Hi Jim,
> Sure, it probably works, but adding support for the non-standard
> %qd specifier seems wrong. Maybe it'd be better to make
> gt_INTTYPES_PRI declare such a PRIdMAX value invalid and to
> provide a replacement definition.
This doesn't work because PRIdMAX must also be understood by the
Simon Josefsson wrote:
> I can reproduce it locally using:
>
> [EMAIL PROTECTED]:~$ gnulib-tool --with-tests --test unictype/category-and
>
> but I don't understand why it is failing. Any ideas? Can others
> reproduce this?
Sure I reproduce this. The link errors come from missing dependencies.
Colin Watson wrote:
> > The set of signals that are supported everywhere is very small, namely:
> > SIGINT
> > SIGILL
> > SIGFPE
> > SIGSEGV
> > SIGTERM
> > SIGABRT
>
> I should have looked at C99 before asking.
That's one half of the research. The other half was to look at what the
s
Bruno Haible <[EMAIL PROTECTED]> writes:
> Ok, if you don't think there are negative side effects of the change, I don't
> want to stand in your way. I only mentioned an alternative change that could
> potentially have less negative side effects. But if there are no downsides
> anyway, your change
Simon Josefsson wrote:
> I reviewed the memmem situation for gnutls, and the recent changes are
> somewhat large. Memmem now pulls in:
>
> ! lgl/m4/eealloc.m4
> ! lgl/m4/malloca.m4
> ! lgl/malloca.c
> ! lgl/malloca.h
> ! lgl/malloca.valgrind
These are no longer pulled in, after Eric's rewrite.
James Youngman wrote:
> ... searching for a constant needle in a large
> number of haystacks. In such cases, the preparation only needs to be
> done once at all. In the specific case of findutils, locate will
> typically search for a single needle in one or two million haystacks.
> Of course, l
Peter Miller wrote:
> yes, I was missing something subtle.
> The attached patch adds comments to explain it to me.
Thanks for suggesting this. Comments about a functions calling convention
should be at the function's definition, not where it is called, therefore
I'm applying this:
2008-01-08 Br
On Tue, Jan 08, 2008 at 02:42:24AM +0100, Bruno Haible wrote:
> Colin Watson wrote:
> > Do you happen to know whether I was overdoing it by adding in #ifdefs
> > for SIGHUP, SIGINT, etc.? I decided better safe than sorry, but didn't
> > know whether there was some basic set of signals that Gnulib's
On Tue, Jan 08, 2008 at 01:56:55AM +0100, Bruno Haible wrote:
> Colin Watson wrote:
> > I use strsignal in man-db, and would like a Gnulib module to cope with
> > its portability problems. Here's one which seems to be doing the right
> > thing for me so far. This is my first attempt at writing a Gn
Paul Eggert <[EMAIL PROTECTED]> wrote:
> Jim Meyering <[EMAIL PROTECTED]> writes:
>
>> I have mixed feelings about this patch.
>> Sure, it probably works, but adding support for the non-standard
>> %qd specifier seems wrong. Maybe it'd be better to make
>> gt_INTTYPES_PRI declare such a PRIdMAX v
Jim Meyering <[EMAIL PROTECTED]> writes:
> I have mixed feelings about this patch.
> Sure, it probably works, but adding support for the non-standard
> %qd specifier seems wrong. Maybe it'd be better to make
> gt_INTTYPES_PRI declare such a PRIdMAX value invalid and to
> provide a replacement def
There are some strange build failures when some unictype modules are
built standalone, see:
http://autobuild.josefsson.org/gnulib/log-200801072354987753000.txt
http://autobuild.josefsson.org/gnulib/log-200801072355046165000.txt
http://autobuild.josefsson.org/gnulib/log-200801080014046171000.txt
ht
Paul Eggert <[EMAIL PROTECTED]> writes:
> Simon Josefsson <[EMAIL PROTECTED]> writes:
>
>> Thus, I think it is appropriate to rename the current efficient 'memmem'
>> gnulib module to 'memmem-fast' and revert to the glibc implementation in
>> the gnulib 'memmem' module.
>
> A minor point: Was the
* Bruno Haible wrote on Tue, Jan 08, 2008 at 03:08:17AM CET:
> Ralf Wildenhues wrote:
> > Bruno, I suppose config.rpath and config.libpath would need adjustment
> > as well. Should I send a patch?
>
> Normally, I update these files after an official libtool release. Once
> a libtool-1.5.26 is rel
Simon Josefsson <[EMAIL PROTECTED]> writes:
> Thus, I think it is appropriate to rename the current efficient 'memmem'
> gnulib module to 'memmem-fast' and revert to the glibc implementation in
> the gnulib 'memmem' module.
A minor point: Was the intent to propagate this into the glibc
version?
I have mixed feelings about this patch.
Sure, it probably works, but adding support for the non-standard
%qd specifier seems wrong. Maybe it'd be better to make
gt_INTTYPES_PRI declare such a PRIdMAX value invalid and to
provide a replacement definition.
