fc2c79ad5f
Add EdgeShape:set/getNext/PreviousVertex. ChainShape:setNext/PreviousVertex can now take no arguments to disable ghost vertices.
...
Resolves issue #1191 .
2016-08-01 10:15:34 -03:00
3157e5df7b
Added ChainShape:getNext/PreviousVertex (see issue #1191 ). Thanks airstruck!
2016-07-31 14:25:12 -03:00
55cb2fad84
Fixture:setCategory and Fixture:setMask now accept a table of categories/masks (resolves issue #1178 ).
2016-07-31 11:51:53 -03:00
0a534d4751
Fixed ChainShape:setPreviousVertex.
2016-07-28 18:08:27 -03:00
f4999bd2c7
Fail gracefully when auto.lua is called on a file not in src/scripts
2016-07-28 17:15:52 +02:00
a11c7278c8
Fix FT_Get_Kerning using char codes instead of glyph indices.
2016-07-20 21:52:56 -03:00
8ea1edbc2a
Cause a Lua error instead of silently failing if a vertex buffer memory allocation fails.
2016-07-17 17:44:38 -03:00
877ebbf9c9
Make WheelJoint/PrismaticJoint:getAxis return a world vector (refs #1176 )
2016-07-15 12:20:58 +02:00
65f53a491f
Make MouseJoint:getBodies() return only one body ( fixes #1179 )
...
Override Joint's Body getters in MouseJoint, to make the second (user-supplied) body the first body returned, and return nil for the second (internal) body
2016-07-14 16:39:29 +02:00
64ee986984
Make LÖVE buildable (barely) with Megasource on Mac.
2016-07-14 16:29:45 +02:00
7266fa4fc5
Fix memory leak in Text:set ( fixes #1181 )
2016-07-14 14:22:38 +02:00
f66011de2f
Return correct size from b64_decode.
...
The current implementation does not take into account padding (and yet it does
actually depend on padding), or skipped whitespace/gibberish characters.
The actual size can be computed by subtracting the padding and gibberish bytes,
but in this case it's easier to just return the actual number of bytes written.
2016-07-12 21:25:23 +02:00
9a2f99dc3d
Added 2-point version of love.physics.newRevoluteJoint ( fixes #1175 )
2016-07-07 21:18:35 +02:00
24da8433d0
Expose axis vector of PrismaticJoint and WheelJoint ( fixes #1176 )
2016-07-07 21:13:49 +02:00
debea1192b
Fixed large burst of particles when dramatically increasing the emission rate of a ParticleSystem. ( fixes #1169 )
2016-06-21 15:48:49 +02:00
b328049517
Fix print coloring bug with empty strings ( fixes #1171 )
2016-06-11 19:51:51 +02:00
9900051647
Improved performance of ImageData:mapPixel when JIT compilation is unavailable (e.g. on iOS and Android)
2016-05-21 22:18:12 +00:00
d6310ca87f
Fixed glyphs being added to partially out-of-bounds positions in Font texture atlases in rare cases (resolves issue #1162 ).
2016-05-11 21:38:12 -03:00
9bebe9e4e8
Updated the version.
2016-05-02 18:03:57 -03:00
d28bc1f42e
Fixed a warning about a missing newline at the end of a file.
2016-05-01 21:04:01 -03:00
67c5eb77ae
Put methods for getTexture* on Quad
2016-04-14 18:36:47 -04:00
b2b02c4f8e
Exposed getTextureDimensions on Quad
2016-04-14 18:32:39 -04:00
aab6dbda5f
Fixed love.event.quit on un-fused iOS games to properly exit to the game list instead of just restarting the current game.
2016-04-10 22:11:28 -03:00
ef5411e8ee
Fix building for Windows XP.
2016-04-10 21:58:03 +00:00
2cbaa7b3ae
Better fix, hopefully
2016-04-10 15:59:03 +00:00
08671ae48e
Fix building for Windows desktop
2016-04-10 15:51:50 +00:00
c12005b4bb
Forgot one LOVE_NOMPG123.
2016-04-10 06:29:58 +02:00
4ef7504c05
Fixed things pointed out by Alex Szpakowski and Bart von Strien
2016-04-09 07:28:03 +02:00
b2291fb62a
Fixes to make Löve compile on Windows 10 UWP
2016-04-07 22:56:03 +02:00
03801e49d6
Fix a potential issue when an OpenAL source that was previously static is reused for a streaming Source.
2016-04-03 09:58:17 -03:00
901cc362df
Added 'shaderswitches' to love.graphics.getStats.
2016-04-02 18:14:52 -03:00
b365574bff
Re-apply enet patch to use autoconf results
2016-04-02 17:25:38 +02:00
62c97381e9
Updated ENet to 1.3.13.
2016-04-02 12:10:44 -03:00
175c3af10b
Updated stb_image to 2.12.
2016-04-02 12:02:39 -03:00
1b1572ad85
love.filesystem.mount accepts DroppedFiles in the first parameter. You could previously pass in DroppedFile:getFilename() and it would work too, this just makes it a bit nicer.
2016-03-31 17:58:27 -03:00
8dd7cc16ab
Fix end-of-stream detection for videos
...
Fixes hang when a video file contains frames past the end of the video stream
2016-03-29 22:43:20 +02:00
3d25416e14
Removed an unused variable.
2016-03-27 11:28:34 -03:00
aa88383793
Fixed per-vertex colors (e.g. spritebatch colors) improperly applying gamma correction when the default shader is used and gamma correction is requested but not supported on OpenGL ES.
2016-03-26 12:01:34 -03:00
829cf5fd06
Fixed ImageFonts and BMFonts with colored bitmap images to proper sRGB conversions when gamma correction is enabled.
2016-03-25 20:13:49 -03:00
090c50795a
Fix love.window.setMode crashing if a Canvas is active.
2016-03-22 18:05:26 -03:00
ea2aadf888
Cleaned up the code for main() a bit.
2016-03-21 20:50:20 -03:00
8b07bad15e
Fixed the build.
2016-03-20 22:36:54 -03:00
9ebd759851
Don't call glGetFloat every time love.graphics.getSystemLimits is called.
2016-03-20 22:00:24 -03:00
bc932aecef
Move spawn.h include, because android doesn't use (or have) posix_spawn
2016-03-20 22:28:14 +01:00
2af327dcf9
Make sure android includes the right headers for System.cpp too
...
Hopefully.
2016-03-20 10:44:36 +01:00
36cc48d306
Move the Timer::sleep implementation up from timer/sdl/Timer.cpp to timer/Timer.cpp, since it doesn't directly use any SDL-specific code.
2016-03-18 23:01:49 -03:00
b983588d2e
We don't need to initialize the SDL timer subsystem to use SDL_Delay, and initializing it causes SDL to create a worker thread (which it uses for SDL_AddTimer and friends, which we don't use).
2016-03-18 22:58:48 -03:00
f7d226d072
Improve reaping of zombie processes on linux ( resolves #1042 )
...
Sadly, it's not a complete fix for #1042 , as there seems to be no good way to
deal with system() in multithreaded applications. At least os.execute can
return values other than -1 now.
2016-03-14 17:14:49 +01:00
d0c0240e09
Very minor channel / variant performance improvements and cleanup.
2016-03-12 19:11:03 -04:00
a0bbcce8ee
The video decoding thread efficiently waits for a video to be created before doing anything, rather than sleeping for 2ms in a loop. Reduces the CPU usage of the video worker thread to ~0% when no videos are loaded.
2016-03-05 21:56:51 -04:00
Alex Szpakowski
Bart van Strien
Leonardo Etcheverry
rude
rude
EntranceJew
T-Bone