Hi Ralf!
Den 2010-06-14 22:40 skrev Ralf Wildenhues:
[ adding automake-patches; this is
http://thread.gmane.org/gmane.comp.gnu.libtool.general/10927/focus=10954 ]
* Peter Rosin wrote on Mon, Jun 14, 2010 at 09:35:45AM CEST:
Den 2010-06-12 10:05 skrev Ralf Wildenhues:
Well, I sort of
Hi Ralf!
Den 2010-06-16 20:57 skrev Ralf Wildenhues:
let's cut the libtool list from replies, ok?
Sure.
* Peter Rosin wrote on Wed, Jun 16, 2010 at 02:30:47PM CEST:
Den 2010-06-14 22:40 skrev Ralf Wildenhues:
* Peter Rosin wrote on Mon, Jun 14, 2010 at 09:35:45AM CEST:
Runnin
Hi Ralf!
Den 2010-06-14 22:40 skrev Ralf Wildenhues:
[ adding automake-patches; this is
http://thread.gmane.org/gmane.comp.gnu.libtool.general/10927/focus=10954 ]
* Peter Rosin wrote on Mon, Jun 14, 2010 at 09:35:45AM CEST:
Den 2010-06-12 10:05 skrev Ralf Wildenhues:
Well, I sort of
Hi Chuck,
Den 2010-06-19 22:09 skrev Charles Wilson:
func_path_conv() assumes that you ALWAYS want to convert from "something
unixish" to win32. If that's the case -- e.g. it appears to only ever
be called from inside func_cl_wrapper, for which that would always be
true -- then it needs to be d
Den 2010-06-19 05:38 skrev Ralf Wildenhues:
* Peter Rosin wrote on Wed, Jun 16, 2010 at 10:30:22PM CEST:
Should I keep attaching the "current" version of the patch?
Sure, but the patch is good to go once copyright papers are through,
with testing added.
Here's the lates
Den 2010-06-22 05:38 skrev Charles Wilson:
What if $CC is specified as 'cl.exe'? IIUC, this means that the compile
script will be invoked as "compile cl.exe " but then this case:
+ cl | *[/\\]cl)
+func_cl_wrapper "$@" # Doesn't return...
+;;
won't fire.
(also folding in the
Hi Chuck,
Den 2010-06-22 14:04 skrev Charles Wilson:
On 6/22/2010 7:53 AM, Peter Rosin wrote:
I added compile5.test. I don't think $build is part of the equation
for the purpose of this test, and the only supported $host at this
point is mingw. Is it what you had in mind?
Well, if $bui
Den 2010-06-22 14:04 skrev Charles Wilson:
On 6/22/2010 7:53 AM, Peter Rosin wrote:
I added compile5.test. I don't think $build is part of the equation
for the purpose of this test, and the only supported $host at this
point is mingw. Is it what you had in mind?
Well, if $build is c
Hi Ralf,
Den 2010-06-22 21:24 skrev Ralf Wildenhues:
* Peter Rosin wrote on Tue, Jun 22, 2010 at 01:53:49PM CEST:
(also folding in the "cl\ * | */cl\ *" cases suggested by Ralf in
another threadlet)
Since the following is way over the top IMHO:
[cC][lL] | *[/\\][cC][lL]
Den 2010-08-01 08:59 skrev Ralf Wildenhues:
Now that assignment has gone through a couple of weeks ago, I went ahead
and committed your version 7 of this patch into a new 'msvc' branch off
of maint and merged that into master. I intend to merge also into
branch-1.11/maint before the 1.11.2 relea
Hi Ralf,
Den 2010-08-01 20:06 skrev Ralf Wildenhues:
* Peter Rosin wrote on Sun, Aug 01, 2010 at 06:54:57PM CEST:
The script still needs @-support, in order to make it
possible for libtool to avoid falling back to "-r -o"
piecewise linking when the command line gets long, so that
Hi Ralf,
and thank you right back for _your_ work on this! :-)
Den 2010-08-04 22:18 skrev Ralf Wildenhues:
thanks for your work on this!
