Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* gst/gsttaginterface.c:
Some documentation fixes (#307394 and #307397).
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* gst/gstvalue.c: (gst_value_intersect_list):
Fix memleak (#309125).
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* docs/manual/advanced-autoplugging.xml:
* gst/gstelementfactory.c:
Documentation patch indicating that factory information may
be overridden after a plugin is loaded (#308766).
Original commit message from CVS:
* gst/gstcaps.c: (gst_caps_from_string_inplace):
* gst/gststructure.c:
* gst/gstvalue.c: (gst_value_deserialize_int_helper):
fix deserialization to have similar semantics to cast's,
and check if the value is fittable into the target size,
instead of doing a G_MIN/G_MAX comparison
* testsuite/caps/Makefile.am:
* testsuite/caps/caps.h:
add caps that specify rgb bit masks using 0xFF...
Original commit message from CVS:
* gst/gstvalue.c:
fix for #308599 - fixes deserializing gint64 types
* gst/parse/grammar.y:
add a log line to show what type we deserialize to
Original commit message from CVS:
* gst/gstformat.c:
* gst/gstquery.c:
Fix two typos in the inline documentation.
* gst/gsttag.c: (gst_tag_register):
Don't spew out an assertion warning if a tag is
already registered, unless the new type differs
from the old one (fixes#308438).
Original commit message from CVS:
* gst/registries/gstlibxmlregistry.c (gst_xml_registry_get_property):
Fix memory leak. g_value_set_string already does g_strdup.
Original commit message from CVS:
* gst/gsturi.c: (gst_element_make_from_uri):
The show must go on, even if the element did not accept
the uri. Fixes endless loop as witnessed in #307804.
* tools/gst-launch.1.in:
Update mp3->ogg transcoding pipeline in examples section
to use rawvorbisenc and oggmux.
Original commit message from CVS:
add more debug statements for element linking
better docs for unlink()
make category column as wide as our widest known (internal) category
fix typo in comment
Original commit message from CVS:
Default values were current values in property output.
Output ranges + defaults also for unreadable values.
Fix output for float and double.
Original commit message from CVS:
* gst/gstevent.c: (gst_event_new):
Set invalid timestamp by default instead of zero on events.
* gst/gstqueue.c: (gst_queue_chain), (gst_queue_get):
Clear interrupt state after handling.
Original commit message from CVS:
* gst/gstvalue.c: (gst_value_list_append_value):
Limit check for whether a list contains a value already to
a normal list only (since this is allowed for a fixed-list).
Should fix testsuite.
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* gst/gstatomic_impl.h:
Add IA64 atomic operations (#305653).
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* gst/elements/gstfilesrc.c: (gst_filesrc_map_region):
Formatting fix (long/longlong), see #305653.
Original commit message from CVS:
* gst/gstvalue.c: (gst_value_list_append_value):
print a g_warning when someone ties to add the same value (as per
gst_value_compare) to a list twice.
Original commit message from CVS:
* configure.ac: Set G_LOG_DOMAIN in config.h to a string literal.
We should have done this all along. Fixes#301646
* gst/Makefile.am:
* gst/gst.c: (init_post):
* gst/gst_private.h:
* gst/parse/grammar.y:
Original commit message from CVS:
Patch by: jonathan at kaolin.hn.org
* gst/gstplugin.c: (gst_plugin_check_module),
(gst_plugin_check_file), (gst_plugin_load_file):
Fixes plugin registering on Mac OSX. Fixes#172526
Original commit message from CVS:
* gst/gstregistry.h:
win32/mingw obviously did not work
* gst/gsttag.h:
remove some whitespace
* libs/gst/dataprotocol/dataprotocol.c: (gst_dp_dump_byte_array):
put back a fix that accidentally got reverted in a backport
Original commit message from CVS:
* gst/gsttrace.c: (read_tsc), (gst_trace_read_tsc):
* gst/gsttrace.h:
tsc's are guint64's.
* tests/seeking/seeking1.c: (main):
so use them that way.
Original commit message from CVS:
2005-05-02 Zaheer Abbas Merali <zaheerabbas at merali dot org>
* testsuite/caps/deserialize.c: (main):
* testsuite/caps/sets.c: (check_caps):
* testsuite/caps/simplify.c: (check_caps):
fix small things to prevent warnings on 64bit
Original commit message from CVS:
* gst/elements/gstfakesink.c: (gst_fakesink_change_state):
don't error when we are in error already
the correct fix would be to not allow erroring in downwards
state changes, but I guess the API police would hunt me down if I
did that
Original commit message from CVS:
* gst/gsttrashstack.h:
Ooohh. a nasty one! After having a failed pop() from the stack,
it's possible that the stack is empty. In that case, don't
follow the NULL pointer.
Original commit message from CVS:
* gst/gstelementfactory.c: (gst_element_factory_create):
* gst/gstobject.c: (gst_object_class_init):
make the name property a construct property, so objects always have
a proper name.
* gst/gstobject.c: (gst_object_set_name_default),
(gst_object_set_name):
fix potential doublefree when resetting a name to its default value
Original commit message from CVS:
* gst/gst.c: (gst_init_check_with_popt_table), (init_post),
(init_popt_callback):
declare gst initialized before eventually rebuilding the registry.
Fixes issues with plugins calling gst_init
* gst/gstelementfactory.c: (gst_element_factory_cleanup),
(gst_element_register), (gst_element_factory_unload_thyself):
warn and fail properly when element details aren't initialized. Also
unref the class when we don't need it anymore instead of never/only
on failure
Original commit message from CVS:
Reviewed by: Ronald S. Bultje <rbultje@ronald.bitfreak.net>
* gst/gstelement.c: (gst_element_unlink):
Allow unlinking ghostpads using gst_element_unlink(), which
fixes#301331.
