Re: It's Huntin' Season
>>"DGMS" == Davide G M Salvetti <salve@debian.org> writes:
DGMS> I, the maintainer, don't want to store the system administrator decision
DGMS> about enabling auctex by default in a configuration file, as I think she
DGMS> has no need to.
You want to affect the behaviour of a program the user
runs. By definition, this is a configurable configuration option.
DGMS> I rather want she to take this decision through debconf; she should be
DGMS> free to do "dpkg-reconfigure auctex" ad any time.
How the configuration file is configured is really irrelevant.
DGMS> What I ask is: how do I tell Emacs to "(require 'tex-site)" on startup
DGMS> if the user answered accordingly to that debconf question?
Emacsen provide a way to set configuration options for emacs
lisp packages. It is called the site-start.d directory mechanism, and
the emacs policy is clear on this.
DGMS> I don't want users to change any file, with regard to this issue. They
DGMS> need not.
Users may decide to change any confiuguration file they wish;
you have to deal with that.
Manoj> If you never change the file once created, you have nothing to
Manoj> worry about. It is only when people are changing files that are
Manoj> already in place in /etc do questions arise.
DGMS> This means that if I need a file with "(require 'tex-site)" I may not
DGMS> put that file in /etc/. Fine.
Wrong. If you wish to have a configuration file that
configures emacs behaviour, it *HAS* to go into /etc. Period.
DGMS> I think you weren't addressing the right question, no matter how clearer
DGMS> you could have been.
Did I do better now?
DGMS> You might tell me where else should I put that file, or what other
DGMS> action should I take to accomplish my above stated goal.
The file _has_ to be under /etc. If you want to use debconf,
it cannot be a conffile. It is, however, a configuration file.
I am beginning to sound like a broken record. If this is still
not clear, please take this offline.
DGMS> User changes WRT this issue, in the opinion of auctex maintainer, should
DGMS> only happen through debconf.
Not your decision to make. Policy is quite clear about this.
MS> You can;t. These are configuration files, and they must be in /etc.
DGMS> Yours Emacs add-on packages ones may well be all configuration files.
DGMS> This one of mine it's not. Really.
`Rubbish.
======================================================================
configuration file
A file that affects the operation of a program, or provides site-
or host-specific information, or otherwise customizes the
behavior of a program. Typically, configuration files are
intended to be modified by the system administrator (if needed or
desired) to conform to local policy or to provide more useful
site-specific behavior.
======================================================================
Hence, since (require 'blah) changes the behaviour of Emacs,
over riding the default handling of TeX files, it is a configuration
file.
DGMS> I hope it's now clear that I see a need for some other scheme
DGMS> to pass lisp form to Emacs on startup which doesn't involve
DGMS> configuration files.
Far from clarifying any such thing, I think it merely points
out the need to actually read policy.
MS> The files are precicely in the place they need to be. These
MS> are configuration files for the package, and hence, as per policy,
MS> they *MUST* be in /etc. Period.
DGMS> I beg to disagree.
DGMS> Configuration files must be in /etc. Files in /etc/ must be
DGMS> configuration files. This is fine with me.
And? What are the reasons behind your disagreemnt? Policy is
crystal clear on this. Your file changes the behaviour of a
program. That means it is a configuration file. Therefore it must
live in /etc/. And user changes in /etc have to be rpeserved. All
this is stated as a _must_ in policy.
DGMS> What I disagree about is the fact, which I think your wordings
DGMS> imply, that emacsen add-on packages can't have some way to pass
DGMS> lisp forms to Emacs on startup other than through configuration
DGMS> files.
Can't do that, by definition of what a configuration file is.
manoj
getting irritated now
--
Before Xerox, five carbons were the maximum extension of anybody's
ego.
Manoj Srivastava <srivasta@debian.org> <http://www.debian.org/%7Esrivasta/>
1024R/C7261095 print CB D9 F4 12 68 07 E4 05 CC 2D 27 12 1D F5 E8 6E
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Reply to: