Bruno Haible wrote:
>> > my suggestion implied to rename the function as well, not
>> > only the module :-)
>>
>> At which point _all_ clients would instantly be exposed to the rename in
>> C code
>
> If we decide to rename get_date() to parse_datetime(), we can add a
> #define get_date parse_da
On 10/01/2010 07:28 AM, Eric Blake wrote:
Done, and pushed in Justin's name. I added (tiny change) in the
ChangeLog to reflect that Justin does not have FSF copyright assignment
on file for gnulib (although technically, it appears that Red Hat has
blanket assignment).
No stress here about doin
Eric,
> > my suggestion implied to rename the function as well, not
> > only the module :-)
>
> At which point _all_ clients would instantly be exposed to the rename in
> C code
If we decide to rename get_date() to parse_datetime(), we can add a
#define get_date parse_datetime
for one year o
On 09/30/2010 03:35 PM, Bruno Haible wrote:
Eric Blake wrote:
I went with get_date, and not Bruno's parse-datetime. Afterall,
parse-duration supplies the parse_duration function, and
get_date supplies the get_date function.
Well, my suggestion implied to rename the function as well, not
only
Eric Blake wrote:
> - if test $HAVE_POLL = 0 || $REPLACE_POLL = 1; then
> + if test $HAVE_POLL = 0 || test $REPLACE_POLL = 1; then
Thanks Eric. The patch series was tested on Linux and mingw, but not on MacOS X.
Bruno
Eric Blake wrote:
> I went with get_date, and not Bruno's parse-datetime. Afterall,
> parse-duration supplies the parse_duration function, and
> get_date supplies the get_date function.
Well, my suggestion implied to rename the function as well, not
only the module :-)
> lib/{getdate.y => get_d
Eric Blake wrote:
> Note: getdate.h is not renamed, to minimize client impact.
> Besides, whenever we eventually add getdate-posix, we'll
> want to use getdate.m4 and getdate.c for the implementation,
> but don't need to worry about getdate.h since getdate() uses
> for its header.
>
> * modules/ge
On 09/30/2010 03:04 PM, Jim Meyering wrote:
Eric Blake wrote:
[moving to bug-gnulib]
On 09/19/2010 01:21 AM, Justin Clift wrote:
...
The next step is to get all required pieces from gnulib,
to run autoreconf, and to invoke ./configure:
diff --git a/bootstrap b/bootstrap
index 2422549..5
Eric Blake wrote:
> * build-aux/bootstrap (check_versions): Convert libtool into
> libtoolize.
> (tool search): Move libtool check earlier, and look for
> glibtoolize for MacOS.
> (gnulib_tool_options): Auto-add --libtool when appropriate.
> Reported by Justin Clift.
Looks good. Thank you both.
Eric Blake wrote:
> * m4/poll.m4 (gl_FUNC_POLL): Add missing test.
> Reported by Justin Clift.
>
> Signed-off-by: Eric Blake
> ---
> ChangeLog |4
> m4/poll.m4 |4 ++--
> 2 files changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/ChangeLog b/ChangeLog
> index af7ba44..8d8d082
Eric Blake wrote:
> [moving to bug-gnulib]
> On 09/19/2010 01:21 AM, Justin Clift wrote:
...
>> The next step is to get all required pieces from gnulib,
>> to run autoreconf, and to invoke ./configure:
>>
>> diff --git a/bootstrap b/bootstrap
>> index 2422549..593e0b2 100755
>> --- a/bootstrap
Bruno Haible writes:
> Thanks for having reported it already [1]. Let's see how they handle it.
They solved it quickly, and I have verified that it works. I'm going to
do a gnulib build using latest GCC 4.5.1 mingw-w64 and latest wine to
see how well it compares against a build on Cygwin...
/S
[moving to bug-gnulib]
On 09/19/2010 01:21 AM, Justin Clift wrote:
On Mac OS X, the "Homebrew" build system (one of the three main ones)
has its m4 macro directory placed in a user configurable location.
This patch adds a simple way to pass this configuration information to
aclocal, using an en
* build-aux/bootstrap (check_versions): Convert libtool into
libtoolize.
(tool search): Move libtool check earlier, and look for
glibtoolize for MacOS.
(gnulib_tool_options): Auto-add --libtool when appropriate.
Reported by Justin Clift.
Signed-off-by: Eric Blake
---
I've tested this with Justin
* m4/poll.m4 (gl_FUNC_POLL): Add missing test.
Reported by Justin Clift.
Signed-off-by: Eric Blake
---
ChangeLog |4
m4/poll.m4 |4 ++--
2 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index af7ba44..8d8d082 100644
--- a/ChangeLog
+++ b/ChangeLo
Note: getdate.h is not renamed, to minimize client impact.
Besides, whenever we eventually add getdate-posix, we'll
want to use getdate.m4 and getdate.c for the implementation,
but don't need to worry about getdate.h since getdate() uses
for its header.
* modules/getdate: Mark obsolete. Move old
[re-adding bug-gnulib]
On 09/29/2010 08:48 PM, Justin Clift wrote:
On 09/30/2010 12:33 PM, Justin Clift wrote:
Any idea how to fix?
Went and found the README file in the .gnulib dir, which said how to
test if a module is compiling ok:
*
You can test that a module
* lib/mbsstr.c (knuth_morris_pratt_multibyte): Assign a raw memory
pointer returned by nmalloca() right to the variable of type
"mbchar_t *" thus removing redundant proxy variable of type "char *".
Signed-off-by: Dmitry V. Levin
---
The issue originally arose during build of vanilla grep-2.7 on
Reuben Thomas writes:
> I've just been puzzling over why the colouring and sorting sometimes
> appeared to be wrong, and discovered that this is because the output
> of pmccabe is sorted on its first column, modified cyclomatic
> complexity, but the figures actually reported by pmccabe2html are f
19 matches
Mail list logo
