bug#40731: Followup

[Lars Ingebrigtsen]

Ilya Ostapyshyn <ilya.ostapyshyn@gmail.com> writes:

> I have found a commit ~6fe661342a24edcaea255c3ba9a37613031554da~ which
> addressed the same issue but in a different way.
>
> However, the commit was reverted
> (hash ~f198a5c5144fdded1400df6e8454e4b1b912c7de~) with the following message:
>
>> The alternate editor may be Emacs, which is useful when you want to
>> eval something in an existing Emacs (if it exists), or in a new Emacs
>> if there's no server running.
>
> I beg to differ. In fact, the --eval is not an optarg parameter, instead
> it tells emacsclient to treat the whole FILE argument as lisp
> expressions for evaluation.
>
> Therefore, invoking
>   emacsclient -a emacs -e '(emacs-version)'
> just opens a file named "(emacs-version)" in a new instance of Emacs if
> server is not running.

Hm.  Unfortunately the reverting commit doesn't refer to the test case
that somebody had for this, but there was a complaint after the change
that the change ruined their work flow.

There was a test case for this, but it wasn't included in the reversion
commit, and I don't recall what it was now...

emacsclient -a "emacs -e" -e '(emacs-version)'

or something?

In any case, we can't regress this work flow, so we can't fix this
somewhat illogical way emacsclient works, so I'm closing this bug report.

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