Commit graph

3399 commits

Author SHA1 Message Date
Thomas Vander Stichele
9b87fae395 gst/gst.c: make the registry cache file HOST_CPU-dependent
Original commit message from CVS:

* gst/gst.c: (init_post):
make the registry cache file HOST_CPU-dependent
2005-12-16 17:34:19 +00:00
Tim-Philipp Müller
8ea948df92 gst/gstminiobject.c: Fix lcopy for mini objects, the mini object needs to be ref'ed.
Original commit message from CVS:
* gst/gstminiobject.c: (gst_value_mini_object_lcopy):
Fix lcopy for mini objects, the mini object needs to be ref'ed.
* tests/check/gst/gstminiobject.c: (my_foo_init),
(my_foo_get_property), (my_foo_set_property), (my_foo_class_init),
(test_value_collection), (gst_mini_object_suite):
Add test to ensure refcounts end up as expected when passing
GstMiniObjects through g_object_get() and g_object_set().
2005-12-15 12:22:38 +00:00
Jan Schmidt
bcdfacaf5d gst/gstpad.c: Don't allow activation of a srcpad in pull_range if it has no getrange function.
Original commit message from CVS:
* gst/gstpad.c: (gst_pad_activate_pull), (gst_pad_link_prepare):
Don't allow activation of a srcpad in pull_range if it has no
getrange function.
Change some debug statements to be a little clearer

* plugins/elements/gsttypefindelement.c:
(gst_type_find_handle_src_query):
Check that we have a peer before executing queries thereupon.

* tests/examples/metadata/read-metadata.c: (message_loop):
Use gst_bus_pop instead of gst_bus_poll when we just want it to
immediately return us any available message with 0 timeout.
2005-12-14 10:09:35 +00:00
Michael Smith
83a815e648 gst/gsttypefindfactory.c: Don't unref factories after calling them.
Original commit message from CVS:
* gst/gsttypefindfactory.c: (gst_type_find_factory_call_function):
Don't unref factories after calling them.
* libs/gst/base/gsttypefindhelper.c: (gst_type_find_helper):
* plugins/elements/gsttypefindelement.c:
(gst_type_find_element_chain):
Free lists of factories after using them. Fixing typefinding memory
leaks.
2005-12-12 19:09:49 +00:00
Stefan Kost
86deff7b5b gst/gstpluginfeature.c: more meaningful debug output
Original commit message from CVS:
* gst/gstpluginfeature.c: (gst_plugin_feature_finalize),
(gst_plugin_feature_load):
more meaningful debug output
* configure.ac:
* tests/Makefile.am:
* tests/old/examples/Makefile.am:
make make distcheck happy again
2005-12-12 18:12:13 +00:00
Wim Taymans
8666c1597d gst/gstsegment.c: Added more documentation.
Original commit message from CVS:
* gst/gstsegment.c: (gst_segment_init),
(gst_segment_set_last_stop), (gst_segment_set_seek),
(gst_segment_set_newsegment), (gst_segment_to_stream_time),
(gst_segment_to_running_time):
Added more documentation.
Make sure the last_pos value is updated properly.
Make sure to_stream_time and to_running_time don't
operate on wrong values.

* tests/check/gst/gstsegment.c: (GST_START_TEST):
Update check.
2005-12-12 12:48:35 +00:00
Thomas Vander Stichele
ef7dedbfc8 gst/gstvalue.c: floor is not needed, F is always positive; this obviates the need for adding -lm when building withou...
Original commit message from CVS:

* gst/gstvalue.c: (gst_value_transform_double_fraction):
floor is not needed, F is always positive; this obviates the
need for adding -lm when building without libxml
2005-12-12 10:59:12 +00:00
Wim Taymans
d16ce65df7 gst/gstbuffer.c: Small docs update.
Original commit message from CVS:
* gst/gstbuffer.c:
Small docs update.

* gst/gstcaps.c: (gst_caps_is_equal):
Don't assert on NULL <--> X. Fixes #323260

* gst/gstminiobject.c: (gst_mini_object_replace):
If we're doing atomic operations, we might just as well use
the proper way to get an atomic pointer.

