bring lisp-eval-defun in line with eval-defun

bring lisp-eval-defun in line with eval-defun

[Sam Steingold]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

eval-defun on defvar resets the variable (as if the evaluated form were
defconst).
lisp-eval-defun does not.
any objection to changing lisp-eval-defun to be in line with eval-defun,
i.e., checking that the form is a defvar and resetting the variable?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFH9ou1Pp1Qsf2qnMcRAlSlAJ9vCqqkuX/ILLu6dB42+Cmd8BD90wCeNrNW
6FMY7dGZSH5prILl53ghZf8=
=2hsJ
-----END PGP SIGNATURE-----

Re: bring lisp-eval-defun in line with eval-defun

[Stefan Monnier]

> eval-defun on defvar resets the variable (as if the evaluated form were
> defconst).
> lisp-eval-defun does not.
> any objection to changing lisp-eval-defun to be in line with eval-defun,
> i.e., checking that the form is a defvar and resetting the variable?

Fine by me, but ... where's the patch?


        Stefan

Re: bring lisp-eval-defun in line with eval-defun

[Sam Steingold]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Stefan Monnier wrote:
|> eval-defun on defvar resets the variable (as if the evaluated form were
|> defconst).
|> lisp-eval-defun does not.
|> any objection to changing lisp-eval-defun to be in line with eval-defun,
|> i.e., checking that the form is a defvar and resetting the variable?
|
| Fine by me, but ... where's the patch?

in the CVS.
do you think this change deserves a NEWS entry?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFH+kWzPp1Qsf2qnMcRAo3DAJkBCgU5a056CIDsUt+5fGy9adeZ3ACdFfte
lssPya11IV1yFdRbwUAf7VE=
=9Zcb
-----END PGP SIGNATURE-----

Re: bring lisp-eval-defun in line with eval-defun

[Stefan Monnier]

> |> eval-defun on defvar resets the variable (as if the evaluated form were
> |> defconst).
> |> lisp-eval-defun does not.
> |> any objection to changing lisp-eval-defun to be in line with eval-defun,
> |> i.e., checking that the form is a defvar and resetting the variable?
> |
> | Fine by me, but ... where's the patch?

> in the CVS.
> do you think this change deserves a NEWS entry?

I don't use lisp-mode, so I don't know how important that is, but if
that change had been made to elisp-mode, I'd want to see it in
etc/NEWS, yes.


        Stefan