On 15 Dec 2015 2:23 pm, "David Kastrup" <dak@gnu.org> wrote:
>
> Sure.
>
> But that commit is not in that class.
>
> -(defconst dir-locals-file ".dir-locals.el"
> +(defconst dir-locals-file ".dir-locals*.el"

I considered this part of the same feature. But I agree it could have been
made on a follow up commit.

> -       (message ".dir-locals error: %s" (error-message-string err))
> +       (message "%s error: %s" dir-locals-file (error-message-string
err))
>
> is a bug fix that is independent from the implementation.

True!

>        (when (and (string-prefix-p (car elt) file
> -                                 (memq system-type
> -                                       '(windows-nt cygwin ms-dos)))
> -                (> (length (car elt)) (length (car dir-elt))))
> -       (setq dir-elt elt)))
> +                                  (memq system-type
> +                                        '(windows-nt cygwin ms-dos)))
> +                 (> (length (car elt)) (length (car dir-elt))))
> +        (setq dir-elt elt)))
>      (if (and dir-elt
>
> is a gratuitous spacing change.

Oops!

> So is most of
>
>      (if (and dir-elt
> -            (or (null locals-file)
> -                (<= (length (file-name-directory locals-file))
> -                    (length (car dir-elt)))))

Hm, I see what you mean. I generally trust emacs to not change
tabs-to-spaces in lines I didn't touch. Maybe the indentation there was
actually wrong (I'm reading this on a variable space font, so I can't
tell), or maybe I changed something that changed the indentation and then
changed it back again.

Anyway, I usually lookout for this before pushing, but in this case the
indentation changes are quite mixed with code changes so it probably
escaped me. Sorry about that.