* Peter Rosin wrote on Wed, Aug 04, 2010 at 12:14:28PM CEST:
Den 2010-08-01 20:06 skrev Ralf Wildenhues:
I wonder whether we can and should call it
Hi Ralf,
Den 2010-08-01 20:06 skrev Ralf Wildenhues:
For actual applicability, it is sufficient if you provide a working
script, post testsuite results with it in place (you might want to
MAKE='make AR=...'
When I do
MAKE='make "AR=/path/to/archive lib"' make "AR=/path/to/archive lib" chec
Hi Ralf,
Den 2010-08-06 07:22 skrev Ralf Wildenhues:
* Peter Rosin wrote on Fri, Aug 06, 2010 at 12:00:53AM CEST:
Den 2010-08-01 20:06 skrev Ralf Wildenhues:
For actual applicability, it is sufficient if you provide a working
script, post testsuite results with it in place (you might want to
Hi Stefano,
Den 2010-08-06 13:54 skrev Stefano Lattarini:
At Friday 06 August 2010, Peter Rosin wrote:
[CUT]
I did get a spurious failure in cond5.test, but it went away when
I ran it separately.
Did you by chance run many tests in parallel? If yes, you might have hit
a known issue:
&quo
Den 2010-08-06 13:43 skrev Peter Rosin:
commit 3d3229ff86bd8194974bcf56948f5c6b0b54d99a
Author: Peter Rosin
Date: Fri Aug 6 13:35:26 2010 +0200
Add new auxiliary 'archive' script, wrapping Microsoft lib.
* lib/archive: New auxiliary script.
* lib/Makefile.am: Add above.
* m4/archi
Den 2010-08-06 15:56 skrev Peter Rosin:
kill success:
=== Running test ./tmp.test
++ pwd
/home/peda/automake/git/automake/tests/tmp.dir
+ set -e
+ cat
+ cat
+ sh -c 'echo %aclocal pid: $$; sleep 2'
%aclocal pid: 12228
+ pid=13504
+ sh -c 'echo %automake pid: $$; sleep 4; echo v
Den 2010-08-06 15:15 skrev Stefano Lattarini:
To help us pinpoint the problem, could you please try to run the
attached "fake" test scripts multiple times, and post the outputs?
Please forgive me the for the ugliness of such scripts, but I'm just
taking wild guesses here.
For tmp.test, I swappe
Den 2010-08-06 15:15 skrev Stefano Lattarini:
At Friday 06 August 2010, Peter Rosin wrote:
Hi Stefano,
Den 2010-08-06 13:54 skrev Stefano Lattarini:
At Friday 06 August 2010, Peter Rosin wrote:
[CUT]
I was not running in parallel and I was running off of the msvc
branch (basically maint I
Hi Stefano,
Den 2010-08-06 20:52 skrev Stefano Lattarini:
At Friday 06 August 2010, Ralf Wildenhues wrote:
Stefano, sorry for not taking your patch; I saw Peter's earlier and
had good success testing it on MinGW; also, I think it is easier
to understand than yours. Hope you don't mind.
Obviou
Den 2010-08-07 17:53 skrev Ralf Wildenhues:
* Ralf Wildenhues wrote on Sat, Aug 07, 2010 at 08:03:56AM CEST:
Darn, we need another name: 'archive' is likely too common when automake
is going to put all files with that name in the distribution of each
package using Automake. No need to send anot
Den 2010-08-07 08:03 skrev Ralf Wildenhues:
Darn, we need another name: 'archive' is likely too common when automake
is going to put all files with that name in the distribution of each
package using Automake. No need to send another patch, but any other
good suggestions that fit in with the exi
Well, the difference in situation is that 'compile' was introduced 11
years ago, when breaking compatibility was frequent; and that 'archive'
is both a verb and a noun, and being used as, well, a name for an
archive of stuff by people. Cf. gettext, for example.
* Peter Rosin wrot
Den 2010-08-10 09:01 skrev Peter Rosin:
BTW, why is ! better than ^ here? Just curious.
I found the answer myself. Quoting the portable shell chapter in the
autoconf manual:
| 11.5 Shell Pattern Matching
|
| Nowadays portable patterns can use negated character classes
| like ‘[!-aeiou]’. The
Den 2010-08-06 15:36 skrev Peter Rosin:
> So, this seems like a bug in the kill builtin in MSYS bash.
