Thunderbird 2.0.0.21 (X11/20090310) |
Hi Bruno,
Bruno Haible wrote:
> Hi Bruce,
>
>> gnulib-izing sharutils
>
> You are welcome to send a one-line patch to gnulib/users.txt!
Sure. Be happy to. First, though, I think I need to get sharutils working....
>>> Notice from module error:
>>> If you are using GNU gettext version 0.16.1 or older, add the following
>>> options
>>> to XGETTEXT_OPTIONS in your po/Makevars:
>>> --flag=error:3:c-format --flag=error_at_line:5:c-format
>
> It cannot be easily suppressed, because gnulib-tool is run before the
> gettext macros are collected by 'aclocal'.
>
> Would it remove the confusion if I would add the following sentence to the
> notice?
>
> "If you are using GNU gettext version 0.17 or newer, you can ignore this
> notice."
I did figure that much out. At issue is whether this is strictly relevant
for the tool manufacturer (e.g. "me" for sharutils), or the builder
(the aforementioned "hapless builder of sharutils"). I also thought
that gtver=$(gettext --version|head -1|sed 's/.*) *//') would readily
answer the question. At least for the tool manufacturer. :)
> The order should be
>
> AC_PROG_CC
> gl_EARLY
> gl_INIT
>
> The doc
> <http://www.gnu.org/software/gnulib/manual/html_node/Initial-import.html>
> suggests
> "invoke gl_EARLY in ./configure.ac, right after AC_PROG_CC,"
> and
> "... the macro gl_INIT. Place it further down in the file, typically where
> you normally check for header files or functions."
OK. I'm convinced. The gnulib (and autoconf!!) need a manual section that
categorizes the various macros and specifies the proper ordering of the
categories. That tidbit was 3/4 the way down a long web page among plenty
of other web pages one has to read and understand. Maybe I'm dumb and lazy,
but I do find it difficult to juggle it all.....
> But there is a known problem with AC_USE_SYSTEM_EXTENSIONS in autoconf <=
> 2.63.
> I'd suggest to use autoconf 2.64 and then remove all invocations of
> AC_GNU_SOURCE.
Since I don't have any at all, that should be fairly simple. However:
> $ autoconf --version
> autoconf (GNU Autoconf) 2.64
and I still see the warning: configure.ac:22: warning: The macro
`AC_GNU_SOURCE' is obsolete.
(and, no, line 22 does not say AC_GNU_SOURCE on it.)
>> * I get a whole mess of messages like this. Since none of this appears in
>> either
>> configure.ac nor m4/sharutils.m4, I'm guessing some macro some where pulls
>> it in.
>> It is hard to trace though:
>>> $ ls ../../../lib/autoconf
>>> ls: cannot access ../../../lib/autoconf: No such file or directory
>> So, to the messages:
>>
>> configure.ac:21: warning: The macro `AC_TRY_RUN' is obsolete.
>
> You can ignore these warnings. AC_TRY_RUN and AC_TRY_LINK cannot go away
> because hundreds of autoconf macros use them.
Then the warnings should go away:
>> configure.ac:21: note: The macro `AC_TRY_RUN' is an obsolescent macro. \
>> please avoid using it.
Such a message would be much less disconcerting. :)
The whole thing ends with this:
> configure: creating ./config.status
> config.status: creating src/remsync
> config.status: creating src/compress-dummy
> config.status: creating contrib/Makefile
> config.status: creating doc/Makefile
> config.status: creating intl/Makefile
> config.status: creating lib/Makefile
> config.status: creating po/Makefile.in
> config.status: creating src/Makefile
> config.status: creating src/mail-files
> config.status: creating src/mailshar
> config.status: creating tests/Makefile
> config.status: error: cannot find input file: `Makefile.in
> $ ls -l Make*
> -rw-r--r-- 2 bkorb dev 1049 Aug 30 11:18 Makefile.am
> $ sed -n '/^AC_CONFIG_FILES(/,/AC_OUTPUT/p' configure.ac
> AC_CONFIG_FILES([
> contrib/Makefile
> doc/Makefile
> intl/Makefile
> lib/Makefile
> po/Makefile.in
> src/Makefile
> src/mail-files
> src/mailshar
> tests/Makefile
> Makefile
> ])
> AC_OUTPUT
Would you like the entire configure run output? Would that give
some more clues? Thank you, Bruno, for your help!!
Regards, Bruce