Re: musl, printf out-of-memory test

> "Jim" == Jim Meyering writes: Jim> That is correct. It is a feature of gdb-7.0 and newer. Jim> You can inspect (watch/break-at/etc.) the same address and expect it Jim> to refer to the same memory location in multiple invocations. Jim> This makes gdb's command-line history even more useful

Re: gnulib-tool and --with-tests

Hi Bruni, >> Yes, please. This is very welcome. OK, I'll add it when I begin to write code for GNULibTests class. It is very good that I didn't miss this letter. :-)

Re: parse-duration-tests: make it build

Bruce Korb wrote: > On 06/21/12 09:22, Bruce Korb wrote: >> On 06/21/12 08:16, Jim Meyering wrote: >>> >>> Maybe just try again? > > make[1]: Leaving directory `/old-home/gnu/proj/gnulib/t' > make: *** No rule to make target `sc_maint'. Stop. That rule is in the top-level Makefile. You must run "

Re: parse-duration-tests: make it build

On 06/21/12 12:41, Bruno Haible wrote: Your commit came without a ChangeLog entry. I added one for you: Woops. Thanks you so much! On 06/21/12 12:41, Bruno Haible wrote: Or, even better in my opinion: All the trouble has been caused my the desire to keep a shell script and a data table in

Re: gnulib-tool and --with-tests

Hello Dmitriy, > If we decide to use such behavior as default, I'll now add this feature to > Python version when I begin to write GNULibTest class. Yes, please. This is very welcome. Note that the change affects only the default value for the 4 modes --create-[mega]testdir and --[mega]test. Not

Re: gnulib-tool and --with-tests

Paul Eggert and Ben Pfaff wrote: > I think the current behavior surprises > pretty much every new gnulib-tool user, and I like the idea of > having --with-tests be the default if --test is used. Thanks for the vote. I think you have a majority. So I am applying this change, followed by a no-op ref

Re: parse-duration-tests: make it build

Bruce Korb wrote: > On 06/21/12 08:16, Jim Meyering wrote: > > > > Maybe just try again? > > My push-ability is still disabled. Can this be fixed, please? Your commit came without a ChangeLog entry. I added one for you: 2012-06-21 Bruce Korb parse-duration test: Avoid spurious outpu

Re: parse-duration-tests: make it build

Jim Meyering wrote: > the change below, either remove the leading TABs (my preference) > and s/-< the leading-TAB-prohibiting syntax check rule. It's not only about the TAB prohibiting syntax check rule. It's also about the ease of editing for those among us who have told their preferred editor to

Re: parse-duration-tests: make it build

On 06/21/12 09:22, Bruce Korb wrote: On 06/21/12 08:16, Jim Meyering wrote: Maybe just try again? make[1]: Leaving directory `/old-home/gnu/proj/gnulib/t' make: *** No rule to make target `sc_maint'. Stop. Also, the way to bypass the test is fairly indecipherable from just looking at cfg.mk

Re: gnulib-tool and --with-tests

Hello everyone! If we decide to use such behavior as default, I'll now add this feature to Python version when I begin to write GNULibTest class.

Re: parse-duration-tests: make it build

Bruce Korb wrote: > On 06/21/12 09:33, Jim Meyering wrote: >> Bruce Korb wrote: >>> On 06/21/12 08:16, Jim Meyering wrote: Maybe just try again? >>> >>> My push-ability is still disabled. Can this be fixed, please? >> >> AFAICT, it's not disabled: >> >>https://savannah.gnu.org/proje

Re: parse-duration-tests: make it build

On 06/21/12 09:33, Jim Meyering wrote: Bruce Korb wrote: On 06/21/12 08:16, Jim Meyering wrote: Maybe just try again? My push-ability is still disabled. Can this be fixed, please? AFAICT, it's not disabled: https://savannah.gnu.org/project/memberlist.php?group=gnulib Ah. Somehow th

Re: gnulib-tool and --with-tests

On 06/21/2012 09:20 AM, Ben Pfaff wrote: > I remember to use --with-tests these > days, but I think that having that be the default would be less > surprising. My sentiments exactly. I think the current behavior surprises pretty much every new gnulib-tool user, and I like the idea of having --wit

[PATCH] maint: disable the strncpy prohibition

Make our "make sc_maint" rule pass once again: >From 6434eb4fc8d1f152ab87e1dd21c60f4bc0d785dd Mon Sep 17 00:00:00 2001 From: Jim Meyering Date: Thu, 21 Jun 2012 18:33:00 +0200 Subject: [PATCH] maint: disable the strncpy prohibition * cfg.mk: Do not prohibit strncpy here. --- ChangeLog | 5 +

