-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Bruno Haible wrote:
> should be includable standalone. (See for example in
> http://www.opengroup.org/susv3xsh/opendir.html.) Therefore I would
> remove the include from the .c file and put it into
> the .h file instead. Actually, I think only size_
Derek R. Price wrote:
> Okay, I've attached the proposed lib/dirent.c, lib/dirent.h, and
> modules/dirent.
Thanks a lot!
> I wasn't sure what to do about the Makefile.am chunk of
> the modules description. Since this is only currently intended to build
> on Windows, the only thing that would re
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Okay, I've attached the proposed lib/dirent.c, lib/dirent.h, and
modules/dirent. I wasn't sure what to do about the Makefile.am chunk of
the modules description. Since this is only currently intended to build
on Windows, the only thing that would rea
Is it okay to just take some code that was
declared to be in the public domain and redistribute it under the GPL?
Yes. (Or use it in a completely proprietary product, or anything else.)
"Derek R. Price" <[EMAIL PROTECTED]> writes:
> Well, there's still the issue that the files started in the public
> domain.
Ah, sorry, I didn't know that. In that case it's OK to incorporate
the public-domain parts into gnulib, but the results (with your fixes)
would be under the LGPL, since thi
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Paul Eggert wrote:
>> I'd be happy to declare my changes to those files also public domain, if
>> it makes things any easier. If this is needed, we might need Jim to
>> declare his tweaks public domain too under the many tiny changes rule,
>
> I'd ra
"Derek R. Price" <[EMAIL PROTECTED]> writes:
> What are the rules about importing public domain code into GNULIB?
Importing public domain code into GNU code is acceptable and has been
done occasionally, but
> I'd be happy to declare my changes to those files also public domain, if
> it make
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Derek R. Price wrote:
> Actually, I made most of the recent non-trivial dirent.h/dirent.c
> changes in CVS.
[...]
> I'd be happy to declare my changes to those files also public domain, if
> it makes things any easier. If this is needed, we might need
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Bruno Haible wrote:
> Derek Price wrote:
>> I believe Bruno said Windows had an , which would imply that
>> there was a readdir() that worked but returned a struct direct *.
>
> I didn't say this. is firmly dead.
Yes, I already went back and reread
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Paul Eggert wrote:
> Thanks. I found a few other instances of checking for dirent.h,
> and a couple of places we need to include before
> (this is for POSIX before 2001). I installed this
> into gnulib, and will propagate it into coreutils shortly.
Paul Eggert wrote:
> Gnulib still uses . How should we kill this off in general?
>
> The simplest thing is to simply omit all references to , which
> is what I'm inclined to do unless someone speaks up.
is obsolete. I agree with you.
Derek Price wrote:
> I believe Bruno said Windows had an , w
"Derek R. Price" <[EMAIL PROTECTED]> writes:
> Patch attached.
Thanks. I found a few other instances of checking for dirent.h,
and a couple of places we need to include before
(this is for POSIX before 2001). I installed this
into gnulib, and will propagate it into coreutils shortly.
Index:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Paul Eggert wrote:
> The simplest thing is to simply omit all references to , which
> is what I'm inclined to do unless someone speaks up.
I'm fine with that. It was what my first pass at this did, actually.
I'm not quite sure what will happen on a
Jim Hyslop <[EMAIL PROTECTED]> writes:
> Sorry about the delay responding. I have committed the patch, with some
> trivial fine-tuning, to the CVS source. CVS's ndir.* are now dead.
OK, thanks.
Gnulib still uses . How should we kill this off in general?
The simplest thing is to simply omit all
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Derek R. Price wrote:
> Jim Hyslop wrote:
>
>>>Can you live with the current solution until I (or someone)
>>>implements a dirent substitute for the Windows build?
>
>
> It would be nice to simplify. Does the attached patch do the trick?
Sorry ab
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Jim Hyslop wrote:
> Can you live with the current solution until I (or someone)
> implements a dirent substitute for the Windows build?
It would be nice to simplify. Does the attached patch do the trick?
2006-07-07 Derek R. Price <[EMAIL PROTECTE
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Paul Eggert wrote:
> No, dirent.h is standardized here:
>
> http://www.opengroup.org/onlinepubs/009695399/basedefs/dirent.h.html
Ah, great information, thanks.
> ndir.h is obsolete and has been obsolete for years.
>
> Sounds like CVS is supplying it
Jim Hyslop <[EMAIL PROTECTED]> writes:
> I get the impression dirent.h and ndir.h are both obsolete.
No, dirent.h is standardized here:
http://www.opengroup.org/onlinepubs/009695399/basedefs/dirent.h.html
ndir.h is obsolete and has been obsolete for years.