Hi,
FYI, this will probably be fixed in a future version of MSYS:
http://sourceforge.net/tracker/index.php?func=detail&aid=3042292&group_id=2435&atid=102435
Cheers,
Peter
Den 2010-08-10 22:18 skrev Ralf Wildenhues:
>>> Wrt. the AM_PROG_AR implementation, __oline__ is a real nono, as it
>>> causes spurious diffs in generated configure scripts. I'll propose
>>> an updated version.
>>
>> Oh, didn't know that. I look forward to it, since I reused some of the
>> code fo
Den 2010-08-12 05:29 skrev Ralf Wildenhues:
> * Peter Rosin wrote on Wed, Aug 11, 2010 at 04:12:10PM CEST:
>> One thing to consider is that the compile script is triggered "by chance"
>> for MSVC. It is just a lucky coincidence that MSVC doesn't support -c -o,
>>
EFI_BOOT_SERVICE_DRIVER|EFI_ROM|EFI_RUNTIME_DRIVER|
NATIVE|POSIX|WINDOWS|WINDOWSCE}[,#[.##]]
/VERBOSE
/WX[:NO]
>From 6545b496ad0868e9d7ffde4372e3b36049ceaf3c Mon Sep 17 00:00:00 2001
From: Peter Rosin
Date: Thu, 12 Aug 2010 16:24:11 +0200
Subject: [PATCH] Enable the
Den 2010-08-12 16:30 skrev Peter Rosin:
> Hi!
>
> lib.exe is simply a wrapper around "link.exe -lib", this patch
> allows the latter form to also be wrapped by ar-lib.
>
> While at it, the patch also allows some other options through
> to the wrapped archiver. I
Den 2010-08-12 23:22 skrev Ralf Wildenhues:
> Hi Peter,
>
> * Peter Rosin wrote on Thu, Aug 12, 2010 at 06:03:03PM CEST:
>> From 8c406f64b66aedee8359ddf1383e087faa2ba233 Mon Sep 17 00:00:00 2001
>> From: Peter Rosin
>> Date: Thu, 12 Aug 2010 18:01:18 +0200
>> S
Den 2010-08-12 19:55 skrev Ralf Wildenhues:
> * Peter Rosin wrote on Thu, Aug 12, 2010 at 01:05:09PM CEST:
>> Or always pass on any '-lib' option to PROGRAM w/o trying to interpret
>> it further. That might be a good addition, since there might be other
>> interesti
.
Ok for the msvc branch?
Cheers,
Peter
>From 8ad9d6ddfb664a527be7e3030a43b822363bc71a Mon Sep 17 00:00:00 2001
From: Peter Rosin
Date: Fri, 13 Aug 2010 15:09:14 +0200
Subject: [PATCH] Support more C++ extensions for MSVC in the compile script.
* lib/compile (func_cl_wrapper): MSVC only recogn
Den 2010-08-07 17:56 skrev Ralf Wildenhues:
> I'm seeing a presumably new failure on Cygwin that I don't remember
> seeing before. The gist is the following: transform2.test tries
> something like
>
> install bla.exe .../inst/bin/foo.exe
> install script.sh .../inst/bin/foo
>
> and the secon
Den 2010-08-13 19:32 skrev Ralf Wildenhues:
> * Peter Rosin wrote on Fri, Aug 13, 2010 at 03:55:54PM CEST:
>> Den 2010-08-07 17:56 skrev Ralf Wildenhues:
>>> install bla.exe .../inst/bin/foo.exe
>>> install script.sh .../inst/bin/foo
>>>
>>> and t
Den 2010-08-13 20:02 skrev Ralf Wildenhues:
> * Peter Rosin wrote on Fri, Aug 13, 2010 at 11:38:43AM CEST:
*snip*
> On the other hand, this means that better support for cl means adjusting
> a number of macros from Autoconf, Gnulib, and maybe Automake as well.
> For example, what abou
Den 2010-08-13 19:08 skrev Ralf Wildenhues:
> Hi Peter,
>
> * Peter Rosin wrote on Fri, Aug 13, 2010 at 03:37:59PM CEST:
>> This is one of the last bits of functionality from the (dwindling)
>> pr-msvc-support branch in Libtool (the remaining bit is triggering
>> t
EOF'
#! /bin/sh
echo "$@"
EOF
That said, don't expect too much from this as using absolute file
names is not really the common case...
Cheers,
Peter
>From 77f0a9fd4265f5b22d5686e3da27c86aee496b6d Mon Sep 17 00:00:00 2001
From: Peter Rosin
Date: Mon, 16 Aug 2010 13:50:32 +
Den 2010-08-16 20:30 skrev Ralf Wildenhues:
> Hi Peter,
>
> [ output reordered to print new vs old ]
>
> * Peter Rosin wrote on Mon, Aug 16, 2010 at 01:59:13PM CEST:
>> time for file in `./filenames 20 / .cc`; do ./compile cl $file; done
>> real0m9.781s
>>
Hi!
A recent discussion on the libtool lists(*) brought this to my
attention.
Ok for the msvc branch?
Cheers,
Peter
(*) http://thread.gmane.org/gmane.comp.gnu.libtool.patches/10366
>From 8372b8c8f82d1a3680dd288da68a786a02329213 Mon Sep 17 00:00:00 2001
From: Peter Rosin
Date: Tue, 31 Aug 2
Hi!
When the compile wrapper is used to link, it needs to also
transform /absolute/object/files.obj to win32 format for
easy consumption by MSVC.
Ok for the msvc branch?
Cheers,
Peter
>From 827d6b341a858b3b9a4c90dc157846e5b242bc4b Mon Sep 17 00:00:00 2001
From: Peter Rosin
Date: Tue, 31
Den 2010-08-31 18:57 skrev Ralf Wildenhues:
> Hi Peter,
>
> * Peter Rosin wrote on Tue, Aug 31, 2010 at 06:47:24PM CEST:
>> Subject: [PATCH] Handle odd winepath failures.
>>
>> * lib/ar-lib (func_file_conv): winepath can exit with zero exit
>> status but empty r
Hi Ralf!
Den 2010-08-31 19:02 skrev Ralf Wildenhues:
> * Peter Rosin wrote on Tue, Aug 31, 2010 at 06:59:25PM CEST:
>> When the compile wrapper is used to link, it needs to also
>> transform /absolute/object/files.obj to win32 format for
>> easy consumption by MSVC.
>>
Hi!
[Automakers, this is from the libtool-patches list...]
Den 2010-09-02 15:06 skrev Peter Rosin:
> 112: Test 24 33 34 45 47 99 100 are "new" failures with low max_cmd_len
> and I think the reason is that the file name conversion adds escapes for
> the backslashes. I.e. .l
Den 2010-09-02 20:37 skrev Ralf Wildenhues:
> * Stefano Lattarini wrote on Thu, Sep 02, 2010 at 06:44:34PM CEST:
>> On Thursday 02 September 2010, Peter Rosin wrote:
>>> $AR -NOLOGO -LIST "$archive" | sed -e 's/\\//g' | while read
>>> mem
Hi Stefano!
Den 2010-09-03 02:48 skrev Stefano Lattarini:
> OK to push the attached patch to msvc as trivial?
No, because it isn't a typo... :-)
Wrappee denotes the thing that is wrapped.
Cheers,
Peter
Den 2010-09-07 23:14 skrev Ralf Wildenhues:
> * Peter Rosin wrote on Tue, Aug 31, 2010 at 09:36:15PM CEST:
>> Den 2010-08-31 18:57 skrev Ralf Wildenhues:
>>> * Peter Rosin wrote on Tue, Aug 31, 2010 at 06:47:24PM CEST:
>>>> Subject: [PATCH] Handle odd winepath fa
rendered obsolete by LT_INIT. Is it also rendered
obsolete by AC_PROG_LIBTOOL? Should I not care about libtool 1.5?
Cheers,
Peter
>From e4810ab05c41461c3b6c5638a41d8d415ebcc40f Mon Sep 17 00:00:00 2001
From: Peter Rosin
Date: Tue, 14 Sep 2010 14:58:17 +0200
Subject: [PATCH] Add new 'A
Hi Stefano,
Den 2010-09-14 20:14 skrev Stefano Lattarini:
> On Tuesday 14 September 2010, Peter Rosin wrote:
>> Hi!
> Hello Peter. Hope you don't mind a quick partial review until we
> hear from Ralf...
