[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [AUCTeX-devel] Re: [AUCTeX-commit] auctex tex.el
From: |
Ralf Angeli |
Subject: |
Re: [AUCTeX-devel] Re: [AUCTeX-commit] auctex tex.el |
Date: |
Tue, 07 Feb 2006 16:40:50 +0100 |
* David Kastrup (2006-02-07) writes:
>>> Anyway, why use a string here as argument for TeX-run-function? Seems
>>> ugly, when actually a list is used.
>>
>> Because it's pumped through `TeX-command-expand'. By using a string
>> we don't have to check for the variable's type in `TeX-command'.
>
> I am not really convinced that is a good reason... Engaging the Lisp
> reader for normal operation seems awkward.
Using a string looked less perverted to me than using a list. Problem
is that we are introducing a dependency between items in
`TeX-command-list'. All the "old" functions which can be specified at
the third position of a command list item expect a string. Allowing a
list instead of a string at the second position of a command list item
without looking at what function is used at the third position looks
problematic to me.
>> And because that way we don't have to insert a choice in the
>> defcustom for `TeX-command-list'.
>
> Is it really worth that?
Apart from the concerns mentioned above, `TeX-command-list' is already
_very_ crowded. I'd like to avoid more confusing options if possible.
I already thought about changing it to a defvar and advising users to
generally use `add-to-list' for augmenting the variable.
I know my approach looks peculiar but I still think it's less
problematic than the alternative.
--
Ralf