[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#28729: 26.0.60; Incorrect Edebug spec for gv-define-setter
From: |
Lars Ingebrigtsen |
Subject: |
bug#28729: 26.0.60; Incorrect Edebug spec for gv-define-setter |
Date: |
Mon, 15 Jul 2019 11:16:24 +0200 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/27.0.50 (gnu/linux) |
Gemini Lasswell <gazally@runbox.com> writes:
> The Edebug spec for gv-define-setter is incorrect. It should use
> :name, so Edebug doesn't confuse it with the getter, and it should use
> def-body instead of body. To reproduce the problem, put this bit of
> code into *scratch*:
>
> (defun obo-ref (a idx)
> (aref a (1- idx)))
>
> (gv-define-setter obo-ref (val a idx)
> `(aset ,a (1- ,idx) ,val))
>
> (defvar my-array [ 10 20 30 40 ])
>
> (defun set-first-elem (n)
> (setf (obo-ref my-array 1) n))
>
> Then:
>
> M-x edebug-all-defs RET
> M-x eval-buffer RET
>
> Result: Eager macro-expansion failure: (wrong-type-argument consp nil)
I can confirm that changing `body' to `def-body' fixes this bug.
> M-: (obo-ref my-array 3) RET
> SPC SPC
>
> Result: Edebug starts stepping through the setter, then quits with
> "Source has changed - reevaluate definition of obo-ref".
This is till a problem. I'm not sure what you mean by "It should use
:name" -- how does that work in edebug specs? (I'm not particularly
familiar with edebug internals.)
--
(domestic pets only, the antidote for overdose, milk.)
bloggy blog: http://lars.ingebrigtsen.no
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- bug#28729: 26.0.60; Incorrect Edebug spec for gv-define-setter,
Lars Ingebrigtsen <=