TestSuite and other Failures in today's CVS head

Hi Robert, David,

If I remove the line 673 of init-cl.lisp:
       (adjust-character-encoding)
from the definition of cl-user::run() function, imaxima works fine.
This happens only when maxima is invoked from imaxima. There is
no error on running maxima from command line.
Also, this does not happen when clisp 2.44 provided by Ubuntu package
system is used.
Also, this does not happen when clisp 2.47 and Maxima cvs head is used
on Mac OS X.

The function adjust-character-encoding() seems to be introduced after
5.17.1 and effectively defined for clisp only in init-cl.lisp .

Is it OK to comment out the above line until this is fully debugged?

Yasuaki Honda, Chiba, Japan



On 2009/02/21, at 1:22, Robert Dodier wrote:

> On Fri, Feb 20, 2009 at 9:09 AM, David Ronis
> <ronis at ronispc.chem.mcgill.ca> wrote:
>
>> SYSTEM::SET-STREAM-EXTERNAL-FORMAT on #<IO TWO-WAY-STREAM
>> #<INPUT UNBUFFERED FILE-STREAM CHARACTER #P"/dev/fd/0" @1>
>> #<OUTPUT UNBUFFERED FILE-STREAM CHARACTER #P"/dev/fd/1">> is illegal
>
> Well, if *debugger-hook* were nil you could backtrace from Clisp's
> debugger prompt, and at least we would know what function that
> originates from. I don't know how to convince emacs to send
> (setq *debugger-hook* nil) to Maxima before anything else.
> I think the -g command line option for Maxima has the same effect.
>
> HTH
>
> Robert Dodier
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima