[sage-devel] Re: Error in I.variety(algorithm='msolve', proof=False)

Hi Dima, Dima Pasechnik wrote: > So this is due to "0 < char <= 2**17 and deg != elim.degree()" - added > by you - which does not make sense to me. > Is this "Criterion" no longer applicable? I don't know. This criterion was suggested to me by Mohab after I complained that msolve -P 1 often retur

Re: [sage-devel] Re: Error in I.variety(algorithm='msolve', proof=False)

Hi Marc, On Sun, Jun 30, 2024 at 8:04 AM Marc Mezzarobba wrote: > > 'Peter Mueller' via sage-devel wrote: > > R. = GF(2)[] > > L = [a^2+a, b^2+b] > > I = ideal(L) > > V = I.variety(algorithm='msolve', proof=False) > > > > raises a `ValueError: positive-dimensional ideal`, which of course is > > n

Re: [sage-devel] Re: Bug in `discrete_log(...,bounds=(1,p.isqrt()))`

I don't want to log to github, but the expected behavior of raising exception if there is no solution in the bound is entirely correct despite the doubtful comment. The running times of some of the algorithms heavily depend on the bounds and it is significantly faster if the bound is small. On Sun

[sage-devel] Re: Bug in `discrete_log(...,bounds=(1,p.isqrt()))`

I opened an issue at github #38316 . Let's continue the discussion there. On Sunday, June 30, 2024 at 12:09:48 PM UTC-4 Georgi Guninski wrote: > I think the problem is abusing the factorization of the group order > and later doing CRT. > > The foll

[sage-devel] Re: Bug in `discrete_log(...,bounds=(1,p.isqrt()))`

I think the problem is abusing the factorization of the group order and later doing CRT. The following code correctly raises exception: dl3=discrete_log_lambda(a,g,bounds=(1,p.isqrt())) ValueError: Pollard Lambda failed to find a log -- You received this message because you are subscribed to th

Re: [sage-devel] Bug in `discrete_log(...,bounds=(1,p.isqrt()))`

> Instead, the algorithm should error out, since number theory shows there's no > non-zero (mod 2^61 - 2) solution to 3^a = 1 mod (2^61 - 1) Why do you reduce (mod p-1)? (p-1) is valid integer solution by Euler's theorem. -- You received this message because you are subscribed to the Google Grou

Re: [sage-devel] Bug in `discrete_log(...,bounds=(1,p.isqrt()))`

I think so too. Here's a more minimal version (note a == 1 in your example): p = 31 g = GF(31)(3) discrete_log(1, g, bounds=(1, 6), algorithm="lambda", operation="*") # 6 g^6 # 16 It seems there's some serious bugs going on, as the return value isn't in the bounds: K = GF(2^61 - 1) d = discrete_

[sage-devel] Bug in `discrete_log(...,bounds=(1,p.isqrt()))`

I think this is bug: p=2**61-1;Kp=GF(p);g=Kp(3);X=p//3;a=g**X dl=discrete_log(a,g,bounds=(1,p.isqrt()),algorithm="lambda",operation="*") dl2=discrete_log(a,g,bounds=(1,p),algorithm="lambda",operation="*") (g**dl==a,g**dl2==a) #(False, True) -- You received this message because you are subscribe

Re: [sage-devel] Re: ask.sagemath.org issues - do we need it?

On Sun, Jun 30, 2024 at 4:14 AM Matthias Koeppe wrote: > > Another migration route that we could consider: > Importing the messages from AskBot into Google Groups (merging with the > sage-support group). Google Groups' interface is pretty bad, one cannot do maths, or properly format things. (and

Re: [sage-devel] Re: Error in I.variety(algorithm='msolve', proof=False)

On Sun, Jun 30, 2024 at 12:08 PM 'Peter Mueller' via sage-devel wrote: > > OK, it apparently was a bug of the previous version of msolve: For `msolve -P > 2 -f file.ms` with msolve 0.6.5-2 yields the wrong result claiming that the > dimension is positive, no matter whether file.ms ends with a ne

[sage-devel] sage-numerical-backends-coin 10.4 release candidate

https://pypi.org/project/sage-numerical-backends-coin/10.4rc2/ -- 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 thi

Re: [sage-devel] Re: Demote brial (= polybori) from standard to experimental

1) Brial is too broken for being an optional package. 2) It should be a separate PR, not one doing 10 other unrelated changes, and touching 600+ files in the process On 29 June 2024 20:57:31 BST, Matthias Koeppe wrote: >Still waiting for review: https://github.com/sagemath/sage/pull/36380 >

[sage-devel] Re: Error in I.variety(algorithm='msolve', proof=False)

OK, it apparently was a bug of the previous version of msolve: For `msolve -P 2 -f file.ms` with msolve 0.6.5-2 yields the wrong result claiming that the dimension is positive, no matter whether file.ms ends with a newline character or or not, while msolve 0.6.6-1 yields the correct result in bo

[sage-devel] Re: Error in I.variety(algorithm='msolve', proof=False)

'Peter Mueller' via sage-devel wrote: > I guess that in your snippet, you manually typed the `enter` key after > the `cat` command in the console. Hmm, no, there is a newline at the end: ~$ hexdump -c /tmp/tmprcz_zw9l 000 a , b \n 2 \n a ^ 2 + a , \n b ^ 2 010

[sage-devel] Re: Error in I.variety(algorithm='msolve', proof=False)

msolve apparently ignores the last line of the input file if it does not end with a newline character. If one adds it, then it yields the correct result. The doc of msolve ("2 Input file format") probably should be clearer about that. I guess that in your snippet, you manually typed the `enter`

[sage-devel] Re: Error in I.variety(algorithm='msolve', proof=False)

'Peter Mueller' via sage-devel wrote: > R. = GF(2)[] > L = [a^2+a, b^2+b] > I = ideal(L) > V = I.variety(algorithm='msolve', proof=False) > > raises a `ValueError: positive-dimensional ideal`, which of course is > nonsense. Exporting the system to an msolve-readable file and using > msolve directl