[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#61310: Eshell modifying and running output regression
From: |
Jim Porter |
Subject: |
bug#61310: Eshell modifying and running output regression |
Date: |
Sun, 5 Feb 2023 21:51:45 -0800 |
On 2/5/2023 8:18 PM, Morgan Smith wrote:
eshell allows you to modify the output of a command and run it. It is a
beautiful thing.
Interesting. I didn't realize this was possible in Eshell (I only
thought you could re-run old *inputs*).
I don't know much about fields but it looks like it is not necessary to
have a command-output field at all. I propose we get rid of that. If
you want to keep the field, can we make it sticky or something?
The output field is actually necessary (or else Eshell would need to be
cleverer about some things). The main issue is that if a command doesn't
output a newline, the command's output can end up on the same line as
the prompt:
~ $ *echo -n [output]
[output]~ $
If the output had no field, C-a would move to the very beginning of the
line, not to the beginning of the input field. Maybe this is a bug in
how fields are handled, but changing field handling in general is
probably too risky.
So instead, let's make the output field sticky as you say. Here's a fix
for that plus regression tests so this won't break in the future.
0001-Ensure-that-Eshell-users-can-run-lines-of-command-ou.patch
Description: Text document