Complexes pr 3

[mikestillman]

This pull request includes the next set of minor package changes to get them working with Complexes (the current plan is to make this the default in the April 2025 M2 release, removing the old ChainComplex type from the Core).

Pinged authors: do these changes look ok to you? (They should all be minor: changing ChainComplex to Complex, etc)

• KustinMiller: @jankoboehm
• EliminationMatrices: @lrtbuse
• EagonResolutions: @eisenbud
• QuillenSuslin: @bbarwick @bstone
• ExteriorIdeals, ExteriorModules: @lucamata
• BettiCharacters, HighestWeights: @galettof

After this pull request, there are not too many packages left to transfer to work directly with the new Complexes package.

[mahrud]

NormalToricVarieties doesn't use Normaliz.

[bbarwick]

@mikestillman (cc: @bstone) The QuillenSuslin.m2 changes look good to me. Thanks!

[lrtbuse]

Dear Mike, Changes in EliminationMatrices seem good to me; thank you very much for taking care of that. Best, Laurent.

…

Le 19 oct. 2024 à 14:50, Mike Stillman ***@***.***> a écrit : This pull request includes the next set of minor package changes to get them working with Complexes (the current plan is to make this the default in the April 2025 M2 release, removing the old ChainComplex type from the Core). Pinged authors: do these changes look ok to you? (They should all be minor: changing ChainComplex to Complex, etc) KustinMiller: @jankoboehm <https://github.com/jankoboehm> EliminationMatrices: @lrtbuse <https://github.com/lrtbuse> EagonResolutions: @eisenbud <https://github.com/eisenbud> QuillenSuslin: @bbarwick <https://github.com/bbarwick> @bstone <https://github.com/bstone> ExteriorIdeals, ExteriorModules: @lucamata <https://github.com/lucamata> BettiCharacters, HighestWeights: @galettof <https://github.com/galettof> After this pull request, there are not too many packages left to transfer to work directly with the new Complexes package. You can view, comment on, or merge this pull request online at: #3526 Commit Summary dee3272 <dee3272> Very minor change to NormalToricVarieties to get it working with Complexes. Also discovered: Complexes doesn't play well with rings with degree length 0. cc90018 <cc90018> changed several res to freeResolution in NormalToricVarieties 4f17a11 <4f17a11> Minor changes to KustinMiller to get it working with Complexes 79d3abe <79d3abe> Minors changes to KustinMiller package to get it working with Complexes db3f902 <db3f902> fixed load paths in Complexes 8c8f76f <8c8f76f> removed calls to length Complex in resolutionMap 8eefd51 <8eefd51> added formation for Ext modules in Complexes 2caeddf <2caeddf> changed map(Complex, Complex, ZZ) to allow multiplication maps 1c121d0 <1c121d0> fixed yonedaExtension for multigraded rings bfa9bd8 <bfa9bd8> Minor changes to EliminationMatrices to get it working with Complexes. 194e81b <194e81b> Minor changes to EliminationMatrices to get it working with Complexes 3b529cc <3b529cc> Minor changes to EagonResolution to get it working with Complexes a4f8c14 <a4f8c14> Minors changes to EagonResolution package to get it working with Complexes f2102aa <f2102aa> Minors changes to QuillenSuslin to get it working with Complexes c3fba6f <c3fba6f> Minor changes to the QuillenSuslin package to get it working with Complexes 5592408 <5592408> minors changes to ExteriorIdeals package to get working with Complexes (added new optional argument to minimalBettiNumbers) 0eefb20 <0eefb20> minors changes to ExteriorModules package to get working with Complexes (added new optional argument to minimalBettiNumbers and bassNumbers) 7a7ed25 <7a7ed25> minor changes to the packages ExteriorIdeals and ExteriorModules to get them working with Complexes b0b3e18 <b0b3e18> minor changes to BettiCharacters package to get working with Complexes. 68f86dc <68f86dc> minor changes to HighestWeights package to get working with Complexes. Also removed definition of global symbol 'Range' cfe3b5d <cfe3b5d> Minor changes to get BettiCharacters and HighestWeights to work with Complexes File Changes (18 files <https://github.com/Macaulay2/M2/pull/3526/files>) M M2/Macaulay2/packages/BettiCharacters.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-97cb56135f07f783888582c0e9876ba72c554de363cfaf1a3ef265572740a8cd> (55) M M2/Macaulay2/packages/Complexes.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-41eef49f83232110d3b7483847dc8caf7fea573766fe3c10d86629e464dc56ff> (18) M M2/Macaulay2/packages/Complexes/ChainComplex.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-1b9d61f4149faf4fd94206d4ce9cf0478295270cdcf0fc19e800f6f411bd653e> (9) M M2/Macaulay2/packages/Complexes/ChainComplexMap.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-28092cb4901a2971c73f23753713d43ec7c1764c359550c21bf58ee26b2d9972> (10) M M2/Macaulay2/packages/Complexes/ChainComplexTests.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-bdcedca072894115717a501afe3de5bc8d5f006def182e47b4036f4966154b33> (6) M M2/Macaulay2/packages/EagonResolution.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-f0114c256b65f5fb3cf2cdc7d15800500cc8ee0c46e8a5abe6681ffd3904ae0b> (92) M M2/Macaulay2/packages/EliminationMatrices.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-e78bd551d80f8ba8a62711b440d7d62305ec588d7b9fdcf18cc464a6ed46d51d> (163) M M2/Macaulay2/packages/ExteriorIdeals.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-a92b487ebc1a9955f04c02d60629dfc48c2831f0e89fa59330a6d40a66b51398> (21) M M2/Macaulay2/packages/ExteriorModules.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-c1130916edee9637791285069512c86f5f01cd1f8d9937f300e89ec4afe4e7c1> (28) M M2/Macaulay2/packages/HighestWeights.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-8257790aade14cad8677bb7c18a5807608e93240b71a8d896be6ee3f709672fc> (2) M M2/Macaulay2/packages/HighestWeights/doc.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-47b6398a05aebee882e272fab7701b54434078dc5e72e38e3d21c542ca39a9d3> (27) M M2/Macaulay2/packages/HighestWeights/examples.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-f97b6fe34819fd987513d659fcd24e9e1d090d6d981766344ffaa49087050b9f> (14) M M2/Macaulay2/packages/HighestWeights/main.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-17fee8e34421362b17d0cf13f7b74b77be4d10a656624e74b87287fc2ea183ad> (6) M M2/Macaulay2/packages/HighestWeights/tests.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-0541338eadf92744dd728cf139f4e991695d635dab0ef7be31b2f2da57a361a5> (4) M M2/Macaulay2/packages/KustinMiller.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-f293dc24fe3697eafe5480ec7052586902a3403cb333bc13715f63b67fcb03b7> (149) M M2/Macaulay2/packages/NormalToricVarieties.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-5789def1ebdcfc513e2834f711ada8211e2e0c57d0efaf5cd0b3c5f39cf6919e> (2) M M2/Macaulay2/packages/NormalToricVarieties/Sheaves.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-e6c5b1aa434242df3a3664feefef50391d57ae74ef37530318cf5eb427340986> (8) M M2/Macaulay2/packages/QuillenSuslin.m2 <https://github.com/Macaulay2/M2/pull/3526/files#diff-b2098b10394cb864fd6cf265488fabbb306c54dc2036a46b0f7a84563f9a7595> (9) Patch Links: https://github.com/Macaulay2/M2/pull/3526.patch https://github.com/Macaulay2/M2/pull/3526.diff — Reply to this email directly, view it on GitHub <#3526>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABISZWAD7L5CB7I3KHN33ZTZ4JIPRAVCNFSM6AAAAABQHOV7KOVHI2DSMVQWIX3LMV43ASLTON2WKOZSGU4TSMJTGQZDEMI>. You are receiving this because you were mentioned.

