[GAP Forum] Fwd: Upgrade anxiety

[John McDermott]

This was posted to the old address (gap-forum at dcs...). Mail to the GAP 
Forum should be posted to the new address forum at gap-system.org.

John.

Begin forwarded message:

> From: Alexander Hulpke <hulpke at math.colostate.edu>
> Date: 23 November 2004 18:01:06 GMT
> To: gap-forum at dcs.st-and.ac.uk
> Subject: Re: Upgrade anxiety
>
> Dear Malka Schaps, Dear GAP-Forum,
>
>>     How compatible are gap 4.2 and gap 4.4?  I have some long programs
>> which work fine in 4.2, and before I ask our administrator to upgrade 
>> to
>> 4.4, I would like to hear that they won't fill up with bugs.
>
> Incompatibilities are minimal and listed in the release announcements. 
> I
> would be surprised if you had a program that would not run correctly 
> in 4.2
> but not in 4.4. (If I am wrong and you upgrade and your program does 
> not run
> any longer, feel free to send me the program and input with which it 
> runs
> under 4.2 but not 4.4 and I will have a look at it.)
>
>>     I am interested in the upgrade because of this 
>> EpimorphismSchurCover
>> function I just heard about from the forum.  It exists already in the
>> 4.3 that I have on my home computer.  How do I ask GAP to save the
>> source, i.e., the Schur cover, so that I can transfer it to the
>> mainframe, which still is running 4.2? ( At least until I get up my 
>> nerve
>
> The source is in lib/schur.gd and lib/schur.gi. However I would advise
> against doing so -- the code has not been tested under 4.2 and might 
> rely on
> other additions. Also 4.4 contains a multitude of bug fixes.
>
>> to order the upgrade.)  It is given as a pc group.  Once it gets do 
>> the
>> mainframe, it will be used to calculate normalizers and centralizers 
>> of
>> various p-subgroups for our internet database of blocks with abelian
>> defect group.  Would it be best to convert to a permutation
>> representation?
>
> No, a pc group will be fine. The group obtained from 
> `EpimorphismSchurCover'
> will be represented as a finitely presented group. You can use
> SolvableQuotient to convert it to a pc group again.
>
> (I believe the `cohomolo' package requires permutation groups. If you 
> have
> large calculations, using this package might be notably faster).
>
> Best wishes,
>
>    Alexander Hulpke
>
> -- Colorado State University, Department of Mathematics,
> Weber Building, 1874 Campus Delivery, Fort Collins, CO 80523-1874, USA
> email: hulpke at math.colostate.edu, Phone: ++1-970-4914288
> http://www.math.colostate.edu/~hulpke
>
>
--
John McDermott
Computing Officer & Purchase Officer
School of Computer Science
University of St Andrews
North Haugh, St Andrews, Fife
KY16 9SS

tel +44 1334 463252
mob +44 7941 507531