Differences From Artifact [6b86d2ab72]:
- File src/OFKernelEventObserver.m — part of check-in [bd05eb8de3] at 2015-04-26 10:40:10 on branch trunk — OFKernelEventObserver: Rename a private method (user: js, size: 9128) [annotate] [blame] [check-ins using]
To Artifact [57b253b8a0]:
- File
src/OFKernelEventObserver.m
— part of check-in
[f9ceddcb7d]
at
2015-04-26 10:42:58
on branch trunk
— OFKernelEventObserver: Keep FD -> object mapping
Removing the FD from the FD -> object mapping results in
OF_processReadBuffers passing nil to objectIsReadyForReading: if there's
still something in the read buffer.Additionally, this also broke when observing the same object for reading
and writing and then removing it for either reading or writing. (user: js, size: 9210) [annotate] [blame] [check-ins using]
︙ | ︙ | |||
325 326 327 328 329 330 331 | count: _maxFD + 1]; } _FDToObject[fd] = object; } if ((action & QUEUE_ACTION) == QUEUE_REMOVE) { | > | | > > | 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 | count: _maxFD + 1]; } _FDToObject[fd] = object; } if ((action & QUEUE_ACTION) == QUEUE_REMOVE) { /* * FIXME: What to do? There might still be an * object that has data in the read * buffer and needs the map! */ } switch (action) { case QUEUE_ADD | QUEUE_READ: [_readObjects addObject: object]; [self OF_addFileDescriptorForReading: fd]; |
︙ | ︙ |