GNU bug report logs -
#11294
[RFC] build: support and require Automake-NG
Previous Next
To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 11294 in the body.
You can then email your comments to 11294 AT debbugs.gnu.org in the normal way.
Toggle the display of automated, internal messages from the tracker.
Report forwarded
to
bug-coreutils <at> gnu.org
:
bug#11294
; Package
coreutils
.
(Sat, 21 Apr 2012 09:50:02 GMT)
Full text and
rfc822 format available.
Acknowledgement sent
to
Stefano Lattarini <stefano.lattarini <at> gmail.com>
:
New bug report received and forwarded. Copy sent to
bug-coreutils <at> gnu.org
.
(Sat, 21 Apr 2012 09:50:02 GMT)
Full text and
rfc822 format available.
Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):
* configure.ac (AM_INIT_AUTOMAKE): Add the 'ng' option, to ensure that
mainstream Automake is not used by mistake when bootstrapping. Also,
bump the required Automake version from '1.11.1' to '1.11e', which is
the latest (and still development-only) version of Automake-NG at the
moment of writing.
* bootstrap (check_versions): Hacked to handle automake and aclocal
from Automake-NG specially. This change should be backported to Gnulib
proper in a later step.
* bootstrap.conf ($buildreq): Require "automake-ng" and "aclocal-ng"
version >= 0.5; don't require mainstream "automake" anymore.
Signed-off-by: Stefano Lattarini <stefano.lattarini <at> gmail.com>
---
I'd like this to be applied to an experimental branch in the coreutils
repository, which will be used to test and experiment with Automake-NG
in a real-world, important, medium-complexity package like GNU coreutils
is.
I hope you'll agree this is a sensible move, which could bring advantages
and improvements to both coreutils and Automake-NG.
Regards,
Stefano
bootstrap | 19 +++++++++++++++++++
bootstrap.conf | 3 ++-
configure.ac | 2 +-
3 files changed, 22 insertions(+), 2 deletions(-)
diff --git a/bootstrap b/bootstrap
index c8ee3cc..cc058bd 100755
--- a/bootstrap
+++ b/bootstrap
@@ -429,6 +429,25 @@ check_versions() {
GZIP) ;; # Do not use $GZIP: it contains gzip options.
*) eval "app=\${$appvar-$app}" ;;
esac
+
+ # Special handling for Automake-NG programs. They are still named as
+ # the mainstream Automake programs ("automake", "aclocal") to avoid
+ # gratuitous incompatibilities with pre-existing usages (by, say,
+ # autoreconf, or custom autogen.sh scripts), but correctly identify
+ # themselves (as being part of "GNU automake-ng") when asked their
+ # version.
+ case $app in
+ automake-ng|aclocal-ng)
+ app=`echo "$app" | sed 's/-ng$//'`
+ if ($app --version | grep '(GNU automake-ng)') >/dev/null 2>&1; then
+ :
+ else
+ echo "$me: Error: '$app' not found or not from Automake-NG" >&2
+ ret=1
+ continue
+ fi ;;
+ esac
+
if [ "$req_ver" = "-" ]; then
# Merely require app to exist; not all prereq apps are well-behaved
# so we have to rely on $? rather than get_version.
diff --git a/bootstrap.conf b/bootstrap.conf
index bb414ef..a2432aa 100644
--- a/bootstrap.conf
+++ b/bootstrap.conf
@@ -307,8 +307,9 @@ gnulib_tool_option_extras="--tests-base=gnulib-tests --with-tests --symlink\
# Build prerequisites
buildreq="\
+automake-ng -
+aclocal-ng -
autoconf 2.62
-automake 1.11.1
autopoint -
bison -
gettext 0.17
diff --git a/configure.ac b/configure.ac
index 5a4860e..dab8b07 100644
--- a/configure.ac
+++ b/configure.ac
@@ -32,7 +32,7 @@ AC_CONFIG_SRCDIR([src/ls.c])
AC_CONFIG_AUX_DIR([build-aux])
AC_CONFIG_HEADERS([lib/config.h:lib/config.hin])
-AM_INIT_AUTOMAKE([1.11.1 no-dist-gzip dist-xz color-tests parallel-tests])
+AM_INIT_AUTOMAKE([1.11e ng no-dist-gzip dist-xz color-tests parallel-tests])
AM_SILENT_RULES([yes]) # make --enable-silent-rules the default.
dnl POSIXCHECK is worthwhile for maintainers, but adds several seconds
--
1.7.9.5
Information forwarded
to
bug-coreutils <at> gnu.org
:
bug#11294
; Package
coreutils
.
