Re: latex-mode hook?

[Emanuel Berg]

Marcin Borkowski <mbork@wmi.amu.edu.pl> writes:

>>> and LaTeX-mode is the AUCTeX's one > (which is
>>> much, much better think autocompletion
>>  I I think I think not
>
> C-c C-e enum TAB RET
>
> and you get
>
> \begin{enumerate} (point) \end{enumerate}

OK, well, that is more of a snippet expansion feature
than autocompletion to my mind. With autocompletion I
mean, say that you want to search for "injury" - when
you type "inj" "injustice" appears. I can be even worse
with a whole list of suggestions (I take it based on a
word list and/or user behavior/probability theory).
That's just disruptive to me - I know what I want to
type, so let me just type it with no visual noise or
"injustices" ever entering my brain.

> How is that not a win? And it even understands your
> \newcommand and \newenvironment invocations.

No, I can see the benefit of that if you are a LaTeX
professional or do it often. In my case it wouldn't be
such a gain because I seldom use LaTeX and when I do it
is for big, "slow" projects. I don't mind taking the
extra time thinking and typing. For a shell language
like Perl or zsh or Python perhaps something like that
would appeal to me, I don't know. (I'm very fond of
typing.)

> texmathp (as the name implies?) returns t if the
> point is in math mode and nil otherwise. It knows
> about $...$, \(...\), \[...\], \begin{equation} ...
> \end{equation}, it handles $...$ inside \text{...}
> inside \[...\] etc. correctly... See here:
> http://mbork.pl/?search=texmathp for two possible
> applications.
>
> In AUCTeX, if I have \usepackage[cp1250]{inputenc}
> near the beginning, the buffer is automatically put
> in the respective encoding. It's not a big deal in
> English-speaking countries, I guess, but here in
> Poland, where Windoze uses CP-1250, old Linux files
> use ISO-8859-2 and everyone else uses UTF-8, it's
> /very/ convenient.

It is a bit difficult following those scenarios to me
but that sounds useful, yes.

-- 
underground experts united