aboutsummaryrefslogtreecommitdiffstats
path: root/lib/http_client.c
Commit message (Collapse)AuthorAgeLines
* Remove the ARCH / CPU defines.Jelmer Vernooij2016-05-15-1/+1
| | | | | | | These only reflect on what arch/cpu bitlbee was built, not on which it is running. This makes the Debian package unreproducible. See e.g. https://tests.reproducible-builds.org/rb-pkg/testing/i386/bitlbee.html
* Use proxy_disconnect() in http, ssl, jabber, oscardequis2015-10-26-1/+1
| | | | | | | | | | Twitter and MSN are all HTTP/SSL, so they don't need it either. The out of tree facebook and steam plugins are also covered by the HTTP/SSL changes. Yahoo is written in a weird way and doesn't seem to need it (it seems it doesn't immediately stop connections when you tell it to logout)
* Allow NULL callback functions in http_dorequestArtem Savkov2015-10-08-3/+7
| | | | | | | | | Check callback function supplied to http_dorequest and only run it if it is not NULL. While it is not the usual case there are some times when there is no need to check the results of a http request. Using a NULL pointer is much more convenient than creating noop functions.
* Reindent everything to K&R style with tabsIndent2015-02-20-428/+411
| | | | | | | Used uncrustify, with the configuration file in ./doc/uncrustify.cfg Commit author set to "Indent <please@skip.me>" so that it's easier to skip while doing git blame.
* Replace isdigit/isalpha/.../tolower/toupper with glib variantsdequis2015-01-16-2/+2
| | | | | | | | | | | This fixes warnings about passing signed chars to them (apparently they are implemented as macros that do array lookups without checks in some platforms, yay) Specifically: functions=isalnum|isalpha|isdigit|isspace|isxdigit|tolower|toupper sed -ir "s/$functions/g_ascii_&/g" **/*.c
* Fix incorrect Free Software Foundation addressMatej Cepl2015-01-16-2/+2
|
* Some more g_source_remove warning fixesdequis2014-10-27-0/+3
| | | | | - http_incoming_data calling itself - jabber_bs_free_transfer calling jabber_si_free_transfer
* Prevent some "Source ID ## was not found..." warningsdequis2014-09-27-0/+3
| | | | | | | These appear in glib 2.40, and, well, are pretty much useless. But people complain about them anyway. Probably fixes trac ticket 1151, at least partially.
* Merging HTTP/1.1 branch. This implements HTTP/1.1 support in http_client.Wilmer van der Gaast2013-06-16-70/+183
|\ | | | | | | | | | | | | Little benefit as I'm not burning my fingers on keepalive connecitons for now, but eventually the Twitter streaming API is going to drop 1.0 support: https://dev.twitter.com/blog/deprecating-http-1.0-streaming-api
| * Use HTTP/1.1 by default in the Twitter module, and stick to the initiallyWilmer van der Gaast2013-06-16-3/+15
| | | | | | | | | | used protocol version when internally handling redirects.
| * Add better handling of HTTP/1.1 and/or keepalive connections. This shouldWilmer van der Gaast2013-06-09-68/+164
|/ | | | | | let me close #641, and more importantly, prepares the Twitter module for an upcoming API change. https://dev.twitter.com/blog/deprecating-http-1.0-streaming-api
* Whoops, remove those debugging printfs as I intended to do before merging.Wilmer van der Gaast2012-12-24-2/+1
|
* Clean up streaming HTTP event handler, which should fix crash-bugs with GLib.Wilmer van der Gaast2012-11-25-0/+3
| | | | | Forgot to do this so far because my libevent code let me get away with it.
* Increasing http_client buffersize to 4KB. You'd think that with things likeWilmer van der Gaast2012-11-24-1/+1
| | | | | | | | | MTU in mind it's not a very useful change, but I seem to be getting data in bigger chunks most of the time. Likely because with SSL you have record (?) boundaries and compression. This should reduce I/O overhead just a little bit.
