- Separate gstsignalprocessor into a separate library (not sure if this
is in the right place, but it works for now anyway)
- Create LV2 element based on LADSPA element, port most discovery
functionality
Use glib defines for searchpath separators. Filter searchpath to avoid scanning
path entries twice. Fix the return in ladspa_plugin_directory_search to return
wheter we found a plugin.
Move the two function from search.c into gstladspa.c and make them static.
Remove non needed arguments from function prototypes. Use glib api for
simplicity and portability.
Original commit message from CVS:
Based on Patch by: Brian Cameron <brian dot cameron at sun dot com>
* ext/ladspa/search.c: (LADSPAPluginSearch):
Avoid searching (null) paths or crashing on platforms where printing a
NULL string segfaults. Fixes#567004.
Original commit message from CVS:
* ext/ladspa/gstladspa.c:
Whitespace.
* ext/ladspa/gstsignalprocessor.c:
Add a FIXME:. not sure if this code does the forwarding correctly.
Original commit message from CVS:
* configure.ac:
* ext/ladspa/Makefile.am:
Link the ladspa plugin with -ldl. It's needed for dlopen() and friends
for loading the ladspa plugins and previously was linked in by
gmodule. Fixes bug #543848.
Original commit message from CVS:
* ext/ladspa/gstladspa.c: (gst_ladspa_base_init),
(gst_ladspa_class_init), (ladspa_describe_plugin), (plugin_init):
Don't use GST_BOILERPLATE as the stuff generated from it is not used
anyway and can't be used.
Store the class struct of the correct type in parent_class.
Pass the LADSPA_Descriptor as class_data to the class_init function
as preparation for the time, when we can add pad templates and friends
in class_init and add a FIXME for that.
Don't use a custom hash table for passing the LADSPA_Descriptors to
base_init but use g_type_set_qdata and g_type_get_qdata.
Original commit message from CVS:
* ext/ladspa/gstsignalprocessor.c:
(gst_signal_processor_class_add_pad_template):
Don't unref the pad template after adding it.
gst_element_class_add_pad_template takes ownership of it.
Original commit message from CVS:
* ext/ladspa/gstladspa.c: (gst_ladspa_get_property):
* ext/sdl/sdlvideosink.c: (gst_sdlvideosink_show_frame):
* gst/mve/gstmvemux.c: (gst_mve_mux_request_new_pad):
* sys/dvb/dvbbasebin.c: (dvb_base_bin_class_init):
Fix 'xyz may be used uninitialized' compiler warnings caused
by broken g_assert_not_reached() macro in GLib-2.15.x and don't
abort() in any case but properly report the error.
Original commit message from CVS:
* ext/ladspa/Makefile.am:
* ext/ladspa/gstladspa.c: (gst_ladspa_class_get_param_spec):
add GstController support to ladspa
Original commit message from CVS:
2007-01-12 Andy Wingo <wingo@pobox.com>
* ext/ladspa/gstsignalprocessor.c (gst_signal_processor_fixate)
(gst_signal_processor_ouija_caps, gst_signal_processor_prepare):
Remove fixate/ouija stuff, thankfully, due to the new
part-negotiation.txt pull-mode negotiation scheme.
(gst_signal_processor_setcaps_pull)
(gst_signal_processor_setcaps): Implement upstream set_caps pull
proxying for pull mode. Now this works: ladspa-sine-fcac !
audioconvert ! alsasink.
Original commit message from CVS:
2007-01-06 Andy Wingo <wingo@pobox.com>
* ext/ladspa/gstsignalprocessor.c
(gst_signal_processor_ouija_caps): Move around in the source
file...
(gst_signal_processor_prepare, gst_signal_processor_do_pulls):
Call ouija_caps in prepare() instead of do_pulls(), a bit earlier.
This allows us to have caps when we do the pad_alloc_buffer().
(gst_pad_alloc_buffer_and_set_caps): Use self->caps instead of the
pad caps, which might not be set yet.
Original commit message from CVS:
2007-01-06 Andy Wingo <wingo@pobox.com>
* ext/ladspa/gstsignalprocessor.c:
(gst_signal_processor_add_pad_from_template)
(gst_signal_processor_fixate): Add a fixate function, to assist in
pathological ladspa-sine-fcac ! fakesink can-activate-pull=true
cases.
(gst_signal_processor_prepare, gst_signal_processor_process): Add
nframes args so that getrange can tell ladspa how many frames to
process.
(gst_signal_processor_ouija_caps): setcaps needs to be called
before processing, which normally happens when chaining a buffer
to a pad. However in getrange mode with no sinks we need to check
explicitly for this condition, guess some caps to use, and use
those to setcaps(). Hence this mystical function.
