Debian Codegraph @

ChangeLog

raw
1 === release 1.28.4 ===
2
3 2026-06-12 13:19:43 +0100 Tim-Philipp Müller <[email protected]>
4
5 * gstreamer.doap:
6 * meson.build:
7 Release 1.28.4
8
9 2026-05-12 16:08:30 +0200 Carlos Falgueras García <[email protected]>
10
11 * tests/check/gst/gstmemory.c:
12 tests: fix support with glib <= 2.67.2
13 These tests use "g_memdup2()" but they don't include
14 "gst/glib-compat-private.h", which makes the compilation fail when linking
15 against Glib <= 2.67.2 as "g_memdup2()" was included at 2.67.3
16 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11618>
17
18 2026-06-03 12:00:58 +0200 Stéphane Cerveau <[email protected]>
19
20 * gst/gstvalue.c:
21 gstvalue: free existing value in list append
22 In gst_value_unique_list_append_and_take_value, the append_value
23 is transfer-full, so if this value is already existing, it should
24 be unset inside the method.
25 Tensor elements append one caps per model output into a unique list
26 (transfer-full). When two outputs share the same caps signature (e.g.
27 YOLOv8-seg), the list deduplicates the second — but the drop path leaked
28 that caps instead of releasing it.
29 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11756>
30
31 2026-06-02 18:55:35 +0200 Stéphane Cerveau <[email protected]>
32
33 * gst/gstcaps.c:
34 * tests/check/gst/gstcaps.c:
35 caps: flag nested caps in static caps as may-be-leaked
36 As static caps are usually leaked, then the nested caps
37 are also leaked and should be set GST_MINI_OBJECT_FLAG_MAY_BE_LEAKED
38 as well.
39 The leak has been spotted with leak tracer and the onnx elements which
40 use for example:
41 static GstStaticCaps scaps = GST_STATIC_CAPS (
42 "video/x-raw,"
43 "tensors=(structure)["
44 "tensorgroups,"
45 "out=(/uniquelist){"
46 "(GstCaps)["
47 "tensor/strided,"
48 "tensor-id=(string)out,"
49 "type=(string)float32"
50 "]"
51 "}"
52 "]");
53 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11756>
54
55 2026-05-28 11:07:24 +0200 Stéphane Cerveau <[email protected]>
56
57 * gst/gstvalue.c:
58 * tests/check/gst/gstvalue.c:
59 gstvalue: fix crash when converting NULL G_TYPE_VALUE_ARRAY to G_TYPE_STRING
60 In the case of Rust binding, converting to G_TYPE_STRING
61 a G_TYPE_VALUE_ARRAY might lead to a crash if the GValue contains NULL.
62 Consider it as an empty string.
63 Reproduce with vp8enc:temporal-scalability-target-bitrate and
64 GstPipelineStudio.
65 Add defensive check in gst_value_transform_any_list_string
66 Add additional tests for array, list and set.
67 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11737>
68
69 2026-05-20 11:21:45 +0800 Qian Hu (胡骞) <[email protected]>
70
71 * gst/gstbufferpool.c:
72 bufferpool: avoid leaking partially preallocated buffers
73 Preallocate buffers into a temporary array in default_start() before handing
74 them to release_buffer(). If allocation fails midway, free the buffers that
75 are still owned by the base class instead of leaving already released buffers
76 behind.
77 This avoids leaking buffers and corrupting cur_buffers when preallocation
78 fails, while still allowing subclasses to take ownership in release_buffer()
79 only after all allocations succeeded.
80 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11664>
81
82 2026-05-15 01:46:30 +0530 Nirbheek Chauhan <[email protected]>
83
84 * gst/gsttask.c:
85 gsttask: Fix racy tests by making unref deterministic
86 gst_check_teardown_element() asserts that the refcount of the element
87 is 1 after its state has been set to NULL which stops all pad tasks.
88 However, after 1179387b9f, the pad task also holds a reference to the
89 element and the finalize hits after the task is stopped and releases
90 the task lock (sync) and before the task thread exits (async).
91 This made the assert flaky, because it would see 1 or 2 depending on
92 how slowly things are scheduled.
93 Co-authored-by: Sebastian Dröge <[email protected]>
94 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11638>
95
96 2026-05-11 16:33:44 +0530 Nirbheek Chauhan <[email protected]>
97
98 * gst/gstinfo.c:
99 gstinfo: Don't use fwrite() on Windows for debug logging
100 When setting GST_DEBUG_FILE, the file was opened with fopen() and
101 written to with fwrite(). This required an expensive fflush() on
102 Windows because fwrite() doesn't support line-buffered I/O.
103 Bypass the CRT userspace buffering by submitting whole lines in
104 a single write to the kernel. This is equivalent to line-buffered FILE
105 writes, and is about 12% faster.
106 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11632>
107
108 2026-05-11 12:49:55 +0530 Nirbheek Chauhan <[email protected]>
109
110 * gst/gstinfo.c:
111 gstinfo: Use stack allocated memory when printing on Win32
112 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11624>
113
114 2026-05-10 03:45:43 +0530 Nirbheek Chauhan <[email protected]>
115
116 * gst/printf/README:
117 * gst/printf/gst-printf.h:
118 * gst/printf/vasnprintf.c:
119 gstinfo: Don't ever use %n to know how much was written
120 The return value of snprintf will let us know that already. %n is only
121 needed for ancient snprintf implementation that didn't do that. We do
122 not support any of those.
123 This allows us to use snprintf instead of sprintf which is marginally
124 faster.
125 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11624>
126
127 2026-05-10 01:58:42 +0530 Nirbheek Chauhan <[email protected]>
128
129 * gst/gstinfo.c:
130 * gst/printf/printf.c:
131 * gst/printf/printf.h:
132 gstinfo: Use stack allocation for <=1KB messages
133 It should be super low-risk to add an extra KB to the stack at all
134 call-sites.
135 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11624>
136
137 2026-05-12 22:46:00 +0300 Sebastian Dröge <[email protected]>
138
139 * gst/gstdatetime.c:
140 datetime: Improve correctness of ISO-8601 string parsing
141 - Use unsigned integers of the correct sizes everywhere for parsing
142 to properly detect errors.
143 - Validate values first before using them.
144 - Actually validate timezone information and don't assert on invalid
145 timezone information.
146 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11616>
147
148 2026-05-09 14:26:55 +0000 L. E. Segovia <[email protected]>
149
150 * gst/gstregistry.c:
151 registry: detect libgstreamer load from Android container and skip canonicalization
152 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11612>
153
154 2026-05-11 18:33:25 +0100 Tim-Philipp Müller <[email protected]>
155
156 * meson.build:
157 Back to development after 1.28.3
158
159 === release 1.28.3 ===
160
161 2026-05-11 18:28:12 +0100 Tim-Philipp Müller <[email protected]>
162
163 * gstreamer.doap:
164 * meson.build:
165 Release 1.28.3
166
167 2026-04-15 16:31:02 +0300 Sebastian Dröge <[email protected]>
168
169 * gst/gstvalue.c:
170 value: On buffer deserialization errors first unmap the buffer and then unref it
171 Also unref the buffer if mapping fails (which is technically impossible).
172 Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/work_items/5042
173 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11590>
174
175 2026-05-08 10:48:38 +0300 Sebastian Dröge <[email protected]>
176
177 * gst/gstallocator.c:
178 allocator: Use g_try_malloc() instead of g_malloc() for sysmem
179 The latter will just abort the process if allocation fails but all the
180 surrounding code is already expecting it to return NULL on failure and this
181 change now allows to recover from impossible allocations.
182 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11569>
183
184 2026-05-07 16:47:47 +0100 Arthur Chan <[email protected]>
185
186 * libs/gst/base/gstbaseparse.c:
187 baseparse: Fix memory leak when subclass returns error
188 Signed-off-by: Arthur Chan <[email protected]>
189 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11567>
190
191 2026-05-07 08:38:01 +0300 Sebastian Dröge <[email protected]>
192
193 * gst/gststreams.c:
194 streams: Add METADATA to the valid stream flags for serialization
195 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11566>
196
197 2026-05-07 08:35:47 +0300 Sebastian Dröge <[email protected]>
198
199 * gst/gststreams.c:
200 streams: Don't assert on empty stream flags during serialization
201 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11566>
202
203 2026-04-07 13:03:19 +0100 Nirbheek Chauhan <[email protected]>
204
205 * gst/gstmacros.h:
206 gstmacros: Remove restrict definition
207 We now require recent GCC, Clang, and MSVC for building, all of which
208 provide the `restrict` keyword. Our define is now breaking external
209 code. F.ex., Qt 6.11 on MSVC uses `__declspec(restrict)` which gets
210 converted to `__declspec(__restrict)` which is not a valid keyword.
211 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11563>
212
213 2026-04-07 13:09:24 +0100 Nirbheek Chauhan <[email protected]>
214
215 * meson.build:
216 meson: Require C std gnu11 or c11
217 When using gcc or clang, gnu11 will be used, and when using MSVC c11
218 will be used which will pass /std:c11. This provides us with the
219 `restrict` keyword on all supported platforms.
220 We can do this now because we require Visual Studio 2019.
221 This feature was added in Meson 1.3
222 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11563>
223
224 2026-05-05 10:18:23 -0400 Xavier Claessens <[email protected]>
225
226 * gst/gstdevicemonitor.c:
227 * tests/check/gst/gstdevice.c:
228 devicemonitor: Wait for start thread to finish when listing devices
229 This fixes a regression introduced by:
230 https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9698
231 Since 1.28.0, gst_device_monitor_start() is not blocking any more, which
232 causes the list of devices returned by gst_device_monitor_get_devices()
233 to be incomplete. For example v4l2 provider would not fill udev
234 properties yet.
235 Solve this by blocking in gst_device_monitor_get_devices() for backward
236 compatibility. That way, applications that wish to not block can wait
237 for %GST_MESSAGE_DEVICE_MONITOR_STARTED before calling get_devices().
238 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11554>
239
240 2026-05-03 11:48:44 +0300 Sebastian Dröge <[email protected]>
241
242 * libs/gst/base/gstaggregator.c:
243 aggregator: Fix documentation
244 Parameter docs for various functions were missing.
245 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11488>
246
247 2026-04-27 20:12:37 +0300 Sebastian Dröge <[email protected]>
248
249 * libs/gst/base/gstbitwriter.c:
250 bitwriter: Zero-initialize uninitialized data
251 Otherwise we will read uninitialized data when combining it with newly
252 written bits.
253 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11471>
254
255 2026-04-27 18:18:22 +0300 Sebastian Dröge <[email protected]>
256
257 * libs/gst/base/gstbitwriter.h:
258 * tests/check/libs/bitwriter.c:
259 bitwriter: Allow unsetting set bits when overwriting them
260 Previously the original bit pattern would always be preserved and new bits
261 only OR-ed into the array. This made it impossible to actually unset bits.
262 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11471>
263
264 2026-04-28 14:52:16 +0100 Tim-Philipp Müller <[email protected]>
265
266 * tools/gst-inspect.c:
267 gst-inspect-1.0: type for string caps fields should be 'string' not 'gchararray'
268 See e.g. gst-inspect-1.0 vorbisenc
269 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11460>
270
271 2026-04-26 18:18:35 +0530 Nirbheek Chauhan <[email protected]>
272
273 * docs/meson.build:
274 meson: Docs build is not supported when gstcheck is disabled
275 subprojects/gstreamer/docs/meson.build:147:13: ERROR: Unknown variable "gst_check_gir".
276 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11444>
277
278 2026-04-24 17:29:42 +0530 Nirbheek Chauhan <[email protected]>
279
280 * libs/gst/check/gstcheck.c:
281 gstcheck: Update 'Since' marker to 1.28.3
282 We backported this new symbol for CI use, but it's a good idea to
283 update the Since marker
284 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11387>
285
286 2026-03-31 11:58:28 +0200 Piotr Brzeziński <[email protected]>
287
288 * libs/gst/check/gstcheck.c:
289 * libs/gst/check/gstcheck.h:
290 check: Add API to run a test suite with forking disabled
291 We already needed to set CK_FORK=no in some tests (e.g. for NVIDIA elements). Turns out Vulkan tests on macOS also need
292 that, so this adds a helper API to use instead of having to override the envvar.
293 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11387>
294
295 2026-02-06 21:42:25 +0000 Thibault Saunier <[email protected]>
296
297 * gst/gstpad.c:
298 gstpad: fix potential buffer leak in get_range_failed error handler
299 In gst_pad_get_range_unchecked(), the get_range_failed error handler
300 did not clean up res_buf when the getrange function returned an error.
301 If a getrange implementation allocates a buffer but then returns a
302 non-OK flow, the buffer would be leaked since it was never written
303 back to the caller's buffer pointer.
304 Add defensive cleanup matching the pattern already used in the
305 probe_stopped_unref handler in the same function.
306 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11284>
307
308 2026-02-06 17:35:26 +0000 Thibault Saunier <[email protected]>
309
310 * plugins/elements/gsttypefindelement.c:
311 typefind: fix buffer leak on pull_range failure during shutdown
312 When gst_pad_pull_range() fails during a racy pipeline teardown, the
313 already-allocated buffer may still be set in the output parameter even
314 though the return value is not GST_FLOW_OK. The code jumps to the
315 pause label without clearing the buffer, leaking it.
316 Use gst_clear_buffer() before the goto to properly release the buffer
317 on all error paths.
318 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11284>
319
320 2026-04-07 20:07:19 +0100 Tim-Philipp Müller <[email protected]>
321
322 * meson.build:
323 Back to development after 1.28.2
324
325 === release 1.28.2 ===
326
327 2026-04-07 20:02:23 +0100 Tim-Philipp Müller <[email protected]>
328
329 * gstreamer.doap:
330 * meson.build:
331 Release 1.28.2
332
333 2026-03-31 14:39:52 +0900 Seungha Yang <[email protected]>
334
335 * plugins/elements/gstqueue.c:
336 queue: Fix potential use-after-free in log function
337 Avoid accessing already pushed event object
338 Regression introduced by
339 https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10068
340 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11192>
341
342 2026-03-18 23:42:43 +0100 Tim-Philipp Müller <[email protected]>
343
344 * po/LINGUAS:
345 * po/kk.po:
346 * po/pt_BR.po:
347 * po/sv.po:
348 * po/zh_CN.po:
349 gstreamer: update translations
350 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11077>
351
352 2026-03-10 23:05:17 +0530 Nirbheek Chauhan <[email protected]>
353
354 * libs/gst/base/gstbaseparse.c:
355 baseparse: Fix out_buffer leak in frame_free and missing ref in frame_copy
356 gst_base_parse_frame_free() only unrefs frame->buffer but not
357 frame->out_buffer. If finish_frame() exits early (e.g. due to a
358 g_return_val_if_fail check on adapter size, or via the scanning path),
359 out_buffer is left non-NULL and leaked when frame_free() zeroes the
360 frame struct via memset without unreffing it.
361 This can be triggered via h264parse (and other video parsers) in the
362 packetized path: gst_h264_parse_parse_frame() takes a buffer from the
363 frame_out adapter and sets frame->out_buffer. If finish_frame() then
364 fails its precondition checks, that buffer is never freed.
365 Also fix gst_base_parse_frame_copy() which uses g_memdup2() to copy
366 the frame struct, giving the copy the same out_buffer pointer as the
367 original, but only adds a ref for buffer and not for out_buffer.
368 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11056>
369
370 2026-03-07 11:47:29 -0500 Xavier Claessens <[email protected]>
371
372 * gst/gstbin.c:
373 bin: iterator is not nullable
374 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11036>
375
376 2026-01-29 17:02:49 +0900 Seungha Yang <[email protected]>
377
378 * libs/gst/base/gstbaseparse.c:
379 baseparse: Preserve upstream buffer duration if possible
380 When upstream buffer timestamps are reliable and subclass has
381 timestamp interpolation disabled, preserve the upstream buffer
382 duration instead of replacing it with GST_CLOCK_TIME_NONE.
383 Subclasses typically derive buffer duration from stream header
384 information, but this can be inaccurate for variable framerate
385 streams or when the signaled framerate itself is unreliable. By
386 forwarding the original buffer duration, this change allows
387 subclasses to handle timing more accurately in such cases.
388 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/11016>
389
390 2026-03-09 00:31:25 +0300 Vitaly Vlasov <[email protected]>
391
392 * plugins/elements/gstfilesink.c:
393 filesink: Fix wrong open() in overwrite mode
394 Fix gst_fopen() when File-sink-file-mode=overwrite is selected on non Windows host
395 Fixes #4946
396 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10990>
397
398 2026-03-03 18:24:41 -0300 L. E. Segovia <[email protected]>
399
400 * gst/gstregistry.c:
401 registry: Skip recursion into and from .dSYM bundles
402 The filtering of dSYM bundles was done *after* an early return that approved
403 non-dot directories.
404 For good measure, let's also add a filtering entry to ban directories when their
405 parent is a dSYM bundle.
406 See https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10850#note_3357038
407 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10970>
408
409 2026-01-29 12:34:46 +0100 Fabian Orccon <[email protected]>
410
411 * gst/gstsystemclock.c:
412 * tests/validate/gst-tester.c:
413 all: GThreadFunc return type fixes
414 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10955>
415
416 2026-02-26 01:53:07 +0000 Tim-Philipp Müller <[email protected]>
417
418 * meson.build:
419 Back to development after 1.28.1
420
421 === release 1.28.1 ===
422
423 2026-02-26 01:44:06 +0000 Tim-Philipp Müller <[email protected]>
424
425 * gstreamer.doap:
426 * meson.build:
427 Release 1.28.1
428
429 2026-02-23 19:42:18 +0900 Seungha Yang <[email protected]>
430
431 * gst/gstutils.c:
432 gst: Add explanatory comment to call_async implementation
433 Clarify why notify is conditionally called for deprecated
434 gst_element_call_async()
435 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10872>
436
437 2026-02-20 09:28:15 +0200 Sebastian Dröge <[email protected]>
438
439 * tools/gst-stats.c:
440 gst-stats: Also allow ANSI colored logs without 0x in front of the thread id
441 The non-ANSI colored regex already handled this correctly and this was required
442 for Windows before already anyway.
443 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10857>
444
445 2026-02-19 13:08:39 -0300 L. E. Segovia <[email protected]>
446
447 * gst/gstregistry.c:
448 registry: Skip .dSYM bundles when loading plugins, try 2
449 If paths within the .dSYM folder are being passed, the current filter won't cover them.
450 See https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10466#note_3330939
451 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10853>
452
453 2026-02-12 02:51:07 +0530 Nirbheek Chauhan <[email protected]>
454
455 * gst/gstregistry.c:
456 * meson.build:
457 gstreamer: Stop using deprecated G_MODULE_SUFFIX
458 It's deprecated on macOS because shared libs can be either .so or
459 .dylib. We already account for that, we might as well stop using
460 G_MODULE_SUFFIX and get rid of deprecation warnings.
461 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10807>
462
463 2026-01-28 11:54:11 +0530 Nirbheek Chauhan <[email protected]>
464
465 * meson.build:
466 gstcheck: Disable fork on tvOS and watchOS
467 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10807>
468
469 2026-01-28 11:45:20 +0530 Nirbheek Chauhan <[email protected]>
470
471 * gst/meson.build:
472 * libs/gst/helpers/ptp/meson.build:
473 * meson.build:
474 meson: Deprecate `system = 'ios'` in cross files, use subsystem
475 Since we require Meson 1.4.0, we can now use subsystem names (added in
476 1.2.0).
477 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10807>
478
479 2023-04-02 16:53:45 +0100 Tim-Philipp Müller <[email protected]>
480
481 * scripts/dist-common-files.py:
482 * scripts/meson.build:
483 modules: dist common files from monorepo root
484 Less noise when making releases, and just need to maintain one copy.
485 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10822>
486
487 2026-02-15 17:45:19 +0000 Tim-Philipp Müller <[email protected]>
488
489 * README.md:
490 modules: remove subproject README.md from git
491 Will be added to the tarballs based on the monorepo README on dist.
492 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10822>
493
494 2026-02-15 17:20:59 +0000 Tim-Philipp Müller <[email protected]>
495
496 * RELEASE:
497 modules: remove RELEASE from git, will be generated from template on dist
498 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10822>
499
500 2026-02-15 15:02:07 +0000 Tim-Philipp Müller <[email protected]>
501
502 * NEWS:
503 modules: Remove NEWS from git which is generated from full release notes
504 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10822>
505
506 2026-02-13 19:50:56 +0900 Seungha Yang <[email protected]>
507
508 * tests/check/libs/bitwriter.c:
509 tests: bitwriter: Fix leak in test
510 Fix leak in test revealed after bugfix. The data returned from
511 gst_bit_writer_reset_and_get_data() must be freed with g_free()
512 as documented.
513 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10805>
514
515 2026-02-13 15:57:13 +0900 Seungha Yang <[email protected]>
516
517 * libs/gst/base/gstbitwriter.c:
518 bitwriter: Steal owned data in reset_and_get_data()
519 Avoid unnecessary data copy/free by returning the owned
520 data directly to the caller
521 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10805>
522
523 2026-02-11 16:49:02 +0900 Seungha Yang <[email protected]>
524
525 * libs/gst/base/gstaggregator.c:
526 aggregator: Handle gap event before segment
527 If gap events are received before segment event, log the situation and
528 gracefully return FALSE instead of relying on the critical warning path
529 in gst_segment_clip()
530 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10789>
531
532 2026-02-05 16:31:30 +0100 François Laignel <[email protected]>
533
534 * plugins/elements/gstmultiqueue.c:
535 multiqueue: reverse playback: use segment stop position as start time
536 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10770>
537
538 2026-02-05 16:24:33 +0900 Seungha Yang <[email protected]>
539
540 * plugins/elements/gstelements_private.c:
541 filesink: Report write error correctly on Windows
542 Propagate write() return value from the writev emulation path
543 as intended
544 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10758>
545
546 2026-02-05 15:48:30 +0900 Seungha Yang <[email protected]>
547
548 * plugins/elements/gstelements_private.c:
549 filesink: Add debug category to file write helper
550 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10758>
551
552 2026-02-02 09:56:59 +0200 Sebastian Dröge <[email protected]>
553
554 * libs/gst/base/gstaggregator.h:
555 aggregator: Various introspection annotations / docs fixes for vfuncs
556 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10723>
557
558 2026-01-26 23:21:06 +0800 He Junyan <[email protected]>
559
560 * gst/gstcaps.c:
561 caps: gst_caps_append_structure should not allow NULL structure
562 This check can also help to fix the possible memory leak of features
563 in gst_caps_append_structure_full() when the structure is NULL.
564 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10689>
565
566 2026-01-28 13:53:28 +0200 Sebastian Dröge <[email protected]>
567
568 * libs/gst/net/gstptpclock.c:
569 ptpclock: Initialize PTP message to avoid invalid uninitialized variable warning
570 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10656>
571
572 2026-01-28 11:57:56 +0200 Sebastian Dröge <[email protected]>
573
574 * libs/gst/helpers/ptp/parse.rs:
575 ptpclock: Fix copy-paste mistake in error messages
576 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10656>
577
578 2026-01-28 11:52:13 +0200 Sebastian Dröge <[email protected]>
579
580 * libs/gst/helpers/ptp/parse.rs:
581 * libs/gst/net/gstptpclock.c:
582 ptpclock: Fix potential out-of-bounds read when parsing ANNOUNCE messages
583 The required length of the message payloader after the common header is 30
584 bytes, not 20 bytes.
585 Also fix the same bug in the Rust code, but here it wouldn't have caused any
586 other problems than a less optimal error message because there is actual bounds
587 checking.
588 Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4871
589 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10656>
590
591 2026-01-28 11:49:41 +0200 Sebastian Dröge <[email protected]>
592
593 * gst/gstdatetime.c:
594 * gst/gsturi.c:
595 gst: Fix a couple of new const-ness warnings around strstr() usage
596 Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4871
597 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10656>
598
599 2026-01-14 23:47:04 +0100 Ruben Gonzalez <[email protected]>
600
601 * libs/gst/base/gsttypefindhelper.c:
602 typefindhelper: EOS case explained in more details
603 Related to https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4848
604 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10632>
605
606 2026-01-09 20:17:52 +0100 François Laignel <[email protected]>
607
608 * plugins/elements/gstinputselector.c:
609 input-selector: don't wait for active pad when eos (active-segment mode)
610 When `sync-mode` is set to `active-segment` and current pad is not active,
611 don't wait depending on the active segment runtime if the active pad is EOS.
612 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10621>
613
614 2026-01-09 18:20:51 +0100 François Laignel <[email protected]>
615
616 * plugins/elements/gstinputselector.c:
617 input-selector: fix regression receiving eos event
618 When a pad receives an eos event, it must first check for if the active pad
619 should be changed. Failure to do that could lead to the event not being
620 forwarded if current pad is the last pad remaining and it was not active
621 before.
622 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10621>
623
624 2026-01-08 11:54:52 +0100 François Laignel <[email protected]>
625
626 * plugins/elements/gstinputselector.c:
627 input-selector: don't consider pad being released checking for eos
628 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10621>
629
630 2026-01-06 15:06:57 +0100 François Laignel <[email protected]>
631
632 * plugins/elements/gstinputselector.c:
633 input-selector: gracefully handle active pad disappearance
634 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10621>
635
636 2026-01-06 11:43:59 +0100 François Laignel <[email protected]>
637
638 * plugins/elements/gstinputselector.c:
639 input-selector: fix race condition selecting next active pad
640 When a pad is being released, the pad is marked as flushing and the selector
641 lock is released before the pad is removed from the element pad list. This is
642 necessary so any in-progress event or buffer handling for this pad terminates
643 before the pad is deactivated.
644 If another thread was trying to select the next active pad at this particular
645 moment, it could end up picking the pad being released before it was removed
646 from the element pad list.
647 This commit adds a flag which is set when the pad is being released, so it won't
648 get picked as the next active pad.
649 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10621>
650
651 2026-01-27 17:10:54 +0000 Tim-Philipp Müller <[email protected]>
652
653 * meson.build:
654 Back to development after 1.28.0
655
656 === release 1.28.0 ===
657
658 2026-01-27 17:02:33 +0000 Tim-Philipp Müller <[email protected]>
659
660 * NEWS:
661 * README.md:
662 * RELEASE:
663 * gstreamer.doap:
664 * meson.build:
665 Release 1.28.0
666
667 2026-01-26 18:51:01 +0530 Sanchayan Maity <[email protected]>
668
669 * libs/gst/base/gstaggregator.h:
670 aggregator: Annotate some GstAggregatorClass methods
671 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10596>
672
673 2026-01-26 14:54:59 +0530 Sanchayan Maity <[email protected]>
674
675 * gst/gstelement.h:
676 element: Annotate request_new_pad
677 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10596>
678
679 2026-01-25 17:17:25 +0000 Tim-Philipp Müller <[email protected]>
680
681 * po/LINGUAS:
682 * po/ar.po:
683 * po/bg.po:
684 * po/de.po:
685 * po/es.po:
686 * po/hr.po:
687 * po/nl.po:
688 * po/pl.po:
689 * po/ro.po:
690 * po/sl.po:
691 * po/sr.po:
692 * po/uk.po:
693 gstreamer: update translations
694 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10598>
695
696 2026-01-08 22:16:19 +0100 Ruben Gonzalez <[email protected]>
697
698 * plugins/elements/gstidentity.c:
699 identity: cosmetic change to mimic fakesink
700 Remove extra space at chain log:
701 ```
702 $ gst-launch-1.0 -v videotestsrc num-buffers=1
703 ! identity silent=false name=i1 ! fakesink name=i2 silent=false
704 ...pts: 0:00:00.000000000, duration: 0:00:00.033333333, offset: 0, offset_end: 1, flags: 00000040..
705 ...pts: 0:00:00.000000000, duration: 0:00:00.033333333, offset: 0, offset_end: 1, flags: 00000040...
706 ```
707 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10518>
708
709 2026-01-07 10:51:59 +0100 Samuel Thibault <[email protected]>
710
711 * gst/gstinfo.c:
712 gstinfo: Fix build on systems where pthread_t is int
713 Otherwise we get
714 error: returning ‘pthread_t’ {aka ‘int’} from a function with return type
715 ‘GstTid’ {aka ‘void *’} makes pointer from integer without a cast [-Wint-conversion]
716 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10508>
717
718 2026-01-05 20:20:51 +0000 Tim-Philipp Müller <[email protected]>
719
720 * meson.build:
721 Back to development after 1.27.90
722
723 === release 1.27.90 ===
724
725 2026-01-05 20:15:10 +0000 Tim-Philipp Müller <[email protected]>
726
727 * NEWS:
728 * RELEASE:
729 * gstreamer.doap:
730 * meson.build:
731 Release 1.27.90
732
733 2026-01-05 18:57:01 +0000 Tim-Philipp Müller <[email protected]>
734
735 * po/af.po:
736 * po/ast.po:
737 * po/az.po:
738 * po/be.po:
739 * po/bg.po:
740 * po/ca.po:
741 * po/cs.po:
742 * po/da.po:
743 * po/de.po:
744 * po/el.po:
745 * po/en_GB.po:
746 * po/eo.po:
747 * po/es.po:
748 * po/eu.po:
749 * po/fi.po:
750 * po/fr.po:
751 * po/fur.po:
752 * po/gl.po:
753 * po/hr.po:
754 * po/hu.po:
755 * po/id.po:
756 * po/it.po:
757 * po/ja.po:
758 * po/ka.po:
759 * po/ko.po:
760 * po/lt.po:
761 * po/lv.po:
762 * po/nb.po:
763 * po/nl.po:
764 * po/pl.po:
765 * po/pt_BR.po:
766 * po/ro.po:
767 * po/ru.po:
768 * po/rw.po:
769 * po/sk.po:
770 * po/sl.po:
771 * po/sq.po:
772 * po/sr.po:
773 * po/sv.po:
774 * po/tr.po:
775 * po/uk.po:
776 * po/vi.po:
777 * po/zh_CN.po:
778 * po/zh_TW.po:
779 gstreamer: update translations
780 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10497>
781
782 2025-12-30 17:53:22 -0500 Doug Nazar <[email protected]>
783
784 * tests/check/elements/queue.c:
785 tests: Fix several memory leaks
786 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10476>
787
788 2025-12-28 17:28:22 +0000 L. E. Segovia <[email protected]>
789
790 * gst/gstregistry.c:
791 registry: Skip .dSYM bundles when loading plugins
792 The .dylibs here aren't actual binaries, they're mach-O files with debug
793 sections only. These will be naturally skipped/blacklisted, but after a
794 long stream of dlopen errors.
