I'll attach my script. It's 26 KB and uses 67 variables. I'm just suggesting checking the number of variables (whatever it is) at the beginning and returning an error message, rather than me checking with gdb, which I've already done!
agape
brent
On Friday, June 9, 2023 at 6:37:44 AM UTC-4 parisse wrote:
> There is code for up to 64 variables. I'm not sure for more. Can you send
> your input? That way I can check with gdb.
>
> On Monday, June 5, 2023 at 9:19:04 PM UTC+2 Brent W. Baccala wrote:
>
>> Hi -
>>
>> I don't think giac can handle more than 15 variables in a Gröbner basis
>> calculation.
>>
>> This limitation isn't really documented anywhere, but if you look in
>> giac's src/cocoa.cc around lines 490-500, function swap_indices and
>> think about it a few minutes, you'll see that it can't handle more than 15
>> variables, I think. I'm looking at giac 1.9.0.
>>
>> I tried a calculation on a ring with 67 variables and the algorithm went
>> into an infinite loop because it couldn't compare monomial exponents
>> properly. It produced a polynomial with two terms, both with the same
>> monomial (different coefficients), and that triggered the infinite loop in
>> the reduction code.
>>
>> I hope somebody working with this code will check my work and verify that
>> giac is so limited. If so, we should add a check to the groebner_basis
>> routine in src/sage/libs/giac/__init__.py
>>
>> agape
>> brent
>>
>>
--
You received this message because you are subscribed to the Google Groups
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sage-devel/7b76d664-cfe6-4204-b146-14e23ceb3edbn%40googlegroups.com.
giac.script3
Description: Binary data
