|
From: | Dmitry Gutov |
Subject: | bug#66211: 30.0.50; C-x v v in vc-diff does not work with lesser versionf of patch |
Date: | Fri, 29 Sep 2023 21:11:26 +0300 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.13.0 |
On 29/09/2023 20:40, Eli Zaretskii wrote:
Date: Fri, 29 Sep 2023 20:27:21 +0300 Cc: ams@gnu.org, 66211@debbugs.gnu.org From: Dmitry Gutov <dmitry@gutov.dev> On 29/09/2023 18:22, Eli Zaretskii wrote:vc-default-checkin-patch calls 'patch'. It's for other backends.Why do we call Patch if we have a way to apply hunks in Lisp? Wasn't that just discussed in the other thread?https://lists.gnu.org/archive/html/emacs-devel/2022-10/msg01353.htmlThat doesn't have a lot in the way of answering my question above.
The discussion regarding that choice is inside the quotes' ladder in that message.
Anyway, it doesn't seem like 'patch' itself is the problem in this case:When Patch is installed, sure. But using our own code is better because it doesn't need Patch and doesn't have to rely on specific options it may or may not support. I'm not insisting, mind you, just asking. It sounds inconsistent with what we do in similar situations in other places, that's all.
My preference was to limit the number of points of unreliability. And/or minor differences (e.g. whether file removals are supported).
'diff-apply-buffer' has only been checked in, we can revisit this a little later, after more testing and stuff.
[Prev in Thread] | Current Thread | [Next in Thread] |