795 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10466>
796
797 2025-12-28 17:27:10 +0000 L. E. Segovia <[email protected]>
798
799 * libs/gst/helpers/ptp/meson.build:
800 ptp: Fix helper test naming convention and gate it
801 This ensures it doesn't get built unless -Dtests=enabled, and that its symbols
802 don't get confused for gst-ptp-helper's.
803 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10466>
804
805 2025-12-26 18:03:19 -0500 Doug Nazar <[email protected]>
806
807 * tests/check/gst/gstinfo.c:
808 info: Increase test interval when running under valgrind
809 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10456>
810
811 2025-12-19 15:40:26 +0530 Nirbheek Chauhan <[email protected]>
812
813 * meson.build:
814 meson: Solve some cyclic dependencies caused by test-only deps
815 gstreamer => gobject-introspection => cairo => fontconfig => freetype2 => harfbuzz => cairo
816 gst-plugins-base => libdrm => cairo => fontconfig => freetype2 => harfbuzz => cairo
817 gst-plugins-good => cairo => librsvg => cairo
818 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10422>
819
820 2025-12-19 15:19:42 +0530 Nirbheek Chauhan <[email protected]>
821
822 * meson.build:
823 meson: Disable introspection option on glib
824 Allows us to update glib and gobject-introspection. We don't use the
825 new introspection infrastructure merged in glib anyway.
826 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10422>
827
828 2025-12-19 18:49:46 +0100 Alicia Boya García <[email protected]>
829
830 * gst/gstutils.c:
831 utils: Fix race in gst_element_link_pads_filtered()
832 gst_element_link_pads_filtered() can occur at the same time the bin
833 is in the middle of a PAUSED->PLAYING transition.
834 Since before this patch the capsfilter state lock bit wasn't being set,
835 in the case of a failure to link it was possible for the state change to
836 NULL in gst_element_link_pads_filtered() to race against a state change
837 to PLAYING inside gst_bin_change_state_func() and lose.
838 This caused errors when the unreffed capsfilter was disposed moments
839 later, as the element dispose function found the element to be
840 accidentally in PLAYING state.
841 As of writing, this particular race reproduces consistently when running
842 check.gst-plugins-good.elements_qtdemux.test_qtdemux_sample_interleaving
843 under valgrind with the the default fair scheduler (valgrind
844 --fair-sched=try), both locally and in the CI. This patch fixes it.
845 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10383>
846
847 2025-12-18 09:19:52 +0100 Carlos Falgueras García <[email protected]>
848
849 * plugins/elements/gstqueue.c:
850 queue: Log the status of the queue at debug level
851 Log the queue status at "GST_LEVEL_DEBUG" when it's empty or full by
852 reusing the current macro "STATUS()".
853 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10403>
854
855 2025-12-15 15:08:02 -0500 Daniel Morin <[email protected]>
856
857 * gst/gstvalue.c:
858 * gst/gstvalue.h:
859 * tests/check/gst/gstcaps.c:
860 * tests/check/gst/gststructure.c:
861 * tests/check/gst/gstvalue.c:
862 * tools/gst-inspect.c:
863 gst: Rename GstValueSet to GstValueUniqueList
864 GObject-Introspection has an issue with GstSet because anything that starts with
865 'gst_value_set' becomes something that belongs to 'GstSet' but we have
866 gst_value_set_bitmask and gst_value_set_SOMETHING (), which all would become
867 methods of GstSet.
868 To avoid this, rename GstSet (aka GstValueSet) to GstUniqueList (aka
869 GstValueUniqueList).
870 Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4813
871 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10384>
872
873 2025-12-13 22:17:41 +0100 François Laignel <[email protected]>
874
875 * plugins/elements/gstinputselector.c:
876 * plugins/elements/gstinputselector.h:
877 input-selector: implement a two-phase sinkpad switch
878 Switching the sinkpad was racy: when a new sinkpad was selected, previous
879 active sinkpad could continue pushing events or buffers, while the new sinkpad
880 already started pushing its sticky events, leading to inconsistencies
881 downstream.
882 This commit implements a a two-phase switch approach:
883 1. When the user selects the active pad, it is first set as pending.
884 2. Functions handling buffers or events first check whether a new pad is pending
885 activation, in which case the new pad activation is commited.
886 3. After completing step 2, the functions handles the buffer or event.
887 The active pad can not change before the buffer or event is pushed.
888 Mutability of the pending pad is protected by SELECTOR_LOCK.
889 Mutability of the active pad is protected by a dedicated RW lock which can be
890 locked independently from SELECTOR_LOCK and can prevent active pad changes when
891 that would interfere with current processing.
892 The RW lock allows preventing the active_pad from being changed while we
893 are processing a buffer or event.
894 Only `gst_input_selector_maybe_commit_active_pad ()` can get this lock in
895 writer mode for a short period of time (committing the `pending_active_pad`
896 to `active_pad`). Other concurrent code paths lock it in reader mode.
897 When a `pending_active_pad` is available (and only in that case),
898 code paths calling `gst_input_selector_maybe_commit_active_pad ()` would
899 hang until the lock is available for writer mode, which is expected since
900 their processing should be performed with respect to the expected `active_pad`.
901 Note that a `GRecMutex` was also considered to prevent possible deadlocks
902 in callbacks, but only a `GRWLock` can deal with the concurrency which occurs
903 during preroll.
904 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10204>
905
906 2025-12-09 19:13:20 +0000 Tim-Philipp Müller <[email protected]>
907
908 * meson.build:
909 Back to development after 1.27.50
910
911 === release 1.27.50 ===
912
913 2025-12-09 19:08:48 +0000 Tim-Philipp Müller <[email protected]>
914
915 * NEWS:
916 * RELEASE:
917 * gstreamer.doap:
918 * meson.build:
919 Release 1.27.50
920
921 2025-12-05 11:19:25 +0200 Sebastian Dröge <[email protected]>
922
923 * libs/gst/base/gstbasesrc.c:
924 basesrc: Fix decide allocation fallback
925 Fallback to generic pool, even if the change to the config was validated. Pools
926 may have other restrictions that may render them unusable.
927 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10298>
928
929 2025-12-04 16:48:37 -0500 Nicolas Dufresne <[email protected]>
930
931 * libs/gst/base/gstbasetransform.c:
932 basetransform: Fix decide allocation fallback
933 Fallback to generic pool, even if the change to the config was validated. Pools
934 may have other restrictions that may render them unusable.
935 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10298>
936
937 2025-11-28 14:12:53 -0600 Olivier Crête <[email protected]>
938
939 * tools/gst-inspect.c:
940 gst-inspect: Pretty print tensor caps
941 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9109>
942
943 2025-08-03 21:39:36 -0400 Daniel Morin <[email protected]>
944
945 * libs/gst/base/gstbasetransform.c:
946 basetransform: Explicitly set sinkpad flag to GST_PAD_FLAG_ACCEPT_INTERSECT
947 - Note this is NOT a breaking change.
948 - Base transform accept caps evaluation is based on intersection between caps
949 and template while the default behaviour should be to verify that caps is a
950 subset of the template. There's a specific pad flag to indicate accept caps
951 should be based on intersect (GST_PAD_FLAG_ACCEPT_INTERSECT). Since alot of
952 elements base based on bastransform and the expected behaviour from the base
953 class to evaluate accept caps is to use intersect we set the flat
954 GST_PAD_FLAG_ACCEPT_INTERSECT inside gst_base_transform_init() on the sinkpad
955 and inside gst_base_transform_acceptcaps_default() we use interect if the flag
956 is set otherwise we use gst_caps_is_subset (caps, template). This way
957 basetransform will honor the flag and allow element that want accept caps to
958 be base on subset to clear this flag.
959 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9109>
960
961 2025-10-30 17:03:35 -0400 Daniel Morin <[email protected]>
962
963 * docs/random/caps_grammar:
964 doc: adding caps grammar doc
965 - Include new grammar for GST_TYPE_SET
966 - Converted doc to markdown
967 - Updated grammar description based on what exist.
968 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10262>
969
970 2025-07-18 15:14:52 -0400 Daniel Morin <[email protected]>
971
972 * tests/check/gst/gstcaps.c:
973 * tests/check/gst/gststructure.c:
974 * tests/check/gst/gstvalue.c:
975 test: Tests for GstSet
976 - tests for GstSet
977 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10262>
978
979 2025-07-29 09:12:46 -0400 Daniel Morin <[email protected]>
980
981 * gst/gststructure.c:
982 gststructure: fix issue when structure contain GST_TYPE_SET
983 - 'set' already use cast annotation for divisor, we don't add another one, to
984 avoid generating '(set)(/set){}'.
985 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10262>
986
987 2025-11-28 14:24:53 -0500 Daniel Morin <[email protected]>
988
989 * gst/gstvalue.c:
990 gstvalue: remove code duplication from GstList
991 - Improve code maintainability by removing duplicated code
992 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10262>
993
994 2025-07-18 15:06:03 -0400 Daniel Morin <[email protected]>
995
996 * gst/gst_private.h:
997 * gst/gststructure.c:
998 * gst/gstvalue.c:
999 * gst/gstvalue.h:
1000 gstvalue: adding GstSet
1001 - Adding GstSet which reuse list annotation `{}`
1002 - Add new annotation to type-cast container type.
1003 `(default_type/container_type){}`. default_type and container_type are both
1004 optional.
1005 Examples:
1006 `(int/set){1}`
1007 `(/set){1}`
1008 `(int)`
1009 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10262>
1010
1011 2025-12-02 18:58:29 +0100 Christoph Reiter <[email protected]>
1012
1013 * plugins/elements/gstfilesink.c:
1014 gstfilesink: fix the build with recent mingw-w64
1015 The build fails with:
1016 /clang64/include/unistd.h:59:5: error: conflicting types for '_chsize'
1017 59 | int ftruncate(int, off_t)
1018 since ftruncate is redefined before unistd.h is included.
1019 Avoid it by including the system headers first.
1020 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10268>
1021
1022 2025-12-01 12:53:46 +0200 Sebastian Dröge <[email protected]>
1023
1024 * gst/gststreams.c:
1025 * gst/gststreams.h:
1026 streams: Add GST_STREAM_TYPE_METADATA for metadata streams
1027 And handle it inside parsebin and tsdemux.
1028 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10249>
1029
1030 2025-11-24 19:42:58 +0900 Haejung Hwang <[email protected]>
1031
1032 * gst/gstpipeline.c:
1033 gstpipeline: Improve resource cleanup logic for clock objects
1034 Clarify and strengthen object release handling by:
1035 -Use gst_clear_object() for NULL-safe unref and to avoid dangling pointers
1036 -Ensure proper refcounting when reusing cur_clock
1037 -Limit clock scope to where it’s needed for base-time and clock recalculation
1038 -Fix memory leak in error path by releasing cur_clock
1039 Reason:
1040 This change makes cleanup more predictable and prevents scenarios where two
1041 pointers could reference the same object, ensuring each resource is released
1042 only once and pointers are invalidated after use. This improves robustness and
1043 maintains consistent object lifetime management
1044 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10179>
1045
1046 2025-11-27 10:23:25 -0500 Nicolas Dufresne <[email protected]>
1047
1048 * gst/gstvalue.c:
1049 gstvalue: Fix GstAllocationParams string convertion on 32bit
1050 The GstAllocationParams structure uses gsize, which is not always 64bit. When
1051 passed to a vararg it must be casted, so it uses the right amount of space.
1052 Not doing so lead to crash when using this feature on 32bit plaforms.
1053 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10222>
1054
1055 2025-11-26 20:49:42 +0530 Nirbheek Chauhan <[email protected]>
1056
1057 * gst/gstdeviceprovider.c:
1058 device-provider: Fix typos in documentation
1059 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9698>
1060
1061 2025-11-25 18:47:48 +0530 Nirbheek Chauhan <[email protected]>
1062
1063 * tests/check/gst/gstdevice.c:
1064 tests/gstdevice: Unbreak the test monitor provider
1065 The test provider with "monitoring support" doesn't emit any devices
1066 when started, but it does emit devices when probed (because it wasn't
1067 started). That's a broken device provider implementation, which caused
1068 the test to break when we started doing async device monitoring.
1069 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9698>
1070
1071 2025-11-26 18:27:23 +0530 Nirbheek Chauhan <[email protected]>
1072
1073 * gst/gstdevicemonitor.c:
1074 device-monitor: Use GSList for started_providers for thread-safety
1075 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9698>
1076
1077 2025-09-16 13:20:08 +0530 Nirbheek Chauhan <[email protected]>
1078
1079 * gst/gstdevicemonitor.c:
1080 * tests/check/gst/gstdevice.c:
1081 device-monitor: Start providers in a separate thread
1082 This avoids blocking when gst_device_monitor_start() is called, which
1083 avoids each app having to spawn a separate thread just to start device
1084 monitoring. This is especially important on Windows, where device
1085 probing can take several seconds.
1086 Calling gst_device_monitor_get_devices() immediately after still does
1087 the right thing; the existing locking still applies.
1088 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9698>
1089
1090 2025-11-25 14:04:19 +0530 Nirbheek Chauhan <[email protected]>
1091
1092 * gst/gstmessage.c:
1093 * gst/gstmessage.h:
1094 gstmessage: Add GST_MESSAGE_DEVICE_MONITOR_STARTED
1095 In the next commit, this will be used to signal that the device
1096 monitor has completed async startup.
1097 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9698>
1098
1099 2025-11-11 14:37:39 +0530 Nirbheek Chauhan <[email protected]>
1100
1101 * tests/check/gst/gstdevice.c:
1102 tests: Move gstdevice from fail_unless to ck_assert
1103 This is better, because ck_assert_*_eq etc will print both arguments
1104 when the comparison fails. fail_unless_equals_int etc exist, but the
1105 there's no equivalent for ck_assert_int_gt etc.
1106 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9698>
1107
1108 2025-10-28 20:09:35 +0530 Nirbheek Chauhan <[email protected]>
1109
1110 * gst/gstinfo.c:
1111 * gst/gstinfo.h:
1112 gstinfo: Add convenience macros around GstLogContext for logging once
1113 The purpose is to avoid having to add boilerplate to every file/plugin
1114 where you want to add a log message which will be printed exactly
1115 once, with the default settings for GstLogContext.
1116 Somehow, this is also faster than using the explicit API:
1117 GST_WARNING_ONCE is 15% faster than GST_CTX_WARNING
1118 The macros are the only publicly-documented API, the functions that
1119 are wrapped by the macros are not meant to be used directly, except by
1120 bindings.
