Here's the original offender (and fix): diff of cl-ppcre-2.0.1/util.lisp:45,46 (defmacro with-unique-names ((&rest bindings) &body body) - "Syntax: WITH-UNIQUE-NAMES ( { var | (var x) }* ) declaration* form* + "Syntax WITH-UNIQUE-NAMES ( { var | (var x) }* ) declaration* form*
Steps to reproduce: * load ECL in Slime * evaluate a form like the following (defmacro break-slime (&rest args) "Syntax: this-is-bad" (list args)) * Slime complains "THIS-IS-BAD is not of type CONS. [Condition of type TYPE-ERROR]"
This is caused by grovel-docstring-for-arglist in slime/swank-ecl.lisp.
By bad luck, Edi happened to use a docstring which *almost* matched the format... Slime/swank should handle it more gracefully.
Later, Daniel
P.S. After tracking that down, I updated to the latest Slime CVS; but slime broke. Another bug report...
On Mon, 29 Dec 2008, Daniel Herring wrote:
Steps to reproduce:
- load ECL in Slime
- evaluate a form like the following
(defmacro break-slime (&rest args) "Syntax: this-is-bad" (list args))
- Slime complains
"THIS-IS-BAD is not of type CONS. [Condition of type TYPE-ERROR]"
Slime patch attached. Please apply; it is a showstopper bug when encountered.
Thanks, Daniel
Applied. Thanks.
On Thu, Jan 29, 2009 at 10:53 PM, Daniel Herring dherring@tentpost.comwrote:
On Mon, 29 Dec 2008, Daniel Herring wrote:
Steps to reproduce:
- load ECL in Slime
- evaluate a form like the following
(defmacro break-slime (&rest args) "Syntax: this-is-bad" (list args))
- Slime complains
"THIS-IS-BAD is not of type CONS. [Condition of type TYPE-ERROR]"
Slime patch attached. Please apply; it is a showstopper bug when encountered.
Thanks, Daniel _______________________________________________ slime-devel site list slime-devel@common-lisp.net http://common-lisp.net/mailman/listinfo/slime-devel
-
Daniel Herring -
Geo Carncross