Suggestions?
2008-01-08 Jim Meyering <[
Hi,
> > > > Another thing:, the relocatable module documentation recommends using:
> > > > bindtextdomain (PACKAGE, relocate (LOCALEDIR));
> > > > however, the result of relocate may be LOCALEDIR itself (no
> > > > relocation) or a newly allocated string.
>
> Yes, this is by design: The relocat
Hi,
test-wcwidth fails on OpenBSD 4.0, because wcwidth returns 0 instead of 2
for many CJK characters. This should detect this bug:
2008-01-08 Bruno Haible <[EMAIL PROTECTED]>
* m4/wcwidth.m4 (gl_FUNC_WCWIDTH): Test also U+3000. Needed to
detect bug on OpenBSD 4.0.
* d
On Tuesday 08 January 2008, Jim Meyering wrote:
> Mike Frysinger <[EMAIL PROTECTED]> wrote:
> > 2008-01-08 Mike Frysinger <[EMAIL PROTECTED]>
> >
> > * m4/acl.m4 (gl_FUNC_ACL): Wrap all ACL logic in a call to
> > AC_ARG_ENABLE(acl).
> >
> > diff --git a/m4/acl.m4 b/m4/acl.m4
> > index 6a1
On Mon, Jan 07, 2008 at 08:49:01PM -0700, Eric Blake wrote:
> Definitely a bug in the resulting fseeko - POSIX requires a successful
> seek to clear the EOF marker. The question now is whether this is the
> native fseeko (in which case, m4/fseeko.m4 needs to be enhanced to detect
> this flaw and w
On Jan 8, 2008 2:58 PM, Benoit Sigoure <[EMAIL PROTECTED]> wrote:
> In this specific case of findutils, wouldn't it be even more
> efficient to use a data structure that orders data according to there
> common prefixes (such as Tries [http://en.wikipedia.org/wiki/Trie])?
While I'm sure there are f
On Jan 8, 2008, at 3:34 PM, James Youngman wrote:
On Jan 8, 2008 1:32 PM, Eric Blake <[EMAIL PROTECTED]> wrote:
(I may later factor out the two_way static functions into a
parameterized
header, in order to share code with strstr, strcasestr, ...;
however, note
that strstr can never achieve
I reviewed the memmem situation for gnutls, and the recent changes are
somewhat large. Memmem now pulls in:
! lgl/m4/eealloc.m4
! lgl/m4/malloca.m4
! lgl/m4/memchr.m4
! lgl/m4/memcmp.m4
! lgl/m4/memmem.m4
! lgl/malloca.c
! lgl/malloca.h
! lgl/malloca.valgrind
! lgl/memchr.c
! lgl/memcmp.c
! lgl/m
On Jan 8, 2008 1:32 PM, Eric Blake <[EMAIL PROTECTED]> wrote:
> (I may later factor out the two_way static functions into a parameterized
> header, in order to share code with strstr, strcasestr, ...; however, note
> that strstr can never achieve sublinear performance, because it costs O(n)
> to f
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 1/7/2008 5:54 PM:
|
| And LONG_NEEDLE_THRESHOLD should better be (1U << (CHAR_BIT - 1))
| rather than (1 << (CHAR_BIT - 1)), for platforms where CHAR_BIT = 32.
| On such platforms, the function two_way_long_needle will not
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
[re-adding bug-gnulib; this patch does not make sense for memmem.c, since
it no longer uses knuth_morris_pratt, but might still be useful for str-kmp.h]
- Original Message
On Tue, 2008-01-08 at 11:15 +1100, Peter Miller wrote:
| Or
> > +TESTS += test-memmem
> > +TESTS_ENVIRONMENT += EXEEXT='@EXEEXT@'
> > +check_PROGRAMS += test-memmem
>
> You can drop the TESTS_ENVIRONMENT line.
I've dropped this line now. TESTS_ENVIRONMENT setting EXEEXT is only needed
when one of the tests is a shell script that calls an executable.
2008
Mike Frysinger <[EMAIL PROTECTED]> wrote:
> 2008-01-08 Mike Frysinger <[EMAIL PROTECTED]>
>
> * m4/acl.m4 (gl_FUNC_ACL): Wrap all ACL logic in a call to
> AC_ARG_ENABLE(acl).
>
> diff --git a/m4/acl.m4 b/m4/acl.m4
> index 6a1951d..2e3a0c5 100644
> --- a/m4/acl.m4
> +++ b/m4/acl.m4
> @
Bruno Haible <[EMAIL PROTECTED]> writes:
> Simon Josefsson wrote:
>> +2008-01-06 Simon Josefsson <[EMAIL PROTECTED]>
>> +
>> +* tests/test-localename.c: Don't include obsolete "setenv.h".
>> +
>> +* modules/localename-tests (Depends-on): Need unsetenv.
>> +
>
> It's good. Thank you; it w
29 matches
Mail list logo