1121 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9939>
1122
1123 2025-11-27 03:33:06 +0530 Nirbheek Chauhan <[email protected]>
1124
1125 * gst/gstinfo.h:
1126 gstinfo: Fix some mistakes and typos in the logging macro docs
1127 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9939>
1128
1129 2025-11-24 14:03:50 -0300 Thibault Saunier <[email protected]>
1130
1131 * gst/gsttaskpool.h:
1132 videoconvertscale: add support for task pool context
1133 This allows applications to provide a shared task pool via context for
1134 multi-threaded video conversion, enabling better control over thread
1135 management across multiple elements.
1136 The element now implements set_context() to receive task pools and uses
1137 them when creating video converters. If no n-threads property is explicitly
1138 set, the element will automatically use the max threads from the provided
1139 shared task pool.
1140 A new test verifies the task pool is actually used by the video converter
1141 during processing.
1142 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10047>
1143
1144 2025-11-11 20:55:41 -0500 Thibault Saunier <[email protected]>
1145
1146 * gst/gsttaskpool.c:
1147 * gst/gsttaskpool.h:
1148 * tests/check/gst/gstcontext.c:
1149 gsttaskpool: add task pool context type and helper API
1150 Add GST_TASK_POOL_CONTEXT_TYPE ("gst.task.pool") well-known context
1151 with helper functions gst_context_set/get_task_pool() for
1152 pipeline-scoped task pool sharing between elements.
1153 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10047>
1154
1155 2025-01-08 09:29:13 -0500 Xavier Claessens <[email protected]>
1156
1157 * gst/gstsystemclock.c:
1158 * gst/gstsystemclock.h:
1159 GstClock: Add gst_clock_is_system_monotonic_clock
1160 It was duplicated in many places and can be useful outside of GStreamer
1161 as well.
1162 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8257>
1163
1164 2025-11-25 18:29:12 +0200 Sebastian Dröge <[email protected]>
1165
1166 * gst/gststreamcollection.c:
1167 streamcollection: Fix race condition between disconnecting notify proxy and notifications
1168 It can happen that a GstStream gets notified at the very same time as a
1169 collection that contains it is being disposed. When timing is bad this can end
1170 up using an already freed GstStreamCollection for proxying the signal.
1171 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10196>
1172
1173 2025-11-25 18:20:37 +0200 Sebastian Dröge <[email protected]>
1174
1175 * gst/gststreamcollection.c:
1176 streamcollection: Use a GstVecDeque instead of a GQueue for the streams
1177 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10196>
1178
1179 2022-06-11 16:19:57 +0900 Camilo Celis Guzman <[email protected]>
1180
1181 * libs/gst/check/libcheck/check_run.c:
1182 libcheck: use SIGABRT instead of SIGKILL on timeout
1183 This allows user-level signal handler to catch this if necessary.
1184 Co-authored-by: Frederik Vestre <[email protected]>
1185 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2585>
1186
1187 2025-11-25 09:33:07 +0900 Seungha Yang <[email protected]>
1188
1189 * docs/plugins/gst_plugins_cache.json:
1190 clocksync: Update plugin docs cache
1191 Update for new rate property and resync signal
1192 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7463>
1193
1194 2025-11-24 15:30:55 +0900 Seungha Yang <[email protected]>
1195
1196 * tests/check/elements/clocksync.c:
1197 tests: clocksync: Add QoS test with custom rate
1198 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7463>
1199
1200 2025-11-20 19:14:25 +0900 Seungha Yang <[email protected]>
1201
1202 * tests/examples/clocksync/clocksync-resync.c:
1203 * tests/examples/clocksync/meson.build:
1204 * tests/examples/meson.build:
1205 examples: Add clocksync's resync signal example
1206 Demonstrates the resync signal usage
1207 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7463>
1208
1209 2025-11-11 16:14:50 +0900 Seungha Yang <[email protected]>
1210
1211 * plugins/elements/gstclocksync.c:
1212 clocksync: Add resync action signal
1213 Adding a new action signal so that clocksync can recalculate
1214 ts-offset dynamically without state change
1215 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7463>
1216
1217 2024-09-06 21:42:18 +0900 Seungha Yang <[email protected]>
1218
1219 * plugins/elements/gstclocksync.c:
1220 * plugins/elements/gstclocksync.h:
1221 * tests/check/elements/clocksync.c:
1222 clocksync: Add rate property
1223 Adding rate property so that clocksync can synchronize
1224 buffer running time against pipeline clock with specified rate factor.
1225 This property can be useful if users want to throttle down pipeline
1226 throughput, such as a non-realtime transcoding pipeline
1227 where the pipeline's CPU and/or hardware resource consumption
1228 needs to be limited.
1229 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7463>
1230
1231 2025-11-20 20:52:08 -0500 Doug Nazar <[email protected]>
1232
1233 * gst/gstinfo.h:
1234 gstinfo: Force comparision to same types
1235 warning: equality comparison between function pointer and void pointer ('::GstLogFunction' (aka 'void (*)
1236 (_GstDebugCategory *, GstDebugLevel, const char *, const char *, int, _GObject *, _GstDebugMessage *, void *)')
1237 and 'void *') [-Wpedantic]
1238 263 | gst_debug_add_log_function (_peel_func, _peel_user_data, _peel_notify);
1239 | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1240 /usr/include/gstreamer-1.0/gst/gstinfo.h:598:14: note: expanded from macro 'gst_debug_add_log_function'
1241 598 | if ((func) == (void *) gst_debug_log_default) { \
1242 | ~~~~~~ ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1243 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10165>
1244
1245 2025-11-07 19:29:24 -0300 L. E. Segovia <[email protected]>
1246
1247 * gst/gst.h:
1248 * gst/gstcpuid.c:
1249 * gst/gstcpuid.h:
1250 * gst/meson.build:
1251 gst: implement Orc-less cpuid routine for selecting asm routines
1252 This commit removes the use of Orc's default target machinery as a way
1253 to do CPUID detection on x86 and Arm. Instead I port xsimd's CPU
1254 detection routine to C, cleaning up the instruction sets we don't use,
1255 and also adding support for GCC/Clang's cpuid and xgetbv builtins.
1256 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10004>
1257
1258 2025-11-17 14:58:16 +0200 Sebastian Dröge <[email protected]>
1259
1260 * gst/gstutils.c:
1261 utils: Annotate temp array in gst_calculate_linear_regression() as such
1262 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10108>
1263
1264 2025-10-28 17:03:30 +0530 Nirbheek Chauhan <[email protected]>
1265
1266 * gst/gstinfo.c:
1267 * gst/gstinfo.h:
1268 gstinfo: Forbid GST_LEVEL_MEMDUMP for GstLogContext logging
1269 This wasn't hooked up correctly, and it's not clear how to implement
1270 this without strong trade-offs. Forbid for now, and we can add it
1271 later if needed.
1272 Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4722
1273 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9936>
1274
1275 2025-11-12 10:12:22 +0200 Sebastian Dröge <[email protected]>
1276
1277 * gst/gstobject.c:
1278 * gst/gstobject.h:
1279 * tests/check/gst/gstobject.c:
1280 object: Add gst_object_get_toplevel() to get the toplevel object
1281 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10075>
1282
1283 2025-11-13 11:00:58 +0000 Philippe Normand <[email protected]>
1284
1285 * gst/gst.c:
1286 * gst/gst.h:
1287 * tests/check/gst/gst.c:
1288 gst: Add gst_check_version() utility function
1289 Useful for version runtime checks.
1290 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9915>
1291
1292 2025-11-12 13:25:16 +0100 Linus Svensson <[email protected]>
1293
1294 * gst/gstcaps.c:
1295 * gst/gstcaps.h:
1296 caps: Correct the cast for const GstCaps
1297 This fixes projects that compile with -Wcast-qual.
1298 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10076>
1299
1300 2025-11-12 13:10:58 +0100 Linus Svensson <[email protected]>
1301
1302 * gst/gstinfo.h:
1303 info: Remove redundant ';'
1304 This fixes projects that compile with -Wpedantic.
1305 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10076>
1306
1307 2025-11-11 20:12:41 +0900 Seungha Yang <[email protected]>
1308
1309 * plugins/elements/gstqueue.c:
1310 queue: Use GST_PTR_FORMAT everywhere
1311 Prints detailed information about the object instead of the pointer
1312 address
1313 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10068>
1314
1315 2025-11-11 11:55:52 +0100 Linus Svensson <[email protected]>
1316
1317 * gst/gstvalue.c:
1318 value: Add missing for symbol for builds without asserts/checks
1319 gst_value_list_or_array_are_compatible needs to be defined if either
1320 g_assert or glib_checks are turned on, and not if both are enabled.
1321 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10069>
1322
1323 2025-06-11 16:21:35 +0900 dongjoo.kim <[email protected]>
1324
1325 * gst/gstelementfactory.c:
1326 factory: Move debug logging to clarify code
1327 Both `factory` and `oclass->elementfactory` are the same pointer here so it will
1328 always be valid, even after the `gst_object_unref()`, but moving the debug logging
1329 before it makes it more obvious that this code is actually correct.
1330 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10049>
1331
1332 2025-11-07 09:14:34 +0200 Sebastian Dröge <[email protected]>
1333
1334 * gst/gstmemory.c:
1335 memory: Clear various fields of GstMapInfo/GstVideoFrame/GstAudioBuffer on unmap
1336 This avoids use-after-frees and allows these functions to be called multiple
1337 times without problems.
1338 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10020>
1339
1340 2025-11-05 15:25:33 +0200 Sebastian Dröge <[email protected]>
1341
1342 * gst/gstbuffer.c:
1343 * gst/gstbuffer.h:
1344 * gst/gstconfig.h.in:
1345 * gst/gstmemory.c:
1346 * gst/gstmemory.h:
1347 * gst/gstminiobject.c:
1348 * gst/gstvalue.c:
1349 * tests/check/gst/gstbuffer.c:
1350 * tests/check/gst/gstmemory.c:
1351 memory: Add gst_map_info_clear() and use GST_MAP_REF_MEMORY for gst_buffer_map()
1352 Also deprecate GstMemoryMapInfo and GstBufferMapInfo, and add g_autoptr support
1353 for GstMapInfo directly.
1354 This simplifies usage of the GstMapInfo API and reduces a bit of
1355 inconsistencies.
1356 For consistency, also add gst_map_info_init().
1357 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10020>
1358
1359 2025-11-05 15:04:09 +0200 Sebastian Dröge <[email protected]>
1360
1361 * gst/gstmemory.c:
1362 * gst/gstmemory.h:
1363 * tests/check/gst/gstmemory.c:
1364 memory: Add new GST_MAP_REF_MEMORY flag
1365 This allows the GstMapInfo to store a strong reference to the memory and
1366 unref it again later on unmap.
1367 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10020>
1368
1369 2025-11-06 17:06:02 +0100 Piotr Brzeziński <[email protected]>
1370
1371 * gst/gststructure.c:
1372 structure: Don't crash if GArray has NULL value
1373 NULL can be a valid value in this case and it would previously cause a segfault here.
1374 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10036>
1375
1376 2025-11-05 09:43:55 +0200 Sebastian Dröge <[email protected]>
1377
1378 * tests/check/gst/gstinfo.c:
1379 info: Fix test pattern to check for an expected debug log line
1380 There's not necessarily a `0` between the sub-second `.` and the debug category.
1381 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/10012>
1382
1383 2025-11-03 18:45:20 +0200 Sebastian Dröge <[email protected]>
1384
1385 * gst/gstinfo.c:
1386 info: Fix format string types for pid/tid on Windows
1387 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9913>
1388
1389 2025-11-03 16:19:30 +0530 Nirbheek Chauhan <[email protected]>
1390
1391 * gst/gstinfo.c:
1392 * gst/meson.build:
1393 info: Use gettid() when available
1394 It is always available on Android, and on Linux with glibc > 2.30 and
1395 recent musl. It is also available on various BSD versions.
1396 This is faster than the syscall since the value will be cached by the
1397 libc and avoids a context switch.
1398 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9913>
1399
1400 2025-11-03 12:19:08 +0200 Sebastian Dröge <[email protected]>
1401
1402 * gst/gstinfo.c:
1403 info: Update PID/TID/pointer format strings for this century
1404 Pointers are generally 32 or 64 bits and use the full range, so use `%16p` /
1405 `%8p` for them.
1406 PID/TID are 32 bit integers on Windows, macOS and Linux and actually use the
1407 full range nowadays too so always use `%10d` / `%10lu`.
1408 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9913>
1409
1410 2025-11-03 12:08:01 +0200 Sebastian Dröge <[email protected]>
1411
1412 * gst/gstinfo.c:
1413 info: Define correct types for the PID between Windows and Linux
1414 And also don't truncate it when using it for the debug log filename.
1415 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9913>
1416
1417 2025-11-03 10:04:55 +0200 Sebastian Dröge <[email protected]>
1418
1419 * gst/gstinfo.c:
1420 info: Print thread ID instead of GThread handle on Linux and otherwise pthread ID
1421 gdb and lldb both print both the pthread_self() thread ID as well as the
1422 LWP (gettid) thread ID. ps/htop print the LWP (gettid) thread ID.
1423 We print the LWP ID here as that covers more applications, and print it as
1424 decimal integer instead of hexadecimal as that's what those tools do too.
1425 On other UNIXes, print the pthread ID hexadecimal as it's at least more useful
1426 than the GThread handle when debugging.
1427 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9913>
1428
1429 2025-10-26 13:56:34 +0000 Seungha Yang <[email protected]>
1430
1431 * gst/gstinfo.c:
1432 info: Print thread ID instead of GThread handle on Windows
1433 Since Visual Studio debugger prints thread id values,
1434 printing the same value here would be much more useful than
1435 logging random GThread address.
1436 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9913>
1437
1438 2024-03-16 03:41:35 -0300 Val Packett <[email protected]>
1439
1440 * gst/gsttaglist.c:
1441 * gst/gsttaglist.h:
1442 gstreamer: capitalize dB consistently
1443 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6369>
1444
1445 2024-03-14 02:26:35 -0300 Val Packett <[email protected]>
1446
1447 * gst/gsttaglist.c:
1448 * gst/gsttaglist.h:
1449 gstreamer: Add EBU R 128 gain tags (RFC 7845)
1450 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6369>
1451
1452 2025-10-25 08:56:00 +0900 Inbok Kim <[email protected]>
1453
1454 * gst/printf/vasnprintf.c:
1455 vasnprintf: free dynamic tmp buffer on error to prevent memory leak
1456 If an error occurs and a dynamically allocated tmp buffer is used,
1457 ensure it is properly freed to prevent a memory leak.
1458 Upstream fix reference:
1459 https://cgit.git.savannah.gnu.org/cgit/gnulib.git/commit/?id=4dd9d8a7fa85e9e9223a6c175b4a3fd46d222ff6
1460 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9893>
1461
1462 2025-06-11 16:02:13 +0900 dongjoo.kim <[email protected]>
1463
1464 * gst/parse/grammar.y.in:
1465 parse: Move g_strfreev() a bit later to avoid use-after-free
1466 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9871>
1467
1468 2025-02-28 19:09:46 +0900 DongJoo Kim <[email protected]>
1469
1470 * gst/gstinfo.h:
1471 gstinfo: Added parentheses to ensure proper evaluation of conditions in logging level checks
1472 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9917>
1473
1474 2025-06-10 14:40:16 +0900 dongjoo.kim <[email protected]>
1475
1476 * gst/gstutils.c:
1477 utils: Fix leak in gst_util_filename_compare
1478 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9916>
1479
1480 2025-10-25 12:08:25 +0100 Xavier Claessens <[email protected]>
1481
1482 * tools/gst-launch.c:
1483 gst-launch: Print details of error msg
1484 It does it already for every message in bus_handler(), which includes
1485 warnings and info, but not for error messages that are instead handled
1486 by bus_sync_handler().
