Some asked in the niri matrix room if this was possible and it seemed quite straight forward and useful. Not entirely sure about the name and documentation, I tried to avoid bikeshedding.
Add keyboard-focus=exclusive|on-demand command line and config option #492
bbb651/fuzzel:ondemand-interactivity into master @ -312,2 +312,4 @@
Default: _top_
*--keyboard-interactivity-on-demand*
Request on demand keyboard focus rather than exclusive.
It took me a while to parse this sentence when reading it for the first time. I think using "on-demand" would make it clearer.
@ -313,1 +313,4 @@
*--keyboard-interactivity-on-demand*
Request on demand keyboard focus rather than exclusive.
This switching keyboard focus to close fuzzel.
It seems like "switching" should be "switches" here, but I'm still not sure what's being expressed.
Oh, I guess the idea was something more like: "This makes fuzzel exit when keyboard focus is lost"?
I think it would be very useful to support the extended protocol, and there's nothing wrong with that part of the implementation.
However, like @craigbarnes, I find the documentation (of fuzzel's options) hard to understand. That needs to be improved.
It might also make more sense to add an enum-like command line/config option, rather than a boolean: keyboard-focus=exclusive|on-demand perhaps? Defaulting to exclusive.
I find the documentation (of fuzzel's options) hard to understand. That needs to be improved.
I'll try, I'm not used to writing man pages. In general I don't like the duplication between fuzzel(1), fuzzel.ini(5) and --help, I know they serve different purposes but especially with flags that just set config options like this one it feels a bit redundant
It might also make more sense to add an enum-like command line/config option, rather than a boolean:
keyboard-focus=exclusive|on-demandperhaps? Defaulting toexclusive.
I agree, even if it's unlikely there'll be additional variants it's just as short and more intuitive to override. Probably also focus over interactivity, I wanted to match the protocol but it could just be mentioned in the man page.
Edit: Unrelated to this PR, I noticed layer defaults to overlay but --help says it defaults to top. This is what I'm talking about, when there's no single source of truth, documentation, completions, etc. eventually go out of date
da29db1b36
72acd91c28
keyboard-interactivity-on-demand command line and config option (削除ここまで)keyboard-focus=exclusive|on-demand command line and config option 72acd91c28
bdaac3041f
bdaac3041f
aa8553b12b
I accidentally force pushed the branch to master while trying to resolve the submodule being removed (some weird interaction with jj and force pushing), and that made codeberg think it's been manually merged... is it possible to reopen the PR? The commit should be good now: bbb651/fuzzel@466abbe0f7
No due date set.
No dependencies set.
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?