Sounds like CVS is supplying its own
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Derek R. Price wrote:
> Paul Eggert wrote:
>
>>>"Derek R. Price" <[EMAIL PROTECTED]> writes:
>>>
>>>
>>>>Thanks, Bruno. I've attached a patch that replaces all the references
>
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Paul Eggert wrote:
> "Derek R. Price" <[EMAIL PROTECTED]> writes:
>
>> Thanks, Bruno. I've attached a patch that replaces all the references
>> to AC_HEADER_DIRENT with calls to AC_CHECK_HEADERS_ONCE([dirent.h
Paul Eggert wrote:
>> Is OpenVMS the only practical VMS target left?
>
> Yes. For now, let's remove the vmsdir.h stuff since I don't think
> anybody uses it now. If it's necessary the VMS guys can supply
> a substitude dirent.h as well.
>
> Also, we should use "__VMS" rather than "VMS" since th
"Derek R. Price" <[EMAIL PROTECTED]> writes:
> Thanks, Bruno. I've attached a patch that replaces all the references
> to AC_HEADER_DIRENT with calls to AC_CHECK_HEADERS_ONCE([dirent.h
> ndir.h]) and makes the accompanying changes in lib/*. This patch does
> s
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Bruno & bug-gnulib, sorry for the double-copy, but I realized belatedly
that it may be useful to CC bug-cvs on this one:
Derek R. Price wrote:
> Bruno Haible wrote:
>> Derek R. Price wrote:
>>> Removing references to AC_HEADER_
Bruno Haible wrote:
> Derek R. Price wrote:
>> Removing references to AC_HEADER_DIRENT (replacing with
>> AC_CHECK_HEADERS_ONCE([dirent.h]))
>
> Yes. All Unix systems nowadays have , OpenVMS too. But native Woe32
> (MSVCRT, not mingw) doesn't.
Thanks, Bruno. I'
Eric Blake <[EMAIL PROTECTED]> writes:
> By the way, my patch added an "-*- Autoconf -*-" modeline, so that it
> was easier to edit in emacs (for example, AC_ macros are colorized by
> font-lock, and emacs looks for [] instead of `' as quoting pairs). Is
> it worth doing this to other m4 files in
Eric Blake <[EMAIL PROTECTED]> wrote:
> By the way, my patch added an "-*- Autoconf -*-" modeline, so that it
> was easier to edit in emacs (for example, AC_ macros are colorized by
> font-lock, and emacs looks for [] instead of `' as quoting pairs). Is
> it worth doing this to other m4 files in g
Paul Eggert CS.UCLA.EDU> writes:
> > only the dirfd.m4 change was necessary to solve the problem at hand.
>
> OK, thanks, could you please apply that?
Committed only the dirfd.m4 changes.
By the way, my patch added an "-*- Autoconf -*-" modeline, so that it
was easier to edit in emacs (for exa
Hi Eric,
* Eric Blake wrote on Tue, Mar 14, 2006 at 06:08:02PM CET:
> CVS findutils currently tickles a bug in its use of AC_HEADER_DIRENT.
> Somehow, the combination of modules pulled in by findutils led to
> the following macro chain (among others):
>
> gl_INIT
&g
Eric Blake <[EMAIL PROTECTED]> writes:
> Huh - you're right. I reverted my getcwd.m4 changes, and everything still
> worked in findutils' ./configure, so it was only the dirfd.m4 change
> necessary to solve the problem at hand. So only half my patch needs to be
> applied.
OK, thanks, could you
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Paul Eggert on 3/14/2006 2:36 PM:
>> * dirfd.m4 (gl_FUNC_DIRFD): Use AC_REQUIRE for AC_HEADER_DIRENT.
>> * getcwd.m4 (gl_FUNC_GETCWD): Require gl_PREREQ_GETCWD outside
>> of shell conditionals.
>
>
Eric Blake <[EMAIL PROTECTED]> writes:
> 2006-03-14 Eric Blake <[EMAIL PROTECTED]>
>
> * dirfd.m4 (gl_FUNC_DIRFD): Use AC_REQUIRE for AC_HEADER_DIRENT.
> * getcwd.m4 (gl_FUNC_GETCWD): Require gl_PREREQ_GETCWD outside
> of shell conditionals.
I can
CVS findutils currently tickles a bug in its use of AC_HEADER_DIRENT.
Somehow, the combination of modules pulled in by findutils led to
the following macro chain (among others):
gl_INIT
+-gl_FUNC_DIRFD
| +-AC_HEADER_DIRENT
+-gl_FUNC_FTS
| +-AC_REQUIRE([gl_FUNC_FTS_CORE])
| +-AC_REQUIRE
32 matches
Mail list logo
