Re: fringe arrow conflict between compile and gud?

[Juri Linkov]

> There could be a problem at the lisp level though as
> overlay-arrow-string is nil unless it has been set explicitly by some
> package.  Perhaps it would make sense to initialize the global value
> of overlay-arrow-string to "=>" which most packages want?
>
> Can you produce a simple, self-contained test case demonstrating the
> problem?

(progn
  (goto-char (point-min))
  (insert "abc\ndef\nghi\n")
  (setq overlay-arrow-position
        (copy-marker (save-excursion (forward-line -1) (point)))
        next-error-overlay-arrow-position
        (copy-marker (save-excursion (forward-line -2) (point)))
        overlay-arrow-variable-list
        '(next-error-overlay-arrow-position overlay-arrow-position)))

No one arrow is visible at this point.  Now evaluate

(setq overlay-arrow-string "=>")

and both arrows are visible.

So there is no need to do

(put 'next-error-overlay-arrow-position 'overlay-arrow-string "=>")

when the global value of overlay-arrow-string is "=>".

I agree that it makes sense to initialize the global value
of overlay-arrow-string to "=>" by default.

-- 
Juri Linkov
http://www.jurta.org/emacs/