74bc0c8468
substantive changes to the logic and safety for naming of (audio/MIDI) sources, especially when created via import
2014-06-02 11:20:37 -04:00
08a1409b1f
back port 6576105 from cairocanvas as fix for data loss/file deletion
2014-05-29 16:28:57 -04:00
Devin J. Pohly
cd0ac04080
don't use llrint on already-integral variables
...
This could trick people into thinking that there is a floating-point
type hiding behind the typedef instead of an integer.
2014-05-14 02:59:25 -04:00
Devin J. Pohly
d067e27762
fix accumulated rounding error in BBT grid
...
Though current_frame is an integral type (framepos_t), it was being
added to as if it were floating-point, leading to an ever-increasing
rounding error when creating the grid for bars/beats/ticks.
2014-05-14 02:54:01 -04:00
2f6065b32c
Fixed problem where importing invalid midi files caused crash.
2014-04-29 06:48:29 +10:00
f5276a104c
add new SMFSource constructor to be used for existing-external files. Fixes #5919 . Needs merging with CC
2014-04-28 10:37:17 -04:00
16a8762dd4
fix minor problem with recent cherry-picking from cairocanvas
2014-04-14 14:14:30 -04:00
be0da8ff5c
merge 5764970709 from cairocanvas as final(?) change related to data loss
2014-04-14 13:07:07 -04:00
2e3a84b15a
check whether a source with the same path exists in session list not just on disk.
...
This is intended to catch a potential wierd condition where a source exists as a stub, but has not yet been written to disk,
and then a new source is created with the same path. Currently it is not understood how/when this might happen
2014-04-14 13:05:35 -04:00
b49bb451d2
further code simplification and rationalization related to MIDI source/file renaming
2014-04-14 13:05:21 -04:00
0d5f4c553a
dramatic change in logic and naming for operations related to adding a MIDI region on demand and cloning/unlinking
...
Existing code would cause data loss due to creation of two Source objects referring the same path, one with removable flags and one without. Careful code review suggested a variety of thinkos, function naming problems and other confusion that caused this. I have tried ot extensively comment what is going on with these operations, because it is one key area in which MIDI differs from audio: with audio, capture is the only way to add a new audio region, but for MIDI there are GUI input events that can add a new region.
2014-04-14 13:05:08 -04:00
384c0a9fac
redesign technique for naming/creating regions for MIDI clone (or other non-capture driven MIDI region creation operations).
...
See comments in Session::new_midi_source_name() for details.
2014-04-14 13:04:14 -04:00
2b3cb052ca
after cloning a MIDI region, mark the source file as non-removable. Fixes reports about missing MIDI files on the forums and IRC
2014-04-14 13:01:23 -04:00
e1bf151ab0
after cloning a MIDI region, mark the source file as non-removable. Fixes reports about missing MIDI files on the forums and IRC
2014-04-14 13:00:24 -04:00
99bf5d9eed
backport 1d85ab27a7 and ba128eea from cairocanvas branch to remove GIO (possible hotfix release)
2014-04-14 12:53:30 -04:00
Sakari Bergen
2482361d88
Fix buffer size mismatch in evoral tests
2014-04-07 21:27:56 +03:00
af0c42ec4f
midi-controller, support log parameters - fixes #5890
2014-03-23 18:59:24 +01:00
713a375011
(backport from cairocanvas branch) do not crash when considering a directory called midi (with any case) as an apparent MIDI file
2014-02-27 20:37:17 -05:00
d8b3e2932f
use new ControlList::add() argument
2014-02-19 13:22:52 -05:00
3e1107c157
do not needlessly keep starting touch/automation watch passes just because a command to start or stop touch was received
2014-02-19 13:22:52 -05:00
a70f914c51
add option to skip guard points when adding new control points to a ControlList; add more debugging statements
2014-02-19 13:22:52 -05:00
d024b500a6
clear up some stupid thinking in Amp regarding the way the underlying AutomationControl/Controllable gets set when Amp::set_gain() is called
2014-02-19 13:22:52 -05:00
29b3c299a1
add a hack to deal with device discovery race, in which the devices were not actually known to be ready before the last update_surfaces() call (which is where binding between strips and routes occurs) ; add support for Mackie Control touch on/off messages
2014-02-19 13:22:52 -05:00
3566fa2d8e
add kxstudio linux vst dirs to LXVST default search path
2014-02-08 11:17:50 -05:00
b9054a1f72
fix missing colon in extension of LXVST search path
2014-02-08 11:16:17 -05:00
d47fe167e3
when cleaning up sources, do not remove "stub" sources from the source list, even though they do not exist on disk yet; remove some debug output
2014-02-07 17:38:42 -05:00
f715cd877c
add FileSource::is_stub() declaration
2014-02-07 17:30:34 -05:00
2c67913245
Fix egregious logic bug in FileSource::removable() and introduce FileSource::is_stub() to hide logic for deciding if a source (file) is a stub
2014-02-07 17:17:30 -05:00
60a9213035
fix utter confusion about session _path in new sessions.
