VSE: using python to remove a scene strip with effect applied causes... weirdness #73828

New Issue

hudson barkley · 2020-02-15T09:18:15+01:00

hudson barkley commented

2020-02-15 09:18:15 +01:00

Blender Version
Broken: 2.81a

Short description of error
when a strip is removed using the python code:

bpy.context.scene.sequence_editor.sequences.remove(strip)

any effect strips that are, or WERE recently applied to this strip will behave in very strange ways -
Single-input effects like speed control can behave as if they were applied to two strips, or just jump around
double-input effects like crossfade can rapidly jump between multiple other strips as if they cant make up their mind which strip it is applied to.

Exact steps for others to reproduce the error
Ive included a test .blend file, there are two python scripts included, both operate on a currently selected strip, run them while various color strips are selected to see the strange things the effects do.

The bottom script will just remove whatever strip is currently selected.
Try using this to remove the red color strip, then grab the green color strip and move it around, blender will crash after a bit of movement (usually).
Try using this to remove the black color strip, the speed control will... uhh, do weird stuff.

The top script will add a new strip, move the effects on the active strip to that, then delete the active strip. Somehow this still causes problems, even tho the strip that was removed now has no effects applied to it.
Try running this on the green color strip, it will cause an instant crash if the new strip is moved
Try running this on the green color strip, then moving the cursor, then moving the new strip, the left crossfade will switch between multiple inputs
Try running this on the black color strip, the speed control will behave very odd, but in a different way.

Some very weird stuff will happen randomly if you replace multiple strips too, ive seen the speed control start thinking its a crossfade...

vse strip remove test.blend

**Blender Version** Broken: 2.81a **Short description of error** when a strip is removed using the python code: ``` bpy.context.scene.sequence_editor.sequences.remove(strip) ``` any effect strips that are, or WERE recently applied to this strip will behave in very strange ways - Single-input effects like speed control can behave as if they were applied to two strips, or just jump around double-input effects like crossfade can rapidly jump between multiple other strips as if they cant make up their mind which strip it is applied to. **Exact steps for others to reproduce the error** Ive included a test .blend file, there are two python scripts included, both operate on a currently selected strip, run them while various color strips are selected to see the strange things the effects do. The bottom script will just remove whatever strip is currently selected. Try using this to remove the red color strip, then grab the green color strip and move it around, blender will crash after a bit of movement (usually). Try using this to remove the black color strip, the speed control will... uhh, do weird stuff. The top script will add a new strip, move the effects on the active strip to that, then delete the active strip. Somehow this still causes problems, even tho the strip that was removed now has no effects applied to it. Try running this on the green color strip, it will cause an instant crash if the new strip is moved Try running this on the green color strip, then moving the cursor, then moving the new strip, the left crossfade will switch between multiple inputs Try running this on the black color strip, the speed control will behave very odd, but in a different way. Some very weird stuff will happen randomly if you replace multiple strips too, ive seen the speed control start thinking its a crossfade... [vse strip remove test.blend](https://archive.blender.org/developer/F8340978/vse_strip_remove_test.blend)

hudson barkley commented

2020-02-15 09:18:15 +01:00

Added subscriber: @snuq

Richard Antalik commented

2020-02-15 18:31:17 +01:00

Changed status from 'Needs Triage' to: 'Confirmed'

Richard Antalik commented

2020-02-17 01:41:43 +01:00

Added subscriber: @iss

Richard Antalik commented

2020-02-17 01:41:43 +01:00

@snuq did you intentionally made cross effects non-overlaping? It shouldn't work like this... Though we can make a feature out of this bug.

In any case, there are 4 bugs here it seems.

@snuq did you intentionally made cross effects non-overlaping? It shouldn't work like this... Though we can make a feature out of this bug. In any case, there are 4 bugs here it seems.

hudson barkley commented

2020-02-17 05:12:59 +01:00

In #73828#873175, @iss wrote:
@snuq did you intentionally made cross effects non-overlaping? It shouldn't work like this... Though we can make a feature out of this bug.

You sure about that not being an intended feature? It has worked that way as long as ive been using the vse. I use that method for crossfades about 90% of the time, keeps the timeline clean

> In #73828#873175, @iss wrote: > @snuq did you intentionally made cross effects non-overlaping? It shouldn't work like this... Though we can make a feature out of this bug. You sure about that not being an intended feature? It has worked that way as long as ive been using the vse. I use that method for crossfades about 90% of the time, keeps the timeline clean

Richard Antalik commented

2020-02-17 05:41:40 +01:00

In 2.79 it turns pink, not sure if it renders same as now...
I am fine with how it is now, but I think it works by accident.. it is still marked as invalid in C I think..

In 2.79 it turns pink, not sure if it renders same as now... I am fine with how it is now, but I think it works by accident.. it is still marked as invalid in C I think..

Sergey Sharybin commented

2020-06-12 13:33:54 +02:00

Added subscriber: @Sergey

Richard Antalik was assigned by Sergey Sharybin

2020-06-12 13:33:54 +02:00

Sergey Sharybin commented

2020-06-12 13:33:54 +02:00

@iss, I see some patches related to this report. What the status here?

blender-admin commented

2020-06-18 05:25:20 +02:00

This issue was referenced by 47f98a38d0

This issue was referenced by 47f98a38d0f8a83648c13a54adc717d49d0898f1

Richard Antalik commented

2020-06-18 05:54:34 +02:00

Changed status from 'Confirmed' to: 'Resolved'

Richard Antalik closed this issue

2020-06-18 05:54:34 +02:00

Richard Antalik commented

2020-06-18 06:07:36 +02:00

Changed status from 'Resolved' to: 'Confirmed'

Richard Antalik reopened this issue

2020-06-18 06:07:36 +02:00

Richard Antalik commented

2020-06-18 06:07:36 +02:00

In #73828#952454, @Sergey wrote:
@iss, I see some patches related to this report. What the status here?

Sorry I missed your question. I was getting patches to be reviewed. There is still 1 crashy aspect to this report so I wil reconfirm and will have to update patch for this.

> In #73828#952454, @Sergey wrote: > @iss, I see some patches related to this report. What the status here? Sorry I missed your question. I was getting patches to be reviewed. There is still 1 crashy aspect to this report so I wil reconfirm and will have to update patch for this.

blender-admin commented

2020-07-26 16:07:34 +02:00

This issue was referenced by 076d033e10

This issue was referenced by 076d033e10eee29bfc8315344bd2ba1f0951ad06

Richard Antalik commented

2020-07-26 16:12:14 +02:00

Changed status from 'Confirmed' to: 'Resolved'

Richard Antalik closed this issue

2020-07-26 16:12:14 +02:00

Thomas Dinges added this to the 2.90 milestone 2023-02-08 16:27:55 +01:00

Sign in to join this conversation.

No Label

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

VSE: using python to remove a scene strip with effect applied causes... weirdness #73828