1487 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9894>
1488
1489 2025-10-10 12:39:20 +0200 Havard Graff <[email protected]>
1490
1491 * gst/gstpad.c:
1492 gstpad: make gst_pad_forward not O(n²)
1493 With many pads, this function grinds to a complete halt.
1494 By instead using a hash-table, and only bother to check if we have
1495 been resynced (because if not, our pads are in order from the iterator),
1496 the algorithm is more like O(n)
1497 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9902>
1498
1499 2025-10-15 11:28:21 +0300 Sebastian Dröge <[email protected]>
1500
1501 * gst/gstbuffer.h:
1502 * gst/gstbufferlist.h:
1503 * gst/gstpad.h:
1504 gst: Mark callback inout parameters as such
1505 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9847>
1506
1507 2025-10-14 09:12:47 -0300 Thibault Saunier <[email protected]>
1508
1509 * gst/gstbin.c:
1510 * gst/gstbin.h:
1511 * gst/gstbuffer.c:
1512 * gst/gstbuffer.h:
1513 * gst/gstbufferlist.c:
1514 * gst/gstcaps.c:
1515 * gst/gstcaps.h:
1516 * gst/gstclock.c:
1517 * gst/gstelement.h:
1518 * gst/gstelementfactory.c:
1519 * gst/gstiterator.h:
1520 * gst/gstmemory.c:
1521 * gst/gstmessage.h:
1522 * gst/gstminiobject.h:
1523 * gst/gstsample.c:
1524 * gst/gststructure.c:
1525 * gst/gststructure.h:
1526 * gst/gsttaglist.c:
1527 * gst/gstvalue.h:
1528 Revert "doc: python: Document PyGObject overrides for core GStreamer"
1529 This reverts commit 81f5440159ca43194e98327e42bdd6a48e946d69.
1530 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9840>
1531
1532 2025-09-28 10:46:19 -0300 Thibault Saunier <[email protected]>
1533
1534 * gst/gstplugin.c:
1535 gstplugin: convert plugin loading mutex to recursive mutex
1536 This fixes a deadlock that occurs when a plugin tries to load another
1537 plugin during its initialization. The deadlock happens when:
1538 1. GES plugin is being loaded and calls load_python_formatters()
1539 2. load_python_formatters() tries to load the Python plugin
1540 3. Both operations try to acquire gst_plugin_loading_mutex
1541 4. Since it's a regular mutex, the second lock attempt deadlocks
1542 Moving to a RecMutex is safe as only one thread is accessing the
1543 internal state.
1544 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9759>
1545
1546 2025-10-04 15:14:31 -0400 Doug Nazar <[email protected]>
1547
1548 * gst/gstbin.h:
1549 * gst/gstghostpad.h:
1550 * gst/gstpad.h:
1551 * gst/gstpadtemplate.h:
1552 * gst/gstpipeline.h:
1553 * libs/gst/controller/gstargbcontrolbinding.h:
1554 * libs/gst/controller/gstdirectcontrolbinding.h:
1555 * libs/gst/controller/gstproxycontrolbinding.h:
1556 * tests/check/gst/gstelement.c:
1557 gst: Add G_GNUC_WARN_UNUSED_RESULT to constructors
1558 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9796>
1559
1560 2025-10-11 09:20:32 -0300 Thibault Saunier <[email protected]>
1561
1562 * libs/gst/controller/controller-prelude.h:
1563 * libs/gst/controller/gsttimedvaluecontrolsource.c:
1564 * libs/gst/controller/gsttimedvaluecontrolsource.h:
1565 controller: Add MT-safe gst_timed_value_control_source_list_control_points()
1566 gst_timed_value_control_source_get_all() is not thread-safe because it
1567 returns transfer-container - the GList contains pointers to internal
1568 GstControlPoint structures that can be freed by another thread while
1569 being accessed, causing use-after-free bugs.
1570 The new list_control_points() returns a full copy of the timed values.
1571 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9830>
1572
1573 2025-10-10 20:56:21 -0300 Thibault Saunier <[email protected]>
1574
1575 * libs/gst/controller/gsttimedvaluecontrolsource.c:
1576 controller: Fix get_all() return type annotation in GIR
1577 The method returns a list of GstControlPoint, not GstTimedValue.
1578 This fixes the type annotation to match the actual return type.
1579 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9828>
1580
1581 2025-09-24 13:58:54 +0530 Nirbheek Chauhan <[email protected]>
1582
1583 * gst/gstvalue.c:
1584 gstvalue: Support 0b/0B prefix for bitmasks
1585 C23, Python, and Rust support binary literals with the 0b or 0B
1586 prefix, which is quite convenient when dealing with bitmasks.
1587 Unfortunately we cannot use this for serializing due to
1588 backwards-compat.
1589 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9732>
1590
1591 2025-09-29 16:52:43 +0900 Seungha Yang <[email protected]>
1592
1593 * gst/gstbin.c:
1594 bin: Port to gst_object_call_async
1595 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/869>
1596
1597 2025-09-29 16:50:54 +0900 Seungha Yang <[email protected]>
1598
1599 * gst/gstelement.c:
1600 * gst/gstelement.h:
1601 * tests/check/gst/gstelement.c:
1602 gst: Deprecate gst_element_call_async method
1603 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/869>
1604
1605 2021-08-19 01:20:01 +0900 Seungha Yang <[email protected]>
1606
1607 * gst/gst_private.h:
1608 * gst/gstelement.c:
1609 * gst/gstobject.c:
1610 * gst/gstobject.h:
1611 * gst/gsttask.c:
1612 * gst/gstutils.c:
1613 * gst/gstutils.h:
1614 * tests/check/gst/gstelement.c:
1615 * tests/check/gst/gstobject.c:
1616 * tests/check/gst/gstutils.c:
1617 gst: Add gst_call_async() and gst_object_call_async() variants
1618 Adding two gst_element_call_async() variant APIs, gst_object_call_async()
1619 and gst_call_async(). gst_object_call_async() is functionally identical to
1620 the gst_element_call_async() but it's allowed to be called against GstObject.
1621 And gst_call_async() can be used for any asynchronous function call
1622 without GstObject dependency.
1623 Not only the case of state change which was mentioned in
1624 the commit which introduced gst_element_call_async() API,
1625 there are various cases where an operation must happen from another thread.
1626 For instance, assume that an object has its own work thread
1627 and the object can notify error via g_object_notify() or it's
1628 variant from internal work thread. As a response of the notify callback,
1629 callee might want to destroy the object. Then, which might
1630 result in an attempt for destroying the object from its work thread
1631 (i.e., try to join thread from self). To avoid the case, notification
1632 must happen from non-work thread, or caller should destroy the
1633 object from other thread.
1634 Although each element/plugin can implement its own thread pool
1635 for the case, expanding the gst_element_call_async() method
1636 would be an easy way to provide a method for such cases.
1637 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/869>
1638
1639 2025-10-05 14:56:35 -0400 Xavier Claessens <[email protected]>
1640
1641 * libs/gst/helpers/meson.build:
1642 meson: Add missing devenv values
1643 Those are the differences spotted between:
1644 - meson devenv -C builddir --dump meson.env
1645 - ./gst-env.py --only-environment > gst.env
1646 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9800>
1647
1648 2025-09-22 12:53:31 -0300 Thibault Saunier <[email protected]>
1649
1650 * gst/gstbin.c:
1651 * gst/gstbin.h:
1652 * gst/gstbuffer.c:
1653 * gst/gstbuffer.h:
1654 * gst/gstbufferlist.c:
1655 * gst/gstcaps.c:
1656 * gst/gstcaps.h:
1657 * gst/gstclock.c:
1658 * gst/gstelement.h:
1659 * gst/gstelementfactory.c:
1660 * gst/gstiterator.h:
1661 * gst/gstmemory.c:
1662 * gst/gstmessage.h:
1663 * gst/gstminiobject.h:
1664 * gst/gstsample.c:
1665 * gst/gststructure.c:
1666 * gst/gststructure.h:
1667 * gst/gsttaglist.c:
1668 * gst/gstvalue.h:
1669 doc: python: Document PyGObject overrides for core GStreamer
1670 Add comprehensive documentation for Python-specific functionality
1671 provided by PyGObject overrides in core GStreamer classes including:
1672 - Bin: make_and_add helper method and multi-element add() support
1673 - Buffer/Memory: context manager support for map operations
1674 - Caps: constructor overrides and container protocol support
1675 - Clock: TIME_ARGS utility function
1676 - Element: link_many static method
1677 - ElementFactory: convenience metadata getters and classmethod make()
1678 - Iterator: Python iteration protocol support
1679 - MiniObject: make_writable and flags property
1680 - Structure: dictionary-like access and constructor overrides
1681 - TagList: container protocol support
1682 Documentation is placed in the appropriate C function documentation
1683 where overrides enhance existing functionality, or in class-level
1684 SECTION documentation for new helper methods.
1685 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9717>
1686
1687 2025-09-20 19:35:34 -0300 Thibault Saunier <[email protected]>
1688
1689 * docs/gst-plugins-doc-cache-generator.py:
1690 doc: Generate hotdoc templates dynamically in build directory
1691 Generate the hotdoc language templates directly in code when building
1692 documentation. These templates enable language-specific documentation
1693 blocks and are now automatically available to all GStreamer subprojects.
1694 This allows markdown documentation to use directives like {{ C.md }},
1695 {{ PY.md }}, {{ JS.md }}, etc. for language-specific content blocks.
1696 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9717>
1697
1698 2025-09-11 09:58:48 -0400 Doug Nazar <[email protected]>
1699
1700 * gst/gstvalue.c:
1701 base: mark items that are unused when checks or asserts are disabled
1702 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9760>
1703
1704 2025-09-25 08:38:47 -0400 Doug Nazar <[email protected]>
1705
1706 * libs/gst/base/gstadapter.h:
1707 * libs/gst/base/gstaggregator.h:
1708 * libs/gst/base/gstbaseparse.h:
1709 * libs/gst/base/gstbasesink.h:
1710 * libs/gst/base/gstbasesrc.h:
1711 * libs/gst/base/gstbasetransform.h:
1712 * libs/gst/base/gstbitwriter.h:
1713 * libs/gst/base/gstbytewriter.h:
1714 * libs/gst/base/gstcollectpads.h:
1715 * libs/gst/base/gstdataqueue.h:
1716 * libs/gst/base/gstflowcombiner.h:
1717 * libs/gst/base/gsttypefindhelper.h:
1718 * libs/gst/check/gstharness.h:
1719 * libs/gst/check/gsttestclock.h:
1720 * libs/gst/controller/gstinterpolationcontrolsource.h:
1721 * libs/gst/controller/gstlfocontrolsource.h:
1722 * libs/gst/controller/gsttimedvaluecontrolsource.h:
1723 * libs/gst/controller/gsttriggercontrolsource.h:
1724 * libs/gst/net/gstnetclientclock.h:
1725 * libs/gst/net/gstnettimepacket.h:
1726 * libs/gst/net/gstnettimeprovider.h:
1727 * libs/gst/net/gstptpclock.h:
1728 gst: Add G_GNUC_WARN_UNUSED_RESULT to funcs with transfer full returns
1729 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9771>
1730
1731 2025-09-24 15:33:02 -0400 Doug Nazar <[email protected]>
1732
1733 * gst/gst.c:
1734 * gst/gstinfo.h:
1735 * tests/check/elements/leaks.c:
1736 * tests/check/gst/gstbuffer.c:
1737 * tests/check/gst/gstevent.c:
1738 * tests/check/gst/gstmessage.c:
1739 * tests/check/gst/gstpad.c:
1740 gst: fixes
1741 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9766>
1742
1743 2025-09-24 15:30:11 -0400 Doug Nazar <[email protected]>
1744
1745 * gst/gstallocator.h:
1746 * gst/gstatomicqueue.h:
1747 * gst/gstbin.h:
1748 * gst/gstbuffer.h:
1749 * gst/gstbufferlist.h:
1750 * gst/gstbufferpool.h:
1751 * gst/gstbus.h:
1752 * gst/gstcaps.h:
1753 * gst/gstcapsfeatures.h:
1754 * gst/gstchildproxy.h:
1755 * gst/gstclock.h:
1756 * gst/gstcontext.h:
1757 * gst/gstcontrolbinding.h:
1758 * gst/gstdatetime.h:
1759 * gst/gstdevice.h:
1760 * gst/gstdevicemonitor.h:
1761 * gst/gstdeviceprovider.h:
1762 * gst/gstdeviceproviderfactory.h:
1763 * gst/gstelement.h:
1764 * gst/gstelementfactory.h:
1765 * gst/gstevent.h:
1766 * gst/gstghostpad.h:
1767 * gst/gstidstr.h:
1768 * gst/gstinfo.h:
1769 * gst/gstiterator.h:
1770 * gst/gstmessage.h:
1771 * gst/gstminiobject.h:
1772 * gst/gstobject.h:
1773 * gst/gstpad.h:
1774 * gst/gstpadtemplate.h:
1775 * gst/gstparse.h:
1776 * gst/gstpipeline.h:
1777 * gst/gstplugin.h:
1778 * gst/gstpluginfeature.h:
1779 * gst/gstpreset.h:
1780 * gst/gstpromise.h:
1781 * gst/gstquery.h:
1782 * gst/gstregistry.h:
1783 * gst/gstsample.h:
1784 * gst/gstsegment.h:
1785 * gst/gststreamcollection.h:
1786 * gst/gststreams.h:
1787 * gst/gststructure.h:
1788 * gst/gsttaglist.h:
1789 * gst/gsttask.h:
1790 * gst/gsttaskpool.h:
1791 * gst/gsttoc.h:
1792 * gst/gsttocsetter.h:
1793 * gst/gsttracerrecord.h:
1794 * gst/gsturi.h:
1795 * gst/gstutils.h:
1796 gst: Mark functions WARN_UNUSED_RESULT which return is full transfer
1797 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9766>
1798
1799 2025-09-11 09:38:48 -0400 Doug Nazar <[email protected]>
1800
1801 * tests/benchmarks/complexity.c:
1802 * tests/benchmarks/controller.c:
1803 * tests/benchmarks/gstbufferstress.c:
1804 * tests/benchmarks/mass-elements.c:
1805 * tests/check/elements/selector.c:
1806 * tests/check/elements/tee.c:
1807 * tests/check/generic/states.c:
1808 * tests/check/gst/gstbus.c:
1809 * tests/check/gst/gstghostpad.c:
1810 * tests/check/gst/gstmeta.c:
1811 * tests/check/gst/gstminiobject.c:
1812 * tests/check/gst/gstutils.c:
1813 * tests/check/gst/gstvalue.c:
1814 * tests/check/libs/basesrc.c:
1815 * tests/check/libs/gsttestclock.c:
1816 * tests/check/libs/transform1.c:
1817 * tests/check/pipelines/cleanup.c:
1818 * tests/check/pipelines/simple-launch-lines.c:
1819 gst: tests: convert g_assert() to g_assert_*() and mark unused items
1820 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9724>
1821
1822 2025-09-11 09:30:56 -0400 Doug Nazar <[email protected]>
1823
1824 * gst/gstmessage.c:
1825 * gst/gstminiobject.c:
1826 * gst/gstvalue.c:
1827 gst: mark items that are unused when checks or asserts are disabled
1828 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9724>
1829
1830 2025-09-11 09:26:25 -0400 Doug Nazar <[email protected]>
1831
1832 * gst/gstmacros.h:
1833 gst: add macros to mark items unused when checks or asserts disabled
1834 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9724>
1835
1836 2025-09-25 08:19:07 +0900 Inbok Kim <[email protected]>
1837
1838 * plugins/elements/gstmultiqueue.c:
1839 multiqueue: Fix object reference handling in signal callbacks
1840 Move gst_object_unref() calls after signal emissions in overrun and
1841 underrun callbacks to ensure the multiqueue object remains valid
1842 during signal handler execution.
