Commit Graph

76 Commits

Author SHA1 Message Date
Cedric BAIL 29226ee49d ecore_con: just a reminder to look at later. 2014-06-27 13:41:20 +02:00
Carsten Haitzler 4b24b386db fix duplicate function call names in ecore and ecore_con that can clash
this fixes T1341. also it removes the pointless return value from
these two functions as the return is never used. also the ifdef in
ecore_main.c seemed wrong as it wasd using fcntl not execvp but the
ifdef was for execvp. this just never was discovered, so it's slid
under the hood for a long time.
2014-06-16 15:57:05 +09:00
Raoul Hecky 06a343afda Fix curl dynamic loading on windows
With win-builds.org curl dll name is libcurl-4.dll. Try to load this
correct filename first
2014-06-03 18:09:14 +02:00
Guillaume Friloux 08a70a113f Fix warning with undefined ntohl(). 2014-05-30 14:15:27 +02:00
Jean-Philippe Andre f4a0c8054f Win64: Fix a bunch of warnings
Fix invalid casts.
Use printf("%z") where appropriate.
Fix unused variables warnings.

Thanks vtorri for the patch.

@fix
2014-05-29 20:02:16 +09:00
Carsten Haitzler 7c00bd5123 fix ecore_con_url using 100% cpu and just let fd handler do work
this fixes T1251 - @fix
2014-05-15 15:48:18 +09:00
Stefan Schmidt 9f7c9f9536 ecore_con: Force unsigned to unsigned comparison
num was changed to size_t which makes it unsigned while count is a normal
(signed) int. Comparing them does not always work as expected but in this
case we already checked if count is < 0 before and would have returned if
that would be true. Thus we can safely cast count to unsigned here as it
will be greater 0 anyway.
2014-04-11 11:50:47 +02:00
Stefan Schmidt b53a68d4c6 ecore_con: Remove superfluous >= 0 check for unsigned
Since num was changed from (signed) int to (unsigned)  size_t in
16d7b981eb this check no longer
makes sense as it will always be true.
2014-04-11 11:26:18 +02:00
Carsten Haitzler 16d7b981eb ecore-con - deal with internal buffer growing over 2g in size
@fix this fixes a corner case where you may buffer 2g or more of data
in ecore-con buffers. this leads to a stall. @fix
2014-04-04 19:36:50 +09:00
Chris Michael 24ac2e278d Fix doxy typo
Signed-off-by: Chris Michael <devilhorns@comcast.net>
2014-01-19 12:06:05 +00:00
ChunEon Park 84cd13bf96 efl/ecore_con - compare same types. 2014-01-18 18:50:51 +09:00
Cedric BAIL 313cd699f2 ecore_con: fix build on windows by matching the #if around the if statement. 2014-01-09 14:30:43 +09:00
Carsten Haitzler e8c13118eb fix mingw build for setuid fix/checks 2014-01-08 22:06:41 +09:00
Carsten Haitzler b95ef3801f setuid safeness - ensure if an app that is setuid doesn't do bad things
this makes efl ignore certain env vars for thnigs and entirely removes
user modules (that no one ever used) etc. etc. to ensure that *IF* an
app is setuid, there isn't a priv escalation path that is easy.
2014-01-08 19:46:23 +09:00
Nicolas Aguirre 503e7a5b03 ecore_con - Fix binary buffer when using ECORE_CON_REMOTE_UDP.
ERR<7807>: lib/eina/eina_binbuf_template_c.x:95 eina_binbuf_append_length() *** Eina Magic Check Failed !!!
This fix a problem where eina_binbuf was used without
calling eina_binbuf_new when ECORE_CON_REMOTE_UDP is used.
2014-01-03 18:57:56 +01:00
Daniel Juyung Seo 76d8532b54 efl: Unified eina critical manro to CRI.
Being annoyed by different types of eina critical macros - CRI, CRIT,
 CRITICAL -, I concluded to unify them to one. Discussed on IRC and
 finally, CRI was chosen to meet the consistency with other macros -
 ERR, WRN, INF, DBG - in terms of the number of characters.
If there is any missing bits, please let me know.
2013-12-26 12:27:13 +09:00
discomfitor dd666a6ffd bugfix: free the entire ecore_con dns struct on info clear to prevent crashes later on 2013-12-20 22:16:59 -05:00
discomfitor 2a143393f9 split ecore_con_dns free functions into one that removes from list and one that does the freeing 2013-12-20 22:16:29 -05:00
discomfitor 0808259eff Revert "workaround for a weird ecore-con crash"
This reverts commit c751a220d6.
2013-12-20 22:04:58 -05:00
discomfitor c751a220d6 workaround for a weird ecore-con crash
kuri ping in case I don't get to this soon it needs to be fixed somewhere else
2013-12-20 21:36:58 -05:00
discomfitor 41fe698fc2 reformat ecore_con so I can read it... 2013-12-20 00:08:43 -05:00
Carsten Haitzler 8cdef1831c ecore-con - fix possible unterminated buffers
this should fix CID 1039725 and CID 1039724
2013-12-13 21:26:05 +09:00
Carsten Haitzler 675309d9d9 ecore_con - dns.c - re-fix coverty double-free. coverity was and is right.
fixes CID 1039909 again (it's back after rebase). it is an actual bug.
2013-12-13 09:41:23 +09:00
discomfitor 5a549c8e97 rebase dns.c from upstream 2013-12-12 09:29:02 -05:00
Carsten Haitzler 1c267deb8f ecore-con - dns.c - fix double free
this fixes CID 1039909
2013-12-11 19:32:54 +09:00
Cedric Bail 8e7921dffa ecore_con: remove unecessary code. 2013-11-28 16:36:21 +09:00
Cedric Bail 1e70d703b3 ecore_con: fix unitialized use of buffer from strncpy.
Fix CID 1039725.
2013-11-27 14:30:03 +09:00
Cedric Bail d0837ee5f5 ecore_con: fix non initialized use of buf with strncpy.
This fix CID 1039723.
2013-11-27 14:27:00 +09:00
Chris Michael 164269a2da Malloc cannot accept a negative number so check 'num' parameter before
calling malloc.

Fixes Coverity CID1039352

Signed-off-by: Chris Michael <cp.michael@samsung.com>
2013-11-18 11:50:42 +00:00
Carsten Haitzler 85a86e77f7 ecore_con - socks support - warning that shows invalid size calc
zmike - i think this was the problem. gcc warning caught it. size was
2013-11-09 10:11:12 +09:00
Guillaume Friloux 838c3bed85 Fixes a bug where when you try to connect to a server using an
inexistant DNS, or having a DNS resolution error, there isnt any event
generated for the application to know.

This is not a complete fix has you need to fix a timeout value for it to
happen, but still, it makes it possible to work around it.

More work is needed on this.
2013-11-06 14:32:11 +01:00
Carsten Haitzler 8ed6324a31 ecore-con - followup commit to curl dynamic load - support win and osx names 2013-10-11 08:13:35 +09:00
Carsten Haitzler 2c1c6b9335 ecore-con: make curl support entirely runtime loaded via eina_module
this makes curl support a pure runtime-only thing. libcurl is loaded by
eina_module (dlopen/dlsym) when curl is actually first needed (when a
url connection/object is created). this means that ecore-con has no
link or compile dependencies on curl, only runtime, AND this saves
memory (due to curl inits using apparently a chunk of private pages).
so this saves memory and moves the dependency to runtime (though still
consider libcurl a dependency of efl - but like a binary executed,
it's at runtime).
2013-10-10 22:31:18 +09:00
Chris Michael cc1a2d4b6b _ecore_con_dns_free already frees the timer and whole dns structure,
so setting dns->timer = NULL here b0rks.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
2013-08-05 13:52:06 +01:00
Chris Michael 0f925220e1 Fix potential double-free of so->answer.
NB: dns_so_submit calls dns_so_reset twice (once at start, once at end
on error). As so_reset will free so->answer the first time through,
add validity check of so->answer for the second iteration.

Signed-off-by: Chris Michael <cp.michael@samsung.com>
2013-08-05 13:48:23 +01:00
Chris Michael d9f8e6b525 Do not call _ecore_con_client_flush with a NULL client.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
2013-08-05 12:39:19 +01:00
Guillaume Friloux 6aab5a9690 This patch is a port of commit 4b885b9b24de9e188c861422d7d2738b49359d22.
It fixes ecore_con_server_timeout_set().
2013-07-17 16:20:45 +02:00
Carsten Haitzler cc400c5c7f fix clang sizeof complaint 2013-07-16 20:45:24 +09:00
Sebastian Dransfeld 336f6ec328 efl: formatting 2013-06-20 13:28:18 +02:00
Sebastian Dransfeld 068f1ddc62 efl: formatting 2013-06-20 13:08:36 +02:00
Daniel Willmann 16cbbb202f ecore_con: Server start time was not initialized correctly
ecore_con_event_server_add() is never called (at least not in the
standard TCP connection case, so initialize svr->start_time in
ecore_con_server_add(). This shouldn't regress any other behaviour as
this is done at the very beginning and any later change to start_time
will still happen.

Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
2013-05-24 16:18:02 +01:00
Guillaume Friloux 7b88439346 cl->start_time isnt initialized, so ecore_con_client_uptime_get() just doesnt work. This patch fixes it. Bug found by asdfuser. 2013-05-24 15:14:09 +01:00
Guillaume Friloux 359c17be56 I missed a timer_reset for the case we did set a server timeout and no client timeout. Thanks again Cravix for the awesome reports. 2013-05-24 15:14:09 +01:00
Guillaume Friloux 996c98f36a Fixes another bug found by Cravix (IRC) where even if data is going over the socket, we still timeout the connection if ecore_con_client_timeout_set and ecore_con_server_timeout_set are used! This is even a worst bug than the previous one. This patch should also be applied and tested on 1.8. Thx Cravix! 2013-05-24 11:33:06 +01:00
Guillaume Friloux 0a293dd981 Fix a bug introduced on 2011-12-10 05:42:32 (GMT) and nobody ever noticed because there isnt enough people that cares about ecore_con (and the lower level libs in general). Big thanks to Cravix (IRC) for finding this issue! 2013-05-24 11:33:06 +01:00
Cedric BAIL 2063e4353d efl: integrate eina_log_timing. 2013-03-27 21:43:45 +09:00
Cedric Bail d4172f555a ecore_con: fix warnings when curl is not available. 2013-03-17 10:34:40 +09:00
Cedric Bail 1959898724 ecore_con: improve portability. 2013-03-16 15:40:26 +09:00
Mike Blumenkrantz 4ab02d7f6b fix writes/flushes with servers for ecore-con 2013-03-14 11:48:05 +00:00
Mike Blumenkrantz bbc7a0bf3e clean up gnutls session init for ecore-con in 1.8 2013-03-11 04:57:12 +00:00