Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

eval (__fzfcmd) hangs indefinitely #6955

Closed
zx8 opened this issue Apr 28, 2020 · 14 comments
Closed

eval (__fzfcmd) hangs indefinitely #6955

zx8 opened this issue Apr 28, 2020 · 14 comments
Assignees
Labels
bug Something that's not working as intended
Milestone

Comments

@zx8
Copy link

zx8 commented Apr 28, 2020

Using 2bc4228


By default, fzf evaluates its executable dynamically (see https://github.com/junegunn/fzf/blob/e6d33f77/shell/key-bindings.fish#L63).

However, this recently stopped working. Running eval (__fzfcmd) or even just eval (echo fzf) seems to hang indefinitely. Hitting Ctrl-C a few times interrupts whatever is hanging and the shell returns to normal, with the following output:

image

@OakNinja
Copy link

OakNinja commented Apr 28, 2020

I was just about to report this, created an issue for jethrokuan/fzf here: jethrokuan/fzf#147

Seems to be related to 3.1.1.

@dive
Copy link

dive commented Apr 28, 2020

It seems that I have the same issue with enhancd in addition to the problem mentioned in the description. Started to happen after the update to 3.1.1.

@zanchey zanchey added the bug Something that's not working as intended label Apr 28, 2020
@zanchey zanchey added this to the fish 3.1.2 milestone Apr 28, 2020
@clason
Copy link

clason commented Apr 28, 2020

I also get similar issues with conda (conda update freezes without output; ctrl-c terminates and shows the previous update).

@ridiculousfish
Copy link
Member

ridiculousfish commented Apr 28, 2020

In hindsight my pushing the eval changes into 3.1.1 may have been a little reckless.

@ridiculousfish ridiculousfish self-assigned this Apr 28, 2020
@OakNinja
Copy link

OakNinja commented Apr 28, 2020

In hindsight my pushing the eval changes into 3.1.1 may have been a little reckless.

Well, I learned a lot while I was still convinced I had totally wrecked my whole environment. It didn’t even work when I created a new user 😁

@dustinfarris
Copy link

dustinfarris commented Apr 28, 2020

I downgraded back to 3.1.0 for now.

@timbedard
Copy link

timbedard commented Apr 28, 2020

For the time being, the Homebrew tap https://github.com/fish-shell/homebrew-fish is still at 3.1b1, for the --devel flag, so that's not a bad option for Homebrew users to downgrade while still letting brew handle the dirty work. That's where I am, since I have to do a coding demo today. Bullet dodged. 😅

@OakNinja
Copy link

OakNinja commented Apr 28, 2020

For the time being, the Homebrew tap https://github.com/fish-shell/homebrew-fish is still at 3.1b1, for the --devel flag, so that's not a bad option for Homebrew users to downgrade while still letting brew handle the dirty work. That's where I am, since I have to do a coding demo today. Bullet dodged. 😅

Wow, thanks! I prefer to stay in brew land on my work machine.

@purge
Copy link

purge commented Apr 28, 2020

FYI, to downgrade to the previous version (which homebrew makes annoyingly hard):
brew install https://raw.githubusercontent.com/Homebrew/homebrew-core/618fcd7b144c0768a01bd4eab8deee960bd86466/Formula/fish.rb

@nicolaturcato
Copy link

nicolaturcato commented Apr 28, 2020

Same issue here. While running a yarn command it doesn't output anything. When I break it with ctrl + c it shows the previous output.
This started happening with 3.1.1 version.

@dustinfarris
Copy link

dustinfarris commented Apr 28, 2020

FYI, to downgrade to the previous version (which homebrew makes annoyingly hard):
brew install https://raw.githubusercontent.com/Homebrew/homebrew-core/618fcd7b144c0768a01bd4eab8deee960bd86466/Formula/fish.rb

don't forget to brew pin as well so homebrew doesn't try to re-upgrade it again later.

ridiculousfish added a commit that referenced this issue Apr 28, 2020
Commit 5fccfd8, with the fix for #6806,
switched eval to buffer its output (like other builtins do). But this
prevents using eval with commands that wants to see the tty, especially
fzf. So only buffer the output if the output is piped to the next process.

This will solve #6955 (which needs to go into a point release).
@fish-shell fish-shell locked as too heated and limited conversation to collaborators Apr 28, 2020
@faho
Copy link
Member

faho commented Apr 28, 2020

Okay, folks, we know about the issue, further discussion isn't all that helpful. I have locked the issue.

@ridiculousfish
Copy link
Member

ridiculousfish commented Apr 28, 2020

This is a serious 3.1.1 regression (mea culpa). I have started Integration_3.1.2 branch with the fix. @zanchey do you have resources to do a 3.1.2 release?

(I originally branched off of master instead of 3.1.1, now fixed.)

@zanchey
Copy link
Member

zanchey commented Apr 28, 2020

Yes, I can do it today.

@zanchey zanchey closed this as completed Apr 29, 2020
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
bug Something that's not working as intended
Projects
None yet
Development

No branches or pull requests