To do before v1.7.0 is released

(Please strike anything that has been completed. And put your handle by it so we know who to thank!)

Development of 1.7.0 has been moved to a Bzr branch. See here for usage details

• Non-coding tasks:
  • Test MP3/OGG/FLAC/WAVs of varying sample rates and report results here
  • Test everything in different versions of Windows, particularly Windows-only bugs Compiling on Windows
  • Test MIDI mappings for correct functionality for as many currently supported controllers as we can
  • Update the Wikipedia entry
  • Finish website update (Albert)
  • Update user manual

• Fix timecode LUT memory corruption (Albert) - https://bugs.launchpad.net/mixxx/+bug/382228
• Fix pasuspender usage in all .desktop files (Albert) - https://bugs.launchpad.net/mixxx/+bug/388657
• Ensure we bundle correct MSVC DLLs with Win32 and Win64 installers (Sean) - https://bugs.launchpad.net/mixxx/1.7/+bug/376801
• Make an executive decision about the output mapping table (Albert)
• Fix qt-opengl and qt-svg dependencies for Ubuntu (Albert) https://bugs.launchpad.net/mixxx/1.7/+bug/365609
• Get Albert to look at possible performance hit: https://bugs.launchpad.net/mixxx/+bug/379317
• If possible, try to figure out Ubuntu ALSA stream freeze - See Mailing list thread and Bug entry Now we just need to nag people to offer a user-installable fix.
• MidiObjectALSASeq is missing a devClose implementation - Pegasus
• Non-C++ task: Get the “res” directory to copy to the proper place (prefix=) on an scons install (python)
• Fix model issue where removing any row removes the last row as well. (midiinputmappingtablemodel.cpp, midimapping.cpp) - Albert
• Fix model issue where completing the learning wizard doesn't list anything in the table until Mixxx is restarted - Albert
• Get rid of MidiType enum, switch MIDI mapping format to use <status> for input blocks as well. - Albert
• Migrate the range hacks out of SoundTouch's bpmdetect.cpp and into bpmdetector.cpp, then update bpmdetect.cpp from the SoundTouch repository - rryan
• Multithreading guru task: Clean up and check MidiScriptEngine thread-safety - rryan…give the man a beer!
• Handle MIDI Pitch input messages (0xE#, LSB, MSB) - Gamegod, Pegasus & Lupin3rd
• Merge the Features_MIDIScriptAutoReaction branch to trunk (currently waiting on the Qt team to see if our solution is a good one) - RRyan and Pegasus
• Fix the version number in defs.h and all MIDI XML files - Pegasus and Mad Jester
• Port old MIDI XML device files to the new format (basically just adding a header block.) - Mad Jester
• Complete MIDI device selection & mapping dialog (see the MidiMapping design spec)
  • Use tree view in preferences and display MIDI devices on left. - gamegod
  • Or select MIDI device, auto select the mapping (combobox) - gamegod
• Reimplement MIDI Learning (100% done - gamegod)
  • Rewrite midiEvent() signal in MidiObject to emit a MidiMessage, and rename the signal to midiLearn()
  • Connect midiLearn() signal to some learn() slot in MidiMapping object
  • Write learn() slot in MidiMapping. learn() should update the mapping and emit(inputMappingChanged()) at the end.
  • Create list of usable controls… - gamegod
• Midi Scripting
  • Clear scripting engine when new MIDI mapping preset is loaded in the Prefs. - gamegod
• MidiInputMappingTableView - add delegate classes - gamegod
  • Display the MidiType (Key/CC/etc) (instead of those crappy enumerated integers) - gamegod
  • Combo boxes for group and control - gamegod
  • Values display in hexadecimal - gamegod
• → MIDI output handling
  • Create MidiOutputMappingTableModel and -View - gamegod
  • Write underlying data structure for output (OutputMidiMapping) - Pegasus
  • Finish adding delegate classes and finish up output table (include correct columns) - targeted for v1.8 instead
  • Fix up <status> and <miditype> mishandling in MidiMessage for outputs - Albert
• Fix the segfault-on-close that's happening on Linux and Windows with Intel graphics drivers. This may not be our problem. See bug #251128.
• Call MIDI script init function when opening the device, not before - Pegasus
• Call MIDI script shutdown function (if one exists) just before closing the device - Pegasus
• Write an upgrade function that moves mixxx.cfg and mixxxtrack.xml to their new homes (ffs) - Pegasus
• → Clear up as many compiler warnings as we can (give preference to the “depreciated function” ones so we don't get bitten later.) (kousu)
• Change version numbers everywhere to “1.7.0-beta1” - Pegasus
• Windows: MIDI feedback on all but the first listed controller doesn't work with more than one connected. It's best to disconnect/remove all (real and virtual) devices except the one you're interested in working with. - Pegasus
• Windows: You may need to tell Windows to send MIDI output to the correct device, especially if you hear random notes instead of see lights on your controller: Start→Settings→Control Panel→Classic View→Sounds and Audio Devices→Audio Tab, set “MIDI Music Playback” at the bottom to the device you're trying to use. - Pegasus

https://launchpad.net/mixxx/+milestone/1.7.0

• MIDI Pitch messages are no longer processed internally and must be mapped to a script function. Fortunately script.pitch can be used very easily from your script.
• Probably regressed like crazy with the M-Audio Xponent. Functionality needs to be rewritten using scripting. - lupin3rd has an input mapping
• The use of multiple MIDI devices simultaneously is not yet supported.
• SelectNextTrack, SelectPrevTrack and LoadSelected do not work in the Browse view. See bug #342120.
• SelectNextTrack & SelectPrevTrack cause the GUI to freeze for a few seconds when you try to scroll down beyond the current page. See bug #361170.
• Certain MIDI controllers' (Stanton SCS devices) scratching is directly tied to latency. As a result, users will need to tune the values at the top of the scripts in question for their particular system.