13
0
Commit Graph

5519 Commits

Author SHA1 Message Date
8a005319b8 clean up aspects of ControlProtocol API 2013-12-21 15:31:28 -05:00
0cf8b9d3ea new ControlProtocolManager API, and proper handling of view/model changes in the RC options (prefs) editor 2013-12-21 15:31:28 -05:00
Hans Baier
611d1ffc29 midi_clock_slave: fix bandwidth to 2/60 for best results 2013-12-19 05:42:52 +07:00
Hans Baier
6e5240c24b interpolation_test: add some necessary assertions 2013-12-19 05:31:39 +07:00
Hans Baier
caac41a9cb midi_clock_slave: fix wrong calculation of loop error
since transport_frame refers to cycle start, the should be position has
to account for position of the midi clock event.
2013-12-18 09:55:15 +07:00
Hans Baier
bc5d23d5fa midi_clock_slave.cc: Better debug messages 2013-12-18 08:26:43 +07:00
Hans Baier
d826de83ae midi clock slave: make bandwidth variable according to period size 2013-12-18 06:19:21 +07:00
f6c9fa4d89 fix a crash caused by not checking if MackieControlProtocol::_surfaces_state exists
I forgot to add this check when i changed the member variable to a pointer.
2013-12-14 11:10:54 -05:00
e82ccd7648 rsynth.lv2: fix note off+on follow up:
special case if the same note is turned off and on in the
same [internal] synth cycle (64samples). x-fade to ADSR attack.
2013-12-14 04:29:12 +01:00
afabc09fbf change Generic MIDI control so that it handles a SendFeedback request synchronously within a process callback 2013-12-13 16:56:34 -05:00
89adeb2a9f make MIDI UI in libardour handle CallSlot requests like it used to (and like it should) 2013-12-13 16:55:57 -05:00
85cd647226 Mackie Control support now saves & restores port connections for devices not using ipMIDI (not required for ipMIDI anyway) 2013-12-13 16:04:34 -05:00
a17bafa964 remove some unused code from mackie support 2013-12-13 16:04:34 -05:00
6c84362959 fix reasonable synth octave 2013-12-13 20:59:08 +01:00
c43a1f39a6 xmas-egg 2013-12-13 20:28:30 +01:00
fd1eb73ef2 adjust LV2 ringbuffer size according to LV2:resize-port
The message-size itself is part of the message which
stored in the ringbuffer. If the rinbuffer overflows
the message is misinterpreted -> segfault.

Choose a more conservative ring-buffer size and take
the requested LV2 size into account.
2013-12-12 14:42:02 +01:00
Nils Philippsen
527b0a78a1 fix const-ness of names array 2013-12-10 23:07:48 +01:00
Nils Philippsen
96947e2f3a add cmp_nocase_utf8()
This is like cmp_nocase(), only that it doesn't use toupper(), tolower()
and therefore is agnostic of the current locale, and attempts to compare
strings in a UTF8-aware way (or falls back to ASCII if one of the
strings isn't UTF8-encoded).
2013-12-10 23:07:47 +01:00
Nils Philippsen
71f6104340 use correct argument order with memset() 2013-12-10 23:07:47 +01:00
Nils Philippsen
27a3f2837a use std::fill_n to fill gain buffer with samples
Using memset fills the buffer with whatever 1.0 as a double or float has
in its LSB.
2013-12-10 23:07:47 +01:00
Nils Philippsen
a86b66181a ignore "unofficial" states in jack_sync_callback() 2013-12-09 17:56:06 +01:00
bbd79b6792 print a warning for every VST >=2.4 plugin
17:29 < rgareus> las: ok. I'll make this translatable, but this warning will show up every time ardour starts (and scans for plugins) which can be annoying.
17:30 < las> rgareus: yeah, well we need to do black/whitelisting of plugins anyway so ...
2013-12-07 18:45:40 +01:00
0c2494595d detect VST >= 2.4 plugins. 2013-12-07 18:04:31 +01:00
ce7bfb489c change "cpu" in cpu load backend functions 2013-12-04 11:48:31 -05:00
ae6b0b9f1a minor tweaks to MIDI clock support, including delivery of position events/messages when stopped 2013-12-02 23:34:12 -05:00
76ea055b85 don't run MIDI clock "tick" if session is silent (e.g. due to a locate) 2013-12-02 23:33:33 -05:00
c19f9f8a47 fix continuous send of MIDI start/position/stop messages introduced by mistake 2013-12-01 12:18:57 -05:00
d1cc7e5a50 fix up a bunch of confusion regarding the size/capacity/allocation of audio & midi buffers 2013-11-29 22:26:33 -05:00
0c4457fa83 drop capacity argument from Buffer constructor, since the abstract class cannot (and should not) do anything with it 2013-11-29 22:25:36 -05:00
146734526c fix indentation 2013-11-29 22:24:59 -05:00
1190f1d0e0 make JACK backend return the right set of values when asked for buffer sizes or sample rate of a device other than the one in use 2013-11-27 17:41:54 -05:00
611d93830a add 0.5 second sleep after closing JACK connection so that next startup/connect is likely to work (allowing JACK time to shutdown). Ugh 2013-11-27 15:25:10 -05:00
56716a1f28 add error check to call to PortManager::reestablish_ports 2013-11-27 15:24:24 -05:00
dc4a984cd6 move AudioBackend::_start() into protected, and expand a bit on documenting ::start() vs ::_start() 2013-11-27 14:27:27 -05:00
4d0caa6ddd add yet more missing parts of previous commit re: libmidi++ and JACK (files not saved in emacs, sigh) 2013-11-27 14:08:22 -05:00
4a71e6f28f add missing parts of previous commit re: libmidi++ and JACK (files not saved in emacs, sigh) 2013-11-27 13:53:58 -05:00
ac4d691fe9 remove unintended linkage between libmidipp and jack 2013-11-27 12:44:23 -05:00
Sakari Bergen
52309c0c4f Fix invalid assertions in AudioGrapher SampleFormatConverter
This fixes an export crash with e.g. 8-bit export
2013-11-23 18:43:54 +02:00
5cbd4ce759 fix crash @ session-restore of plugins windows
The question is: why is _owner not set [yet], when the
window is restored. This rather smells like a race condition.
2013-11-22 18:22:55 +01:00
2eb79acefd lv2: increase scratch buffer size to request port minimum size 2013-11-17 00:33:32 +01:00
982c38dda1 allow to include In/Disk button on Meterbridge 2013-11-09 13:16:08 +01:00
3ce39cc410 update .po files with --no-fuzzy-matching in use 2013-11-05 11:15:46 -05:00
0f33baf32f back-port two actual bug fixes from cairocanvas branch 2013-11-05 10:17:40 -05:00
7aa46da32f fix incorrect use of sizeof() in snprintf (function never actually used in ardour,but ... ) 2013-11-05 10:17:22 -05:00
Julien de Kozak
6992153ae9 fix tests compilation 2013-11-05 07:51:04 +01:00
Sakari Bergen
3fca030603 Fix invalid silence trimmer end-of-input behavior (multiple EndOfInputs)
This caused an export bug when:
a) normalizing
b) adding silence to end
c) having more than one channel

The sound was corrupted by repeating each jack frame as many times
as there were channels.
2013-11-03 21:22:00 +02:00
7fec991077 fix thinko in declaration of ARDOUR::PortEngine::get_port_by_name() 2013-10-28 09:46:59 -04:00
Nils Philippsen
5f00d2f3a7 allow linking unbundled versions of some libraries
(libltc, rubberband, taglib, vamp-sdk)
2013-10-28 09:06:09 +01:00
17dc72d08e silence warning for ancient libc users&devs :)
string.h 'index' is a global declaration
marked legacy in 2001 and removed in posix.1-2008
2013-10-27 17:03:49 +01:00
2cad99bf39 add notes about inlined lv2 atom/util.h functions 2013-10-27 09:56:07 +01:00