Search in search results. Searcheption.

[hkdobrev]

Discourse thread: http://discuss.atom.io/t/searchception-missing/1779

It expands on #67.

In Sublime when you search in a project/folder you get the search results in a normal buffer. You could perform any text manipulation in there and also search. Atom should do that as well.

[benogle]

Project search results are not text in atom. I'm not sure about the manipulation part, but the searching part seems interesting

We would need to think about exactly how it integrates into the ui. Maybe the find-in-buffer interface works in the results pane. However, the current FindModel only works with pane items that have a buffer property.

This is not something the core team plans to work on in the short term. If f you feel strongly about it, I can look at pull requests. I have a feeling it is a non-trivial change.

[park9140]

👍 This would be great.

[townie]

This is a must have feature!

[alimoosavi15]

This is the only feature I'm missing from Komodo IDE.

[rahlzel]

It's sad to find this request started back in 2014 with nothing to show for it now in 2016. We do need this functionality. As @hkdobrev mentioned, it's one of the big reasons I use Sublime. With massive projects, sometimes we have a vague clue to find the code we're looking for, but that clue gives us a few hundred results that we then need to dig through. Being able to search the search results would help us narrow things down tremendously.

Thanks!

[eliveikis]

Same here, the one feature that keeps me using Sublime.

[voltuer]

Any news on this?? @damieng @travco @lee-dohm @codingbelief

[damieng]

I think what Ben said still stands. It's not on the core team's roadmap so it would need to be a community PR to be considered.

[travco]

@sebfindling Expanding on Ben's 2014 answer (on the technical side):

Project search results are not text in atom. ...

... the current FindModel only works with pane items that have a buffer property.

As far as I know, to make this viable, one of two things needs to be done. Either adapt entire find model (as it is) to make it able to read in non-buffer content (and figure out how to differentiate what we can and can't handle searchwise). Or change the output of search results to be a buffer - and still be as useful as it was (not sacrificing/backpedaling on features).

Unless there's a trivial third option, I'm personally not comfortable with atom's internals (or this plugin's for that matter) to dive into something like this with any degree of confidence myself - as I think the second option is the only viable one.