Inexplicable flaw: non-overridable .emacs

Inexplicable flaw: non-overridable .emacs

[kj]

Please correct me if I'm wrong, it is my understanding that one
cannot use a command-line switch to tell emacs to read a config
file *other* than the standard ~/.emacs.

If this is correct, I seems to me an inexplicable design flaw.  Is
there some fundamental reason for it?

Of course, if my understanding is incorrect, please toss me a
cluebrick!

TIA!

kj

-- 
NOTE: In my address everything before the first period is backwards;
and the last period, and everything after it, should be discarded.

Re: Inexplicable flaw: non-overridable .emacs

[David Kastrup]

kj <socyl@987jk.com.invalid> writes:

> Please correct me if I'm wrong, it is my understanding that one
> cannot use a command-line switch to tell emacs to read a config
> file *other* than the standard ~/.emacs.
>
> If this is correct, I seems to me an inexplicable design flaw.  Is
> there some fundamental reason for it?
>
> Of course, if my understanding is incorrect, please toss me a
> cluebrick!

With the option -q it does not read any .emacs file, and with the
option -l somefile.el it reads a different file.  With the option -u
username it reads the startup file of the given username.

Did you actually read the info page concerning startup options?

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Inexplicable flaw: non-overridable .emacs

[Maarten Bergvelt]

In article <etc2fj$7dl$1@reader2.panix.com>, kj wrote:
> Please correct me if I'm wrong, it is my understanding that one
> cannot use a command-line switch to tell emacs to read a config
> file *other* than the standard ~/.emacs.
> 
> If this is correct, I seems to me an inexplicable design flaw.  Is
> there some fundamental reason for it?

There was some discussion about this in gnu.emacs.help last month, see
the thread "change location of .emacs" especially the postings by 
Eli Zaretskii. 

I don't understand why people do object to using ~/.emacs, though.

-- 
Maarten Bergvelt		

Re: Inexplicable flaw: non-overridable .emacs

[Eli Zaretskii]

> From: kj <socyl@987jk.com.invalid>
> Date: Thu, 15 Mar 2007 18:15:15 +0000 (UTC)
> 
> Please correct me if I'm wrong, it is my understanding that one
> cannot use a command-line switch to tell emacs to read a config
> file *other* than the standard ~/.emacs.

You are only partly wrong: Emacs has a --load command-line switch that
would load any file you name.  But such loading is not 100% equivalent
to how Emacs loads a .emacs file, because .emacs is read at a certain
point during the Emacs session startup, while files given via --load
are read at a different point.  So the effects could be subtly
different, especially with respect to display setup.

> If this is correct, I seems to me an inexplicable design flaw.  Is
> there some fundamental reason for it?

Unix programs that support .something initialization files all have
this behavior: they read the init file of the current user.  Emacs
just behaves like Unix users expect it to.

Could you please tell why you don't want to have a .emacs file?

Re: Inexplicable flaw: non-overridable .emacs

[kj]

In <mailman.972.1173989687.7795.help-gnu-emacs@gnu.org> Eli Zaretskii <eliz@gnu.org> writes:

>> From: kj <socyl@987jk.com.invalid>
>> Date: Thu, 15 Mar 2007 18:15:15 +0000 (UTC)
>> 
>> Please correct me if I'm wrong, it is my understanding that one
>> cannot use a command-line switch to tell emacs to read a config
>> file *other* than the standard ~/.emacs.

>You are only partly wrong: Emacs has a --load command-line switch that
>would load any file you name.  But such loading is not 100% equivalent
>to how Emacs loads a .emacs file, because .emacs is read at a certain
>point during the Emacs session startup, while files given via --load
>are read at a different point.  So the effects could be subtly
>different, especially with respect to display setup.

That's good to know, thanks, but, again, I'm puzzled: why would
Emacs treat ~/.emacs differently from a user-specified alternative?
It seems perversely unaccommodating on Emacs' part (even if only
in a passive-agressive sort of way :)  )

>Could you please tell why you don't want to have a .emacs file?

Nothing against an .emacs file, but occasionally I want to run
Emacs with a drastically different set of customizations from what's
in my .emacs file.  Since a lot of these customizations pertain
display, I need to do some experimentation to determine whether
the --load option will do what I need. 

kj
-- 
NOTE: In my address everything before the first period is backwards;
and the last period, and everything after it, should be discarded.

Re: Inexplicable flaw: non-overridable .emacs

[Joost Kremers]

kj wrote:
> Nothing against an .emacs file, but occasionally I want to run
> Emacs with a drastically different set of customizations from what's
> in my .emacs file.

if that is what you're after, you may be able to put an if-statement in
your .emacs to test for the special circumstances in which you need to use
this different set of customisations. if these are circumstances that emacs
itself cannot test for, you can always call emacs with something like

emacs --eval "(setq load-my-other-dot-emacs t)"

and test for the value of load-my-other-dot-emacs.

-- 
Joost Kremers                                      joostkremers@yahoo.com
Selbst in die Unterwelt dringt durch Spalten Licht
EN:SiS(9)

Re: Inexplicable flaw: non-overridable .emacs

[Kevin Rodgers]

kj wrote:
> In <mailman.972.1173989687.7795.help-gnu-emacs@gnu.org> Eli Zaretskii <eliz@gnu.org> writes:
> 
>>> From: kj <socyl@987jk.com.invalid>
>>> Date: Thu, 15 Mar 2007 18:15:15 +0000 (UTC)
>>>
>>> Please correct me if I'm wrong, it is my understanding that one
>>> cannot use a command-line switch to tell emacs to read a config
>>> file *other* than the standard ~/.emacs.
> 
>> You are only partly wrong: Emacs has a --load command-line switch that
>> would load any file you name.  But such loading is not 100% equivalent
>> to how Emacs loads a .emacs file, because .emacs is read at a certain
>> point during the Emacs session startup, while files given via --load
>> are read at a different point.  So the effects could be subtly
>> different, especially with respect to display setup.
> 
> That's good to know, thanks, but, again, I'm puzzled: why would
> Emacs treat ~/.emacs differently from a user-specified alternative?

Because ~/.emacs is loaded between default.el and site-start.el, after
the so-called initial command line options are processed but before the
so-called action options.  If your installation doesn't have a
site-start.el file, you should be OK -- modulo display setup, as Eli
points out.

> It seems perversely unaccommodating on Emacs' part (even if only
> in a passive-agressive sort of way :)  )

-- 
Kevin Rodgers
Denver, Colorado, USA

Re: Inexplicable flaw: non-overridable .emacs

[David Hansen]

On Sat, 17 Mar 2007 13:43:58 +0000 (UTC) kj wrote:

> Nothing against an .emacs file, but occasionally I want to run
> Emacs with a drastically different set of customizations from what's
> in my .emacs file.  Since a lot of these customizations pertain
> display, I need to do some experimentation to determine whether
> the --load option will do what I need.

What always works with all UNIX programs is

$ HOME=$OTHERHOME emacs

Just create $OTHERHOME/.emacs and this will be loaded.

David