Turned the pipe getter into a method to be able to reuse it. So now operators will check for pipe before the main execution, and try to update it (if audacity has been opened since blender launched)

In the addon prefs, there is now a property to point on the audacity executable to allow the addon to automatically launch it. A downside is to have a waiting time, between calling audacity and having it open and ready. This waiting time, if too short, can cause error. The default seems ok (5s) but it can be set by user in prefs if needed. An improvement could be to have an operator processing this automatically, but for now, seems ok.

Here is a little demo video : https://youtu.be/o2HQz504-ls

A try to make the play operator more consistent and predictable, here is a little demo video : https://youtu.be/e3l50GfcjGQ

There are a few changes in the behavior

• The play operator has a poll to not be able to work if animation is playing, if audacity has not received anything yet during this session and from this scene, if in strip mode with no strip name in send_strip property
• The play operator is now modal, so the operator will work until Space/Esc is pressed, or animation is stopped. The mouse actions are still possible ("PASSTHROUGH") to navigate in blender
• The operator can be either cancel (ESC or animation stopped) or finish (SPACE), if cancelled, the playhead will be reset to original location, if finished, the playhead will stay in current location
• Send operators now set the audacity_send property to True to be able to track if something has been sent to Audacity during this session
• The modal draw on sequence editor a "report" for the user to have feedback operator is working and hints on shortcuts
• When playing in sequence mode, the play starts from the begining of the active frame range (normal or preview) and read the entire range
• the play command sent to audacity is now "PlayLooped:" to mimic blender behavior and stays synced with it
• [edit] remove undo of this operator, seems weird for a play operator to have undo ability, i think it could be confusing for user to have this undo step instead of, for example, their last cut in the edit
• [edit] added a startup handler to reset the audacity_send properties of every scenes when loading a blend

I hope you will find these tweaks coherent, of course tell me if anything bothers you ! cheers !

Multi file addon, very few changes in the behavior and the code :

• added addon prefs class for further additions
• grouped custom properties in a pointer property in the bpy.types.Scene for better organisation (scene.audacity_tools_props)
• changing properties values through the operators instead of type values (scene.audacity_tools_props.record_start=something instead of bpy.types.Scene.record_start = something)
• adding a startup handler to reset properties on blend file loading
• renaming the play operator (SEQUENCER_OT_stop_in_audacity->SEQUENCER_OT_play_stop_in_audacity sequencer.stop_in_audacity->sequencer.play_stop_in_audacity)
• fix the issue with unique naming suggestiong in receive operator
• add "Audacity Tools --- " at the start of the pipe print statements for better readability
• [edit] add doc_urland tracker_url links in the addon infos to be able to go to the readme or create an issue from the addon prefs

after testing a little bit this awesome addon, i tried to do some small fixes :

• Add gitignore for easier forking
• Fix an error of assignment for fade_curvevariable (to me, fade_curve and enveloppe processing seems not useful if there is no animation in the sequencer
• Unique naming when receiving sound from audacity in the name field in the exportHelper panel (i arbitrarily put *name of the blend file*_from_audacity.wav but of course, it could be other thing

Environment:

• Audacity opened with mod_strip_pipe enabled
• Blender opened in VSE with an audio strip selected
• push "Send Strip" button

Result Error: location: :-1 Error: Python: Traceback (most recent call last): File "/home/menda/.config/blender/2.92/scripts/addons/audacity_tools_for_blender-main/init.py", line 339, in execute do_command("SelectAll") File "/home/menda/.config/blender/2.92/scripts/addons/audacity_tools_for_blender-main/init.py", line 73, in do_command response = get_response() File "/home/menda/.config/blender/2.92/scripts/addons/audacity_tools_for_blender-main/init.py", line 59, in get_response line = FROMPIPE.readline() NameError: name 'FROMPIPE' is not defined

location: :-1

Hello, thank you for such an awesome addon :D I like it very much even though I usually use Audition :D

It appears that tracks are being summed to mono for some reason. Is this a bug or feature?

• Strip sending works just fine and everything is stereo as it should be.

• But after clicking "receive" and then "send" again - strip appears to be mono.

There was a problem with handling spaces in filepath in export command, needed extra " . I renamed the filepath variable for the audacity command, and use the self variable for the rest cheers

Some small improvements i made while trying to read the code and catch up with your additions (great ones, the play operator works way better for now !)

Just a little thing, the explanatory gif hosted in this repo is quite heavy, i just cloned the repo to test the addon and waited 15mb download, no big deal but pasting url from giphy for example would be more efficient Cheers !

The first time you push "receive" it renders the audio in Audacity and imports it as a strip. But the next time, it does the same and creates a new strip. Perhaps, only force update file of strip in Blender?

I see the license is gpl-3.0-or-later. If you create a license file, this information will be displayed within the first page view for the project. If you want help with this, I could make a PR for you.

Encountering what might be a file name limit. When I send a 33 character-named file audactity reports it can't be found, and the first letter of the reported name was missing. I edited the file name to be below 32 and that worked.

Blender 3.0.1; Audacity 3.1.3.

Blender waiting, and he doesn't start. Command line show:

Audacity Tools --- Looking for Audacity pipe

But if the audacity is open, this works perfectly.

Hi, I get an error when trying to receive from Audacity in 2.92 official

Python: Traceback (most recent call last):
  File "addons\audacity_tools_for_blender\operators\receive_from_audacity.py", line 107, in execute
    channel=find_completely_empty_channel(),
  File "C:\Program Files\WindowsApps\BlenderFoundation.Blender_2.92.100.0_x64__ppwjx1n5r4v9t\Blender\2.92\scripts\modules\bpy\ops.py", line 132, in __call__
    ret = _op_call(self.idname_py(), None, kw)
RuntimeError: Error: File 'voiceover\voiceover_from_audacity.wav' could not be loaded


location: <unknown location>:-1

When sending a single strip from blender to audacity, if user does not do anything to it, and get it back into blender, it seems (according to the waveform), that this strip audio has been levelled up in audacity. You can see it in this screenshot (the up one is from audacity, the lower one is the source)