1843 This prevents potential race conditions where signal handlers might
1844 access a freed object, improving thread safety in multiqueue operations.
1845 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9738>
1846
1847 2025-09-25 09:46:23 +0900 Inbok Kim <[email protected]>
1848
1849 * libs/gst/net/gstnetclientclock.c:
1850 netclientclock: Fix memory leak in error paths
1851 Add missing g_object_unref() calls for servaddr in error handling
1852 blocks to prevent memory leaks when socket creation, binding, or
1853 address resolution fails.
1854 The servaddr object was not being properly unreferenced in the
1855 no_socket, bind_error, and getsockname_error error paths, leading
1856 to memory leaks in failure scenarios.
1857 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9740>
1858
1859 2025-09-24 12:55:28 -0400 Xavier Claessens <[email protected]>
1860
1861 * tools/gst-launch.c:
1862 gst-launch: Do not assume error messages have a src element
1863 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9737>
1864
1865 2025-09-18 18:39:47 +0300 Sebastian Dröge <[email protected]>
1866
1867 * libs/gst/base/gstbytereader.c:
1868 * libs/gst/base/gstbytewriter.c:
1869 bytereader: Add various missing annotations
1870 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9709>
1871
1872 2025-09-18 18:23:46 +0300 Sebastian Dröge <[email protected]>
1873
1874 * libs/gst/base/gstadapter.c:
1875 adapter: Annotate size parameters of `take()` and `map()` as in-parameters
1876 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9709>
1877
1878 2025-09-15 14:29:23 -0700 Xavier Claessens <[email protected]>
1879
1880 * libs/gst/base/gstadapter.c:
1881 GstAdapter: Add doc clarification regarding timestamps
1882 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9690>
1883
1884 2025-09-15 15:22:59 +0200 Piotr Brzeziński <[email protected]>
1885
1886 * libs/gst/base/gstaggregator.c:
1887 aggregator: Prevent start time selection before reaching PLAYING
1888 If this code was hit before aggregator reached PLAYING, we'd get garbage
1889 start_time because the base time was still 0.
1890 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9689>
1891
1892 2025-09-09 12:56:26 +0300 Sebastian Dröge <[email protected]>
1893
1894 * gst/gstmetafactory.c:
1895 metafactory: Use same parameter name in header and source file to make g-i happy
1896 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9664>
1897
1898 2025-09-08 15:11:45 +0300 Sebastian Dröge <[email protected]>
1899
1900 * gst/gstmemory.c:
1901 * gst/gstmemory.h:
1902 memory: Add accessor for the `GstMapInfo` `data field
1903 This has clearer ownership semantics compared to the plain struct member.
1904 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9661>
1905
1906 2025-09-08 11:06:16 +0300 Sebastian Dröge <[email protected]>
1907
1908 * gst/gstpad.c:
1909 * gst/gstpad.h:
1910 pad: Add more accessors for `GstPadProbeInfo` fields
1911 Including setters for the fields that are valid to modify.
1912 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9661>
1913
1914 2025-09-08 11:05:50 +0300 Sebastian Dröge <[email protected]>
1915
1916 * gst/gstpad.c:
1917 * gst/gstpad.h:
1918 pad: Replace some `allow-none` annotations with `nullable`
1919 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9661>
1920
1921 2025-09-05 12:28:30 -0400 Xavier Claessens <[email protected]>
1922
1923 * gst/gstmeta.c:
1924 meta: throttle down warning about unregistered meta
1925 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9652>
1926
1927 2025-09-07 20:39:44 +0100 Tim-Philipp Müller <[email protected]>
1928
1929 * meson.build:
1930 Back to development after 1.27.2
1931
1932 === release 1.27.2 ===
1933
1934 2025-09-07 20:34:55 +0100 Tim-Philipp Müller <[email protected]>
1935
1936 * NEWS:
1937 * RELEASE:
1938 * gstreamer.doap:
1939 * meson.build:
1940 Release 1.27.2
1941
1942 2025-09-06 11:08:47 +0300 Sebastian Dröge <[email protected]>
1943
1944 * gst/gstbin.c:
1945 * gst/gstelement.c:
1946 * gst/gstmessage.c:
1947 * libs/gst/base/gstbasesink.c:
1948 * plugins/elements/gstdataurisrc.c:
1949 * plugins/tracers/gstlog.c:
1950 * tests/check/libs/aggregator.c:
1951 * tests/check/pipelines/simple-launch-lines.c:
1952 * tools/gst-launch.c:
1953 gst: Change usage of gst_element_state_*() to gst_state_*()
1954 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9655>
1955
1956 2025-09-06 11:04:16 +0300 Sebastian Dröge <[email protected]>
1957
1958 * gst/gstutils.c:
1959 * gst/gstutils.h:
1960 element: Deprecate gst_element_state_*() API
1961 It's misnamed: the enums in question are called GstState* and not GstElementState*.
1962 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9655>
1963
1964 2025-08-25 15:22:20 -0400 Xavier Claessens <[email protected]>
1965
1966 * gst/gst.h:
1967 * gst/gstbuffer.c:
1968 * gst/gstmeta.c:
1969 * gst/gstmetafactory.c:
1970 * gst/gstmetafactory.h:
1971 * gst/gstregistrybinary.c:
1972 * gst/gstregistrychunks.c:
1973 * gst/meson.build:
1974 meta: Add GstMetaFactory to dynamically register GstMetaInfo
1975 This allows plugins to register their GstMetaInfo so
1976 gst_meta_deserialize() can deserialize them automatically.
1977 This helps making unixfd common cases work out of the box instead of
1978 relying on application to do it.
1979 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9598>
1980
1981 2025-09-05 13:33:04 +0300 Sebastian Dröge <[email protected]>
1982
1983 * gst/gststructure.c:
1984 caps: Mark out parameter of gst_structure_get_caps() as `transfer none`
1985 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9646>
1986
1987 2025-09-05 11:27:41 +0300 Sebastian Dröge <[email protected]>
1988
1989 * gst/gstcaps.c:
1990 caps: Mark gst_caps_new_full() parameters as `transfer full`
1991 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9524>
1992
1993 2025-08-19 18:00:49 +0300 Sebastian Dröge <[email protected]>
1994
1995 * gst/gstmessage.c:
1996 * gst/gstmessage.h:
1997 message: Don't mark transfer-full structure for the redirect message as const
1998 This is just confusing.
1999 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9524>
2000
2001 2025-08-11 17:29:21 +0300 Sebastian Dröge <[email protected]>
2002
2003 * gst/gstiterator.c:
2004 iterator: Mark master cookie parameter to constructor as gpointer
2005 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9524>
2006
2007 2025-08-11 12:36:45 +0300 Sebastian Dröge <[email protected]>
2008
2009 * gst/gst.c:
2010 gstreamer: Mark parameters to gst_init() as optional
2011 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9524>
2012
2013 2025-08-11 11:24:29 +0300 Sebastian Dröge <[email protected]>
2014
2015 * gst/gstcapsfeatures.c:
2016 * gst/gststructure.c:
2017 gstreamer: Mark refcount parameter to structure/capsfeatures set_parent_refcount() as pointer
2018 Otherwise gobject-introspection considers it a plain integer.
2019 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9524>
2020
2021 2025-08-09 18:51:42 +0300 Sebastian Dröge <[email protected]>
2022
2023 * gst/gstutils.c:
2024 utils: Mark first parameter of gst_calculate_linear_regression() as array
2025 The length is provided in a complicated way that can't be handled by
2026 gobject-introspection but this at least marks it as array.
2027 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9524>
2028
2029 2025-08-09 18:49:57 +0300 Sebastian Dröge <[email protected]>
2030
2031 * gst/gstutils.c:
2032 utils: Mark parameters of gst_util_simplify_fraction() as inout
2033 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9524>
2034
2035 2025-08-09 18:48:42 +0300 Sebastian Dröge <[email protected]>
2036
2037 * gst/gsttypefind.c:
2038 typefind: Mark return value of gst_type_find_peek() as array
2039 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9524>
2040
2041 2025-08-09 18:36:14 +0300 Sebastian Dröge <[email protected]>
2042
2043 * gst/gstmeta.c:
2044 meta: Mark data parameter of gst_meta_deserialize() as array
2045 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9524>
2046
2047 2025-08-09 18:31:04 +0300 Sebastian Dröge <[email protected]>
2048
2049 * gst/gstcapsfeatures.c:
2050 * gst/gststructure.c:
2051 gstreamer: Mark structure/capsfeatures set_parent_refcount() as non-introspectable
2052 The pointer parameter can't be made use of automatically by bindings.
2053 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9524>
2054
2055 2025-09-02 19:22:02 -0400 Doug Nazar <[email protected]>
2056
2057 * tests/check/gst/gstbuffer.c:
2058 buffer: Add test for _memset() & _memcmp() with various offsets
2059 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9628>
2060
2061 2025-08-21 08:38:03 -0400 Xavier Claessens <[email protected]>
2062
2063 * gst/gstbuffer.c:
2064 * gst/gstbuffer.h:
2065 * gst/gstbufferlist.c:
2066 * gst/gstbufferlist.h:
2067 * gst/gstcaps.c:
2068 * gst/gstcaps.h:
2069 * gst/gstmemory.c:
2070 * gst/gstmemory.h:
2071 * gst/gstmessage.c:
2072 * gst/gstmessage.h:
2073 * gst/gstquery.c:
2074 * gst/gstquery.h:
2075 gst: Add _take and _steal to more mini objects
2076 Those are simple wrappers around mini object API, similar to
2077 what GstEvent does.
2078 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9595>
2079
2080 2025-08-21 09:02:45 +0300 Sebastian Dröge <[email protected]>
2081
2082 * libs/gst/base/gstbaseparse.c:
2083 baseparse: Try harder to fixate caps based on upstream in default negotiation
2084 Upstream might provide a width/height while downstream has the field but accepts
2085 a range. gst_caps_fixate() would select the minimum value of that range later
2086 but it would be more accurate to take the upstream value, at least if it's a
2087 subset of what downstream accepts.
2088 Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4608
2089 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9593>
2090
2091 2025-08-20 15:32:15 +0300 Sebastian Dröge <[email protected]>
2092
2093 * gst/gstbuffer.c:
2094 * gst/gstbuffer.h:
2095 * gst/gstbufferlist.c:
2096 * gst/gstbufferlist.h:
2097 * gst/gstcaps.c:
2098 * gst/gstcaps.h:
2099 * gst/gstcontext.c:
2100 * gst/gstcontext.h:
2101 * gst/gstevent.c:
2102 * gst/gstevent.h:
2103 * gst/gstmemory.c:
2104 * gst/gstmemory.h:
2105 * gst/gstmessage.c:
2106 * gst/gstmessage.h:
2107 * gst/gstquery.c:
2108 * gst/gstquery.h:
2109 * gst/gstsample.c:
2110 * gst/gstsample.h:
2111 * gst/gsttaglist.c:
2112 * gst/gsttaglist.h:
2113 * gst/gsturi.h:
2114 gst: Convert `is_writable()` / `make_writable()` macros to inline functions
2115 Plus actual functions that are exported from the library.
2116 Apart from improving type-safety, this also makes bindings more happy.
2117 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9588>
2118
2119 2025-08-07 17:09:22 +0200 Vivienne Watermeier <[email protected]>
2120
2121 * libs/gst/base/gstbaseparse.c:
2122 * tests/check/libs/baseparse.c:
2123 baseparse: don't clear most sticky events after a FLUSH_STOP event
2124 Clearing sticky events - besides EOS, STREAM_GROUP_DONE, and SEGMENT -
2125 risks losing them if a flush occurs before we get another buffer.
2126 Also adds a unit test: parser_sticky_events_after_flush
2127 Fixes #4193
2128 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9506>
2129
2130 2025-08-12 11:10:55 -0400 Thibault Saunier <[email protected]>
2131
2132 * gst/gstdebugutils.c:
2133 * gst/gsttracerutils.c:
2134 debugutils: Use SHOW_FULL_PARAMS when GstDotsTracer is active
2135 When the dots tracer is detected among active tracers, switch from
2136 SHOW_ALL to SHOW_FULL_PARAMS to prevent parameter ellipsization.
2137 The dots tracer is designed to work with gst-dots-viewer which can
2138 handle long parameter values, so we preserve the full text in dot files.
2139 This improves the viewing experience when using the dots tracer with
2140 the gst-dots-viewer tool.
2141 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9547>
2142
2143 2025-08-12 10:59:18 -0400 Thibault Saunier <[email protected]>
2144
2145 * gst/gsttracerutils.c:
2146 * gst/gsttracerutils.h:
2147 tracerutils: keep tracers that do not register to any hook alive as others
2148 This checks if a tracer has registered to any hook after
2149 initialization and automatically registers it to the "none" hook if not
2150 so the lifetime of those tracer is the same as tracer who registers to hooks.
2151 For example the `dots` tracer was being freed right after initialization
2152 which was unexpected.
2153 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9547>
2154
2155 2025-08-15 16:44:34 +0200 Havard Graff <[email protected]>
2156
2157 * gst/gstbuffer.c:
2158 buffer: Fix gst_buffer_memcmp() / gst_buffer_memset() using wrong memory index
2159 Regression from 160205b483dc7a1ae5da60d80a722837cf7c01d0.
2160 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9565>
2161
2162 2025-08-09 17:10:32 +0300 Sebastian Dröge <[email protected]>
2163
2164 * gst/gstvalue.c:
2165 value: Consider NULL caps in array a fixed value
2166 There's no requirement for caps in arrays to be non-NULL and NULL is clearly a
2167 fixed, single value.
2168 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9522>
2169
2170 2025-08-13 16:18:22 +0200 Ruben Gonzalez <[email protected]>
2171
2172 * tools/gst-launch.c:
2173 gst-launch: clean unused local variable
2174 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9551>
2175
2176 2025-07-14 20:05:34 +0200 Mathieu Duponchelle <[email protected]>
2177
2178 * libs/gst/base/gstaggregator.c:
2179 * libs/gst/base/gstaggregator.h:
2180 aggregator: implement start-time-selection=now
2181 This is actually the only correct mode when live, let's at least expose
2182 an option for it.
