Re: [Emacs-diffs] scratch/gnus-decoded 548c778: Don't add encoding cookie in Gnus' active files

Re: [Emacs-diffs] scratch/gnus-decoded 548c778: Don't add encoding cookie in Gnus' active files

[Stefan Monnier]

>     * lisp/gnus/gnus-util.el (gnus-write-active-file): This doesn't do
>       what I thought it would do.
[...]
> -      (insert (format ";; -*- encoding: %s; -*-\n\n" coding-system-for-write))

Not sure what you expect it to do, but the above cookie indeed likely
doesn't do anything at all.  If you change `encoding` to `coding` then
it should have the effect of telling Emacs which encoding to use when
reading the file.


        Stefan

Re: [Emacs-diffs] scratch/gnus-decoded 548c778: Don't add encoding cookie in Gnus' active files

[Eric Abrahamsen]

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

>>     * lisp/gnus/gnus-util.el (gnus-write-active-file): This doesn't do
>>       what I thought it would do.
> [...]
>> -      (insert (format ";; -*- encoding: %s; -*-\n\n" coding-system-for-write))
>
> Not sure what you expect it to do, but the above cookie indeed likely
> doesn't do anything at all.  If you change `encoding` to `coding` then
> it should have the effect of telling Emacs which encoding to use when
> reading the file.

Well that's embarrassing! I was wrong on two levels. It was later
pointed out to me (this is in the later discussion on #35383) that the
cookie wouldn't even influence how Emacs read the file in this case, so
there wasn't much point. Right now the code assumes that it will be able
to read files written as 'raw-text using 'utf-8-emacs.