bug#38370: 26.2; Confusing message 'Hunk already applied' in diff-mode

bug#38370: 26.2; Confusing message 'Hunk already applied' in diff-mode

[ndame]

[-- Attachment #1: Type: text/plain, Size: 880 bytes --]


Recently I turned on next-error-follow-minor-mode in diff buffers, so
the source context is shown automatically as I move around in the diff
buffer.

It works well, but after a while I noticed there is always a 'Hunk
already applied' message in the echo area.

I investigated and turns out the message comes from here:

    next-error
    next-error-function
    diff-next-error
    diff-goto-source
    diff-hunk-status-msg


So goto source prints this message every time, even if one just
presses RET in a diff buffer.

I found it confusing, because I was just moving around, I didn't try
to apply anything, so this message wasn't really informative, it
stopped my flow with its unwanted appearance and reference to an other
operation.

Shouldn't this 'Hunk already applied' message appear only if the user
actually tries to apply something?

[-- Attachment #2: Type: text/html, Size: 1078 bytes --]

bug#38370: 26.2; Confusing message 'Hunk already applied' in diff-mode

[Lars Ingebrigtsen]

ndame <emacsuser@freemail.hu> writes:

> Recently I turned on next-error-follow-minor-mode in diff buffers, so
> the source context is shown automatically as I move around in the diff
> buffer.
>
> It works well, but after a while I noticed there is always a 'Hunk
> already applied' message in the echo area.

[...]

> I found it confusing, because I was just moving around, I didn't try
> to apply anything, so this message wasn't really informative, it
> stopped my flow with its unwanted appearance and reference to an other
> operation.
>
> Shouldn't this 'Hunk already applied' message appear only if the user
> actually tries to apply something?

Yeah, I've always found that messaging to be confusing and worrisome, as
if I had entered a command to change the source code instead of just
moving around in it.

I've now removed it.

Looking at the code in diff-mode, as far as I can see, the actual
"apply" functions will still output the hunk status.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#38370: 26.2; Confusing message 'Hunk already applied' in diff-mode

[Stefan Monnier]

> Yeah, I've always found that messaging to be confusing and worrisome, as
> if I had entered a command to change the source code instead of just
> moving around in it.

Hmm... I liked having that information without having to remember how to
run `diff-test-hunk`.  I agree it was more harmful than useful when moving
around as in the OP's case, but it was handy when using RET.


        Stefan

bug#38370: 26.2; Confusing message 'Hunk already applied' in diff-mode

[Lars Ingebrigtsen]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>> Yeah, I've always found that messaging to be confusing and worrisome, as
>> if I had entered a command to change the source code instead of just
>> moving around in it.
>
> Hmm... I liked having that information without having to remember how to
> run `diff-test-hunk`.  I agree it was more harmful than useful when moving
> around as in the OP's case, but it was handy when using RET.

Hm...  perhaps we should just change the message instead so it doesn't
sound so much like a warning?  For instance...  Uhm...  No, I can't
think of anything short that can't be interpreted as warning right
now...

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#38370: 26.2; Confusing message 'Hunk already applied' in diff-mode

[Juri Linkov]

>>> Yeah, I've always found that messaging to be confusing and worrisome, as
>>> if I had entered a command to change the source code instead of just
>>> moving around in it.
>>
>> Hmm... I liked having that information without having to remember how to
>> run `diff-test-hunk`.  I agree it was more harmful than useful when moving
>> around as in the OP's case, but it was handy when using RET.
>
> Hm...  perhaps we should just change the message instead so it doesn't
> sound so much like a warning?  For instance...  Uhm...  No, I can't
> think of anything short that can't be interpreted as warning right
> now...

Maybe add this message to the default value of 'ignored-messages'
from bug#42865.  Then it could be easily removed with

  (delete-from-list 'ignored-messages "^Hunk already applied")