2183 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9394>
2184
2185 2025-08-09 19:00:00 +0300 Sebastian Dröge <[email protected]>
2186
2187 * gst/gstcaps.c:
2188 * gst/gstchildproxy.c:
2189 * gst/gstelementfactory.c:
2190 * gst/gststructure.c:
2191 * gst/gsttaglist.c:
2192 * gst/gsttracerrecord.c:
2193 gstreamer: Make sure to zero-initialize the GValue before G_VALUE_COLLECT_INIT
2194 G_VALUE_INIT does not zero-initialize the data member as automatic
2195 zero-initialization only happens for non-union types. For union types the
2196 initialized value is undefined.
2197 Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4595
2198 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9525>
2199
2200 2025-08-11 10:42:06 +0300 Sebastian Dröge <[email protected]>
2201
2202 * libs/gst/helpers/ptp/net.rs:
2203 ptp: Fix a new Rust 1.89 compiler warning on Windows
2204 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9527>
2205
2206 2025-08-08 17:50:03 +0300 Sebastian Dröge <[email protected]>
2207
2208 * gst/meson.build:
2209 gstreamer: Disable miniobject inline functions for gobject-introspection for non-subprojects too
2210 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9518>
2211
2212 2025-08-08 10:13:55 +0300 Sebastian Dröge <[email protected]>
2213
2214 * libs/gst/helpers/ptp/net.rs:
2215 ptp: Fix new compiler warning with Rust 1.89
2216 This still compiles with Rust 1.48 and newer.
2217 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9512>
2218
2219 2025-08-06 17:54:47 +0530 Nirbheek Chauhan <[email protected]>
2220
2221 * gst/gstinfo.h:
2222 docs: Initialize the debug category in class_init when possible
2223 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9496>
2224
2225 2025-07-31 17:52:03 -0400 Olivier Crête <[email protected]>
2226
2227 * docs/random/plan-0.11.txt:
2228 docs: Remove 0.11 plan
2229 We're 10 years past.
2230 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9476>
2231
2232 2025-07-31 17:50:33 -0400 Olivier Crête <[email protected]>
2233
2234 * AUTHORS:
2235 AUTHORS: Remove outdated files
2236 They only contained historical contributors, the modern version is
2237 to look at the git logs.
2238 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9476>
2239
2240 2025-07-31 17:44:21 -0400 Olivier Crête <[email protected]>
2241
2242 * MAINTAINERS:
2243 MAINTAINERS: Update to reflect current maintainership
2244 Instead of listing everyone, just point to GitLab
2245 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9476>
2246
2247 2025-07-24 20:20:39 +0100 Nirbheek Chauhan <[email protected]>
2248
2249 * meson.build:
2250 meson: Pass sysprof=disabled to glib
2251 sysprof cannot be built on Windows, and this causes the build to fail
2252 on Windows.
2253 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9438>
2254
2255 2025-07-13 17:30:20 +0100 Nirbheek Chauhan <[email protected]>
2256
2257 * gst/gstevent.h:
2258 * gst/gstquery.h:
2259 * gst/gstsegment.h:
2260 gstreamer: Disable C5287 warning on MSVC
2261 ```
2262 ../subprojects/gstreamer/gst/gstpad.c(3866): warning C5287:
2263 operands are different enum types 'GstQueryType' and 'GstQueryTypeFlags';
2264 use an explicit cast to silence this warning
2265 ```
2266 We abuse these enums, and MSVC emits a warning for this specific case:
2267 https://learn.microsoft.com/en-us/cpp/error-messages/compiler-warnings/compiler-warnings-c5200-through-c5399
2268 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9431>
2269
2270 2025-07-11 12:49:47 -0400 Daniel Morin <[email protected]>
2271
2272 * gst/gstvalue.c:
2273 * tests/check/gst/gstcaps.c:
2274 caps: fix nested caps
2275 - Fixes #4534
2276 - Fix gst_caps_is_fixed(), fix gst_caps_fixate () for caps-in-caps
2277 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9378>
2278
2279 2025-06-30 02:14:59 -0400 Doug Nazar <[email protected]>
2280
2281 * gst/gstbuffer.c:
2282 buffer: Find initial memory block without unnecessary mapping
2283 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9354>
2284
2285 2025-07-08 20:00:07 +0100 Tim-Philipp Müller <[email protected]>
2286
2287 * meson.build:
2288 Back to development after 1.27.1
2289
2290 === release 1.27.1 ===
2291
2292 2025-07-08 19:55:15 +0100 Tim-Philipp Müller <[email protected]>
2293
2294 * NEWS:
2295 * RELEASE:
2296 * gstreamer.doap:
2297 * meson.build:
2298 Release 1.27.1
2299
2300 2025-07-08 17:06:15 +0100 Tim-Philipp Müller <[email protected]>
2301
2302 * po/hr.po:
2303 * po/ka.po:
2304 gstreamer: update translations
2305
2306 2025-05-30 15:05:30 -0400 Doug Nazar <[email protected]>
2307
2308 * gst/gsttracerutils.c:
2309 tracerutils: Fix a few memory leaks
2310 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9151>
2311
2312 2025-06-23 14:24:57 -0400 Doug Nazar <[email protected]>
2313
2314 * libs/gst/net/gstnetclientclock.c:
2315 * libs/gst/net/gstnetclientclock.h:
2316 * tests/check/libs/gstnetclientclock.c:
2317 netclientclock: Add deinitialize function for testing
2318 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9115>
2319
2320 2025-05-28 08:25:45 -0400 Doug Nazar <[email protected]>
2321
2322 * tests/check/libs/gstnetclientclock.c:
2323 * tests/check/meson.build:
2324 netclientclock: test: add delay under valgrind to free clock from cache
2325 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9115>
2326
2327 2025-05-28 08:23:25 -0400 Doug Nazar <[email protected]>
2328
2329 * gst/gstinfo.c:
2330 info: free dwarf cache during shutdown
2331 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9115>
2332
2333 2025-05-28 08:22:05 -0400 Doug Nazar <[email protected]>
2334
2335 * plugins/elements/gstidentity.c:
2336 identity: unref clock after usage
2337 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9115>
2338
2339 2025-06-23 23:42:42 -0400 Daniel Morin <[email protected]>
2340
2341 * tests/check/gst/gstcaps.c:
2342 gstcaps: add test for caps
2343 - if a subset contain identical element in superset it is still a subset it they
2344 are note equal. e.g is_subset (<1, 1000>, <1, [1, 1000])
2345 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9220>
2346
2347 2025-06-13 12:02:25 -0400 Daniel Morin <[email protected]>
2348
2349 * tests/check/gst/gstvalue.c:
2350 gstvalue: increase test coverage for _is_subset()
2351 - Identify _is_subset () general/specials cases.
2352 - Add multiple test cases for general cases.
2353 - Add multiple test cases for unfixed value.
2354 - Add mutliple test cases for fixed array value.
2355 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9220>
2356
2357 2025-06-23 23:40:13 -0400 Daniel Morin <[email protected]>
2358
2359 * gst/gstvalue.c:
2360 gstvalue: generalize gst_value_is_subset()
2361 - Add case where we have element in subset that are equal to element in
2362 superset. In this case subset is still a subset if at least one element is a
2363 subset of the corresponding element in superset.
2364 - clarify gst_value_is_subset inline documentation
2365 - Add helper VALUE_TYPE_SINGLETON (type) to identify GType that represent a
2366 value that is not a collection.
2367 - Fixed issue in gst_value_is_subset_array_array (<1, 2, 3>, <0, 1, 4, 2, 3>)
2368 would have returned TRUE because because the alignment of the subset could be
2369 done multiple times. Now once alignment of set as been done once, everything
2370 else in the superset need to be a subset of corresponding element without
2371 interuption. is_subset (<1, 2, 3>, <0, 1, 2, 3>) => TRUE, but is_subset (<1,
2372 2, 3>, <0, 1, 4, 2, 3>) => FALSE.
2373 - If both array are fixed value and we perform a is_subset on them, both value
2374 can't be equal to be considered subset. (For consistency with other fixed
2375 values)
2376 - Define gst_value_subtract (singleton, array) and
2377 gst_value_subtract (array, singleton), to allow a gst_value_is_subset
2378 (singleton, array) and gst_value_is_subset (array, singleton). General
2379 case of gst_value_is_subset (v1, v2) use gst_value_subtract (v1, v2) and
2380 gst_value_subtract (v2, v1) to evaluate _is_subset(), therefore we need these
2381 for is_subset() operation that involve singleton and array.
2382 Note gst_value_subtract (array, array) is not implemented but this case is not
2383 require by gst_value_is_subset () and as it has a direct handling. Warned if
2384 the case is used.
2385 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9220>
2386
2387 2025-07-01 20:53:47 +0200 Mathieu Duponchelle <[email protected]>
2388
2389 * libs/gst/base/gstaggregator.c:
2390 aggregator: add sub_latency_min to pad queue size
2391 It should be possible for a subclass to let data accumulate on any of its input
2392 pads for as long as it has introduced latency.
2393 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9320>
2394
2395 2025-06-22 03:07:58 -0400 Doug Nazar <[email protected]>
2396
2397 * tests/check/elements/selector.c:
2398 selector: tests: Fix initialization of segment
2399 If GST_DEBUG is enabled and it dumps the segment, valgrind will warn about
2400 uninitialized memory.
2401 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9273>
2402
2403 2025-06-22 03:05:20 -0400 Doug Nazar <[email protected]>
2404
2405 * tests/check/gst/gstinfo.c:
2406 gstinfo: tests: Ensure that the target msg is one of the ones seen
2407 If GST_DEBUG is enabled we will get multiple log messages. Signal
2408 success if one of messages is correct.
2409 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9273>
2410
2411 2025-06-22 03:01:52 -0400 Doug Nazar <[email protected]>
2412
2413 * tests/check/gst/gstmemory.c:
2414 memory: tests: Fix test to ensure all logs are less severe than WARN
2415 If GST_DEBUG is enabled the existing test will fail as it gets INFO
2416 log messages.
2417 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9273>
2418
2419 2025-06-22 02:59:06 -0400 Doug Nazar <[email protected]>
2420
2421 * tests/check/libs/transform1.c:
2422 * tests/check/libs/transform2.c:
2423 transform: test: TestTransData is not an object
2424 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9273>
2425
2426 2025-06-23 13:42:32 +0200 David Maseda Neira <[email protected]>
2427
2428 * tools/gst-inspect.c:
2429 gst-inspect-1.0: Added type info for caps fields
2430 gst-inspect-1.0 now shows the expected GType for each field of the
2431 capabilities on a pad.
2432 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9270>
2433
2434 2025-06-26 10:27:33 +0300 Sebastian Dröge <[email protected]>
2435
2436 * tests/check/gst/gstutils.c:
2437 gstutils: Mark times array as static to avoid symbol conflict with the POSIX function
2438 It works fine but there can be linker warnings, e.g. with mold.
2439 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9288>
2440
2441 2025-06-18 18:22:18 +0300 Sebastian Dröge <[email protected]>
2442
2443 * gst/gstbuffer.c:
2444 * gst/gstbuffer.h:
2445 * tests/check/gst/gstbuffer.c:
2446 buffer: Add optional info structure to GstReferenceTimestampMeta
2447 This allows carrying additional per-timestamp information if needed.
2448 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9245>
2449
2450 2025-06-23 12:44:52 +0300 Sebastian Dröge <[email protected]>
2451
2452 * gst/gstpad.c:
2453 pad: Only remove TAG events on STREAM_START if the stream-id actually changes
2454 Missing part from 88a36b53c5064d186acb317b0b72633ef5d886e3
2455 See also https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4097
2456 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9268>
2457
2458 2025-06-19 09:47:32 -0400 Doug Nazar <[email protected]>
2459
2460 * tests/check/libs/baseparse.c:
2461 baseparse: test: Fix race on test start
2462 It's possible that the sink pad is still flushing when the first buffer
2463 is processed causing the test to timeout when we activate the sink after
2464 starting the pipeline.
2465 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9256>
2466
2467 2025-05-02 10:23:18 +0200 Alicia Boya García <[email protected]>
2468
2469 * gst/gstpad.c:
2470 tracers: Fix deadlock in latency tracer
2471 Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4317
2472 This patch fixes a regression in
2473 a35bf1e3847351703542869755b6b9e536b9d2fd.
2474 The new calls to tracers in the affected change were being done while
2475 holding the pad object lock, which wasn't the case in the old ones,
2476 leading to the latency tracer deadlocking in gst_object_get_parent().
2477 The dependency on the pad lock for those calls was accidental. This
2478 patch removes it by temporarily unlocking during the affected calls,
2479 not unlike how it's done when calling a probe callback or the
2480 send_event() function of the downstream element.
2481 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8925>
2482
2483 2025-06-06 08:59:20 -0400 Daniel Morin <[email protected]>
2484
2485 * gst/gstvalue.c:
2486 * tests/check/gst/gstcaps.c:
2487 value: fix subset between arrays
2488 - Adding gst_value_is_subset_array_array to test if an array is a subset of
2489 another array.
2490 - Check gst_caps_is_subset() between containing arrays work.
2491 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9184>
2492
2493 2025-06-03 22:32:17 -0400 Daniel Morin <[email protected]>
2494
2495 * gst/gstcaps.h:
2496 * gst/gststructure.c:
2497 * gst/gststructure.h:
2498 gststructure: add gst_structure_get_caps
2499 - Get field of type #GstCaps from #GstStructure
2500 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9184>
2501
2502 2025-06-10 14:41:22 +0200 Víctor Manuel Jáquez Leal <[email protected]>
2503
2504 * scripts/gen-changelog.py:
2505 * tools/gst-inspect.c:
2506 gstreamer-vaapi: remove subproject
2507 It's almost superseded by va plugin in gst-plugins-bad.
2508 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9200>
2509
2510 2025-06-03 22:11:59 +0100 James Cowgill <[email protected]>
2511
2512 * gst/gstvecdeque.c:
2513 vecdeque: Use correct index type in gst_vec_deque_drop_struct
2514 Fixes some `-Wsign-compare` warnings. These two indices should be
2515 `gsize` like the other variables in this function.
2516 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9171>
2517
2518 2025-06-04 12:47:55 +0200 Thibault Saunier <[email protected]>
2519
2520 * gst/gstinfo.c:
2521 gstinfo: Add missing log context stub functions when debugging disabled
2522 When GST_DISABLE_GST_DEBUG is defined, the log context functions were
2523 missing their stub implementations, causing link errors. Add the missing
2524 stub functions to the disabled debug section.
2525 Also add (nullable) annotation to gst_log_context_get_category since
2526 it can return NULL when debugging is disabled.
