Eli Zaretskii <eliz@gnu.org> writes:

Hi Eli & Jordan,

>> As you see, it requires two changes, in esh-proc.el and tramp.el.
>
> I think both can be cherry-picked to emacs-26.

Done. The change in tramp.el needed some massage.

Jordan, could you pls test whether the changes I have applied to the
emacs-26 branch work for you? See attachment for the changes.

> But do you know why we need expand-file-name here at all?

Well, in Emacs 26 we couldn't adapt exec-path for remote
processes. This is available since Emacs 27 only.

> At the very least, the comment about start-process should be removed,
> I think.

Done.

>> > Btw, isn't it confusing that start-file-process needs only the "local"
>> > part of COMMAND?  Why cannot its handler DTRT internally instead?
>> 
>> What do you do, if COMMAND is another remote location than
>> default-directory?
>
> Why is that a problem?  We always run the process in
> default-directory, right?

I said "another *remote* location". I don't want to see

(let ((default-directory "/ssh:user@host:directory"))
  (start-file-process "foo" buffer
  "/ssh:another-user@another-host:foo"))

Of course one could raise an error. But usually, this construct doesn't
appear as direct as in the example.

I believe using a local name for PROGRAM isn't such a burden.

>> And more general, there could also be file names in the PROGRAM-ARGS
>> part of start-file-process. Who decides, which of them is a remote file
>> name to be stripped to the local part, and which offers remote file name
>> syntax to be used literally?
>
> the caller, of course.  But I'm not asking about arguments, I'm asking
> about PROGRAM, and only about it.

See above. I don't believe it would make the life easier for callers of
`start-file-process', if we would allow a remote file name for PROGRAM. 

>> That's why we have decided (long ago), to not allow remote arguments for
>> both start-file-process and process-file.
>
> At the very least, this should be prominently mentioned in the
> respective doc strings and in the ELisp manual.  As written now, this
> is entirely undocumented.  Moreover, the part about "the local part of
> default-directory" in the doc string and in the manual is confusing,
> because we have no description of what that means.  The only attempt
> of describing it, in file-local-name's doc string, viz.:
>
>   It returns a file name which can be used directly as argument of
>   ‘process-file’, ‘start-file-process’, or ‘shell-command’.
>
> is IMO unsatisfactory, because it describes how results could be used,
> not what they are.

The docstring speaks about.

--8<---------------cut here---------------start------------->8---
PROGRAM and PROGRAM-ARGS might be file names.  They are not
objects of file name handler invocation.
--8<---------------cut here---------------end--------------->8---

The Elisp manual speaks about. (info "(elisp) Asynchronous Processes")

--8<---------------cut here---------------start------------->8---
This function does not try to invoke file name handlers for PROGRAM
or for the rest of ARGS.
--8<---------------cut here---------------end--------------->8---

Best regards, Michael.