On Thu, Jan 03, 2013 at 04:40:02PM -0200, Lucas De Marchi wrote:
> or you use "#include_next " inside Escape's libgen.h
> and don't bother with never ever creating Escape.h
>
> http://gcc.gnu.org/onlinedocs/cpp/Wrapper-Headers.html
>
>
> Not sure if other compilers support this, though
It'
On Tue, Dec 18, 2012 at 09:24:46AM +0900, Carsten Haitzler wrote:
> a file on disk is specified to survive reboots.. UNLESS it is specifically in
> part of the fs that is discarded on reboot. vast difference between them.
Well, most systems clean up /tmp on boot, so they discard the content as
wel
On Mon, Dec 17, 2012 at 07:57:36PM -0200, Gustavo Sverzut Barbieri wrote:
> On Mon, Dec 17, 2012 at 7:43 PM, Joerg Sonnenberger > wrote:
>
> > On Mon, Dec 17, 2012 at 07:10:01PM -0200, Gustavo Sverzut Barbieri wrote:
> > > Jokes aside, we appreciate to have E to
On Mon, Dec 17, 2012 at 07:10:01PM -0200, Gustavo Sverzut Barbieri wrote:
> Jokes aside, we appreciate to have E to run well on that, but it
> cannot block development because features that is in the standard for
> 12 years without being implemented.
First of all, shm_open is an optional part of P
On Thu, Dec 06, 2012 at 11:02:26AM -0200, Gustavo Sverzut Barbieri wrote:
> I disagree here, given that we never use the first operand as ! in any
> of our checks, just yes/no, there is no harm. The benefit is reducing
> the number of calls to test binary and also the resulting code size as
> it's
On Thu, Dec 06, 2012 at 12:57:39PM +0100, Vincent Torri wrote:
> so it's a problem with clang. Fix your CC, then
Clang defaults to C99, so this should only happen if it was explicitly
instructed otherwise...
Joerg
--
Log
On Fri, Nov 23, 2012 at 09:37:55PM +0100, Vincent Torri wrote:
> please revert. All that stuff is not correct, as I said in one of the ticket
>
> see my previous mail. Also, ecore_getopt.c **already** have alloca
> support... Please review the patches instead of appplying them blindly
More import
On Wed, Nov 14, 2012 at 03:25:35PM +0900, Carsten Haitzler wrote:
> On Tue, 13 Nov 2012 21:43:45 -0800 Dave Ray said:
>
> wtf? somehting ADDED underscores in front of the symbols.. that is so wrong...
> the src as it reads in svn atm anyway doent have these and i dont see how a
> macro could do t
On Wed, Nov 07, 2012 at 01:58:56PM -0200, Paulo Alcantara wrote:
> 2) Since CPUID instruction was available on older x86 CPUS (like 486
> as demarchi just said), should we still keep NOT using CPUID at all ?
This is not really true. CPUID was introduced with the Intel Pentium,
but some later gener
On Thu, Oct 25, 2012 at 07:32:19PM +0200, rustyBSD wrote:
> III. There is a problem with _e_fm_op_random_char().
> When wewant to randomize a string we do srand(time())
> foreachcharacter, so if the file is small enough
> to be deleted inless than a second, it's not randomized.
> And even if it's b
On Thu, Sep 27, 2012 at 07:10:29PM +0900, Carsten Haitzler wrote:
> here's the question... this is pretty lame security-wise. shouldn't we be
> military/cia/nsa spec and overwrite it at least 7 times? :) oh and this will
> probably/possible get screwed by logging fs's or flash media that may shuffl
On Fri, Sep 07, 2012 at 11:49:18PM +0900, Carsten Haitzler wrote:
> On Fri, 7 Sep 2012 11:45:35 -0300 Gustavo Sverzut Barbieri
> said:
>
> > On Friday, September 7, 2012, Enlightenment SVN wrote:
> >
> > > Log:
> > > vincent - try this on windows.
> >
> >
> > If this fixes his problems, we may
On Wed, Sep 05, 2012 at 04:39:23PM +0300, Tom Hacohen wrote:
> On 05/09/12 16:26, Daniel Willmann wrote:
> > Hi,
> >
> > On 09/05/2012 02:12 PM, Tom Hacohen wrote:
> >> free, strlen and strcpy are just a few of the many examples that don't
> >> accept NULL. Having sanity checks everywhere is not "s
On Mon, Aug 06, 2012 at 09:51:04PM +0900, Carsten Haitzler wrote:
> why MUST an option with a single - have a single char following? what rulebook
> states this? it's an arbitrary distinction that is very gnu-ish, but it is not
> a rule. who declares the law of what a long option is or is not when
On Mon, Jul 30, 2012 at 07:11:48PM +0900, Carsten Haitzler wrote:
> On Mon, 30 Jul 2012 02:38:18 -0400 (EDT) david.obo...@aliceadsl.fr said:
>
> > Hi,
> >
> > I wonder why the evas_blit_main.c, the copy is made by a loop instead of a
> > memcpy. This is a problem with an architecture or with pe
On Tue, May 29, 2012 at 02:07:24PM +0900, Daniel Juyung Seo wrote:
> but it breaks applications when they use -Werror option.
You can use -Wno-deprecated to selectively disable the warning or
-Wno-error=deprecated to keep the warning, but don't assert on it.
Make sure to put that after -Werror tho
On Wed, May 23, 2012 at 05:56:10PM +0900, Bluezery wrote:
> 2012/5/23 Hyoyoung Chang :
> > Maybe it's best time to talk about sizeof(char)
> >
> > Do we need to sizeof(char)? I just that is problem of coding convention.
> > such as (a == 0) and (!a).
> > Is there any predefined convention about thi
On Mon, May 21, 2012 at 12:15:10AM -0300, Raphael Kubo da Costa wrote:
> After talking about the previous version of the patch with raster on
> IRC, I'm sending a more conservative version that does not change the
> Linux behavior -- /dev/shm is expected to work on Linux, and the
> non-Linux case w
On Sat, May 05, 2012 at 01:18:48PM +1000, David Seikel wrote:
> Just ignoring the failed allocation and trying to use a NULL pointer
> will likely crash you anyway, but that's just being lazy. Failing
> gracefully is generally better than failing disgracefully.
Part of the problem here is that it
On Fri, May 04, 2012 at 10:17:29AM +0900, Cedric BAIL wrote:
> On Thu, May 3, 2012 at 8:57 PM, Vincent Torri wrote:
> > is it normal that no check is done when calloc is called ?
>
> It's a good question. In E17, there isn't much handling of this
> situation. I am wrong in assuming that on all un
On Mon, Apr 23, 2012 at 03:05:09PM +0200, thomasg wrote:
> XZs liblzma is LGPLv2.1 though.
No, only the fallback support for limited platforms using gnulib is.
The rest is PD.
Joerg
--
For Developers, A Lot Can Happen In
On Thu, Mar 08, 2012 at 12:51:19AM +0900, Carsten Haitzler wrote:
> On Wed, 7 Mar 2012 12:37:42 -0300 Gustavo Sverzut Barbieri
> said:
>
> > On Wed, Mar 7, 2012 at 6:05 AM, Enlightenment SVN
> > wrote:
> > > Log:
> > > to be paranoid - ensure umask for mkstemp allows no other uids oir
> > > grp
On Tue, Feb 21, 2012 at 11:33:51AM +0100, Vincent Torri wrote:
> On Tue, Feb 21, 2012 at 9:50 AM, Joerg Sonnenberger
> wrote:
> > On Tue, Feb 21, 2012 at 07:23:36AM +0100, Vincent Torri wrote:
> >> On Tue, Feb 21, 2012 at 1:18 AM, Joerg Sonnenberger
> >> wrote:
>
On Tue, Feb 21, 2012 at 07:23:36AM +0100, Vincent Torri wrote:
> On Tue, Feb 21, 2012 at 1:18 AM, Joerg Sonnenberger
> wrote:
> > On Tue, Feb 21, 2012 at 12:10:47AM +0100, Vincent Torri wrote:
> >> On Mon, Feb 20, 2012 at 11:38 PM, Joerg Sonnenberger
> >> wrote:
>
On Tue, Feb 21, 2012 at 12:10:47AM +0100, Vincent Torri wrote:
> On Mon, Feb 20, 2012 at 11:38 PM, Joerg Sonnenberger
> wrote:
> > On Mon, Feb 20, 2012 at 11:28:28PM +0100, Lionel Orry wrote:
> >> - eina: the __gnu_printf__ format attribute is only valid since GCC
> >
On Mon, Feb 20, 2012 at 11:28:28PM +0100, Lionel Orry wrote:
> - eina: the __gnu_printf__ format attribute is only valid since GCC
> 4.4. However, we can find __printf__ format before. I'm not sure they
> are equivalent though, so any hints on the fix appreciated... (tested
> with GCC 4.3.4)
They
On Wed, Dec 07, 2011 at 02:29:17AM -0500, Christopher Michael wrote:
> On 12/06/11 20:10, Enlightenment SVN wrote:
> > Log:
> > NO! you break api. you made my e sit here with a segv in a getenv.
> >because now many libraries and api's don't have prototyopes for
> >malloc/calloc and much mor
On Sat, Dec 03, 2011 at 08:08:26AM +0900, Carsten Haitzler wrote:
> On Fri, 2 Dec 2011 15:16:37 +0100 Joerg Sonnenberger
> said:
>
> > On Fri, Dec 02, 2011 at 08:40:26AM +1000, David Seikel wrote:
> > > On Thu, 1 Dec 2011 23:26:25 +0100 Joerg Sonnenberger
> > >
On Fri, Dec 02, 2011 at 07:33:15PM -0200, Iván Briano (Sachiel) wrote:
> 2011/12/2 Joerg Sonnenberger :
> > On Fri, Dec 02, 2011 at 10:16:17PM +0100, Vincent Torri wrote:
> >> so why not just do a & 0xff instead of casting ?
> >
> > The cast is IMO cleaner
On Fri, Dec 02, 2011 at 10:16:17PM +0100, Vincent Torri wrote:
> so why not just do a & 0xff instead of casting ?
The cast is IMO cleaner in the intentions. It also has the theoretical
advantage of working independent of CHAR_BIT==8.
Joerg
On Fri, Dec 02, 2011 at 08:05:43PM +0100, Sebastian Dransfeld wrote:
> On 12/02/2011 04:33 AM, Enlightenment SVN wrote:
> > Log:
> > more trivial fixes from Joerg Sonnenberger
>
> This must be rubbish. isspace() takes int as argument, not unsigned char
You are wrong. Please
On Fri, Dec 02, 2011 at 08:40:26AM +1000, David Seikel wrote:
> On Thu, 1 Dec 2011 23:26:25 +0100 Joerg Sonnenberger
> wrote:
>
> > On Mon, Nov 28, 2011 at 09:54:18PM +0900, Carsten Haitzler wrote:
> > > http://download.enlightenment.org/releases/edje-1.1.
On Fri, Dec 02, 2011 at 09:51:12PM +0900, Carsten Haitzler wrote:
> > > 2. all the chasting and changing to unsigned char for passing into isspace
> > > () - isspace() actually takes an int, not unsigned char, so this just
> > > doesn't make sense. :( (well it does according to the manual page i ha
On Fri, Dec 02, 2011 at 03:37:30PM +0900, Carsten Haitzler wrote:
> On Thu, 1 Dec 2011 17:41:17 +0100 Joerg Sonnenberger
> said:
>
> this is really late... and a lot of these patches i find questionable. i'll
> question here (things not questioned seem ok):
>
> 1. _
On Fri, Dec 02, 2011 at 04:00:46PM +0900, Carsten Haitzler wrote:
> On Thu, 1 Dec 2011 17:41:17 +0100 Joerg Sonnenberger
> said:
>
> > Hi all,
> > sorry for not finding the time for doing this earlier. Attached is a
> > patch that sorts out a number of smaller iss
On Thu, Dec 01, 2011 at 10:29:12PM -0500, Michael Blumenkrantz wrote:
> looks reasonable, but this doesn't exist in pulseaudio's code. is it patched
> during build on bsd?
The part in iochannel.c is conditionalized on HAVE_CREDS in pulseaudio.
Joerg
--
On Mon, Nov 28, 2011 at 09:54:18PM +0900, Carsten Haitzler wrote:
> http://download.enlightenment.org/releases/ecore-1.1.0-beta.tar.gz
> http://download.enlightenment.org/releases/ecore-1.1.0-beta.tar.bz2
First patch comments out a bunch of seemingly unused static functions.
Second and third patch
Self-explaining.
--- src/modules/mixer/pa.c.orig 2011-12-02 00:47:27.0 +
+++ src/modules/mixer/pa.c
@@ -350,7 +350,9 @@ con(Pulse *conn, int type __UNUSED__, Ec
INF("connected to %s", ecore_con_server_name_get(ev->server));
conn->fd = dup(ecore_con_server_fd_get(ev->server));
On Mon, Nov 28, 2011 at 09:54:18PM +0900, Carsten Haitzler wrote:
> http://download.enlightenment.org/releases/edje-1.1.0-beta.tar.gz
> http://download.enlightenment.org/releases/edje-1.1.0-beta.tar.bz2
Default is installing the examples. This introduces a dependency on
$(SRCDIR)/Vera.ttf, which i
On Mon, Nov 28, 2011 at 09:54:18PM +0900, Carsten Haitzler wrote:
> http://download.enlightenment.org/releases/eet-1.5.0-beta.tar.gz
> http://download.enlightenment.org/releases/eet-1.5.0-beta.tar.bz2
Copyright question for this. Unlike eet 1.2, this now installs a bunch
of example files by defaul
Hi all,
sorry for not finding the time for doing this earlier. Attached is a
patch that sorts out a number of smaller issues with eina on NetBSD and
Solaris. Some are noise, some are real bugs. This brings it down to one
compiler warning for iconv, which is expected (and messy).
Joerg
Use the SunP
On Sun, Nov 13, 2011 at 07:25:48PM +0100, Vincent Torri wrote:
> Albin Tonerre (Lutin) told me that there is (maybe, i don't know much
> about that) an ABI break, with a move a function from a source code to an
> inline function. I can't remember the function, but Albin can retrieve it,
> I thin
On Tue, Nov 01, 2011 at 06:11:29PM +0900, Kim Yunhan wrote:
> But it already sent to map server, and it already made useless HTTP
> connections.
What about using HTTP pipelining?
Joerg
--
RSA® Conference 2012
Save $700 b
On Mon, Oct 31, 2011 at 12:09:48AM +0200, Tom Hacohen wrote:
> On 30/10/11 23:45, Mike Blumenkrantz wrote:
> > but it won't matter, because once there is a byte which differs it will stop
> > comparing
>
> Never assume an internal implementation, especially when you are wrong.
> :) The biggest adv
On Sun, Oct 30, 2011 at 06:28:03AM -0200, Gustavo Sverzut Barbieri wrote:
> Tom is right. You just want to compare the prefix, but you're not sure
> icon_path is >= 7 bytes thus strncmp() and not memcmp(). strncmp() is
> to be used and not the traditional (strcmp()) as you just want the
> prefix.
On Fri, Apr 29, 2011 at 11:58:43AM +0200, Vincent Torri wrote:
> I'm still wondering aboutefficiency : for example, for pdf rendering,
> shouldn't the use of a pdf generic loader be slower than wht i can achieve
> in eyesight ?
Only if you fork once per page or something like that. Give it a min
On Sat, Apr 23, 2011 at 03:11:22PM -0400, Mike Blumenkrantz wrote:
> On Sat, 23 Apr 2011 21:00:11 +0200 (CEST)
> Vincent Torri wrote:
>
> >
> > Hey
> >
> > On Sat, 23 Apr 2011, Mike Blumenkrantz wrote:
> >
> > > configure.ac:429, we add --as-needed and probably shouldn't?
> >
> > what do you
On Wed, Apr 13, 2011 at 11:14:43PM +0200, Cedric BAIL wrote:
> >> > and that MAP_POPULATE does not need a too recent linux kernel ?
> >
> > This should be just conditionalized.
>
> Agreed. Both should be conditionalized with configure detection.
#ifndef MAP_POPULATE
#define MAP_POPULATE 0
#endif
On Wed, Apr 13, 2011 at 10:18:27PM +0200, Cedric BAIL wrote:
> On Wed, Apr 13, 2011 at 8:36 PM, Vincent Torri wrote:
> > On Wed, 13 Apr 2011, Vincent Torri wrote:
> >> are you sure that O_CLOEXEC and ACCESSPERMS are portable ?
>
> I hope they are. It's a pretty basic feature, being able to
> auto
On Mon, Mar 14, 2011 at 10:15:07AM +0100, Mathieu Taillefumier wrote:
> - gnutls_transport_set_ptr(cl->session, (gnutls_transport_ptr_t)cl->fd);
> + gnutls_transport_set_ptr(cl->session,(gnutls_transport_ptr_t)(long)
> cl->fd);
s/long/uintptr_t/ you mean.
Joerg
--
On Mon, Feb 28, 2011 at 02:49:06PM -0300, Gustavo Sverzut Barbieri wrote:
> How many HTML pages do you see declaring new entities? Of course
> parsing HTML with it it's better to do using the SAX so you can handle
> close-tags automatically as most people don't close things like
> or .
Actually,
On Mon, Feb 28, 2011 at 10:20:16AM -0300, Gustavo Sverzut Barbieri wrote:
> I've worked with expat before and it's way more complex to use and
> heavy. Sure, it will work and will handle namespace, and encoding...
> but many times you don't want or need it.
The point is that if you don't do that,
On Sun, Feb 27, 2011 at 08:25:06PM -0300, Gustavo Sverzut Barbieri wrote:
> Again, this is far from standards compliant as libxml2, but it's even
> far from the bloat these xml libs carry. I'd propose it to be
> integrated into Eina as it's very useful and small.
The problem with most adhoc XML pa
On Wed, Feb 23, 2011 at 08:52:04PM +0100, Vincent Torri wrote:
> ../../src/modules/mp/chained_pool/eina_chained_mempool.c: In function
> 'eina_chained_mempool_free':
> ../../src/modules/mp/chained_pool/eina_chained_mempool.c:324: error: label at
> end of compound statement
>
> we can replace "go
On Sun, Dec 19, 2010 at 02:20:21PM -0800, Michael Jennings wrote:
> On Sunday, 19 December 2010, at 22:30:18 (+0100),
> Joerg Sonnenberger wrote:
>
> > "grep -o" is not portable. Can't you do:
> >
> > svnversion -n "${SVN_REPO_PATH:-.}"
On Sun, Dec 19, 2010 at 03:36:57PM -0500, Mike Blumenkrantz wrote:
> changed:
> svnversion -n "${SVN_REPO_PATH:-.}"|grep -Eo "^[0-9]+"
"grep -o" is not portable. Can't you do:
svnversion -n "${SVN_REPO_PATH:-.}"| sed 's,^\([0-9]*\).*$,\1,'
for much the same result?
Joerg
--
On Sat, Oct 30, 2010 at 01:24:31PM +0200, Cedric BAIL wrote:
> In fact lstat will do the trick.
d_ino / d_fileno exist for the same reasons as d_type -- to avoid having
to stat in first place.
Joerg
--
Nokia and AT&T pre
On Tue, Oct 19, 2010 at 12:17:43AM +0200, Vincent Torri wrote:
>
> which means that emotion is GPL, iiuc.
No, it just means that it has to be a GPL compatible license. The normal
2-clause BSD license or LGPL is fine.
Joerg
On Wed, Aug 25, 2010 at 10:22:19PM +0200, Vincent Torri wrote:
> On Wed, 25 Aug 2010, Raphael Kubo da Costa wrote:
> > FWIW, on FreeBSD at least, limits.h includes both sys/limits.h and
> > sys/syslimits.h (the latter only if __POSIX_VISIBLE is defined).
>
> All versions ?
At least down to FreeBS
On Wed, Aug 25, 2010 at 08:32:28PM +0200, Vincent Torri wrote:
>
>
> On Mon, 23 Aug 2010, Joerg Sonnenberger wrote:
>
> >On Sun, Aug 22, 2010 at 06:08:38AM +0200, Vincent Torri wrote:
> >>>-#if defined (__MacOSX__) || defined (__FreeBSD__) || (defined (__MACH_
On Sun, Aug 22, 2010 at 06:08:38AM +0200, Vincent Torri wrote:
> > -#if defined (__MacOSX__) || defined (__FreeBSD__) || (defined (__MACH__)
> > && \
> > - defined (__APPLE__))
> > -# include
> > -#endif
> > +#include
>
> Did Joerg try that on all the platforms that were in the #define ? If n
On Sat, Aug 21, 2010 at 01:02:43PM +1000, Carsten Haitzler wrote:
> releases of Eina, Eet, Evas, Ecore, Embryo, Edje, Efreet, E_Dbus and Eeze
> from:
ecore seems to have lost ECore_X_Screen_Size. It is still used in the
xcb binding, but not defined anymore.
Joerg
---
On Mon, Aug 16, 2010 at 07:41:01AM +1000, Carsten Haitzler wrote:
> On Mon, 16 Aug 2010 00:33:33 +0200 Joerg Sonnenberger
>
> said:
>
> > On Mon, Aug 16, 2010 at 07:13:13AM +1000, Carsten Haitzler wrote:
> > > > edje:
> > > > The second paragraph is
On Mon, Aug 16, 2010 at 07:13:13AM +1000, Carsten Haitzler wrote:
> > edje:
> > The second paragraph is non-OSI and some parts are weird, e.g. the file
> > doesn't actually contain a copyright notice. I think the 2-clause BSD
> > license covers the intention and is OSI approved, but that's your cal
Hi all,
I'm just going over the various COPYING files to prepare tagging the
appropiate licenses for pkgsrc.
e_dbus:
I'm not sure if this is intentional, but this is certainly not
the MIT license and it looks like it is functionally equal or stronger
than the original BSD license. E.g. it might no
Hi all,
attached patch fixes two issues:
(1) If using a non-default prefix, share/pixmaps should be checked
before the hard-coded /usr/share/pixmaps fallback. This basically
restores symmetry with the other cases like share/icons.
(2) Do a fallback lookup if the theme is not valid or faked. This
On Sat, Aug 14, 2010 at 11:35:13PM +0200, Albin Tonnerre wrote:
> - If efreet returns an SVG icon when SVG rendering is not compiled in evas,
>then you get no icon where an xpm icon (which could have been rendered
>correctly) might have existed.
OK
> - Since there is no way to ask evas
hi all,
in src/bin/e_main.c there is this wonderful gem _e_main_test_svg_loader.
Writting a hard-coded XML file to a known location is just asking for
trouble. It basically means that anyone with write access to /tmp can
make the E17 user overwrite a file with permissions of the user. It also
creat
Hi all,
attached is a small bugfix for ecore. If the to-be-monitored path
doesn't exist, the poll monitor currently tries to remove an item from a
list which was never hooked up.
Joerg
$NetBSD$
--- src/lib/ecore_file/ecore_file_monitor_poll.c.orig 2010-08-14
16:48:51.0 +
+++ src/li
On Sat, Aug 14, 2010 at 08:36:57PM +1000, Carsten Haitzler wrote:
> What: the march to EFL 1.0.0 - first stage "ALPHA"
> Covers: eina evas ecore embryo edje efreet e_dbus
> Extras: eet (going to 1.4.0 to be in sync with efl 1.0.0)
>
> it's been a while coming. but it's pretty much about time to go
On Fri, Jul 09, 2010 at 06:57:49PM +0200, Cedric BAIL wrote:
>First to put some background around this proposal. We want to push
> more the use of Edje for application layout (thanks to the always
> improving Elementary external support). This mean that a lot of string
> end up inside the Edje
On Mon, Mar 01, 2010 at 12:13:03AM +0100, Vincent Torri wrote:
> and, btw, no, it can't be achieve this way : vsnprintf does not
> return the number of char if the buffer is not sufficiently arge
> (contrary to the libc implentation on linux). It returns -1 in that
> case
If you mean the function
On Sun, Feb 28, 2010 at 10:02:25PM +, Rui Miguel Silva Seabra wrote:
> The attached patch would seem to add http basic auth to ecore_con_url
Does it work with usernames or passwords containing ":"?
Joerg
--
Download
On Sat, Feb 20, 2010 at 05:25:38PM +0100, Massimiliano Calamelli wrote:
> Hello, i removed the permission to create a new page in Trac for
> registered users to temporary stop the spam in Trac.
Does Trac allow ACLs for that? E.g. pull all known-to-be-good users in a
group that has the privilege an
On Tue, Jan 19, 2010 at 12:31:26PM -0200, Gustavo Sverzut Barbieri wrote:
> This is not happening for reasons already said in this list. Summary
> is that some cases it is worth to share installations amongst
> different systems and architectures.
I should add that it doesn't work as well for ano
On Tue, Jan 19, 2010 at 12:31:26PM -0200, Gustavo Sverzut Barbieri wrote:
> Why do you care about these names anyway? Distros will handle this for
> you, and they should not have much work into getting these things
> correct (debian allow wildcards in their *.install, gentoo handles
> automatically
On Tue, Jan 19, 2010 at 08:01:38AM +0900, Carsten Haitzler wrote:
> if there are patches pending - bring attention to them. if there are bugs -
> speak up.
Can you please drop OS name and platform from the module pathes in EVAS
and co? It doesn't really buy anything as you can't trivially mix
arch
On Thu, Dec 31, 2009 at 11:03:33AM +1100, Brett Nash wrote:
> > %ld and unsigned int? You must be joking... Defining PRIdMAX and
> > friends can be easily automated.
>
> Unfortunately although PRIdMAX is useful, I don't believe there is one
> for size_t. Instead you use the %z modifier, which lea
On Wed, Dec 30, 2009 at 10:44:23AM -0200, Gustavo Sverzut Barbieri wrote:
> > Unfortunately there is no safe way of printing it in C89 (ie windows).
> > The best way is:
> > printf("... %ld..",(unsigned int)blah);
%ld and unsigned int? You must be joking... Defining PRIdMAX and friends
can
On Sun, Nov 15, 2009 at 06:22:45AM +0100, Vincent Torri wrote:
> what they won't like is:
>
> int i1;
> i1 = 4;
> int i2;
>
> that is, not having all the variable declaration at the beginning of the
> block
Are pre-C99 compilers a desired platform for E17?
Joerg
--
On Thu, Oct 29, 2009 at 08:53:47AM +0800, Brian Wang wrote:
> On Wed, Oct 28, 2009 at 9:11 PM, Joerg Sonnenberger
> wrote:
> > On Wed, Oct 28, 2009 at 01:41:01PM +0100, Peter Wehrfritz wrote:
> >> Your patch assumes that an UTF8 character initiates a new word. That is
&g
On Wed, Oct 28, 2009 at 01:41:01PM +0100, Peter Wehrfritz wrote:
> Your patch assumes that an UTF8 character initiates a new word. That is
> wrong for the most European language, if not for all. So there must be
> another solution.
The best bet you can do is likely iswblank(x) || wcwidth(x) == 0
On Wed, Oct 07, 2009 at 08:50:35PM +0200, Marc Andre Tanner wrote:
> Does anybody know how to correctly set up the Sans alias?
Take a look at examples from fontconfig, e.g. 30-urw-aliases.conf.
In fact, 45-latin.conf should do what you want.
Joerg
On Tue, Sep 08, 2009 at 08:52:39PM -0300, Gustavo Sverzut Barbieri wrote:
>- ultra-dirty hack to remove extra pointer. The hack bases on the
> fact that we just use ->last from the first node, that always have the
> ->prev == NULL. So we could have the first ->prev == ->last, but how
> to know
On Wed, Aug 19, 2009 at 02:36:54PM +0200, Cedric BAIL wrote:
> It's not gcc specific other decent compiler support this feature. But
> apparently not suncc, so you can patch if needed.
void arithmetic is NOT a feature. It is a historic accident and bug.
Code using it is simply broken. Nothing less
On Fri, Jun 26, 2009 at 12:30:02PM +0200, Vincent Torri wrote:
>> The other (and IMO preferable option) is to use AC_SYS_LARGEFILE.
>
> i've just tried on my distro (ubuntu 32 bits), and that macro defines
> only _FILE_OFFSET_BITS to 4. No _LARGEFILE_SOURCE or _LARGEFILE64_SOURCE.
>
> Is it a prob
On Fri, Jun 26, 2009 at 03:27:45PM +1000, Carsten Haitzler wrote:
> #define _LARGEFILE_SOURCE
> #define _LARGEFILE64_SOURCE
> #define _FILE_OFFSET_BITS 64
> #include
> #include
> #include
> #include
[snip]
The other (and IMO preferable option) is to use AC_SYS_LARGEFILE.