(gst_signal_processor_do_pulls): Pull in bytes, not samples.
Divine the caps if necessary.
(gst_signal_processor_getrange): Interpret the length as bytes,
not samples.
(gst_signal_processor_chain): nframes=G_MAXUINT, will be limited
by incoming buffer sizes.
Original commit message from CVS:
* ext/ladspa/gstsignalprocessor.c: (gst_signal_processor_setcaps),
(gst_signal_processor_event):
Reset flow_state back to _OK after a flush stop so that we exit our
error state after the flush. Fixes#374213
Original commit message from CVS:
* ext/ladspa/gstladspa.c: (gst_ladspa_base_init):
Convert ' ' into '_'. Try to keep as many characters in the padtemplate
names as possible.
Original commit message from CVS:
* ext/ladspa/gstsignalprocessor.c: (gst_signal_processor_flush),
(gst_signal_processor_do_pushes):
A push() gives away our refcount so we should not use the buffer on the
pen anymore.
Original commit message from CVS:
2006-08-04 Andy Wingo <wingo@pobox.com>
* ext/ladspa/gstsignalprocessor.h: Add infrastructure for storing
whether a processor can work in place or not, and for keeping
track of its state. Change the FlowReturn instance variable from
"state" to "flow_state", all callers changed.
* ext/ladspa/gstsignalprocessor.c (gst_signal_processor_setup)
(gst_signal_processor_start, gst_signal_processor_stop)
(gst_signal_processor_cleanup): New functions to manage the
processor's state.
(gst_signal_processor_setcaps): start() as well as setup() here.
(gst_signal_processor_prepare): Respect CAN_PROCESS_IN_PLACE.
(gst_signal_processor_change_state): Stop and cleanup the
processor as we go to NULL.
* ext/ladspa/gstladspa.c (gst_ladspa_base_init): Reuse buffers if
INPLACE_BROKEN is not set.
* ext/ladspa/gstsignalprocessor.c (gst_signal_processor_prepare):
Do the alloc_buffer in bytes, not frames.
Original commit message from CVS:
2006-08-04 Andy Wingo <wingo@pobox.com>
* ext/ladspa/gstsignalprocessor.c (gst_signal_processor_setcaps)
(gst_signal_processor_prepare)
(gst_signal_processor_update_inputs)
(gst_signal_processor_process, gst_signal_processor_pen_buffer)
(gst_signal_processor_flush)
(gst_signal_processor_sink_activate_push)
(gst_signal_processor_src_activate_pull)
(gst_signal_processor_change_state): Remove the last of the code
that assumes that we process whole buffers at a time. Fix some
debugging. Seems to work now in some cases.
Original commit message from CVS:
2006-08-01 Andy Wingo <wingo@pobox.com>
* ext/ladspa/gstsignalprocessor.c (gst_signal_processor_process):
Fix nframes-choosing.
(gst_signal_processor_init): Init pending_in and pending_out.
Original commit message from CVS:
2006-08-01 Andy Wingo <wingo@pobox.com>
* ext/ladspa/gstsignalprocessor.c (gst_signal_processor_init): No
more default sample rate, although we never check that the sample
rate actually gets set. Something for the future.
(gst_signal_processor_setcaps): Some refcount fixes, flow fixes.
(gst_signal_processor_event): Refcount fixen.
(gst_signal_processor_process): Pull the number of frames to
process from the sizes of the buffers in the input pens.
(gst_signal_processor_pen_buffer): Remove an incorrect FIXME :)
(gst_signal_processor_do_pulls): Add an nframes argument, and use
it instead of buffer_frames.
(gst_signal_processor_getrange): Refcount fixen, pass nframes on
to do_pulls.
(gst_signal_processor_chain)
(gst_signal_processor_sink_activate_push)
(gst_signal_processor_src_activate_pull): Refcount fixen.
* ext/ladspa/gstsignalprocessor.h: No more buffer_frames, yay.
Original commit message from CVS:
* ext/ladspa/gstsignalprocessor.c: (gst_signal_processor_setcaps),
(gst_signal_processor_process):
don't query buffer-frames from caps, add lots of debug-log,
try fix for assert (#349189)
Original commit message from CVS:
* ext/ladspa/gstsignalprocessor.c: (gst_signal_processor_event),
(gst_signal_processor_process):
Fix compilation of LADPSA. It doesn't seem to work, and isn't
enabled for the build, but it helps me win the feature-count
competitions ooh yeah.
Original commit message from CVS:
2005-10-07 Andy Wingo <wingo@pobox.com>
* ext/raw1394/gstdv1394src.c: Make interruptible, so it won't
block forever in a read().
