I take it back, that override doesn't fully prevent it after all. 😢 On Mon, Dec 5, 2022 at 6:57 AM Ryan B wrote: > Confirmed, probably the same root cause. When I brute force override > python-nav-end-of-statement like below, it doesn't happen any more. > > (defun python-nav-end-of-statement (&optional noend) > (interactive "^") > (forward-line 1)) > > > On Mon, Dec 5, 2022 at 6:44 AM Ryan B wrote: > >> I think I have a repro for this with a single quote ' , as opposed to >> triple string. I think I started seeing this hang after upgrading to Emacs >> 28.2. I now hit it very often, it's pretty painful. I don't know for sure >> that it's the same bug, but it seems likely. Reproduces with emacs -Q: >> >> ====== >> class Foo(): >> >> def __init__(self): >> ' >> >> def bar(self): >> """Fetches posts and converts them to ActivityStreams activities. >> See source.Source.get_activities_response for details. app_id is >> ignored. min_id is translated to Twitter's since_id. >> """ >> pass >> ====== >> >> The single quote and multiple lines in the docstring are necessary for >> the repro. No extra newline at the end needed though. >> >> JD, do you have a workaround for this? I may look into overriding >> python-nav-end-of-defun until it's fixed. Any other ideas? >> >> My emacs-version: GNU Emacs 28.2 (build 1, aarch64-apple-darwin21.1.0, NS >> appkit-2113.00 Version 12.0.1 (Build 21A559)) of 2022-09-12 >> >> -- >> https://snarfed.org/ >> > > > -- > https://snarfed.org/ > -- https://snarfed.org/