EDL support transition writes

[mikemahony]

Generates EDLs with dissolve transitions that look like:

001 AX V C     00:00:00:01 00:00:01:13 00:00:00:00 00:00:01:12
002 AX V C     00:00:01:13 00:00:01:13 00:00:01:12 00:00:01:12
002 AX V D 036 00:00:00:01 00:00:02:13 00:00:01:12 00:00:04:00
003 AX V C     00:00:00:01 00:00:02:01 00:00:04:00 00:00:06:00

[mikemahony]

Hi @jminor, I'm not 100% on this but I wanted to get it on your radar.

This line, for example, doesn't seem quite right yet: https://github.com/mikemahony/OpenTimelineIO/blob/bab455428c065840160a908c5748f038c363ad4f/opentimelineio/adapters/cmx_3600.py#L646 - I do need to account for the handles in some way like this, yes? I think I'm hardly accounting for them at all at this point... :-/

I'd love to hear your thoughts when you have some time.

PS. I hope you don't mind that this MR points to your aaf_adapter branch; I'm currently using that branch for all its excellent AAF goodness.

[mikemahony]

Added a new commit that's more correct but still not right. Is it possible that track.range_of_child_at_index is somehow miscalculating?

[mikemahony]

Ok. test_nucoda_edl_write_with_transition_2 feels like it's on the right track. Will reconcile with test_nucoda_edl_write_with_transition next.

[mikemahony]

Sorry for all the craziness! I made myself some diagrams and re-inspected the AVID-generated EDL I received from CZuber, confirming that it imports into Nucoda correctly. I'm now pretty certain that the following test will indeed need to pass for the EDL to import into Nucoda: https://github.com/mikemahony/OpenTimelineIO/blob/be2f4808e8bdad55cf6992bd10cf582dfcec2025/tests/test_cmx_3600_adapter.py#L428

It might be best to talk in person, but just in case, here are my diagram notes:

Data from original CrossDissolve_Day-Night_Long_1.aaf (via OTIO)
v330_21f    130 275     145 frames                               00:00:05:10 00:00:11:11
trans       -57 -43
v330_23e    222 422     200 frames                               00:00:09:06 00:00:17:14

Source ranges, no transition
A |-------------------------|                               130 275     00:00:05:10 00:00:11:11                   
B               145         |-------------------------|     222 422     00:00:09:06 00:00:17:14
                                        200

Visualized on generic timeline
A |---------------------|\--|                               130 232     00:00:05:10 00:00:09:16
B                 |----\|-------------------|               165 422     00:00:06:21 00:00:17:14
                  |-----|---|
                      57 43

Visualized EDL style                                                                                            Cumulative time
A |---------------|xxxxxxxxx|                               130 175     00:00:05:10 00:00:07:07                 45         00:00:01:21
B       45        |\------------------------|               222 422     00:00:09:06 00:00:17:14                 245        00:00:10:05
                              200
                  |---------|
                      100


Generated by this adapter code
TITLE: CrossDissolve_Day-Night_Long_1
001  AX       V     C        00:00:05:11 00:00:07:08 00:00:00:00 00:00:01:21 (45)
002  AX       V     C        00:00:07:08 00:00:07:08 00:00:01:21 00:00:01:21 (0)
002  AX       V     D 100    00:00:09:07 00:00:17:15 00:00:01:21 00:00:14:09 (300)   <--- This seems wrong, like it's including the transition duration?

The track.range_of_child_at_index value for each child:
(0, 145.0)
(95.0, 195.0)
(145.0, 345.0)

  0              95        145
A |--------------|----------|        195 
D                |--------------------|              345             
B                           |---------|---------------|    
                                         200

I confirmed that this is what it needs to look like:
001  v330_21f V     C        00:00:05:11 00:00:07:08 00:00:00:00 00:00:01:21 (45)
002  v330_21f V     C        00:00:07:08 00:00:07:08 00:00:01:21 00:00:01:21 (0)
002  v330_23e V     D    100 00:00:09:07 00:00:17:15 00:00:01:21 00:00:10:05 (200)

[jminor]

Hey @mikemahony did all of this get merged into master? This old PR onto my repo is still here. I wanted to make sure it all got in before I close this.