On 8/20/2024 10:18 PM, Jim Porter wrote:
> On 8/20/2024 8:15 PM, Jim Porter wrote:
>> Hopefully the attached reproducers help make sense of this. I've also 
>> updated my patch to handle 'min-width' in what I think is a simpler 
>> way. This implementation relies on the fact that you can't nest 
>> 'min-width' specs (the iterator struct can only hold one spec at a 
>> time). I'm guessing on some of these parts, so I may be totally 
>> off-base, but the test cases do what I expect anyway...
> 
> Whoops, I'd uploaded the wrong patch. Here's the correct one.

... and here's a new test case (overlay-min-width). This shows that now 
you can apply 'min-width' as an overlay property, which would be great 
for 'visual-wrap-prefix-mode'. I could then convert that code to use 
overlays (at least for 'min-width', maybe for 'wrap-prefix' too), which 
would make it easier to clean up when deactivating the mode.

Of course, if there's a way to write regression tests for this, I'm 
happy to do so, but I've never looked into how (or if) one can write 
regression tests for display engine features.