* libs/gst/base/gstbasesink.c: (gst_base_sink_get_position):
Clean up debugging.
2005-12-07 15:16:43 +00:00
Michael Smith
4e8cd3c843 gst/parse/grammar.y: Remove handling of { } for threads.
Original commit message from CVS:
* gst/parse/grammar.y:
Remove handling of { } for threads.
2005-12-07 11:52:05 +00:00
Thomas Vander Stichele
288f61d21f expand tabs
Original commit message from CVS:
expand tabs
2005-12-06 19:29:15 +00:00
Thomas Vander Stichele
ccd5fa2866 remove deprecation guard from docstring
Original commit message from CVS:
remove deprecation guard from docstring
2005-12-05 14:14:02 +00:00
Thomas Vander Stichele
4477b7090b fix docs
Original commit message from CVS:
fix docs
2005-12-05 13:06:37 +00:00
Wim Taymans
4f28ca8331 libs/gst/base/gstbasetransform.c (gst_base_transform_prepare_output_buf)
Original commit message from CVS:
2005-12-05  Andy Wingo  <wingo@pobox.com>

patch by: Wim Taymans <wim@fluendo.com>

* libs/gst/base/gstbasetransform.c
(gst_base_transform_prepare_output_buf)
(gst_base_transform_buffer_alloc):
* plugins/elements/gstqueue.c (gst_queue_bufferalloc): Call
alloc_buffer_and_set_caps.

* gst/gstpad.c (gst_pad_alloc_buffer): Changed to not call
set_caps on the source pad.
(gst_pad_alloc_buffer_and_set_caps): New function, does what
alloc_buffer used to do. Fixes #322874.

* docs/gst/gstreamer-sections.txt:
* docs/design/part-negotiation.txt:
* docs/pwg/advanced-negotiation.xml: Update for the alloc_buffer
changes.
2005-12-05 13:01:35 +00:00
Wim Taymans
0aa5cea288 Back out previous code changes, leave doc updates, file bugs instead.
Original commit message from CVS:
* gst/gstcaps.c: (gst_caps_is_equal):
* plugins/elements/gstfakesrc.c: (gst_fake_src_class_init),
(gst_fake_src_create):
Back out previous code changes, leave doc updates, file bugs
instead.
2005-12-05 10:59:17 +00:00
Wim Taymans
39e0ac7465 gst/gstcaps.c: Comparing NULL against !NULL yields different caps, not a failure.
Original commit message from CVS:
* gst/gstcaps.c: (gst_caps_is_equal):
Comparing NULL against !NULL yields different caps, not a
failure.
2005-12-05 10:22:55 +00:00
Wim Taymans
9d8037eed8 gst/gstpipeline.c: Fix small typo in docs.
Original commit message from CVS:
* gst/gstpipeline.c:
Fix small typo in docs.
2005-12-05 10:18:27 +00:00
Thomas Vander Stichele
9def80ce05 gst/gst.c (init_post): remove hard-coded 0.9 location for registries/plugins with a MAJORMINOR one.
Original commit message from CVS:
2005-12-05  Andy Wingo  <wingo@pobox.com>

patch by: Thomas Vander Stichele  <thomas at apestaart dot org>

* gst/gst.c (init_post): remove hard-coded 0.9 location for
registries/plugins with a MAJORMINOR one.
(plugin_desc): Rename library from gstcoreleements to
staticelements. Fixes #323222.
2005-12-05 09:53:54 +00:00
Thomas Vander Stichele
057de1f845 add tests that seem to show that the guint64/gdouble conversions are correct.
Original commit message from CVS:
* gst/gstutils.c: (gst_util_guint64_to_gdouble),
(gst_util_gdouble_to_guint64), (gst_util_uint64_scale_int64):
* gst/gstutils.h:
* tests/check/gst/gstutils.c: (GST_START_TEST), (gst_utils_suite):
add tests that seem to show that the guint64/gdouble conversions
are correct.
2005-12-04 09:57:11 +00:00
Wim Taymans
381742562f gst/: Fix docs again.
Original commit message from CVS:
* gst/gstregistry.c: (gst_registry_add_path):
* gst/gstregistry.h:
* gst/gstregistryxml.c:
Fix docs again.
Removed old code from gstregistryxml.c
2005-12-02 12:34:47 +00:00
Wim Taymans
db2b4d916c gst/gstutils.c: Small cleanup.
Original commit message from CVS:
* gst/gstutils.c: (gst_util_uint64_scale_int64),
(gst_util_uint64_scale_int):
Small cleanup.

* libs/gst/base/gstbasesink.c: (gst_base_sink_handle_object):
Add debug log line.

* libs/gst/base/gstbasetransform.c: (gst_base_transform_event):
Add FIXME.
2005-12-02 11:36:50 +00:00
Michael Smith
b32059fdc6 gst/gst.c: Escape a % to make gtkdoc happier; bug 322958.
Original commit message from CVS:
* gst/gst.c:
Escape a % to make gtkdoc happier; bug 322958.
2005-12-01 23:57:07 +00:00
Andy Wingo
a5820445bf gst/gstevent.h (struct _GstEvent): Meant to remove the extra padding here before, but it missed the commit.
Original commit message from CVS:
2005-12-01  Andy Wingo  <wingo@pobox.com>

* gst/gstevent.h (struct _GstEvent): Meant to remove the extra
padding here before, but it missed the commit.
2005-12-01 12:00:56 +00:00
Thomas Vander Stichele
9c28102415 fix the docs build
Original commit message from CVS:
fix the docs build
2005-11-30 19:36:54 +00:00
Thomas Vander Stichele
870761131b various fixes to make
Original commit message from CVS:

* configure.ac:
* gst/Makefile.am:
* gst/gst.c:
* gst/gstplugin.h:
* gst/gstregistry.h:
* tests/benchmarks/complexity.c:
* tests/benchmarks/mass-elements.c:
* tests/check/Makefile.am:
* tools/Makefile.am:
* tools/gst-inspect.c:
* tools/gst-xmlinspect.c:
various fixes to make
--disable-nls --disable-registry --disable-loadsave           --disable-parse --disable-gst-debug
work and get the core .so down to 360444 bytes after stripping
2005-11-30 19:01:53 +00:00
Thomas Vander Stichele
9055674a03 add some explicit casts update dsp files; also installs the debug build in \gstreamer\debug to separate it from the n...
Original commit message from CVS:
add some explicit casts
update dsp files; also installs the debug build in \gstreamer\debug to separate it from the non-debug build
2005-11-30 10:13:54 +00:00
Thomas Vander Stichele
6802a8b4ba docs/gst/gstreamer-sections.txt: remove double entries from the docs
Original commit message from CVS:

* docs/gst/gstreamer-sections.txt:
remove double entries from the docs
* gst/gst_private.h:
* gst/gstinfo.c: (_gst_debug_init):
remove the THREAD debug category
* gst/Makefile.am:
* gst/gstqueue.c:
* gst/gstqueue.h:
* docs/gst/gstreamer.types:
* plugins/elements/gstqueue.c: (gst_queue_get_type),
(gst_queue_init), (gst_queue_finalize), (gst_queue_change_state):
completely move queue and fix up debugging categories
2005-11-30 09:59:04 +00:00
Thomas Vander Stichele
aa56b376a7 gst/gstclock.c: remove crack non-portable handrolled DEBUG macro
Original commit message from CVS:
2005-11-30  Thomas Vander Stichele  <thomas at apestaart dot org>

* gst/gstclock.c: (do_linear_regression):
remove crack non-portable handrolled DEBUG macro
2005-11-30 09:18:15 +00:00
Edward Hervey
0db5df88fc gst/: Fix memory leak in GstTypeFindFactory.
Original commit message from CVS:
* gst/gsttypefind.c: (gst_type_find_register):
* gst/gsttypefind.h:
* gst/gsttypefindfactory.c: (gst_type_find_factory_init),
(gst_type_find_factory_dispose):
* gst/gsttypefindfactory.h:
Fix memory leak in GstTypeFindFactory.
2005-11-29 23:56:20 +00:00
Thomas Vander Stichele
2011345b96 move queue from core to the elements plugin ç
Original commit message from CVS:

* gst/gst.c:
* plugins/elements/Makefile.am:
* plugins/elements/gstelements.c:
* plugins/elements/gstqueue.c:
move queue from core to the elements plugin

ç
2005-11-29 19:47:00 +00:00
Andy Wingo
a08a8d8fea libs/gst/base/: en-LARGE the padding.
Original commit message from CVS:
2005-11-29  Andy Wingo  <wingo@pobox.com>

* libs/gst/base/gstbasetransform.h:
* libs/gst/base/gstbasesrc.h:
* libs/gst/base/gstbasesink.h: en-LARGE the padding.