Original commit message from CVS:
2005-08-28 Andy Wingo <wingo@pobox.com>
* Updates for two-arg init from GST_BOILERPLATE.
* ext/ladspa/gstsignalprocessor.c (gst_signal_processor_init): Use
the second arg for the class, because G_OBJECT_GET_CLASS (self)
returns the wrong thing.
(gst_signal_processor_add_pad_from_template): Make pads of the
right type.
* ext/ladspa/gstladspa.c (gst_ladspa_class_get_param_spec): Make
writable param specs G_PARAM_CONSTRUCT so default values work.
(gst_ladspa_init): Use the second arg for the class.
Original commit message from CVS:
2005-08-26 Andy Wingo <wingo@pobox.com>
* ext/ladspa/gstladspa.c:
* ext/ladspa/gstladspa.h: Finish porting, still doesn't work but
it does compile and register. I have more features than you.
* ext/ladspa/gstsignalprocessor.h:
* ext/ladspa/gstsignalprocessor.c: Updates, bug fixen.
Original commit message from CVS:
2005-08-25 Andy Wingo <wingo@pobox.com>
* ext/ladspa/gstladspa.h:
* ext/ladspa/gstladspa.c: Halfway-ported. Doesn't compile yet.
* ext/ladspa/gstsignalprocessor.h:
* ext/ladspa/gstsignalprocessor.c: New files, the start of a base
class for DSP elements.
* configure.ac: Sort the external libs checks, add a ladspa check,
output the ladspa makefile.
Original commit message from CVS:
second batch :
remove ',' at end of enums as they could confuse older gcc, foreign compilers (forte) and gtk-doc
(in gst-plugins/ext/ this time)
Original commit message from CVS:
Remove all config.h includes from header files, add it to each source file and remove duplicate config.h includes from several source files
Original commit message from CVS:
Fix this too. Note that the per-plugin licensing is gone because of company's new per-plugin (rather than per-element) API. We basically need both... Ohwell, I just want this stuff to compile.
Original commit message from CVS:
merge TYPEFIND branch. Major changes:
- totally reworked type(find) system
- all typefind functions are in gst/typefind now
- more typefind functions then before
- some plugins might fail to compile now because I don't have them installed and they
a) require bytestream or
b) haven't had their typefind fixed.
Please fix those plugins and put the typefind functions into gst/typefind if they don't have dependencies
Original commit message from CVS:
New typefind system:
* bytestream is now part of the core
* all plugins have been modified to use this new typefind system
* asf typefinding added
* mpeg video stream typefiding removed because it's broken
* duplicate typefind entries removed
* extra id3 typefinding added, because we've seen 4 types of files
(riff/wav, flac, vorbis, mp3) with id3 headers and each of these needs
to work. Instead, I've added an id3 element and let it redo typefiding
after the id3 header. this needs a hack because spider only typefinds
once. We can remove this hack once spider supports multiple typefinds.
* with all this, mp3 typefinding is semi-rewritten
* id3 typefinding in flac/vorbis is removed, it's no longer needed
* fixed spider and gst-typefind to use this, too.
* Other general cleanups
Original commit message from CVS:
* actually recurse into sndfile if we are able
* big ladspa cleanups, mainly to comply with the buffer-frames caps property, but also general
cleanups
- the samplerate prop is gone, if you want to set it explicitly (as in for get-based plugins)
you need to use a filtered connection, just like with buffer-frames
* big float2int and int2float changes for buffer-frames compatibility - I think it's quite a bit
simpler
* make the ossclock general, add it to gstaudio, and use it in sndfile as well
i need to update mimetypes, but that's coming soon. there are some other plugins that don't
support buffer-frames, i guess i need to get around to fixing them as well.
Original commit message from CVS:
New mimetypes gone into effect today - this commit changes all old mimetypes over to the new mimetypes spec as described in the previous commit's document. Note: some plugins will break, some pipelines will break, expect HEAD to be broken or at least not 100% working for a few days, but don't forget to report bugs
Original commit message from CVS:
compatibility fix for new GST_DEBUG stuff.
Includes fixes for missing includes for config.h and unistd.h
I only ensured for plugins I can build that they work, so if some of them are still broken, you gotta fix them yourselves unfortunately.
Original commit message from CVS:
another batch of connect->link fixes
please let me know about issues
and please refrain of making them yourself, so that I don't spend double
the time resolving conflicts
Original commit message from CVS:
* fix to adder: don't pull on unusable pads
* fix to ladspa: get output buffers first, so we can know how many frames to
process
- this breaks inplace buffer stuff, hmm
* fix to float2int: add a bufferpool with the same nframes as the source pad's
pool -> no extra latency