(Tue, 08 May 2012 15:07:02 GMT)
Full text and
rfc822 format available.
Message #8 received at submit <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
On 04/21/2012 11:48 AM, Stefano Lattarini wrote:
> * configure.ac (AM_INIT_AUTOMAKE): Add the 'ng' option, to ensure that
> mainstream Automake is not used by mistake when bootstrapping. Also,
> bump the required Automake version from '1.11.1' to '1.11e', which is
> the latest (and still development-only) version of Automake-NG at the
> moment of writing.
> * bootstrap (check_versions): Hacked to handle automake and aclocal
> from Automake-NG specially. This change should be backported to Gnulib
> proper in a later step.
> * bootstrap.conf ($buildreq): Require "automake-ng" and "aclocal-ng"
> version >= 0.5; don't require mainstream "automake" anymore.
>
> Signed-off-by: Stefano Lattarini <stefano.lattarini <at> gmail.com>
> ---
>
> I'd like this to be applied to an experimental branch in the coreutils
> repository, which will be used to test and experiment with Automake-NG
> in a real-world, important, medium-complexity package like GNU coreutils
> is.
>
> I hope you'll agree this is a sensible move, which could bring advantages
> and improvements to both coreutils and Automake-NG.
>
I've updated the patch to take into account:
- the recent bump in the Automake-NG version (1.11e => 1.12a);
- the bump in the minimal Autoconf version required by Automake-NG
(2.62 => 2.65);
- the support for Automake-NG recently added to the Gnulib-provided
bootstrap script
- the removal of the m4 macro 'AM_PROG_MKDIR_P' from the master
branch of the Automake repository (and thus from Automake-NG,
where 'master' is regularly merged).
So, OK to apply this patch to a new branch in the coreutils official
repository? Or it's better if I clone the coreutils repo on GitHub
and work there, to have more freedom while experimenting?
Regards,
Stefano
[0001-build-support-and-require-Automake-NG.patch (text/x-diff, attachment)]
Information forwarded
to
bug-coreutils <at> gnu.org
:
bug#11294
; Package
coreutils
.
(Tue, 08 May 2012 15:07:02 GMT)
Full text and
rfc822 format available.
Information forwarded
to
bug-coreutils <at> gnu.org
:
bug#11294
; Package
coreutils
.
(Tue, 08 May 2012 20:00:02 GMT)
Full text and
rfc822 format available.
Message #14 received at 11294 <at> debbugs.gnu.org (full text, mbox):
Stefano Lattarini wrote:
> On 04/21/2012 11:48 AM, Stefano Lattarini wrote:
>> * configure.ac (AM_INIT_AUTOMAKE): Add the 'ng' option, to ensure that
>> mainstream Automake is not used by mistake when bootstrapping. Also,
>> bump the required Automake version from '1.11.1' to '1.11e', which is
>> the latest (and still development-only) version of Automake-NG at the
>> moment of writing.
>> * bootstrap (check_versions): Hacked to handle automake and aclocal
>> from Automake-NG specially. This change should be backported to Gnulib
>> proper in a later step.
>> * bootstrap.conf ($buildreq): Require "automake-ng" and "aclocal-ng"
>> version >= 0.5; don't require mainstream "automake" anymore.
>>
>> Signed-off-by: Stefano Lattarini <stefano.lattarini <at> gmail.com>
>> ---
>>
>> I'd like this to be applied to an experimental branch in the coreutils
>> repository, which will be used to test and experiment with Automake-NG
>> in a real-world, important, medium-complexity package like GNU coreutils
>> is.
>>
>> I hope you'll agree this is a sensible move, which could bring advantages
>> and improvements to both coreutils and Automake-NG.
>>
> I've updated the patch to take into account:
>
> - the recent bump in the Automake-NG version (1.11e => 1.12a);
> - the bump in the minimal Autoconf version required by Automake-NG
> (2.62 => 2.65);
> - the support for Automake-NG recently added to the Gnulib-provided
> bootstrap script
> - the removal of the m4 macro 'AM_PROG_MKDIR_P' from the master
> branch of the Automake repository (and thus from Automake-NG,
> where 'master' is regularly merged).