Not at all, he's got enough to do as it is I would think...
>&
bing?
>> + test $exit_status = 76 && exit 77
> I'm not sure I understand the semantic here; an exit status of 76
> means you have the expected `lib' program... and in this case you
> skip the test because it requires `lib' itself? (BTW, sorry for
>
Den 2010-09-14 20:14 skrev Stefano Lattarini:
> On Tuesday 14 September 2010, Peter Rosin wrote:
>> However, I have only updated tests/ar.test to cope with the new
>> reality. So, a lot of tests (100?) are likely to fall over due to
>> the new portability warnings. The reason
Den 2010-09-15 12:47 skrev Stefano Lattarini:
> On Wednesday 15 September 2010, Peter Rosin wrote:
>> BTW, I forgot to ask before, but can you please point out such a
>> fake merge, so that I can see what I should aim for?
> Honestly, I've never had to do such a "fa
Den 2010-09-15 16:35 skrev Stefano Lattarini:
> Hello automakers.
>
> The "msvc" branch still has configure and aclocal.m4 generated
> with autoconf 2.65, while the "maint" and "master" branches use
> autoconf 2.67.
>
> This can cause annoying situations when switching among branches,
> such as
Den 2010-09-15 16:57 skrev Eric Blake:
> On 09/15/2010 08:52 AM, Peter Rosin wrote:
>> Can we please wait with that until Cygwin releases autoconf 2.67?
>
> It already has:
> http://cygwin.com/ml/cygwin-announce/2010-09/msg3.html
Oh boy, that's quick!
No objections then...
Cheers,
Peter
Den 2010-09-15 17:13 skrev Stefano Lattarini:
> That considered: Peter, do you now think it's ok to apply my patch?
Whatever. If you think it will be so much easier for you, then go
ahead. I guess I'll have to cope somehow...
Cheers,
Peter
Den 2010-09-15 23:56 skrev Ralf Wildenhues:
> For now, I didn't get any further than just some comments upon comments
> (a review-review, if you like):
>
> * Stefano Lattarini wrote on Wed, Sep 15, 2010 at 01:45:17AM CEST:
>> On Tuesday 14 September 2010, Peter Rosin wrote
Hi Ralf,
Den 2010-09-16 20:03 skrev Ralf Wildenhues:
> Hi Peter,
>
> I've looked over the non-testsuite part of this now. Comments below.
> I'll try to get the rest done this weekend.
Ok, great, and thanks!
> * Peter Rosin wrote on Thu, Sep 16, 2010 at 10:50:05AM C
Den 2010-09-17 11:58 skrev Stefano Lattarini:
> On Friday 17 September 2010, Peter Rosin wrote:
>> Den 2010-09-16 20:03 skrev Ralf Wildenhues:
>>>> +
>>>> + if (! $seen_ar)
>>>> + {
>>>> +msg ('portability&#
Den 2010-09-21 12:46 skrev Stefano Lattarini:
> Hi Peter.
>
> On Tuesday 21 September 2010, Peter Rosin wrote:
>> Den 2010-09-17 11:58 skrev Stefano Lattarini:
>>> Or what about doing somethins similar to what gcc does, and add a
>>> new `-Wextra' catego
Den 2010-09-21 14:35 skrev Stefano Lattarini:
> On Tuesday 21 September 2010, Peter Rosin wrote:
>> Den 2010-09-21 12:46 skrev Stefano Lattarini:
>>> Now, many projects might not care at all to support this building
>>> environment, while still wanting to use `-Wall&#x
:00 2001
From: Peter Rosin
Date: Tue, 21 Sep 2010 16:55:25 +0200
Subject: [PATCH] compile: implement library search to support MSVC static
linking
* lib/compile (func_cl_wrapper): Implement library search and
-static option so that the user can select whether to prefer
dll import libraries or sta
Den 2010-09-21 19:56 skrev Ralf Wildenhues:
> Hi Peter,
>
> * Peter Rosin wrote on Tue, Sep 21, 2010 at 05:11:26PM CEST:
>> This is for the msvc branch. This idea was previously discussed here:
>> http://lists.gnu.org/archive/html/libtool-patches/2010-09/msg00044.html
Hi Ralf,
Den 2010-10-03 12:02 skrev Ralf Wildenhues:
> This test fails when echo interprets backslashes:
>
> + touch fake2.lib
> + ./ar-lib ./lib x fake2.lib
> + opts=lib -NOLOGO -EXTRACT:dir^Lake2.obj fake2.lib
> + test xlib -NOLOGO -EXTRACT:dir^Lake2.obj fake2.lib = xlib -NOLOGO
> -EXTRACT:dir
Den 2010-09-21 19:02 skrev Stefano Lattarini:
> On Tuesday 21 September 2010, Peter Rosin wrote:
>>>> is if it doesn't show up with -Wall and that it doesn't trip up
>>>> -Wall -Werror.
