summaryrefslogtreecommitdiff
path: root/libs
AgeCommit message (Collapse)Author
2015-07-31Implement AudioBackend::driver_name() in PortaudioBackendTim Mayberry
This was missing and is necessary for EngineControl dialog to work properly
2015-07-31Add extra DEBUG_TRACE output to PortaudioBackendTim Mayberry
2015-07-31WinMME based midi input/output for portaudio backendTim Mayberry
TODO: Use MMCSS to elevate thread priorities Enable/test and fix SYSEX related code
2015-07-31Fix for PortAudioBackend::available_sample_ratesTim Mayberry
Was being called without initializing PA. PA should probably be initialized in ctor but PA backend also needs to support hot-plugging devices at some point so this will do for now
2015-07-31Support selecting separate input and output devices in portaudio backendTim Mayberry
2015-07-31Refactor PortAudioIO::discover into three private methodsTim Mayberry
2015-07-31Set default input/output device based on selected host api rather than ↵Tim Mayberry
global default
2015-07-31Fix crash when unable to open portaudio streamTim Mayberry
Calling Pa_Terminate after failing to open stream causes an assertion
2015-07-31Add host api/driver selection to PortaudioBackendTim Mayberry
2015-07-30Replace "%z" modifier with glib macro for portability.Robin Gareus
2015-07-30fix file metadata tagging on windows.Robin Gareus
This allows TagLib to open the file and create a Tag object (see 2a0365c) Tagging still fails later (taglib seek/SetFilePointer fails when writing)
2015-07-29Don’t create MTC debug string if it’s not needed.Robin Gareus
Removes periodic call to operator<< (and eventually loadlocale) in RT-thread in the debug version.
2015-07-28Revert "debug/fix LV2 state loading on OSX 10.5/PPC"Robin Gareus
This reverts commit 22bc62ed4360aaea29c131cc7dead87fa25ef228. forwarded upstream: http://dev.drobilla.net/ticket/1073
2015-07-28don't crash if taglib cannot open fileRobin Gareus
2015-07-27Clean up extra white space added by editor.Len Ovens
2015-07-27Handle the 4 common encoder types.Len Ovens
2015-07-27Add enc to midi map for mcp style encoders.Len Ovens
2015-07-27Make program change toggle or 1 only.Len Ovens
2015-07-27Allow any one midi event to control only one thing.Len Ovens
2015-07-27fix LV2 freewheeling port, explicit cast bool to float.Robin Gareus
http://lv2plug.in/ns/lv2core/#freeWheeling mentions "true" and http://lv2plug.in/ns/lv2core/#toggled defines true as > 0. false as <= 0.
2015-07-27debug/fix LV2 state loading on OSX 10.5/PPCRobin Gareus
2015-07-27add Session::routes_solo_changed(), a plural method which makes changing ↵Paul Davis
many route solo states much faster. Code originally by Grygorii Zharun in Tracks
2015-07-27fixes/updates for move to ::open()Robin Gareus
2015-07-26Move ARDOUR::touch_file to pbd/file_utils.hTim Mayberry
2015-07-26Use g_open instead of ::open in ARDOUR::touch_file for portabilityTim Mayberry
2015-07-25Use g_open and sf_open_fd in SndFileSource for portabilityTim Mayberry
2015-07-25Use g_open and sf_open_fd in SndFileImportable instead of sf_open and ↵Tim Mayberry
Glib::locale_from_utf8
2015-07-25Use g_open instead of ::open in translation related utility functionsTim Mayberry
This is for portability on Windows
2015-07-25Use g_open instead of ::open in AudioSource for portabilityTim Mayberry
This is only really for Windows as g_open and ::open are the same thing on other platforms
2015-07-25Use GStatBuf in AudioSource::initialize_peakfile for portabilityTim Mayberry
This was initially part of f89a976f that got reverted
2015-07-25Revert "When peakfiles get opened using a non-glib function (in a ↵Tim Mayberry
non-English locale) make sure that we pass locale-specific paths" This reverts commit f89a976f1a0477fba30bd7117310e6d04b370d68.
2015-07-25Revert "When saving a session (in a non-English locale) make sure that we ↵Tim Mayberry
use a locale-specific path" This reverts commit cf5a8651d848fa5333e1c567286fc0eec2b0a0f7.
2015-07-25Revert "A few more instances of non-glib 'open()' getting used without ↵Tim Mayberry
locale-specific input paths" This reverts commit 52ef02b387d8267315547129f7f100048a5c1166.
2015-07-25Add unit test to check that libxml2 expects utf-8 encoded file paths on WindowsTim Mayberry
2015-07-25Only create one test output directory in file copy testTim Mayberry
Rather than one directory for each file
2015-07-25Add unit test to assert that using locale_from_utf8/::open will fail for ↵Tim Mayberry
some file paths on Windows
2015-07-25Remove sse optimization sources from test applicationsTim Mayberry
These are already built and included in libardour
2015-07-24VST info: use g_fopen() for Windows compat.Robin Gareus
2015-07-24coreaudio aggregate device fix for 10.5Robin Gareus
2015-07-23A few more instances of non-glib 'open()' getting used without ↵John Emmas
locale-specific input paths (noticed by Todd. probably more to come...)
2015-07-23When saving a session (in a non-English locale) make sure that we use a ↵John Emmas
locale-specific path
2015-07-23When peakfiles get opened using a non-glib function (in a non-English ↵John Emmas
locale) make sure that we pass locale-specific paths
2015-07-23signed/unsigned comparisonRobin Gareus
2015-07-23amend a73a039a3a, separate VST cache for 32/64bitRobin Gareus
2015-07-23avoid llabs ambiguityRobin Gareus
(old gcc has a built-in)
2015-07-22fix typo in old (pre 10.6) coreaudio API wrapperRobin Gareus
fixes aggregate device support for PPC builds.
2015-07-21VST plugin title fix #6467Robin Gareus
2015-07-21bye bye Unix.Robin Gareus
An IPv4 or IPv6 address query for the name localhost must always resolve to the respective loopback address [rfc6761]. According to several users, apparently chromebook and OSX `host localhost` fails but configuring 127.0.0.1 works. (maybe a DNS/DHCP issue?! or ipv6)
2015-07-20do not call playhead priority (auto-return) code after selections change, ↵Paul Davis
unless doing the Tracks playhead priority thing
2015-07-20fix auditioning on systems with larger disk read chunk sizes.Paul Davis
This really brings into focus the question of whether any of the "varifill" code should really remain. This was probably one of the only remaining places where a partial-fill operation was done and the code was broken for this case. What we know about disk i/o streaming doesn't support varifill much at all. Something to think about.