Joerg
-
On Wed, Jun 24, 2009 at 11:40:06PM +1000, Carsten Haitzler wrote:
> > > unmapping should help give a hint... it also looks better in top/ps :)
> >
> > The first part is already done by using madvise. It doesn't forcefully
> > trash the VM space though.
>
> neither does munmap() - it sits in cache
On Sat, Jun 20, 2009 at 09:25:59PM +1000, Carsten Haitzler wrote:
> > Is it really usefull to unmap ? can't we trust the kernel to drop it
> > when under presure ?
>
> unmapping should help give a hint... it also looks better in top/ps :)
The first part is already done by using madvise. It doesn'
On Fri, Jun 12, 2009 at 11:36:08AM -0300, Gustavo Sverzut Barbieri wrote:
> On Fri, Jun 12, 2009 at 11:08 AM, Joerg
> Sonnenberger wrote:
> > Hi all,
> > attached patch avoids the deprecated getopt_long_only functions. They
> > only exist in GNU libc, while getopt_long is
On Fri, Jun 12, 2009 at 03:11:57PM -0300, Gustavo Sverzut Barbieri wrote:
> > About ecore_getopt:
> >
> > Why is it using argv[0] in the help texts? That is (IMO) confusing,
> > wouldn't basename(argv[0]) be better?
>
> just because most tools do that, i can change that if you wish.
Please. It mi
On Fri, Jun 12, 2009 at 11:36:08AM -0300, Gustavo Sverzut Barbieri wrote:
> On Fri, Jun 12, 2009 at 11:08 AM, Joerg
> Sonnenberger wrote:
> > Hi all,
> > attached patch avoids the deprecated getopt_long_only functions. They
> > only exist in GNU libc, while getopt_long is
On Fri, Jun 12, 2009 at 04:56:29PM +0200, Vincent Torri wrote:
>>> That would be sys/time.h, but I am not sure how it works on Windows (or
>>> if it does at all). This is the version of minimal namespace pollution.
>>
>> let's Vincent say for windows :-)
>
> I don't think that entrance can be used
Hi all,
attached patch avoids the deprecated getopt_long_only functions. They
only exist in GNU libc, while getopt_long is also present e.g. in the
various BSD libcs.
Joerg
Index: src/daemon/spawner.c
===
--- src/daemon/spawner.c
Hi all,
the attached patch avoids a GNU-makeism. $< are only allowed in suffix
rules.
Joerg
Index: data/config/Makefile.am
===
--- data/config/Makefile.am (revision 41009)
+++ data/config/Makefile.am (working copy)
@@ -11,4 +1
Hi all,
e17 will curretly set LC_ALL as well as LANGUAGE and LANG. That is quite
annoying as at least on NetBSD LC_ALL overrides all other variables.
I am not aware of a system that doesn't use LANG or LANGUAGE if LC_* is
not specified, so the most sane behavior is to just drop LC_ALL, IMO.
Can thi
On Thu, Apr 23, 2009 at 12:27:22AM +1000, Carsten Haitzler wrote:
> Eet 1.2.0 release.
>
> Eina, Evas, Ecore, Embryo, Edje, E_Dbus, Efreet and Enlightenment 060 snap.
> This is svn rev 40284. These are clean and pass ditcheck with proper sonames
> and module versioning.
>
> see http://www.enlight
On Sun, Apr 19, 2009 at 08:46:31PM -0300, Gustavo Sverzut Barbieri wrote:
> And freeze is over! Packagers should use revision 40244, I documented
> this in http://trac.enlightenment.org/e/wiki/ReleaseSchedule
Could you also just role a set of snapshot tarballs, please?
Joerg
On Fri, Apr 10, 2009 at 05:47:43AM -0300, Andre Dieb wrote:
> Here's a patch for handling EBADF (Bad file descriptor) error after select()
> (following the thread "random abort caused by ecore").
I'm not sure F_GETFD exists everywhere, but what about using F_GETFL?
Joerg
On Fri, Feb 13, 2009 at 12:22:14AM +1100, Carsten Haitzler wrote:
> technically you should memset() the entire sizeof(struct) to 0 first to ensure
> it is 0.
Technically that is only silencing Valgrind and redundant from the PoV
of C. If that is the goal, memset is the right approach; packed is ve
1 - 100 of 118 matches
Mail list logo
