bug#23585: 25.0.93; Win32: spurious ^M pop up at byte offset 65536

bug#23585: 25.0.93; Win32: spurious ^M pop up at byte offset 65536

[Anders Lindgren]

[-- Attachment #1: Type: text/plain, Size: 644 bytes --]

Hi!

I've noticed that sometimes an extra carriage return character pops up in
my source files, when running Emacs under Windows. They are always located
at the end of a line. I use auto-revert mode and I've only seen this in
files already open in Emacs.

Today I noticed that one such carriage return character was placed at
buffer position 62879 in line 2658. This corresponds to byte position 65536
(or hex 0x10000) in the file. Clearly, this can't be a coincidence.

By the way, I'm not sure Emacs is the culprit here -- I use subversion as a
version control system and Visual Studio for debugging and sometimes minor
edits.

    -- Anders

[-- Attachment #2: Type: text/html, Size: 803 bytes --]

bug#23585: 25.0.93; Win32: spurious ^M pop up at byte offset 65536

[Eli Zaretskii]

> Date: Fri, 20 May 2016 12:16:51 +0200
> From: Anders Lindgren <andlind@gmail.com>
> 
> I've noticed that sometimes an extra carriage return character pops up in my source files, when running
> Emacs under Windows. They are always located at the end of a line. I use auto-revert mode and I've only
> seen this in files already open in Emacs.
> 
> Today I noticed that one such carriage return character was placed at buffer position 62879 in line 2658. This
> corresponds to byte position 65536 (or hex 0x10000) in the file. Clearly, this can't be a coincidence.
> 
> By the way, I'm not sure Emacs is the culprit here -- I use subversion as a version control system and Visual
> Studio for debugging and sometimes minor edits.

If you can come up with a reproducible recipe, we could debug this.

Thanks.

bug#23585: 25.0.93; Win32: spurious ^M pop up at byte offset 65536

[Anders Lindgren]

[-- Attachment #1: Type: text/plain, Size: 581 bytes --]

>
> If you can come up with a reproducible recipe, we could debug this.
>
>
I wish I could. Whoever, they seem to pop up about once every few weeks,
and I only discover them after I've continued working with the file. The
main reason why I posted the bug without a recipe was to see if anybody
else had seen anything similar, or if anybody would have a hunch on what
the cause is.

Of course, one could try to provoke this by creating and appending to files
with a carriage return/newline at the critical location and see if Emacs
loads or reverts them incorrectly.

    -- Anders

[-- Attachment #2: Type: text/html, Size: 876 bytes --]

bug#23585: 25.0.93; Win32: spurious ^M pop up at byte offset 65536

[Glenn Morris]

Without a recipe, I can't see this going anywhere, so closing.