Hello Chris,

I ran the labs in the following sequence:
1) Polynomials
2) Files

At step ((3) 3 of 9) of Files lab, the following happens:
CRLF ss fread (jpath '~system\extras\config\stdcfg.ijs');0 80  NB. This line
attempts to get executed

wdhandler error pops up:
error in jijs_jctrl_fkey
length error ss
CRLF ss fread (jpath '~system\extras\config\stdcfg.ijs');0 80
[-0]

After this, it is not possible to proceed with the lab anymore - the
statement keeps getting executed and the error keeps occuring!

Suspecting the ss defintion, I find:
  ss
(([EMAIL PROTECTED] - [EMAIL PROTECTED]) % [)"0

So, I rerun the lab on a fresh session...
  ss
I. @ E.

Clearly ss has been overwritten with a different definition. And I find that
in (20 of 25) of Polynomial lab, this is indeed the case!

Thinking that "ss" is a lab local variable led me to this post.

Is "ss" a standard definition that has been overwritten in the Polynomial
lab?

Regards,
Yuva




On 10/3/06, Chris Burke <[EMAIL PROTECTED]> wrote:

Yuvaraj Athur Raghuvir wrote:
> When I tried running different labs one after another, I got into
conflicts
> of definition. Clearly some of the letters are used differently in the
> different labs and the sequence of use determines the current definition
in
> the session.
>
> So, I was wondering if any of the following options could be applied to
the
> current session:
> a) Define a locale for the Lab that gets reset or is indicated to be
reset
> when the lab terminates or abandoned respectively.
> b) Labs could "refresh" the session memory in a sense that the required
> definitions for the lab could be selectively overwritten. This way the
> assumptions of the Lab author is reflected in the session memory state
as
> well.

A lab is intended to be run in a normal session, and it is expected that
the user will experiment while running the lab, or after it is
completed. Running them in a special locale, or erasing all definitions
after the last lab section is run, would make them less useful.

In theory, labs should not redefine standard utilities, and each lab
should ensure that required definitions are available. But there is no
easy way to enforce that after running a lab, the system is in a clean
enough state to run any other lab. The only reliable way to handle this
is to restart J when running a new lab.

If you know of specific problems when running one lab after another, let
me know and I will try to get them fixed.

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to