bug#12791: 24.2; An option to load user init file with -batch

bug#12791: 24.2; An option to load user init file with -batch

[Leo]

-batch implies -q and I cannot find an option to negate it. For example,
I want to add some tests to my emacs init files and be able to run
`ert-run-tests-batch' in batch mode.

Any objection if I add an option such as '--load-init-file'?

Leo

bug#12791: 24.2; An option to load user init file with -batch

[Eli Zaretskii]

> From: Leo <sdl.web@gmail.com>
> Date: Sat, 03 Nov 2012 15:54:44 +0800
> 
> -batch implies -q and I cannot find an option to negate it. For example,
> I want to add some tests to my emacs init files and be able to run
> `ert-run-tests-batch' in batch mode.
> 
> Any objection if I add an option such as '--load-init-file'?

What's wrong with "-l ~/.emacs"?  The difference in the loading order
is a non-issue in batch mode anyway.

bug#12791: 24.2; An option to load user init file with -batch

[Leo]

On 2012-11-03 18:02 +0800, Eli Zaretskii wrote:
> What's wrong with "-l ~/.emacs"?  The difference in the loading order
> is a non-issue in batch mode anyway.

not portable.

Leo

bug#12791: 24.2; An option to load user init file with -batch

[Eli Zaretskii]

> From: Leo <sdl.web@gmail.com>
> Date: Sat, 03 Nov 2012 22:41:41 +0800
> 
> On 2012-11-03 18:02 +0800, Eli Zaretskii wrote:
> > What's wrong with "-l ~/.emacs"?  The difference in the loading order
> > is a non-issue in batch mode anyway.
> 
> not portable.

Sorry, I don't understand.  Please elaborate: when does it not work
and why?

bug#12791: 24.2; An option to load user init file with -batch

[Glenn Morris]

Eli Zaretskii wrote:

>> > What's wrong with "-l ~/.emacs"? The difference in the loading
>> > order is a non-issue in batch mode anyway.
>> 
>> not portable.
>
> Sorry, I don't understand.  Please elaborate: when does it not work
> and why?

One possibility: maybe the init file is actually ~/.emacs.el, or
~/.emacs.d/init.el, etc. So the caller would have to do (a small amount
of) actual work to locate it, rather than just letting Emacs handle it
transparently as normal.

(It sure would waste less time if reports were less vague, particularly
when asking for new features to be implemented.)

bug#12791: 24.2; An option to load user init file with -batch

[Lars Ingebrigtsen]

Leo <sdl.web@gmail.com> writes:

> On 2012-11-03 18:02 +0800, Eli Zaretskii wrote:
>> What's wrong with "-l ~/.emacs"?  The difference in the loading order
>> is a non-issue in batch mode anyway.
>
> not portable.

(I'm going through old bug reports that unfortunately weren't resolved
at the time.)

I think people know where their ~/.emacs files are, so "-Q -l ~/.emacs"
covers virtually all use cases.  I don't think adding a --load-init-file
option would be sufficiently useful as a feature (and it's ambiguous
what init file we're referring to), so I'm closing this bug report.

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

bug#12791: 24.2; An option to load user init file with -batch

[Juanma Barranquero]

On Sat, Nov 3, 2012 at 8:54 AM, Leo <sdl.web@gmail.com> wrote:

> -batch implies -q and I cannot find an option to negate it. For example,
> I want to add some tests to my emacs init files and be able to run
> `ert-run-tests-batch' in batch mode.

Another small difference between runing an interactive emacs and doing
"emacs -batch" and loading your .emacs is that under "emacs -batch" it
is possible that the directory pointed out by user-emacs-directory
does not exist, and locate-user-emacs-file quite deliberately takes
steps not to create it (because otherwise you'll end with .emacs.d/
directories at unexpected places, while bootstrapping,etc.)

    Juanma