[ESS-bugs] Variable name affects indentation

[Martin Maechler]

On Sat, Aug 18, 2012 at 11:27 AM, Vitalie Spinu <spinuvit at gmail.com> wrote:
>   >> Martin Maechler <maechler at stat.math.ethz.ch>
>   >> on Sat, 18 Aug 2012 11:13:51 +0200 wrote:
>
>   >            ;; modified by shiba at isac 7.3.1992
>   >            (cond ((and (numberp ess-expression-offset)
>   >                        (re-search-backward "[ \t]*expression[ \t]*" bol t))
>   >                   ;; This regexp match every "expression".
>   >                   ;; modified by shiba
>   >                   ;;(forward-sexp -1)
>   >                   (beginning-of-line)
>   >                   (skip-chars-forward " \t")
>   >                   ;; End
>   >                   (+ (current-column) ess-expression-offset))
>
>   > and it is clearly why it leads to the wrong indentation here.
>
> But, cannot we just remove that? Why would we treat "expression"
> differently from other functions? There are other functions with the
> same semantics (quote, bquote, substitute) and we are not treating them
> separately.

they have a very different semantic:  expression is very much  list( ...)

and the other reason to not remove it is back compatibility.
Really, Vitalie!
Just the fact that you  (or one or two people on the ESS-help list or
github threads) don't use a feature or don't particularly find it
compelling,
does not mean that it should be removed.

We maintain one of the most long term and stable open source projects
in statistics,
probably *the* most long term...
and we gain new users by nice features and using new Emacs features
and facilities as those that you provided or contributed to... thank
you very much ..
BUT we will lose or at least irritate long time users, by changing
behavior that just a few people
have sought would be better than the decennial old, often subtly
optimized (or at least compromized between conflicting targets or
tastes), behavior  that we have been endearing.

Apropos: New bug report (of new behavior):

M-$ (ispell-word)  inside a simple word in  *R*  now fails
with
        Invalid function: ("^<<.*>>=" . "^@")

Martin