|
net.server_select: Restore checks for wantread/timeout errors on receive
|
Matthew Wild |
Thu, 03 Dec 2009 19:18:18 +0000 |
|
net.server_select: Make bufferlen() method return, of all things, the buffer length
|
Matthew Wild |
Thu, 03 Dec 2009 14:08:54 +0000 |
|
net.server_select: Fix possibility of losing data sent from a socket just before it closes (thanks Sjoerd Simons)
|
Matthew Wild |
Thu, 03 Dec 2009 02:28:34 +0000 |
|
net.server: Remove redundant commented code and some trailing whitespace
|
Matthew Wild |
Mon, 15 Feb 2010 22:53:32 +0000 |
|
net.server: When libevent backend is in use, override signal.signal to use server.hook_signal
|
Matthew Wild |
Sun, 10 Jan 2010 15:26:36 +0000 |
|
Merge with tip.
|
Tobias Markmann |
Sun, 29 Nov 2009 21:33:37 +0100 |
|
net.server: Log an error when libevent is requested, but luaevent is unavailable, and don't load luaevent when not requested.
|
Waqas Hussain |
Tue, 24 Nov 2009 14:37:14 +0500 |
|
net.server: Add some comments to explain to waqas how it all works :)
|
Matthew Wild |
Sun, 22 Nov 2009 14:42:56 +0000 |
|
net.server: Small fix for addtimer() compatibility code
|
Matthew Wild |
Sat, 21 Nov 2009 17:13:21 +0000 |
|
net.server: New net.server to choose the appropriate library from server_select/server_event based on the availability of luaevent and the use_libevent config option
|
Matthew Wild |
Sat, 21 Nov 2009 02:39:08 +0000 |
|
server.lua: Standardise on 'client' variable to refer to the socket, makes the code a bit cleaner
|
Matthew Wild |
Mon, 26 Oct 2009 23:35:48 +0000 |
|
server.lua: Add addsocket(), refactor all code to use it, fixes nasty socket object leak - moral of the story is to not duplicate code, and not prematurely optimise. But you knew that already didn't you?
|
Matthew Wild |
Mon, 26 Oct 2009 23:29:04 +0000 |
|
net.server: Allow changing the sslctx of a connection after it has been established
|
Matthew Wild |
Sat, 17 Oct 2009 16:21:27 +0100 |
|
Merge with 0.5
|
Matthew Wild |
Wed, 14 Oct 2009 14:12:14 +0100 |
|
net.server: Better handling of closing sockets during send
|
Matthew Wild |
Wed, 14 Oct 2009 14:09:51 +0100 |
