2fafab9a0e
Add luax_checkstring/pushstring functions that handle inline \0 characters
2011-11-15 21:37:11 +01:00
cf2bfcc4a2
Fix more memory leaks (issue #327 )
2011-11-12 15:00:26 +01:00
57f9cf439e
Add LOVE_UNUSED macro to stamp out all those "unused parameter" warnings
2011-10-12 14:03:38 -05:00
f9084c4625
Thread:getKeys needs to return a 1-indexed table
2011-08-31 12:28:56 +02:00
efb04ab27c
Add love.thread.getKeys (issue #276 )
2011-08-22 18:05:09 +02:00
77ccb93506
Fix regression due to thread backends commit
2011-07-05 15:30:55 +02:00
ce5927379d
Use a lua_type switch/case in thread.send, to prevent casts (bug #252 )
...
Push non-love userdata as lightuserdata, will probably break it, but shouldn't crash love (bug #243 )
2011-07-03 18:02:53 +02:00
1cf6a6fee1
update the project for OS X and make pthreads work properly
2011-06-28 22:02:53 -05:00
8c5db2a92f
Fix up posix threading backend
2011-06-28 17:01:35 +02:00
da4c199467
Hopefully the last fixes to the win32 thread backend
2011-06-09 12:23:11 +02:00
994ad74e22
Additional cleanup
2011-06-09 11:01:11 +02:00
709263b032
Some fixes for the new threading backends
2011-06-09 10:56:13 +02:00
3460bf918c
Enabled headless build in autotools
2011-06-08 17:14:41 +02:00
1d5fca6f46
Multiple thread backends.
2011-06-08 16:54:35 +02:00
3fe9b2ff16
Reverted module selection feature. This is not really desired.
...
Users (or lovers) do not get to choose which implementation of
a module they would like to use.
The abstraction is for dealing with portability issues, not
for giving people the ability to choose an alternative
(inferior) backend. The best choice for any platform/situation
should at all times be the *default* for that platform/situation.
If this is not the case; it's a bug (that is easily fixed).
Any module should provide the functionality promised in the docs.
The lover (or user) should not care how, and if [s]he does, [s]he
does still not get to choose.
Also, a related note: duplication of wrapper code is something I
especially wanted to avoid when I first (or should I say ...
"eventually") formed the current architecture.
--HG--
branch : minor
2011-03-26 17:06:05 +01:00
b4139ffab5
Rename loaders to fit their locations
...
--HG--
branch : minor
2011-03-13 20:25:35 +01:00
2115cf19b9
Fixed a couple of compiler warnings in minor branch
...
--HG--
branch : minor
2011-03-06 17:09:38 +01:00
b0a07c7a4b
the merge screwed up the new thread API, fixed
...
--HG--
branch : minor
2011-03-05 23:18:54 -05:00
ee282759e6
merge default into minor
...
--HG--
branch : minor
2011-03-05 22:13:48 -05:00
453593c9b0
Catch some more file-opening exceptions and make them lua errors instead
2011-02-18 16:10:20 +01:00
31f8c68b4b
Fixed more locking issues in love.thread
2011-02-11 20:41:03 +01:00
fd34409bd4
Peek gets locks too
2011-02-11 19:50:34 +01:00
28c49d8fb4
getThreads now indexes with the names again
2011-02-11 18:53:23 +01:00
677df8d4f0
Merge 'default' into 'minor'
...
--HG--
branch : minor
2011-01-17 11:08:58 +01:00
6acacffb66
Threads get/set instead of send/receive (issue #166 )
...
--HG--
branch : minor
2011-01-15 22:26:16 +01:00
fde708160e
bringing in the new year by changing every 2010 to 2011
2011-01-15 14:40:27 -06:00
9d108d2bde
Added missing lua_strlen to ThreadVariant construction. MSVC coerced the 'const char*'-argument into the bool version of the constructor, causing the wrong function to be called.
2011-01-09 16:36:13 +01:00
089da5b08d
More race conditions, what fun
2011-01-03 22:19:24 +01:00
ece02577a5
More race conditions (bug #131 )
2010-12-31 11:37:50 +01:00
cfeb01db42
Prevent race condition in thread:receive (bug #131 )
2010-12-31 00:44:32 +01:00
c4fe77694b
Applied (and fixed) vrld's thread patch
2010-12-27 14:34:43 +01:00
e840a8f96c
Another embedded zero issue in threads (isue #128 )
2010-12-20 19:08:29 +01:00
11ab5600d3
Correctly handle embedded zeros in thread names and ThreadVariants (bug #128 )
2010-12-07 19:37:59 +01:00
8a7e2a0629
Thread wrapper checks for failed thread creation. ( fixes #90 )
2010-10-23 18:16:20 +02:00
fe373075a4
Thread now has access to mutex and cond as well, fixes bug where errors wouldn't get picked up
2010-09-20 00:56:37 +02:00
48a94f32d9
newThread now accepts files
2010-08-04 14:22:30 +02:00
140d3866a3
Removed ineffective and error-prone killthread in thread destructor
2010-05-23 15:36:49 +02:00
200d4bc653
It looks like a dirty hack, it probably is, but this fixes the GC
2010-05-20 21:53:50 +02:00
7c33786100
fixed a ridiculous number of gcc warnings, nearly all of which were complaints about unused parameters - the remaining 199 warnings I get are from code that isn't ours (box2d is responsible for all but 8 of them)
2010-02-26 03:55:21 -05:00
dbf8301f24
Merge (the hardest merge evahr, thanks rude)
2010-02-25 20:13:50 +01:00
051f6f6eaa
Switched global thread to love._curthread and made love.thread.getThread() return that
2010-02-25 19:52:00 +01:00
16bf591c96
Fixed a MSVC warning.
2010-02-24 21:35:21 +01:00
5fca0e9b33
Changed getThreads so that the caller must allocate the memory.
2010-02-24 21:34:55 +01:00
368765820e
Changed unnecessary pass-by-value to pass-by-reference.
2010-02-24 21:21:45 +01:00
bdafe0e455
Globals?! When did Bartbes the Sex Machine abandon reason for MADNESS?
2010-02-24 20:47:56 +01:00
b4ad36cc51
Objects extending love::Object need to have virtual destructors. They are garbage collected as love::Object.
2010-02-24 20:43:05 +01:00
4ba513b679
Bye-bye ThreadModuleRegistrar. An abstract ThreadModule will do that job.
2010-02-24 20:40:59 +01:00
4ad7634a78
updated the Xcode project with love.thread, and fixed a few gcc warnings
2010-02-22 00:25:51 -05:00
75f651c84b
getThreads works now (love.thread)
2010-02-16 16:02:07 +01:00
8457d461f8
Added a 'main' thread (love.thread)
2010-02-14 19:14:07 +01:00
Bart van Strien
Bill Meltsner
Przemysław Grzywacz
rude
Bart van Strien