Editor and Preview, sequence with different sized images
The spine editor 4.1.09 will hit an infinite dialog opening loop when changing a sequence from region to mesh if images in sequence does not all have the exact same image size. It seems to be connected to the preview window. And can be triggered by playing the animation in the previewer.

Running an animation with a sequence which has images of different sizes, has a side effect of flashing the image size differences in the animation window.

My question?
How is the sequence feature meant to be used?

The editor has different behavior for sequences (with images of different sizes) if the sequence is a region or a mesh.

If the sequence is a mesh and the sequence consists of different sized images, the editor will show the entire sequence as being the same size. Effectively changing the resolution while keeping the size the same.

However if the sequence is a region and the sequence consists of different sized images, the editor will show the size differences sequence as being the same size.

JSON Export
The json export does not contain information about the size of the individual images in a sequence. Which makes it a bit hard for us to throw an error if all the images in a sequence does not have the same size.

Further more the size exported to the json seem to be from the last image in the sequence.

"sequence": {
   "images/sequence/sequence_": {
      "width": 1000,
      "height": 250,
      "sequence": { "count": 21, "start": 0, "digits": 5 }
   }
},
"not_sequence": {
   "images/image": { "x": 0.5, "width": 127, "height": 26 }
}

So how is the sequence feature meant to be used?

Are different sized images in a sequence not supposed to be supported?

Should it work like mesh, where images size result in resolution change?
or region where image size result in a visual size change in the sequence animation.

Being able to have images in different sizes could optimise space consumption in sprite maps etc, but is hard to use in practice since all images in the sequence have the same anchor point and can not have individual transform offset...

Thanks for the snappy answer.

Please do let me know when you have had time to look into the other issues and questions.

Furthermore when trying to replicate your spineboy example, we experience inconsistencies with the "Frame" field in the sequence settings. Sometimes with a region sequence it appears to be greyed out and cannot be edited... is this somehow intentional or a bug?