Friday, 2015-12-04

*** ddom has quit IRC00:01
*** ricardocrudo has quit IRC00:14
*** artfwo has quit IRC02:27
*** Magnus_RM has quit IRC03:35
*** LAbot has joined #lv205:43
-wolfe.freenode.net- [freenode-info] channel trolls and no channel staff around to help? please check with freenode support: http://freenode.net/faq.shtml#gettinghelp05:43
*** aombk has quit IRC06:14
*** edogawa has joined #lv207:57
*** falktx|work has joined #lv208:55
*** ricardocrudo has joined #lv209:28
*** ssj71 has quit IRC09:48
*** ddom has joined #lv209:57
falktx|workdrobilla: have you seen MPE yet?12:03
falktx|workthis http://expressiveness.org/2015/04/24/midi-specifications-for-multidimensional-polyphonic-expression-mpe12:03
falktx|workdrobilla: juce is getting support for that very soon. several companies are backing the spec12:04
falktx|workdraft here https://docs.google.com/document/d/1vpjxoPHw82X3xyNvE6_hsDeL86vloNQZC83NHD8edow/12:04
falktx|worksomething nice to have in lv2 I think12:04
falktx|worka base helper struct or class that handles the binary midi stuff and just calls different functions will be nice to have12:05
* falktx|work will try this once the juce mpe code is public12:06
falktx|workdrobilla: I imagine the plugin will have a new optionalFeature saying it supports MPE12:09
rgareusfalktx|work: we came across this before.   It ended in a rant how MIDI won't ever die and just be hacked-up :)12:16
rgareusbut yes, we should add LV2 support for this.12:16
rgareuswhile we're at this,  are there URIS for NRPN (host side parsed)  or is this left for plugins to decode?12:17
*** drobilla` has joined #lv214:24
drobilla`I'm not really sure what could be added to "support" this that'd help much.14:24
drobilla`If we do add some kind of note handling callback thingie, it should definitely be at least theoretically capable of supporting not-MIDI, IMO14:24
*** drobilla has quit IRC14:26
drobilla`falktx|work: Yes, feature14:28
drobilla`rgareus: Not sure what you mean by that14:29
rgareusdrobilla`: the host can parse the NRPN  and send the plugin a 14 bit value.14:31
rgareusinstead of a few CCs14:31
rgareus(move complexity to the host, rather than require plugins to implement parsing it)14:31
drobilla`rgareus: The closest you can get with actually MIDI would be to add normalization rules about the event containing all the necessary "sub-events"14:35
drobilla`If it's not (sent via) MIDI, well, then it's pretty much just a piece of the "make easy to use utility thingie for handling note and control events in general" thing14:36
*** drobilla` is now known as drobilla14:38
drobillaWhich'd be great, but is a biggish thing14:38
*** falktx|work has quit IRC14:46
*** falktx|work has joined #lv214:48
*** aombk has joined #lv214:49
*** ventosus has joined #lv216:16
*** ddom has quit IRC16:27
*** colock has joined #lv216:38
*** colock has quit IRC16:40
*** colock has joined #lv216:45
*** deva has joined #lv216:50
rgareusfalktx:  still at falktx|work ?17:42
rgareusshall we start an new inquistion initiative?    if Carla and Ardour would refuse to load plugins that don't validate. Do you think that'll make plugin Authors listen?17:44
rgareusor would that just drive them away to   VST or the likes of that?17:44
falktx|workrgareus: we can patch the broken plugins17:53
falktx|workdevs will learn from the good examples17:53
falktx|workI made it easy on kxstudio by providing a 'sord_validate_lv2' that already includes all the lv2 ttl data (including meta and schemas17:54
falktx|workand the mod + kxstudio stuff)17:54
*** NickSB has quit IRC17:58
*** NickSB has joined #lv218:09
drobillaA warning is probably good enough18:13
drobillaBe nice if there was API for sord_validate, but it depends on some stuff sord doesn't (regexp)18:14
*** rncbc has joined #lv218:14
rgareusdrobilla: we've had warnings for years.  they clutter up every bug report on Ardour I've seen in recent years (there are more messages about CURIE than anything else).18:27
falktx|workthose messages need to print the URI that caused it18:28
falktx|workotherwise they're useless18:28
rgareusbah. those messages should not be there in the first place.  just stop reading the ttl and drop the whole plugin :)18:29
rgareusbut yes, if the warning stays. it should point to the source18:30
rgareushas anyone tried http://bipscript.org/ yet?18:32
drobillaYes.  API thing, that.18:36
drobillaAt least, in one case.  Another is more troublesome, since you'd have to pay to keep that information around in a model18:37
drobillaThe bad prefix ones can probably be caught earlier and rejected outright, though18:37
*** falktx|work has quit IRC18:39
*** deva has quit IRC18:46
*** ricardocrudo has quit IRC19:26
*** ssj71 has joined #lv219:54
*** ricardocrudo has joined #lv220:31
* drobilla => Toronto20:40
*** drobilla has quit IRC20:40
*** ventosus has left #lv220:45
*** ricardocrudo has quit IRC21:19
*** ricardocrudo has joined #lv221:58
*** ricardocrudo has quit IRC22:19
*** ricardocrudo has joined #lv222:23
*** Magnus_RM has joined #lv223:16
*** ssj71 has quit IRC23:22
*** drobilla has joined #lv223:47
drobillaTurns out the parse error was just ignored23:48
drobillaI can't get the message itself to have a line number without some breakey changes, but got to this, anyway: http://pastebin.com/rKi7e0pW23:49
drobilla(lilv refuses to instantiate if there were parse errors)23:49
drobillargareus: So you'll get a different sort of noise now, at least :)23:53
rgareusw00t!23:53
drobillaActually unloading the data loaded so far from the model proved slightly annoying, so I think I just won't care about that part until I have a reason to :)23:58
rgareusdrobilla: next step is to refuse loading plugins that don't define mandatory   (such as lv2:microVersion)23:59

Generated by irclog2html.py 2.13.0 by Marius Gedminas - find it at mg.pov.lt!