* gst/gstconfig.h.in (GST_PADDING_LARGE): New define, the number
of pointers by which to pad very extensible base classes (like the
ones in libs/gst/base).
2005-11-29 19:37:49 +00:00
Thomas Vander Stichele
a0c4c4812c CVS surgery + support to move base, check, and net out of gst and into libs/gst
Original commit message from CVS:
CVS surgery + support to move base, check, and net out of gst
and into libs/gst
2005-11-29 19:12:30 +00:00
Andy Wingo
65ce5ed080 gst/gstevent.h (struct _GstEvent): Only one pointer of padding.
Original commit message from CVS:
2005-11-29  Andy Wingo  <wingo@pobox.com>

* gst/gstevent.h (struct _GstEvent): Only one pointer of padding.

* gst/gststructure.h (struct _GstStructure): Only one pointer of
padding.

* gst/gstquery.h (struct _GstQuery): Only one pointer of padding.

* gst/gstpluginfeature.h: Remove a comment in PluginFeature.

* gst/gstplugin.h (struct _GstPluginClass): Add some padding.

* gst/gstobject.h: (struct _GstObject): Only one pointer of
padding; reduces object size by about 30%. We don't expect
anything else to go into gstobject.

* gst/gstminiobject.h (struct _GstMiniObject)
(struct _GstMiniObjectClass): Only one pointer of padding; the
payload is only a pointer and two ints anyway. For the class there
are only two methods as well.

* gst/gstelement.h (struct _GstElementClass): Removed
the state_changed signal callback, it is not used.
2005-11-29 18:57:59 +00:00
Thomas Vander Stichele
9389a2f8b3 check/Makefile.am: look in the right place for elements, a lot more chance of success
Original commit message from CVS:

* check/Makefile.am:
look in the right place for elements, a lot more chance of
success
* gst/Makefile.am:
remove indexers and elements subdirs
* plugins/Makefile.am:
make indexers conditional
2005-11-29 18:14:35 +00:00
Thomas Vander Stichele
8c96632081 remove moved dirs
Original commit message from CVS:
remove moved dirs
2005-11-29 18:03:52 +00:00
Thomas Vander Stichele
233e55299e remove empty dirs
Original commit message from CVS:
remove empty dirs
2005-11-29 18:02:45 +00:00
Andy Wingo
41f439b0a1 Rename gstnet-tempname to gstnet. Fixes #322257.
Original commit message from CVS:
2005-11-29  Andy Wingo  <wingo@pobox.com>

* check/Makefile.am:
* pkgconfig/gstreamer-net-uninstalled.pc.in:
* pkgconfig/gstreamer-net.pc.in:
* gst/net/Makefile.am: Rename gstnet-tempname to gstnet. Fixes
#322257.
2005-11-29 17:47:06 +00:00
Thomas Vander Stichele
3ea45adb90 gst/gstevent.*: fix #320529 - clean up new_segment API and structure.
Original commit message from CVS:

* gst/gstevent.c: (gst_event_new_new_segment),
(gst_event_parse_new_segment):
* gst/gstevent.h:
fix #320529 - clean up new_segment API and structure.
Let's hope everyone was using the methods, and not the structure.
2005-11-29 17:23:56 +00:00
Edward Hervey
c4d1ffa5c7 gst/base/gstbasesink.c: Properly handle non GST_FORMAT_TIME segment
Original commit message from CVS:
* gst/base/gstbasesink.c: (gst_base_sink_handle_object),
(gst_base_sink_event), (gst_base_sink_do_sync),
(gst_base_sink_activate_pull), (gst_base_sink_change_state):
Properly handle non GST_FORMAT_TIME segment
* gst/elements/gstidentity.c: (gst_identity_transform_ip):
Properly handle non GST_FORMAT_TIME segment
* gst/gstsegment.c:
This function is valid if the accumulator is 0 and the format
is different from the requested format.
2005-11-29 17:13:44 +00:00
Jan Schmidt
aff648ab66 gst/base/gstbasetransform.c: Treat a pad alloc with new caps the same as if we were not negotiated, in order to allow...
Original commit message from CVS:
* gst/base/gstbasetransform.c: (gst_base_transform_buffer_alloc):
Treat a pad alloc with new caps the same as if we were not
negotiated, in order to allow a changing upstream output
to produce a new format of data.
2005-11-29 15:12:22 +00:00
Edward Hervey
33a15d5e67 gst/base/gstbasetransform.c: The event virtual method is now properly implemented, with a default handler
Original commit message from CVS:
* gst/base/gstbasetransform.c: (gst_base_transform_class_init),
(gst_base_transform_event), (gst_base_transform_eventfunc):
The event virtual method is now properly implemented, with a default
handler
Sub classes should call the parent_class event method. They should
return FALSE if they had a problem handling the given event, or don't
want GstBaseTransform to send that even downstream
* gst/elements/gstidentity.c: (gst_identity_class_init),
(gst_identity_init), (gst_identity_event),
(gst_identity_transform_ip), (gst_identity_set_property),
(gst_identity_get_property):
* gst/elements/gstidentity.h:
Added the single-segment boolean property.
If set to TRUE, it will output a single segment of data, starting from
0, will eat up all incoming newsegment, and modify the timestamp of the
buffers accordingly
2005-11-29 14:47:07 +00:00
Tim-Philipp Müller
696cf53af5 gst/gstghostpad.c: Don't ref NULL target pad (#322751). Improve docs.
Original commit message from CVS:
* gst/gstghostpad.c: (gst_proxy_pad_get_target):
Don't ref NULL target pad (#322751). Improve docs.
2005-11-29 13:10:38 +00:00
Michael Smith
9bc0e85dce gst/gstregistryxml.c: Don't crash if we failed to load a feature from a plugin.
Original commit message from CVS:
* gst/gstregistryxml.c: (load_plugin):
Don't crash if we failed to load a feature from a plugin.
2005-11-29 11:07:54 +00:00
Thomas Vander Stichele
eaf17081ca Makefile.am: don't run checks if we don't have check
Original commit message from CVS:
* Makefile.am:
don't run checks if we don't have check
* common/check.mak:
remove the registry when running make torture
* docs/gst/gstreamer-sections.txt:
remove second multiply
* gst/gstqueue.c: (gst_queue_loop):
fix a compile warning when disabling debug
2005-11-28 19:58:56 +00:00
Jan Schmidt
9f5ea35423 gst/gstinfo.h: Hey! Let's print the pad name if the pointer != NULL instead of when it == NULL :-)
Original commit message from CVS:
* gst/gstinfo.h:
Hey! Let's print the pad name if the pointer != NULL instead
of when it == NULL :-)
2005-11-28 19:43:27 +00:00
Wim Taymans
2dc0ef57fd check/gst/gstutils.c: Updated check, add some scaling accuracy checking code.
Original commit message from CVS:
* check/gst/gstutils.c: (GST_START_TEST):
Updated check, add some scaling accuracy checking code.

* gst/gstutils.c: (gst_util_div128_64),
(gst_util_uint64_scale_int64), (gst_util_uint64_scale),
(gst_util_uint64_scale_int):
Fix 6 times faster division code. Optimize for common
1/1 and less common X/1 cases.
2005-11-28 18:44:11 +00:00
Thomas Vander Stichele
597335c574 doc updates
Original commit message from CVS:
doc updates
2005-11-28 17:59:25 +00:00
Wim Taymans
870d272630 check/gst/gstutils.c: More checks.
Original commit message from CVS:
* check/gst/gstutils.c: (GST_START_TEST), (gst_utils_suite):
More checks.

* gst/gstclock.c: (gst_clock_finalize), (gst_clock_set_master),
(do_linear_regression), (gst_clock_add_observation):
Cleanups.
Release lock when the clock cannot be slaved.
Catch the case where the regression returned an invalid denominator.

* gst/gstutils.c: (gst_util_div128_64_iterate),
(gst_util_div128_64), (gst_util_uint64_scale_int64),
(gst_util_uint64_scale), (gst_util_uint64_scale_int):
Add protentially more performant non-iterative 128/64 divide function
that unfortunatly does not work yet.
Shortcut the trivial 0/X = 0 case.
Remove the warnings on overflow.
2005-11-28 16:05:35 +00:00
Thomas Vander Stichele
faa56ab3a8 gst/gstplugin.c: everything causing a plugin not to load should be at least a WARNING
Original commit message from CVS:
* gst/gstplugin.c: (gst_plugin_register_func):
everything causing a plugin not to load should be at least a WARNING
2005-11-28 14:18:22 +00:00
Thomas Vander Stichele
275498ed35 log caps
Original commit message from CVS:
log caps
2005-11-28 14:02:54 +00:00
Thomas Vander Stichele
b3ac5a0f4e fix docs
Original commit message from CVS:
fix docs
2005-11-28 14:01:52 +00:00