log libpurple/proxy.c @ 22819:42dfa1139b5c

age author description
Fri, 21 Mar 2008 17:26:20 +0000 Daniel Atallah Fix SOCKS5 CHAP authentication to be more robust; the current stuff depends on the attributes coming in in a specific order and the whole authentication having a maximum length of 20 bytes. Big thanks to Jonathan Rudolph ("simple" on IRC) for helping figure out what was wrong.
Sun, 16 Mar 2008 20:36:52 +0000 Mark Doliner Fix a bunch of compiler warnings caused by my addition of G_GNUC_PRINTF()
Wed, 05 Mar 2008 18:17:25 +0000 Mark Doliner Bring back the lines that free info.whatever on subsequent calls to
Wed, 05 Mar 2008 18:13:11 +0000 Mark Doliner merge of 'e4cc3b79eb04e4937c5fa764f6839a8e07397651'
Wed, 05 Mar 2008 10:00:45 +0000 Mark Doliner Improve our error handling of Gnome proxy settings.
Wed, 05 Mar 2008 09:43:31 +0000 Mark Doliner Short circuit and return out of the function early if gconftool-2
Wed, 05 Mar 2008 18:11:02 +0000 Mark Doliner disapproval of revision 'd5ef30aafa0c0ecff773017ab7ac542f9ba7b4b4'
Wed, 05 Mar 2008 09:40:16 +0000 Mark Doliner These things are always NULL, so these if statements are never triggered
Wed, 20 Feb 2008 03:04:47 +0000 Daniel Atallah Patch from eperez to plug a leak in the ntlm proxy authentication. Fixes #4702
Sat, 03 Nov 2007 17:52:28 +0000 Nathan Walp replace most calls to strerror with calls to g_strerror. strerror will return
Fri, 26 Oct 2007 04:59:23 +0000 Ka-Hing Cheung Patch from Andrew Gaul to constify a bunch of static variables to reduce
Sun, 21 Oct 2007 05:02:09 +0000 Luke Schierer applied changes from e33c17e58da161b8cffaf13e9d79e9bb4b9ce4ce
Sun, 21 Oct 2007 04:46:33 +0000 Luke Schierer applied changes from 2762c6075c0dc52a96098c5478c5bf68cfd890a3
Sun, 16 Sep 2007 01:05:22 +0000 Sean Egan remove gpl boilerplate from doxygen docs
Fri, 14 Sep 2007 18:26:20 +0000 Daniel Atallah Apply a similar fix to the proxy prefs loading fix to the status scores.
Fri, 14 Sep 2007 17:55:05 +0000 Daniel Atallah Fix proxy settings to be loaded correctly when libpurple is initialized. This was broken because it relied on the prefs being loaded after proxy initialization so that the pref callbacks would be triggered and I changed the prefs loading to happen right at the beginning to fix other issues. Fixes #3092.
Thu, 13 Sep 2007 01:25:22 +0000 Daniel Atallah This is parts 1 and 2 of patch #2910 from Vladimir (thecrux) to improve NTLM proxy handling. Specifically, it specifies that we're using Unicode strings and uses the hostname of the local machine instead of incorrectly using the proxy server's hostname for authentication. References #2910.
Sat, 08 Sep 2007 03:09:35 +0000 John Bailey The FSF changed its address a while ago; our files were out of date.
Thu, 06 Sep 2007 06:59:23 +0000 Kevin Stange merge of '906281124efc7c5fcab1d5a7b8066c6a99d7927d'
Wed, 05 Sep 2007 22:33:22 +0000 Sadrul Habib Chowdhury Patch from QuLogic. Fixes #2903 ('Missing newlines in debug messages.')
Sun, 02 Sep 2007 00:03:43 +0000 Sadrul Habib Chowdhury disapproval of revision 'c672802b647f1230fbd9b0edf383fb3d558cf719'
Sat, 01 Sep 2007 23:50:22 +0000 Sadrul Habib Chowdhury Tooltips can be for the buddylist or for conversation windows.
Sat, 02 Jun 2007 06:26:24 +0000 Richard Laager Eliminate some type punning warnings in proxy.c. This was inspired by a
Wed, 30 May 2007 03:01:02 +0000 Nathan Walp fix socks5 from over-reading the headers, and throwing away data (fixes jabber file receiving)
Sat, 12 May 2007 14:28:59 +0000 Stu Tomlinson Fix for detecting gnome proxy settings from ticket #935
Wed, 25 Apr 2007 23:55:56 +0000 Sean Egan Works for me! Renames prefs: /core to /purple, /gaim/gtk to /pidgin, /gaim/gnt to /finch
Sun, 01 Apr 2007 02:17:06 +0000 Evan Schoenberg I commonly see a crash in which socket_ready_cb(), shortly after a laptop wakes from sleep, is passed invalid (previously freed) connect_data. It looks like this:
Mon, 19 Mar 2007 07:01:17 +0000 Sean Egan sed -ie 's/gaim/purple/g'
Fri, 23 Feb 2007 05:17:20 +0000 Evan Schoenberg Added a new GaimEventLoopUiOps item, input_get_error(). This function allows the UI to return the current error status on a socket/input. If the UI does not implement it (as Pidgin and gntgaim do not, since glib's handling of sockets is sane), it is just a wrapper around getsockopt(). Implemented or not, its return values should match those of getsockopt() with a level of SOL_SOCKET and an option of SO_ERROR. For curious souls, Adium will be using this to provide a working version of getsockopt(); the CoreFoundation CFSocket class which is used for socket read/write calls getsockopt() with SO_ERROR itself, thereby clearing the error flag [as documented in getsockopt()'s behavior], so depending upon it for determining if an error occurred leads to significant misbehavior.
Fri, 02 Feb 2007 23:14:20 +0000 Nathan Walp clean up some compile warnings