>
> So, OK to apply this patch to a new branch in the coreutils official
> repository? Or it's better if I clone the coreutils repo on GitHub
> and work there, to have more freedom while experimenting?
Hi Stefano,
The git commit hooks that we use (e.g., to prohibit pushing merge commits)
might well slow you down: I presume they'd have to be adjusted to permit
merge commits or non-ff messiness) on that new branch.
As you suggest, using another repository may be better.
Information forwarded
to
bug-coreutils <at> gnu.org
:
bug#11294
; Package
coreutils
.
(Tue, 08 May 2012 23:11:01 GMT)
Full text and
rfc822 format available.
Message #17 received at 11294 <at> debbugs.gnu.org (full text, mbox):
Hi Jim, thanks for the feedback.
On 05/08/2012 09:57 PM, Jim Meyering wrote:
> Stefano Lattarini wrote:
>>
>> So, OK to apply this patch to a new branch in the coreutils official
>> repository? Or it's better if I clone the coreutils repo on GitHub
>> and work there, to have more freedom while experimenting?
>
> Hi Stefano,
>
> The git commit hooks that we use (e.g., to prohibit pushing merge commits)
>
Ugh, why do you do that?
> might well slow you down: I presume they'd have to be adjusted to permit
> merge commits or non-ff messiness) on that new branch.
> As you suggest, using another repository may be better.
>
OK. Here it is:
<http://github.com/slattarini/coreutils-am-ng>
If you can confirm that works correctly for you, feel free to close this
bug report.
Thanks,
Stefano
Reply sent
to
Jim Meyering <jim <at> meyering.net>
:
You have taken responsibility.
(Wed, 09 May 2012 12:00:02 GMT)
Full text and
rfc822 format available.
Notification sent
to
Stefano Lattarini <stefano.lattarini <at> gmail.com>
:
bug acknowledged by developer.
(Wed, 09 May 2012 12:00:02 GMT)
Full text and
rfc822 format available.
Message #22 received at 11294-close <at> debbugs.gnu.org (full text, mbox):
Stefano Lattarini wrote:
> Hi Jim, thanks for the feedback.
> On 05/08/2012 09:57 PM, Jim Meyering wrote:
>> Stefano Lattarini wrote:
>>>
>>> So, OK to apply this patch to a new branch in the coreutils official
>>> repository? Or it's better if I clone the coreutils repo on GitHub
>>> and work there, to have more freedom while experimenting?
>>
>> Hi Stefano,
>>
>> The git commit hooks that we use (e.g., to prohibit pushing merge commits)
>>
> Ugh, why do you do that?
That update/push hook is to avoid accidental same-branch merges.
Eventually I might relax it to allow cross-branch merges, though
probably only with a small, fixed set of branch names.
>> might well slow you down: I presume they'd have to be adjusted to permit
>> merge commits or non-ff messiness) on that new branch.
>> As you suggest, using another repository may be better.
>>
> OK. Here it is:
>
> <http://github.com/slattarini/coreutils-am-ng>
>
> If you can confirm that works correctly for you, feel free to close this
> bug report.
Thanks. Confirmed and done.
Message #23 received at 11294-close <at> debbugs.gnu.org (full text, mbox):
On 05/09/2012 01:57 PM, Jim Meyering wrote:
> Stefano Lattarini wrote:
>> Hi Jim, thanks for the feedback.
>> On 05/08/2012 09:57 PM, Jim Meyering wrote:
>>> Stefano Lattarini wrote:
>>>>
>>>> So, OK to apply this patch to a new branch in the coreutils official
>>>> repository? Or it's better if I clone the coreutils repo on GitHub
>>>> and work there, to have more freedom while experimenting?
>>>
>>> Hi Stefano,
>>>
>>> The git commit hooks that we use (e.g., to prohibit pushing merge commits)
>>>
>> Ugh, why do you do that?
>
> That update/push hook is to avoid accidental same-branch merges.
>
Ah, like a "git pull master" issues with the local master having diverged from
the remote one. Got it now, makes definite sense.
>> If you can confirm that works correctly for you, feel free to close this
>> bug report.
>
> Thanks. Confirmed and done.
>
Good :-)
Best regards,
Stefano
bug archived.
Request was from
Debbugs Internal Request <help-debbugs <at> gnu.org>
to
internal_control <at> debbugs.gnu.org
.
(Thu, 07 Jun 2012 11:24:03 GMT)
Full text and
rfc822 format available.
This bug report was last modified 13 years and 74 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.