[Openmcl-devel] setting up Clozure CL on Windows - XEmacs/SLIME

[Gary Byers]

On Wed, 14 Oct 2009, Mark H. David wrote:

> Hi.  I'm trying to set up SLIME with Clozure CL on Windows with Xemacs.
>
>
> I think it's set up right, but it's obviously not, since when I do m-x
> slime, I get:
>
>
>> (progn (load "c:\\Documents and
>> Settings\\mhd\\.xemacs\\slime\\swank-loader.lisp" :verbose t) (funcall
>> (read-from-string "swank-loader:init")) (funcall (read-from-string
>> "swank:start-server") "C:\\DOCUME~1\\mhd\\LOCALS~1\\Temp\\slime.4424"
>> :coding-system "iso-latin-1-unix"))
>>
>>> Error: "C" is not a defined logical host
>>> While executing: CCL::PATHNAME-HOST-SSTR, in process listener(1).
>>> Type :GO to continue, :POP to abort, :R for a list of available
>> restarts.
>>> If continued: Skip loading init file.
>>> Type :? for other options.
>> 1 > > Error: "c" is not a defined logical host
>>> While executing: CCL::PATHNAME-HOST-SSTR, in process listener(1).
>>> Type :POP to abort, :R for a list of available restarts.
>>> Type :? for other options.
>> 2 >
>
>
> Any advice?
>
>
> Versions:
>
>
> XEmacs 21.4 (patch 21) "Educational Television" [Lucid] (i586-pc-win32)
> of Sun Oct 07 2007 on VSHELTON-PC2t
>
>
> Clozure CL: Version 1.3-RC1-r11837S  (WindowsX8632)

Why are you using a "release candidate" rather than an up-to-date release
(or the trunk) ?

Windows namestrings can contain an initial device component (drive letter) 
followed by a colon; a logical pathname namestring can also contain one or
more characters followed by a colon.  CCL's namestring parsing code tries
to resolve the ambiguity via a heuristic: if the namestring contains
physical directory separators (#\/ characters), it's treated as a physical
pathname namestring (and a leading "c:/" is parsed as a device specifier);
otherwise, it's assumed to be a logical pathname namestring and whatever
precedes a colon is assumed to be a logical  host.  (Other plausible 
heuristics are about as bad.)

Windows can use a backslash character as well as a forward slash as a
directory separator. CCL's namestring-parsing code only recognizes
forward slash (in the context described above and elsewhere); -most-
callers of that code canonicalized the namestring (replacing
backslashes with forward slashes) before calling it, but there were
exceptions and it was possible to pass a physical-pathname namestring
that used backslashes to separate directories to pathname operations
that ... well, basically got confused by them.  At some point in the
last several months, the parsing functions were changed to do that
canonicalization themselves, and the confusion went away.

I know that those changes have been in the trunk for a while and they'll
be in the 1.4 release; I don't know if they were propagated to 1.3's svn
tree, and I'm quite sure that they weren't in 1.3RC1.

>
>
> SLIME: /slime.el/1.1148/Mon Mar  9 22:51:24 2009//
>
>
>
> _______________________________________________
> Openmcl-devel mailing list
> Openmcl-devel at clozure.com
> http://clozure.com/mailman/listinfo/openmcl-devel
>
>