* Detect and handle streaming connection loss.Wilmer van der Gaast2012-11-11-0/+2
|
* Mostly finished HTTP streaming support: Shrink the buffer and add aWilmer van der Gaast2012-11-11-3/+25
| | | | | http_close().
* Very immature code for reading from the streaming API. It reads from aWilmer van der Gaast2012-11-10-0/+1
| | | | | | fixed URL and tried to parse individual JSON objects. Not doing anything useful with it.
* Reworked http_client a little bit to support streaming besides just bufferingWilmer van der Gaast2012-11-10-47/+110
| | | | | a complete response before giving it back to the caller.
* Since I can't figure out where the stalls are coming from at this point, atWilmer van der Gaast2012-09-22-1/+1
| | | | | | | | least have a work-around for it. After hitting the Twitter timer a few times while a previous fetch still seems to be running, close the connection. Auto-reconnect will do the rest.
* Solve a whole bunch of Twitter module crashes: Twitter responses seem toWilmer van der Gaast2012-09-22-2/+15
| | | | | | be getting truncated sometimes. Treat this as an error in http_client already instead of returning partial data.
* Merging SSL certificate verification for GnuTLS, with help from AopicieR.Wilmer van der Gaast2011-12-23-5/+16
|\
| * Add verify argument to ssl_connect() so HTTPS-based stuff is also secure.Wilmer van der Gaast2011-12-19-3/+14
| | | | | | | | | | (Think of Twitter, but also MSN/Yahoo! authentication.)
| * Initial merge of tls_verify patch from AopicieR.Wilmer van der Gaast2011-12-19-2/+3
|/
* Support HTTP/1.1 redirect status codes and use HTTPS for OAuth setup. ThisWilmer van der Gaast2011-12-11-6/+22
| | | | | | is required for identi.ca and really should be done for Twitter as well. Twitter OAuth is still broken though, it seems to disagree about signatures.
* Somewhat improve debug logging of HTTP/SOAP stuff.Wilmer van der Gaast2011-12-02-0/+10
|
* GnuTLS now also needs ssl_pending() implemented. Bug #860.Wilmer van der Gaast2011-11-14-1/+4
|
* Cleanup of http_client fix. Use g_strdup_printf and completely avoid strcat,Wilmer van der Gaast2011-11-13-14/+21
| | | | | truncate request if we switched from POST to GET.
* Undoing old workaround for MSN troubles and added more proper fix. TheWilmer van der Gaast2011-11-12-4/+7
| | | | | | http_client module needs some refactoring though. Will do that later.. This should hopefully fix bug #850.
* Send Connection: close headers so webservers won't try keepalive connectionsWilmer van der Gaast2011-04-18-0/+1
| | | | | which http_client really can't deal with.
* Merging mainline.Wilmer van der Gaast2010-07-17-4/+6
|\
| * No idea why http_dorequest() ever returned void*. Don't hide the type, it'sWilmer van der Gaast2010-07-17-3/+4
| | | | | | | | | | not a secret (the pointer is shared with a type later anyway).
| * Don't be a dumbass and stop following redirects if there doesn't seem toWilmer van der Gaast2010-07-16-1/+2
| | | | | | | | | | be an end.
* | Make purple use BitlBee's event handling API. Since the APIs never reallyWilmer van der Gaast2009-10-11-3/+3
|/ | | | | | | | | | diverged too much this is fairly transparent. I did rename and redefine GAIM_INPUT_* variables to really make it work without adding another stupid layer in between. One problem left, the new libpurple input API doesn't care about return values. Fixing that in the next CL.
* Fixed crappy memory management in http_client.Wilmer van der Gaast2008-09-28-9/+10
|
* Moving all generic files to lib/ instead of having some in / and some inWilmer van der Gaast2006-06-25-0/+453
protocols/, and adding RC4 code.
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-26 23:07:02 +0000