...
Yikes!
2014-02-07 13:56:44 -05:00
7bc3e0dab4
don't add a [shared] panner for the sends to monitoing section
...
fixes issue with 'stuck' mono panners (when using a stereo
monitoring section and auto-connect): the send to monitor
section is added after the main panner. At this point the
channel-count is different, but the send shared panner() with
the main route.
Here: mono-panner on track, stereo-panner; and they shared control
settings. since stereo-panner's width is 1, the position cannot be
changed.
2014-02-07 19:30:43 +01:00
7c69b0ab82
add debug-message to track down missing Sources
2014-02-05 18:49:32 +01:00
371e8bdb55
VBAP nomenclature s/Direction/Azimuth/
2014-02-03 19:12:29 +01:00
591ff9ceb6
update audio-buffer assert, take offset into account
2014-02-03 16:59:59 +01:00
74385d7267
remove cruft - unused _size in audio-buffers
2014-02-03 16:59:51 +01:00
592be26a24
fix stem-export buffer-size check
2014-02-03 16:51:08 +01:00
9c0b62e2f3
fix processor_lock
...
Only WriterLock requires to hold process_lock() as well
otherwise Route::process_output_buffers() may deadlock
2014-02-03 15:09:17 +01:00
12576a9f29
forward port lv2_evbuf update (from jalv)
2014-02-01 23:07:23 +01:00
ce2feeac56
even though man alloca says to include <alloca.h>, it apparently isn't necessary and no such header exists in the msvc world
2014-01-30 11:18:58 -05:00
66d0241118
move check for n_speakers so that we avoid needless work
2014-01-29 15:50:17 -05:00
2d8352123a
fix vbap-speaker calculations to work with alloca()
2014-01-29 21:45:13 +01:00
a754a7cc0d
fix up the abomination caused by moving from variable length (multidimensional) arrays to alloca'ed arrays, specifically access to arr[a][b].
...
This needs checking by an actual VBAP+multispeaker user.
2014-01-27 20:37:17 -05:00
f8737f7012
add new folders to linux vst search path to help out AVLinux and some other specialized distros
2014-01-27 14:55:58 -05:00
10d577146a
replace standards-wobbling variable-length-arrays with alloca()
2014-01-27 14:20:19 -05:00
241c1aaaed
fix #5840 ; redefinition of typedef 'VstTimeInfo'
2014-01-26 19:11:54 +01:00
9a190b8d2b
fix crash when removing synth from midi-track
...
(and there are bypassed audio-plugins after said synth)
2014-01-26 00:46:39 +01:00
40d8c5ae01
change VBAP visual parameter name again :)
2014-01-23 15:32:26 +01:00
b0b71782e9
VBAP change visual parameter name. Diffusion -> Spread
2014-01-23 14:57:19 +01:00
0da004202e
LV2 plugin save/delete -- fixes 5835
2014-01-23 13:14:25 +01:00
d005e8866b
midi-audition: make "no synth" selection work.
2014-01-23 00:22:02 +01:00
8ad30bb76e
configurable midi audition synth
2014-01-22 23:59:24 +01:00