>>>
>>> Yes, sorry for not pointing it out explicitly. T
ate. The
patch is based on the msvc branch. The old MSVC depmodes are still
needed for MSVC 6 so can't be deleted without causing a regression.
The testsuite seems happy (but I only ran the depcomp*.test tests).
Cheers,
Peter
>From 9d0f654afbddb82da01c09c2c22f8a18a0cf0e8b Mon Sep 17 00:
Den 2010-10-06 17:55 skrev Peter Rosin:
> @@ -90,10 +90,18 @@ if test "$depmode" = msvcmsys; then
> # This is just like msvisualcpp but w/o cygpath translation.
> # Just convert the backslash-escaped backslashes to single forward
> # slashes to satisfy depend.m4
cally carrying this
burden all by yourself. As long as it's within a week or two
you are doing a *great* job!
> * Peter Rosin wrote on Wed, Oct 06, 2010 at 05:55:44PM CEST:
>> I noticed that MSVC 7 and above sports the option -showIncludes
>
>> cl -showIncludes foo.c
>
Den 2010-11-14 23:00 skrev Stefano Lattarini:
> On Sunday 14 November 2010, Ralf Wildenhues wrote:
>> Hello Stefano,
>>
>> I think it is time to reevaluate some of the work that has been done in
>> the last few weeks. I'd like to ask you to postpone pushes of pending
>> stuff that has a clock tick
Den 2010-11-15 01:39 skrev Stefano Lattarini:
> On Monday 15 November 2010, Peter Rosin wrote:
>> Running MSVC on osfX.Y is not interesting. I'd write a patch to skip
>> the test
> Yeah, I agree this is the best option.
>> but I can't then test the result.
>
Den 2010-11-15 09:19 skrev Peter Rosin:
> diff --git a/tests/defs b/tests/defs
> index 82c3b5b..e37bb96 100644
> --- a/tests/defs
> +++ b/tests/defs
> @@ -290,6 +290,15 @@ do
>echo "$me: running texi2dvi -o /dev/null --version"
>( texi2dvi -o /
Hi!
I found this little gem in compile while reading the code. Pushed as
obvious to the msvc branch.
Cheers,
Peter
>From 89bf665808e6ed7a68957342e6343cf3261ebf14 Mon Sep 17 00:00:00 2001
From: Peter Rosin
Date: Mon, 15 Nov 2010 10:27:55 +0100
Subject: [PATCH] compile: clear the `
Den 2010-11-15 10:33 skrev Peter Rosin:
> Hi!
>
> I found this little gem in compile while reading the code. Pushed as
> obvious to the msvc branch.
Obviously I tripped up while committing something obvious. Sorry for
being sloppy. So, I also pushed this follow-up commit to the
Den 2010-11-15 10:40 skrev Peter Rosin:
> Den 2010-11-15 10:33 skrev Peter Rosin:
>> Hi!
>>
>> I found this little gem in compile while reading the code. Pushed as
>> obvious to the msvc branch.
>
> Obviously I tripped up while committing something obvious. Sorr
Den 2010-11-15 11:44 skrev Stefano Lattarini:
>> Should I fixup the ChangeLog entry?