2527 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9122>
2528
2529 2025-06-04 10:38:12 +0200 Thibault Saunier <[email protected]>
2530
2531 * plugins/tracers/gstdots.c:
2532 dots: Do not WARN when a file can't be removed
2533 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9122>
2534
2535 2025-06-04 10:07:58 +0200 Thibault Saunier <[email protected]>
2536
2537 * gst/gstpad.c:
2538 pad: Change incompatible caps warning to GST_INFO
2539 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9122>
2540
2541 2024-09-23 10:31:13 -0300 Thibault Saunier <[email protected]>
2542
2543 * libs/gst/base/gstaggregator.c:
2544 aggregator: Do not set event seqnum to INVALID
2545 This might happen when we get EOS without any data flow
2546 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9128>
2547
2548 2025-05-16 13:32:08 +0200 Thibault Saunier <[email protected]>
2549
2550 * gst/gstinfo.h:
2551 general: Stop checking `G_HAVE_GNUC_VARARGS` now that we depend on c99
2552 Cleaning up a bit the code now that we can rely on C99 which specifies
2553 varargs for macros.
2554 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8990>
2555
2556 2025-05-21 10:01:24 -0400 Doug Nazar <[email protected]>
2557
2558 * gst/gstdebugutils.c:
2559 gstreamer: A few small memory cleanups
2560 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9044>
2561
2562 2025-05-23 13:07:34 +0200 Thibault Saunier <[email protected]>
2563
2564 * gst/gststructure.c:
2565 * gst/gststructure.h:
2566 gst: Add a gst_structure_is_writable method
2567 There are cases (in the gst-python bindings for example) where
2568 it is interesting to know that the structure is not writable
2569 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9027>
2570
2571 2025-05-22 14:17:34 +0300 Sebastian Dröge <[email protected]>
2572
2573 * gst/gstmeta.c:
2574 meta: Add g_return_val_if_fail() for NULL valid_tags in gst_meta_api_type_tags_contain_only()
2575 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9054>
2576
2577 2025-05-18 16:02:52 +0300 Jordan Petridis <[email protected]>
2578
2579 * gst/gsttracerutils.c:
2580 gsttracerutils: Fix leak in gst_tracer_utils_create_tracer()
2581 Co-authored-by: Alicia Boya García <[email protected]>
2582 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9028>
2583
2584 2025-04-09 01:51:49 -0400 Thibault Saunier <[email protected]>
2585
2586 * gst/gstelement.h:
2587 * gst/gstinfo.c:
2588 * gst/gstinfo.h:
2589 * tests/check/gst/gstinfo.c:
2590 gst: info: Add a GstLogContext API
2591 Add a new API to control logging behavior, particularly for implementing
2592 "log once" functionality and periodic logging. This helps avoid spamming
2593 logs with repetitive messages.
2594 The API provides:
2595 - Static and dynamic context creation
2596 - Configurable message identity calculation
2597 - Periodic reset capability
2598 - Context-aware logging macros
2599 - Element message variants with context support
2600 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6855>
2601
2602 2025-05-21 10:54:48 +0200 Guillaume Desmottes <[email protected]>
2603
2604 * gst/gstvalue.c:
2605 * tests/check/gst/gstvalue.c:
2606 core: gstvalue: fix ANY/EMPTY caps (features) hash
2607 They should be special cases as both do not have any actual caps/features.
2608 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9037>
2609
2610 2025-05-18 11:28:29 +0300 Sebastian Dröge <[email protected]>
2611
2612 * gst/gstmeta.c:
2613 * gst/gstmeta.h:
2614 meta: Add gst_meta_api_type_tags_contain_only()
2615 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/9004>
2616
2617 2025-04-04 04:41:24 +0900 Seungha Yang <[email protected]>
2618
2619 * libs/gst/base/gstbaseparse.c:
2620 baseparse: Add disable-clip property
2621 Adding a property to allow pushing buffers that are out of segment,
2622 and do not drop out of segment buffers by default
2623 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8773>
2624
2625 2025-05-13 09:54:07 +0300 Sebastian Dröge <[email protected]>
2626
2627 * gst/gstpipeline.c:
2628 pipeline: Store the actual latency even if no static latency was configured
2629 Previously the latency was only stored if a static latency was configured on the
2630 pipeline, which caused gst_pipeline_get_configured_latency() to always return
2631 GST_CLOCK_TIME_NONE in that case.
2632 Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4429
2633 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8972>
2634
2635 2025-05-16 13:11:33 +0300 Sebastian Dröge <[email protected]>
2636
2637 * gst/gstelement.c:
2638 element: ref-sink the correct pad template when replacing an existing one
2639 templ is the new one that is being stored and that needs to be ref-sinked,
2640 padtempl is the old one that just needs to be unreffed.
2641 Fixes leaking the old template, and also makes sure that the new template is not
2642 floating which can cause use-after-frees with bindings as they might wrongly
2643 take ownership of a still floating template.
2644 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8988>
2645
2646 2025-05-13 16:28:04 +0300 Jordan Petridis <[email protected]>
2647
2648 * tests/check/gstreamer.supp:
2649 core: suppress glib_init_ctor as well
2650 We already suppress gobject_init_ctor and this
2651 is the same.
2652 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8979>
2653
2654 2025-05-13 19:47:37 -0400 Doug Nazar <[email protected]>
2655
2656 * tests/check/gst/gstcontroller.c:
2657 controller: Free various props before being set
2658 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8978>
2659
2660 2025-05-13 19:46:34 -0400 Doug Nazar <[email protected]>
2661
2662 * libs/gst/controller/gstdirectcontrolbinding.c:
2663 directcontrolbinding: Free various props before being set
2664 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8978>
2665
2666 2025-05-13 19:15:21 -0400 Doug Nazar <[email protected]>
2667
2668 * gst/gstpadtemplate.c:
2669 all: Annotate *_set_property() contructor only props without free
2670 Properties that are marked constructor only aren't required to be freed
2671 before g_value_dup_*() as they can only be called once during construction.
2672 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8978>
2673
2674 2025-05-07 10:48:25 -0400 Xavier Claessens <[email protected]>
2675
2676 * gst/gstmessage.c:
2677 gstmessage: Debug error message is nullable
2678 When debug is NULL, gst-launch-1.0 won't print
2679 "Additional debug info:" line.
2680 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8943>
2681
2682 2025-04-10 11:22:29 +0530 Santosh Mahto <[email protected]>
2683
2684 * gst/gstbuffer.c:
2685 gstanalytics: Add transform function to copy the tensor meta
2686 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8825>
2687
2688 2025-03-15 20:56:17 +0100 Tim-Philipp Müller <[email protected]>
2689
2690 * meson.options:
2691 meson: rename meson_options.txt to meson.options
2692 Which is supported since Meson 1.1:
2693 https://mesonbuild.com/Release-notes-for-1-1-0.html#support-for-reading-options-from-mesonoptions
2694 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8651>
2695
2696 2025-04-01 18:22:18 +0530 Nirbheek Chauhan <[email protected]>
2697
2698 * libs/gst/helpers/ptp/meson.build:
2699 gst-ptp-helper: Fix meson warning about rust_crate_type
2700 WARNING: Project targets '>= 1.4' but uses feature deprecated since
2701 '1.3.0': rust_crate_type arg in static_library. Use rust_abi or
2702 rust.proc_macro() instead.
2703 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8753>
2704
2705 2025-04-25 12:18:19 +0300 Sebastian Dröge <[email protected]>
2706
2707 * libs/gst/base/gstaggregator.c:
2708 aggregator: Don't produce buffers when live but not in PLAYING yet
2709 Especially in force-live=true mode it was possible to produce buffers before the
2710 element was set to PLAYING as long as a clock was available already.
2711 This could easily lead to outputting buffers too early, and e.g. before the
2712 correct base time is set and available.
2713 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8897>
2714
2715 2025-04-25 12:15:17 +0300 Sebastian Dröge <[email protected]>
2716
2717 * libs/gst/base/gstaggregator.c:
2718 aggregator: Check after waiting if we're still running and otherwise stop
2719 Previously we might've produced a buffer needlessly.
2720 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8897>
2721
2722 2025-04-10 17:32:19 -0400 Thibault Saunier <[email protected]>
2723
2724 * gst/gstdebugutils.c:
2725 gst: debug: Add information about active tracers in dot files
2726 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8820>
2727
2728 2025-04-16 11:06:07 -0400 Thibault Saunier <[email protected]>
2729
2730 * gst/gstmacos.h:
2731 * gst/gstmacos.m:
2732 macos: Move macos function documentation to the .h so the introspection has the information
2733 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8854>
2734
2735 2025-04-09 12:10:33 -0400 Thibault Saunier <[email protected]>
2736
2737 * gst/gsttracerutils.c:
2738 tracerutils: Do not warn on empty string as tracername
2739 It doesn't matter if there is an "empty tracer" specified.
2740 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8860>
2741
2742 2025-04-01 11:02:58 +0800 Shengqi Yu (喻盛琪) <[email protected]>
2743
2744 * gst/gstpluginloader.c:
2745 pluginloader: fix pending_plugins Glist use-after-free issue
2746 When plugin_loader_load_and_sync returns false in plugin_loader_replay_pending,
2747 the cur Glist l->pending_plugins will be added to the blacklist.
2748 However, the l->pending_plugins might have already been loaded and freed in handle_rx_packet,
2749 so causing a use-after-free issue.
2750 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8747>
2751
2752 2025-03-31 15:39:14 +0200 Mathieu Duponchelle <[email protected]>
2753
2754 * libs/gst/base/gstaggregator.c:
2755 aggregator: expose current-level-* properties on sink pads
2756 As aggregator internally queues data (up to latency), those properties
2757 are helpful to monitor queue levels in the complete pipeline.
2758 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8731>
2759
2760 2025-04-03 09:13:31 -0300 L. E. Segovia <[email protected]>
2761
2762 * cmake/FindGStreamer.cmake:
2763 cmake: Fix using pkgconf for Windows cross builds
2764 pkgconf has its own default logic on Windows for setting up a prefix,
2765 which does not match the default behaviour of pkg-config (blindly
2766 respecting modules' ${prefix} variable).
2767 See
2768 https://github.com/pkgconf/pkgconf/commit/dcf529b83d621ed09e99e41fc35fdffd068bd87a
2769 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8768>
2770
2771 2025-04-03 09:02:59 -0300 L. E. Segovia <[email protected]>
2772
2773 * cmake/FindGStreamer.cmake:
2774 cmake: Fix PKG_CONFIG_PATH formatting for Windows cross-builds
2775 The PKG_CONFIG_PATH use of semicolons must match the host system, not
2776 the build system. This fixes calling pkg-config for Windows to Android
2777 cross builds.
2778 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8768>
2779
2780 2025-04-01 02:04:48 +0900 Seungha Yang <[email protected]>
2781
2782 * gst/gstpluginloader-win32.c:
2783 pluginloader-win32: Fix helper executable path under devenv
2784 lpApplicationName argument of CreateProcessW should be complete path
2785 of executable.
2786 Fixing regression introduced by
2787 https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8614
2788 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8746>
2789
2790 2025-02-13 16:04:29 +0100 Guillaume Desmottes <[email protected]>
2791
2792 * gst/gstvalue.c:
2793 * gst/gstvalue.h:
2794 * tests/check/gst/gstvalue.c:
2795 gstvalue: add hashing
2796 Will be used to implement the Hash trait in gstreamer-rs, see
2797 https://gitlab.freedesktop.org/gstreamer/gstreamer-rs/-/merge_requests/1639
2798 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8518>
2799
2800 2025-02-17 16:07:37 +0100 Guillaume Desmottes <[email protected]>
2801
2802 * gst/gst_private.h:
2803 * gst/gsttaglist.c:
2804 taglist: add _gst_tag_list_structure() as internal API
2805 Will be needed to implement hash on GValue.
2806 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8518>
2807
2808 2025-02-13 16:22:18 +0100 Guillaume Desmottes <[email protected]>
2809
2810 * gst/gstvalue.c:
2811 gstvalue: rename internal table to gst_value_hash_table
2812 Will use the gst_value_hash symbol to implement a new method.
2813 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8518>
2814
2815 2025-03-14 22:15:38 -0400 Doug Nazar <[email protected]>
2816
2817 * gst/gsttracer.c:
2818 tracer: Free various props before being set
2819 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8648>
2820
2821 2025-03-14 19:14:43 -0400 Doug Nazar <[email protected]>
2822
2823 * gst/gstcontrolbinding.c:
2824 * gst/gstdevice.c:
2825 * gst/gstpadtemplate.c:
2826 all: Annotate *_set_property() contructor only props without free
2827 Properties that are marked constructor only aren't required to be freed
2828 before g_value_dup_string() as they can only be called once during construction.
2829 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8648>
2830
2831 2025-03-26 15:38:20 +0100 Guillaume Desmottes <[email protected]>
2832
2833 * plugins/tracers/gstdots.c:
2834 tracers: dots: fix debug log
2835 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8699>
2836
2837 2025-03-06 17:16:20 +0100 David Smitmanis <[email protected]>
2838
2839 * gst/gstpluginloader-win32.c:
2840 pluginloader-win32: Correctly handle whitespace paths when executing gst-plugin-scanner
2841 On Windows, if the path to gst-plugin-scanner.exe contained
2842 whitespace, gstreamer would via CreateProcessW attempt to execute
2843 several files "up" the path tree; e.g. if the scanner path was
2844 "C:\Program Files\gstreamer app\gst-plugin-scanner.exe", it would try
2845 to execute C:\Program, C:\Program.exe, C:\Program Files\gstreamer.exe"
2846 and so on.
2847 This is how CreateProcessW behaves with unquoted whitespace arguments
2848 in lpCommandLine if lpApplicationName is NULL.
2849 By passing the binary path as lpApplicationName instead, the problem
2850 is avoided.
2851 Also quote arguments to gst-plugin-scanner.exe as they are paths as well.
2852 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8614>
2853
2854 2025-01-22 15:02:03 +0100 Marc Leeman <[email protected]>
2855
2856 * meson.build:
2857 meson.build: test for and link against libatomic if it exists
2858 It's needed on some platforms for some subset (or all) atomic operations and
2859 checking for the cases when it's actually needed is quite complex.
2860 Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/4300
2861 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8637>
2862
2863 2025-03-13 16:41:14 +0100 Branko Subasic <[email protected]>
2864
2865 * gst/gsttracerutils.h:
2866 tracer: Make it compile when tracer hooks are disabled
2867 The GST_TRACER_POOL_BUFFER_(DE)QUEUED macros used when tracer hooks are
2868 disabled took only one argument, causing compile errors when building
2869 with tracer hooks disabled.
2870 Change-Id: I0958c0b018f1fc4a6d84ab0bdd9e42895ae1fe77
2871 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8638>
2872
2873 2025-03-10 11:52:21 -0300 Thibault Saunier <[email protected]>
2874
2875 * gst/gstbufferpool.c:
2876 * gst/gsttracerutils.c:
2877 * gst/gsttracerutils.h:
2878 tracer: Add a hook to track when buffers are queued/dequeued in pools
2879 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8617>
2880
2881 2025-03-12 13:59:45 +0100 Tim-Philipp Müller <[email protected]>
2882
2883 * README.md:
2884 * RELEASE:
2885 * meson.build:
2886 Back to development in main branch after 1.26.0
2887 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8621>
2888
2889 === release 1.26.0 ===
2890