|
From: | J.P. |
Subject: | bug#64855: 30.0.50; ERC 5.6: Make scrolltobottom less erratic |
Date: | Wed, 13 Sep 2023 07:05:52 -0700 |
User-agent: | Gnus/5.13 (Gnus v5.13) |
"J.P." <jp@neverwas.me> writes: > v5. Make new behavior hinge on new option, disabled by default. Assign > some module functions explicit hook depths. Add news and tests. I've added a version of this as commit 617ddb808999a71c925b68f5369d77aebfcd9254 Consider all windows in erc-scrolltobottom-mode But it comes with a few known problems. The first can be observed during the rapid insertion of newly arriving messages, like you get with history playback or large swaths of help text. Basically, when point is away from the prompt and you issue a command that changes `window-start' significantly (e.g., M-<), point hardly moves at all, maybe half a screenful at most. Another issue is only noticeable if you have code running that suppresses message insertion after prompt submissions (see bug#49860's `echo-message' for one example). Basically, if you've got `erc-scrolltobottom-relaxed' enabled and you stare long enough, you'll notice that the prompt drifts downward with every submitted round of multi-line input. Attached is a patch that attempts to address both of these issues, along with a couple more to get at some unrelated odds and ends. I'm also sort of thinking we ought to temporarily change the default of the new option `erc-scrolltobottom-all' to t for a few weeks to help flush out any other glaring bugs introduced by this feature. If anyone thinks that's a bad idea, please say so. Thanks.
0001-doc-misc-erc.texi-Fix-display-buffer-example.patch
Description: Text Data
0002-5.6-Simplify-erc-fill-module-docstring.patch
Description: Text Data
0003-5.6-Skip-erc-ignored-user-p-when-erc-ignore-list-is-.patch
Description: Text Data
0004-5.6-Run-erc-scrolltobottom-on-pre-insert-uncondition.patch
Description: Text Data
[Prev in Thread] | Current Thread | [Next in Thread] |