On Sat, Sep 19, 2009 at 2:27 AM, Simon King <[email protected]> wrote:
>
> Hi William!
>
> On 19 Sep., 10:45, William Stein <[email protected]> wrote:
> [...]
>> > 1.  Those control characters are there because GAP is put in a special
>> > "control" mode when Sage communicates with it.    See the command used
>> > to start GAP.
>>
>> > 2. Regarding what the logfile option is for, well.. it is just
>> > something built into pexpect.  There are is certainly no guarantee
>> > that it can be used to exactly reproduce a session for debugging.
>
> Difficult. So, perhaps the other approach (let gap.eval() and gap.set
> () dump the commands) would be better, although when I tried then some
> errors remained.
>
>> It occurred to me that GAP does have a "workspace" save feature, which
>> might be useful.  It makes it possible to save the complete state of
>> GAP.  Maybe you can use sage to get GAP into some state of interest to
>> you, then save the workspace, then load it into another GAP session
>> and interact with it.
>
> This might work. However, part of the problem is that I don't know
> exactly at what command the error occurs. I only know: It produces
> "recursion depth trap(5000)"; when doing anything else afterwards, a
> "recursion depth trap(0)" results; and if this is done, a repetition
> of the critical command works without problem.
> Also, it seems that the 10,000s of commands preceding the "critical"
> command are needed to trigger the error. Therefore I thought a
> complete log might be better.

Maybe you should try asking on the GAP list what those error messages mean?

William

--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---

Reply via email to