> I'd say just to merge the "update scriptversion" message with the
> previous entry in the ChangeLog, like this:
>
> +2010-11-15 Peter Rosin
> +
> + co
Den 2010-11-15 23:23 skrev Ralf Wildenhues:
> Hi Peter,
>
> * Peter Rosin wrote on Mon, Nov 15, 2010 at 09:19:44AM CET:
>> This is based off of master (but I guess it also applies to tests-init)
>> but the backport to the msvc branch is trivial. I "stole" the xsi-te
Den 2010-11-16 12:15 skrev Stefano Lattarini:
> On Tuesday 16 November 2010, Peter Rosin wrote:
>> Den 2010-11-15 23:23 skrev Ralf Wildenhues:
>>> Hi Peter,
>>>
>>> * Peter Rosin wrote on Mon, Nov 15, 2010 at 09:19:44AM CET:
>>>> This is based off
Den 2011-01-09 11:41 skrev Ralf Wildenhues:
> @@ -384,7 +385,7 @@ sub reset()
>
> =item C
>
> -Register a suffix rules defined on C<$where> that transform
> +Register a suffix rules defined on C<$where> that transforms
> files ending in C<$src> into files ending in C<$dest>.
>
> This upgra
Den 2011-02-15 21:19 skrev Ralf Wildenhues:
> It's a bit of a shame this uses a temp file though. awk should be able
> to cope without (but be sure to try old Solaris awk if you go for it).
Ahh, who needs awk? :-)
unindent ()
{
sed -n '
x
:useit
/./{
G
h
s/\n.*//
Den 2011-02-17 12:15 skrev Peter Rosin:
> Den 2011-02-15 21:19 skrev Ralf Wildenhues:
>> It's a bit of a shame this uses a temp file though. awk should be able
>> to cope without (but be sure to try old Solaris awk if you go for it).
>
> Ahh, who needs awk? :-)
>
Den 2011-02-17 19:18 skrev Stefano Lattarini:
> Hello Peter.
>
> On Thursday 17 February 2011, Peter Rosin wrote:
>> Den 2011-02-15 21:19 skrev Ralf Wildenhues:
>>> It's a bit of a shame this uses a temp file though. awk
>>> should be able to cope without (
Den 2011-02-18 13:12 skrev Peter Rosin:
> Yes, I will try my best not to spam the list with more versions of
> this baby...
Famous last words. Sorry.18 lines -> 15 lines -> 12 lines
Will be AFK the next week, so with a bit of luck I will have forgotten all
about this whe
the first invocation of
>> 'unindent'; e.g.:
>
> Good idea.
How about this?
Should the sed command to strip comments be "s/ *#.*//", or does that
fall into this bucket:
Unicos 9 sed loops endlessly on patterns like ‘.*\n.*’.
Cheers,
Peter
>From cf34
Den 2011-02-28 11:04 skrev Peter Rosin:
> * tests/unindent.test: New test.
> * tests/Makefile.am (TESTS): Update.
I continued reading and noticed the tests proposed by Stefano with
superior coverage, consider my feeble tests dropped.
Cheers,
Peter
Den 2011-02-28 21:36 skrev Ralf Wildenhues:
> Hello,
>
> * Stefano Lattarini wrote on Mon, Feb 28, 2011 at 01:54:20PM CET:
>> On Monday 28 February 2011, Peter Rosin wrote:
>>> How about this?
>>>
>>> Should the sed command to strip comments be "s/
Den 2011-03-01 06:12 skrev Ralf Wildenhues:
> * Peter Rosin wrote on Mon, Feb 28, 2011 at 10:01:35PM CET:
>> Den 2011-02-28 21:36 skrev Ralf Wildenhues:
>>> It would be nice your sed script would allow sed scripts like
>>> s/#//
>>> s/ #//
>>>
>
Den 2011-03-01 10:40 skrev Stefano Lattarini:
> On Tuesday 01 March 2011, Peter Rosin wrote:
>> Good, I pushed with that squashed in. I didn't dare a merge to
>> master as the conflicts looked scary.
>>
> I've taken a look and done the merge. Luckily, the
Den 2011-03-01 11:37 skrev Stefano Lattarini:
> On Tuesday 01 March 2011, Peter Rosin wrote:
>> Den 2011-03-01 10:40 skrev Stefano Lattarini:
>>> On Tuesday 01 March 2011, Peter Rosin wrote:
>>>> Good, I pushed with that squashed in. I didn't dare a merge to
Den 2011-03-01 10:40 skrev Stefano Lattarini:
> I'd also like to enhance `self-check-unindent.test' to ensure that
> `unindent' remains usable in parallel in the future. OK to commit
> the patch below (to master)? I'll push in 72 hours if there is no
> objection.
