wine/gecko
0
0
Fork 0
mirror of https://gitlab.winehq.org/wine/wine-gecko.git synced 2024-09-13 09:24:08 -07:00
Code Issues Packages Projects Releases Wiki Activity
120,346 Commits 1 Branch 165 Tags 1.2 GiB
35629f6f28
Commit Graph

8 Commits

Author SHA1 Message Date
Ehsan Akhgari
aab1f242be Bug 834869 - AudioBuffer's Unlink method should drop js objects; r=mccr8 
--HG--
extra : rebase_source : 3ee8bc593cdffff135a7094eacee7c5d34c70848
 2013-01-25 16:21:22 -05:00
Trevor Saunders
b98e34aed0 bug 822289 - remove NS_IMPL_CYCLE_COLLECTION_CLASS and friends r=mccr8 2013-01-12 07:40:33 -05:00
Olli Pettay
f62ac2c85e Bug 811206, Fix JSHolder drop handling, part 3 (fixes), r=mccr8,khuey 
--HG--
extra : rebase_source : 7b2726d2aa7f5764a3c4cb750e5a735728069687
 2012-11-28 03:37:57 +02:00
Benoit Jacob
5a8b82a1cb Bug 807437 - Eliminate redundant Traverse/Unlink CC macros - part 1: automatic changes - r=mccr8 2012-11-15 02:32:40 -05:00
Ehsan Akhgari
1f14e63f36 Bug 808374 - Balance holding and dropping the JS object references for AudioBuffers initialized with 0 channels; r=mmcr8 2012-11-05 13:49:35 -05:00
Peter Van der Beken
0b7bff61e0 Fix for bug 799465 (Add complete support for non-nsISupports objects in new DOM bindings) - remove nsISupports from AudioContext. r=ehsan. 
--HG--
extra : rebase_source : f37b849c020bd9e4bc72e16df71a0241fafb0829
 2012-10-09 20:50:26 +02:00
Ehsan Akhgari
f8e3b2e4eb Bug 793294 follow-up: Hold on to JS objects early in AudioBuffer::InitializeBuffers 2012-09-26 11:52:01 -04:00
Ehsan Akhgari
5366306b68 Bug 793294 - Implement AudioBuffer; r=bzbarsky,smaug 
This is the full implementation of the AudioBuffer object.  There are
two ways to create these objects from an audio context and this patch
implements only one of them.

The construction of the AudioBuffer object is a two step process: the
object should be created with operator new first, and then
InitializeBuffers should be called on it.  InitializeBuffers is
fallible, because it uses the JS API to create the underlying typed
arrays, but that's fine, since the length of the buffers comes from web
content, and we don't want to use infallible allocations for those
anyways.

We hold on to the JS objects from the C++ implementation, and trace
through all of those objects, so that a GC does not kill those object
without us knowing.

The buffer should be possible to manipulate from both C++ and JS, and
the C++ object probably needs to support a set of methods for the C++
callers at some point.
 2012-09-25 17:58:50 -04:00