Comments
-
Greg Hurrell
Yeah, I can repro this. I'm guessing that you have hit on a very pathological degenerate case that causes the matching algorithm to spend a lot of time recursively searching for the best possible score.
(The way it works on finding match is to bump the cursor to the right one position every time it finds a match, just to see if there is an alternative way of finding a match which is higher scoring. Given that there are so many ways of matching that search string given the files in your project, it seems to go a bit crazy. Actually, even with a lot of recursion, I am still surprised at how long it's taking. There may be something else at play here which I haven't figured out yet.)
-
Greg Hurrell
I might have to put a short-circuit in the recursion so that at pathological depths (to be empirically determined, but could be anywhere between 5 and 20 levels, or perhaps more...) it will just bail and say, "I don't care if a better score is possible, I'm stopping right here".
-
Greg Hurrell
Status changed:
- From: new
- To: open
-
Greg Hurrell
Kind changed:
- From: feedback
- To: bug
-
Greg Hurrell
Summary changed:
- From: Hanging on particular folder names/structure when searching for file (Command-T)
- To: Hanging on particular folder names/structure when searching for file
-
Greg Hurrell
I think I've fixed this by adding memoization, which I'll be pushing out soon. I no longer have access to your sample files, however.
I'm going to close this issue for now. Feel free to re-open if you think it's still a problem after testing the next Command-T release.
-
Greg Hurrell
Status changed:
- From: open
- To: closed
Add a comment
Comments are now closed for this issue.