One other thing that could conce
Den 2011-05-04 19:17 skrev Stefano Lattarini:
> + lex tests: require just `lex' instead of not `flex'
FWIW, s/not //
Cheers,
Peter
Den 2011-05-29 16:26 skrev Stefano Lattarini:
> Currently, every decent non-GNU make program makes it possible
> to remake out-of-date autotools-generated files with a simple
> "make Makefile" issued from the top-level directory. And while
> having to run "make Makefile" by hand isn't as convenien
Den 2011-05-29 19:32 skrev Stefano Lattarini:
> Hi Peter, thanks for the review.
I wouldn't call it review, it was more of a knee-jerk reaction
(and I haven't even read the actual patch).
> On Sunday 29 May 2011, Peter Rosin wrote:
>> Den 2011-05-29 16:26 skrev Stefano La
Den 2011-05-30 22:22 skrev Stefano Lattarini:
> On Sunday 29 May 2011, Peter Rosin wrote:
>> Den 2011-05-29 19:32 skrev Stefano Lattarini:
>>> Hi Peter, thanks for the review.
>>
>> I wouldn't call it review, it was more of a knee-jerk reaction
>>
> Well
Den 2011-06-02 17:36 skrev Stefano Lattarini:
> This should offer greater testsuite coverage for those developers
> that override CONFIG_SHELL at configure time in order to test more
> shells on a single system, instead of just the default `/bin/sh'.
Do we not want to test these scripts in the sam
Den 2011-06-04 11:24 skrev Stefano Lattarini:
> On Saturday 04 June 2011, Peter Rosin wrote:
>> Den 2011-06-02 17:36 skrev Stefano Lattarini:
>>> This should offer greater testsuite coverage for those developers
>>> that override CONFIG_SHELL at configure time in order t
Den 2011-06-05 12:57 skrev Stefano Lattarini:
> On Sunday 05 June 2011, Peter Rosin wrote:
>> Den 2011-06-04 11:24 skrev Stefano Lattarini:
>>> On Saturday 04 June 2011, Peter Rosin wrote:
>>>> Den 2011-06-02 17:36 skrev Stefano Lattarini:
>>>>> Thi
Den 2011-06-05 14:56 skrev Stefano Lattarini:
> On Sunday 05 June 2011, Peter Rosin wrote:
>> Den 2011-06-04 11:24 skrev Stefano Lattarini:
>>> Yes, and you're absolutely right complaining about the LIB= -> LIB=''
>>> useless change, and the cat ->
Den 2011-06-06 11:06 skrev Stefano Lattarini:
>> [SNIP]
Ditto
and the reason is a dubious argument to increase testsuite coverage,
see more below.
We both think the testsuite should execute the script as they are executed
when they are used. And I think that it should onl
es on Solaris systems which also had GNU Bash installed.
>
> From a suggestion by Peter Rosin. See discussion at:
> <http://lists.gnu.org/archive/html/automake-patches/2011-06/msg00016.html>
>
> * tests/defs (xsi-shell): Now check that $SHELL, rather than the
> shell cu
Den 2011-06-06 21:48 skrev Stefano Lattarini:
> * tests/ar-lib.test: If the variable `$test_prefer_config_shell'
> is set to "yes", run the script under test with configure-time
> determined $SHELL, rather than with /bin/sh.
> The `$test_prefer_config_shell' variable defaults to empty, but
> can be
Den 2011-06-06 21:49 skrev Stefano Lattarini:
> With the previous commit, the user could prefer the use of $SHELL
> over /bin/sh in some tests checking the Automake-provided shell
> scripts by manually exporting `test_prefer_config_shell' to "yes"
> in the environment. With this commit, we ensure
Nothing more to say on 1/3.
Den 2011-06-07 12:29 skrev Stefano Lattarini:
> On Tuesday 07 June 2011, Peter Rosin wrote:
>> Den 2011-06-06 21:48 skrev Stefano Lattarini:
>>> * tests/ar-lib.test: If the variable `$test_prefer_config_shell'
>>> is set to "
1 - 100 of 337 matches
Mail list logo
