Eric Blake <[EMAIL PROTECTED]> wrote:
> According to Bruno Haible on 8/18/2007 5:12 PM:
>>>
>>> results in calling close_stdin first. From there, if close_stdin detects
>>> an error, there is the potential for stdout/stderr to be unflushed (and
>>> thus output data lost) when close_stdin calls _e
Bruno Haible <[EMAIL PROTECTED]> wrote:
> Eric Blake wrote:
>> m4 already calls close_stdin always, whether or not it read from stdin, so
>> the current implementation is correct even when stdin is still in the
>> initial state. However, if you want to optimize it further, such patches
>> are wor
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/18/2007 5:12 PM:
>>
>> results in calling close_stdin first. From there, if close_stdin detects
>> an error, there is the potential for stdout/stderr to be unflushed (and
>> thus output data lost) when close_stdin calls
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/18/2007 4:36 PM:
> Eric Blake wrote:
>> Now that POSIX 200x is standardizing getline and getdelim as part of
>> , should we get rid of getline.h and beef up stdio_.h, then
>> adjust all the callers (such as yesno.c) to as
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/18/2007 5:02 PM:
> Hi Eric,
>
> lseek on BeOS has another deficiency: In some cases it can modify the
> file offset although the same call fails. POSIX says this should not happen.
> But I don't think it's worth working
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/18/2007 4:59 PM:
> Hi Eric,
>
> The lseek configure-time test fails not only on mingw, but also on BeOS.
> Here is a proposed fix.
>
> 2007-08-18 Bruno Haible <[EMAIL PROTECTED]>
>
> * lib/lseek.c: Include .
>
Eric Blake wrote:
> m4 already calls close_stdin always, whether or not it read from stdin, so
> the current implementation is correct even when stdin is still in the
> initial state. However, if you want to optimize it further, such patches
> are worth considering.
Here is a primitive function t
Eric Blake wrote:
> > To make this API more easy to understand, how about this?
> > 1) Rename the function 'close_stdin' to 'close_stdin_stdout',
> > 2) Add a function 'close_stdin', which closes stdin only. In the
> > implementation, arrange that when 'close_stdin' and
> > 'close_stdin_s
Hi Eric,
lseek on BeOS has another deficiency: In some cases it can modify the
file offset although the same call fails. POSIX says this should not happen.
But I don't think it's worth working around it in lib/lseek.c. Here is
instead a proposed workaround for the unit test:
2007-08-18 Bruno Hai
Hi Eric,
The lseek configure-time test fails not only on mingw, but also on BeOS.
Here is a proposed fix.
2007-08-18 Bruno Haible <[EMAIL PROTECTED]>
* lib/lseek.c: Include .
(rpl_lseek): Add workaround code also for Unix platforms.
Needed for BeOS.
* doc/functi
On 8/18/07, Eric Blake <[EMAIL PROTECTED]> wrote:
> handler. But that is more invasive, as there are already a number of
> programs using atexit(close_stdout).
As a user of these functions, if a gnulib change would produce a bug
in my code unless I change it, that's likely to cause a problem. T
Eric Blake wrote:
> Now that POSIX 200x is standardizing getline and getdelim as part of
> , should we get rid of getline.h and beef up stdio_.h, then
> adjust all the callers (such as yesno.c) to assume this?
Hmm, can't we wait until the POSIX standard with these changes is official?
Arguments pr
> > 2007-08-18 Bruno Haible <[EMAIL PROTECTED]>
> >
> > * lib/rpmatch.c: Include langinfo.h.
> > (N_): New macro.
> > (try): Copy the pattern into safe memory before caching it.
> > (localized_pattern): New function/macro.
> > (rpmatch): Use it. Add translator comments.
> >
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/18/2007 11:38 AM:
> Hi Eric,
>
>> How about the following patch?
>> * lib/closein.c (close_stdin_only): New function.
>> (close_stdin): Make closing stdout optional.
>> * lib/closein.h (close_stdin_only):
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Now that POSIX 200x is standardizing getline and getdelim as part of
, should we get rid of getline.h and beef up stdio_.h, then
adjust all the callers (such as yesno.c) to assume this?
- --
Don't work too hard, make some time for fun as well!
Eric B
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/18/2007 1:25 PM:
> Eric Blake wrote:
>>> + translated_pattern = nl_langinfo (nl_index);
>> Not reliable. nl_langinfo is entitled to use static storage that can be
>> overwritten by later calls to nl_langinfo or by
Eric Blake wrote:
> > + translated_pattern = nl_langinfo (nl_index);
>
> Not reliable. nl_langinfo is entitled to use static storage that can be
> overwritten by later calls to nl_langinfo or by subsequent changes to the
> locale. You will need to do something like strdup to avoid problems
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Eric Blake on 8/18/2007 10:34 AM:
> According to Paul Eggert on 8/17/2007 5:43 PM:
>> Eric Blake-1 <[EMAIL PROTECTED]> writes:
>
>>> is it reasonable to have yesno install an atexit
>>> handler on first invocation? If the handler is not
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/18/2007 11:53 AM:
> Paul Eggert replied to Eric Blake:
>>> is it reasonable to have yesno install an atexit
>>> handler on first invocation? If the handler is not present,
>>> then stdin was never used (at least not by y
Paul Eggert replied to Eric Blake:
> > is it reasonable to have yesno install an atexit
> > handler on first invocation? If the handler is not present,
> > then stdin was never used (at least not by yesno).
>
> Yes, that sounds reasonable to me.
Still, I haven't understood what is so special abo
Hi Eric,
> How about the following patch?
> * lib/closein.c (close_stdin_only): New function.
> (close_stdin): Make closing stdout optional.
> * lib/closein.h (close_stdin_only): Add declaration.
This API of closein.h can easily lead to bugs: Say, one piece of code does
at
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Paul Eggert on 8/17/2007 5:43 PM:
> Eric Blake-1 <[EMAIL PROTECTED]> writes:
>
>> is it reasonable to have yesno install an atexit
>> handler on first invocation? If the handler is not present,
>> then stdin was never used (at least not
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/18/2007 9:42 AM:
> Eric Blake wrote:
>> The yesno module also has a potential POSIX-compliance problem
>> on all platforms. Inside rpmatch, the code uses _("^[yY]") to
>> get the locale-specific regex to use as the yesex
Eric Blake wrote:
> The yesno module also has a potential POSIX-compliance problem
> on all platforms. Inside rpmatch, the code uses _("^[yY]") to
> get the locale-specific regex to use as the yesexpr. However,
> this is not necessarily the same as using nl_langinfo(YESEXPR).
Why is this a POSIX
Paul Eggert wrote:
> I can do
> the other part too, if you like, but I figured you'd rather do that.
Yes. Just committed this:
2007-08-18 Bruno Haible <[EMAIL PROTECTED]>
* modules/fstrcmp: New file, from GNU gettext with modifications.
* lib/fstrcmp.h: New file, from GNU gette
Paul Eggert wrote:
> I installed the following patch into gnulib to accomplish the first part.
Thanks. Good to see that we can finally achieve code sharing here, through
the use of parametrizable include files.
> One piece (a diffseq module) just for diffseq.h, and the other piece
> (fstrcmp, say
26 matches
Mail list logo
