bug#33054: 25.2; diff-model: M-k kills wrong hunk

bug#33054: 25.2; diff-model: M-k kills wrong hunk

[Dima Kogan]

[-- Attachment #1: Type: text/plain, Size: 434 bytes --]

I'm using a very recent build: e4896856172

Steps to reproduce bug:

- Open attached diff. Emacs should open it in diff-mode
- Put point on 'd' in the second "diff": (goto-char 98)
- M-k

This should kill the hunk we're on. We are unambiguously sitting on the
second hunk, but the latest diff-mode is confused, thinks we're on the
first hunk, and kills that instead. Kinda. Moving the point even one
character forward makes it work.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: bug.patch --]
[-- Type: text/x-diff, Size: 229 bytes --]

diff --git a/111.py b/111.py
--- a/111.py
+++ b/111.py
@@ -154,3 +154,3 @@
 
-    aaa
+    bbb
 
diff --git a/222.py b/222.py
--- a/222.py
+++ b/222.py
@@ -224,3 +230,3 @@ writemodel(m_to)
     asdf
-    21234
+    345
     zxcv

bug#33054: 25.2; diff-model: M-k kills wrong hunk

[Alan Mackenzie]

Hello, Dima:

In article <mailman.2215.1539636970.1284.bug-gnu-emacs@gnu.org> you wrote:
> [-- text/plain, encoding 7bit, charset: US-ASCII, 14 lines --]

> I'm using a very recent build: e4896856172

Very recent?  That was from September 8th, over a month ago.  ;-)

> Steps to reproduce bug:

> - Open attached diff. Emacs should open it in diff-mode
> - Put point on 'd' in the second "diff": (goto-char 98)
> - M-k

> This should kill the hunk we're on.

It does that for me.

> We are unambiguously sitting on the second hunk, but the latest
> diff-mode is confused, thinks we're on the first hunk, and kills that
> instead. Kinda. Moving the point even one character forward makes it
> work.

It works for me anyway (With a build updated at most a day ago).  I've
looked at the git log for diff-mode.el, and nobody's changed anything
relevant, that I can see, since early September.

What's your environment?  Operating system, and so on?

Or, maybe something really wierd is happening, and it would stop
happening on an up to date snapshot of Emacs.

-- 
Alan Mackenzie (Nuremberg, Germany).

bug#33054: 25.2; diff-model: M-k kills wrong hunk

[dima]

Alan Mackenzie <acm@muc.de> writes:

>> Steps to reproduce bug:
>
>> - Open attached diff. Emacs should open it in diff-mode
>> - Put point on 'd' in the second "diff": (goto-char 98)
>> - M-k
>
>> This should kill the hunk we're on.
>
> It does that for me.

Hi Alan. Thanks for the response. I was consistently seeing the problem,
but poked around a bit more after seeing your comment. I found the
issue: I was doing something wrong (specifics completely uninteresting).
There is no issue, and I'm closing the bug. Sorry for the noise.

bug#33054: 25.2; diff-model: M-k kills wrong hunk

[Alan Mackenzie]

Hello, Dima

On Tue, Oct 16, 2018 at 12:06:10 -0700, dima@secretsauce.net wrote:
> Alan Mackenzie <acm@muc.de> writes:

> >> Steps to reproduce bug:

> >> - Open attached diff. Emacs should open it in diff-mode
> >> - Put point on 'd' in the second "diff": (goto-char 98)
> >> - M-k

> >> This should kill the hunk we're on.

> > It does that for me.

> Hi Alan. Thanks for the response. I was consistently seeing the problem,
> but poked around a bit more after seeing your comment. I found the
> issue: I was doing something wrong (specifics completely uninteresting).
> There is no issue, and I'm closing the bug. Sorry for the noise.

That's OK, glad you got it sorted out!

-- 
Alan Mackenzie (Nuremberg, Germany).