Trimming #1
Description
This is the tracking issue for completing the experimental-trim feature.
The idea here is that we can get quite far with trimming/slicing without having to worry about how data is encoded and thus avoid the cost of re-encoding (which is out of scope for this project).
Status
Experimental and incomplete. Currently only supports a single audio track and isn't thoroughly tested yet.
TODO
- Take edit lists into account
- Construct new edit list adjustments for un-trimmable ranges
- Ensure correctness for tracks where one or more samples correspond to multiple points in time (e.g. chapter, still image, etc.)
- Support key-frames via
stssand/orsdtpatoms - Evaluate trim snapping (grow/shrink trim range by up to x amount) to avoid edit list adjustments (e.g. snap to video track key frames if they're close enough)