login  home  contents  what's new  discussion  bug reports     help  links  subscribe  changes  refresh  edit

Edit detail for #271 )show Expression Boolean gives a list revision 7 of 7

1 2 3 4 5 6 7
Editor: gdr
Time: 2008/10/17 16:52:24 GMT-7
Note: Fixed in OpenAxiom


From gdr Fri Oct 17 16:52:24 -0700 2008
From: gdr
Date: Fri, 17 Oct 2008 16:52:24 -0700
Subject: Fixed in OpenAxiom
Message-ID: <20081017165224-0700@axiom-wiki.newsynthesis.org>

Submitted by : (unknown) at: 2007-11-17T22:19:14-08:00 (15 years ago)
Name :
Axiom Version :
Category : Severity : Status :
Optional subject :  
Optional comment :

As I was looking for boolean simplification, I type )sh Expression Boolean. I get the list of the real expressions. So I type

sin (A::Expression Boolean) or true.
Line 1: sin (A::Expression Boolean) or true. ...................................A Error A: syntax error at top level Error A: Improper syntax. 2 error(s) parsing

Axiom disapproves

>> System error: Caught fatal error [memory may be damaged]?...

And the axiom process died the command-line after.

cannot reproduce axiom process died --Bill Page, Mon, 27 Feb 2006 12:37:15 -0600 reply
I can not reproduce your observation that:
the axiom process died the command-line after.

This does not happen for me on at least the Windows version of Axiom.

But I do find it very odd that Axiom allows the construction of the domain Expression Boolean without complaint:

x:Expression Boolean:=true

\label{eq1} \mbox{\rm true} (1)
Type: Expression(Boolean)

The scary sounding message: 'Caught fatal error [memory may be damaged]?' occurs for many operations applied to this object.

The definition of the constructor Expression R: OrderedSet seems wrong to me since there are may operations exported by Expression that are not naturally interpreted over a domain that is just an OrderedSet. For example Boolean is considered ordered as follows:

Boolean has OrderedSet

\label{eq2} \mbox{\rm true} (2)
Type: Boolean
true > false

\label{eq3} \mbox{\rm true} (3)
Type: Boolean

although we might also question the wisdom of considering Boolean ordered in this way.

I've been running thousands of integrals thru axiom recently. There is a problem related to interrupting Axiom while garbage collection is running (I'm guessing) and, very occasionally, Axiom will generate an error or die completely (only once so far). However if Axiom is not interrupted I have yet to see it fail.

The "memory may be damaged" message occurs quite often but I have yet to see any surface damage on the DIMM packages :-). I'm not sure why Bill thought this was a reasonable error message but it does not seem to be a problem. Only console interrupts have caused failure so far and those seem to be related to some race condition in garbage collection.


Status: open => fix proposed

Status: fix proposed => fixed somewhere

no patch available

Patch from Fricas --alfredo, Thu, 16 Oct 2008 03:38:56 -0700 reply