bug: better handling of folds

[folke]

Did you check the docs?

• I have read all the flash.nvim docs

Is your feature request related to a problem? Please describe.

currently matches inside folds are labeled

Describe the solution you'd like

need to think this through :)

Describe alternatives you've considered

not really

Additional context

No response

[tssm]

Could this be an option? Jumping to folds is quite useful sometimes

[folke]

Whats the use-case? Using incremental mode?

[tssm]

No, just navigation. Sometimes I wanna jump to a fold and open it because the code I must read/modify is folded

[folke]

It doesn't seem to be possible to set extmarks on the folded line.

With something like nvim-ufo, you will see the markers

[MariaSolOs]

With something like nvim-ufo, you will see the markers

@folke what do you mean? I'm using nvim-ufo and I still cannot use flash to jump to folded lines or have any other indicators that would allow me to do so.

[folke]

@MariaSolOs I mean you can jump to the summary shown on the fold for a word in that fold.

Still need to make it so that a folded line also gets a label (one) when the text inside the fold matches the pattern.

[folke]

Added proper support for folds.

The fold line will now show the label of the first match inside a fold. See example below:

When hitting the label, the cursor will open the fold and jump to the first result. Also works in visual and operator pending modes.

[MariaSolOs]

The fold line will now show the label of the first match inside a fold.

Why limit the label to the first match? Not asking you to change this BTW, I'm just curious :)

[folke]

Since you can't see the targets you are jumping to, I don't think it makes sense to show all the labels for items in the folds?

[MariaSolOs]

@folke What I had in mind is that say I'm jumping to the following line with se:

> fn folded_method () {

I thought something like this would happen (adding the labels to both occurrences of e):

> fn foldad_mxthod () {

[folke]

Oh you mean on the foldtext? That's not how it currently works. It's also not straightforward to map actual cursor positions to a position inside the foldtext.

The feature I did add is say you have a multiline fold and the pattern matches something in line 3 of your fold, then you'd still be able to jump to that

[MariaSolOs]

That makes sense, thanks for explaining!