Re: parse-duration-tests: make it build

Bruce Korb wrote: > On 06/21/12 08:16, Jim Meyering wrote: >> >> Maybe just try again? > > My push-ability is still disabled. Can this be fixed, please? AFAICT, it's not disabled: https://savannah.gnu.org/project/memberlist.php?group=gnulib Maybe you're having some sort of local network diffi

Re: parse-duration-tests: make it build

On 06/21/12 08:16, Jim Meyering wrote: Maybe just try again? My push-ability is still disabled. Can this be fixed, please? >From d8ef24cb52c23fb136d52325e36af70dead06ebb Mon Sep 17 00:00:00 2001 From: Bruce Korb Date: Thu, 21 Jun 2012 09:20:27 -0700 Subject: [PATCH] testing: fix typo in here

Re: gnulib-tool and --with-tests

Bruno Haible writes: >> What do you think of making --with-tests a no-op, and add a >> --without-tests to disable it? This would save us all some typing... > > You mean to make this change only in combination with --test, > --create-testdir, and --create-megatestdir? Not with --import. > > I'm u

Re: Libvirt / GNULIB failures using Mingw64 toolchain

On 06/21/2012 03:42 AM, Bruno Haible wrote: > +/* Enable this function definition only of gnulib's has prepared > it. only of -> only if I pushed a fix for this (and for the other occurrences of the problem).

Re: parse-duration-tests: make it build

Bruce Korb wrote: > What happened to my GIT access? > > $ git push > fatal: The remote end hung up unexpectedly > > This should be "obvious". Hi Bruce, That appears to have been introduced by me when I changed from TAB-based indentation to space-based. I didn't take into consideration the possibi

Re: parse-duration-tests: make it build

What happened to my GIT access? $ git push fatal: The remote end hung up unexpectedly This should be "obvious". >From 054ba400bcc664216ba988eb979f868665a5f98b Mon Sep 17 00:00:00 2001 From: Bruce Korb Date: Thu, 21 Jun 2012 07:49:16 -0700 Subject: [PATCH] testing: fix typo in here doc * tests/

Re: Libvirt / GNULIB failures using Mingw64 toolchain

On Thu, Jun 21, 2012 at 12:42:17PM +0200, Bruno Haible wrote: > Daniel P. Berrange wrote: > > Here's what i see in config.status for my libvirt build: > > > > $ grep -i vscanf config.status > > S["GNULIB_VSCANF"]="0" > > > > $ grep STDIO config.status > > S["NEXT_AS_FIRST_DIRECTIVE_STDIO_H"]="" >

Re: Libvirt / GNULIB failures using Mingw64 toolchain

Daniel P. Berrange wrote: > Here's what i see in config.status for my libvirt build: > > $ grep -i vscanf config.status > S["GNULIB_VSCANF"]="0" > > $ grep STDIO config.status > S["NEXT_AS_FIRST_DIRECTIVE_STDIO_H"]="" > S["NEXT_STDIO_H"]="" > S["REPLACE_STDIO_WRITE_FUNCS"]="1" > S["REPLACE_STDIO_

Re: [PATCH] Do use readlink if available

Bernd Jendrissek wrote: > Running under dash, type -p readlink fails because dash doesn't > understand -p. Indeed, I confirm that - under bash, "type -p foo" and "type foo" produce the same exit code for various values of foo, - POSIX specifies the 'type' command only without '-p'

Re: gnulib-tool and --with-tests

Hi Simon, > > $ ./gnulib-tool --test --with-tests parse-duration > > Using '--test --with-tests' or '--create-testdir --with-tests' is quite > common. Is there any reason why --with-tests isn't the default? It's for consistency with --import. Many packages like to have --import without tests, a

Re: posixtm: doesn't like -1 for 32-bit time_t's?

On 06/20/2012 07:54 PM, Simon Josefsson wrote: > Changing the line into > > { "190112132045.51", 13, 1, -2147483649LL}, /* Fri Dec 13 20:45:51 1901 > */ > > makes the self-test work. Thanks, I pushed the following, which I hope fixes the problem in a slightly more-general way. --- Change

Re: [musl] Re: musl bugs found through gnulib

On 06/20/2012 07:21 PM, Rich Felker wrote: >>> Replacement of mktime, because of >>> checking for working mktime... no > > This test is buggy; it goes into an infinite loop due to integer > overflow UB, because the condition to break out of the loop is only > checked when the test does not fail

Re: gnumakefile: usable on its own without maintainer-makefile?

Simon Josefsson wrote: > Jim Meyering writes: >> Simon Josefsson wrote: >>> Building the 'gnumakefile' module fails: >>> >>> http://autobuild.josefsson.org/gnulib-mega/log-201206201113065857000.txt >>> >>> Looking at that module, I wonder if that module is useful at all without >>> maintainer-make