[galettof]

Hi Mike,

From what I can tell, the changes are fine for the HighestWeights and BettiCharacters packages.

Thank you!

[mikestillman]

@ggsmith K3Carpets is failing: it is using CompleteIntersectionResolutions. Unfortunately, it doesn't look as trivial to modify as the other packages we have done. It uses also NonminimalComplexes, which we have not yet modified to work with Complexes.

[mahrud]

Do you want to revert this PR to commit cfe3b5d, which passed all tests? I think changes to every package up to that point except for KustinMiller and EagonResolutions have been approved by the authors.

The remainder can be added in the next PR perhaps.

[ggsmith]

Do you want to revert this PR to commit cfe3b5d, which passed all tests? I think changes to every package up to that point except for KustinMiller and EagonResolutions have been approved by the authors.

The remainder can be added in the next PR perhaps.

Not yet—we're stilling working on it.

[jankoboehm]

Hi Mike, The changes look very good with respect to the KustinMiller package, provided the tests pass of course. Thanks a lot!

[mikestillman]

@ggsmith I'm not sure I can find the bug in time, before the release gets made. We might need to back off to the last pr-3 pull request...

[mikestillman]

@d-torrance What do you think about this? I'm also concerned that I won't be able to quickly fix the conflict errors after David's PR for PencilsOfQuadrics...

[d-torrance]

Yeah, holding off on this PR until after the release sounds like a good idea.

[mahrud]

In the current version rawMinimalBetti receives a raw computation that hasn't started yet, so here is a fix:

diff --git a/M2/Macaulay2/packages/Complexes/FreeResolutions.m2 b/M2/Macaulay2/packages/Complexes/FreeResolutions.m2
index 27f0bd9aa8..a472fffe0f 100644
--- a/M2/Macaulay2/packages/Complexes/FreeResolutions.m2
+++ b/M2/Macaulay2/packages/Complexes/FreeResolutions.m2
@@ -202,6 +202,7 @@ resolutionObjectInEngine = (opts, M, matM) -> (
         complex maps
         );

+    if not opts.StopBeforeComputation then
     RO.compute(opts.LengthLimit, opts.DegreeLimit);
     RO.complex(opts.LengthLimit)
     )
@@ -691,7 +692,8 @@ minimalBetti Module := BettiTally => opts -> M -> (
        nvars := # generators(R, CoefficientRing => A);
        lengthlimit = nvars + if A === ZZ then 1 else 0;
         );
-    C = freeResolution(M, DegreeLimit => degreelimit, LengthLimit => lengthlimit + 1, Strategy => Nonminimal);
+    C = freeResolution(M, DegreeLimit => degreelimit, LengthLimit => lengthlimit + 1,
+       Strategy => Nonminimal, StopBeforeComputation => true);
     rC := M.cache.ResolutionObject.RawComputation;
     B := unpackEngineBetti rawMinimalBetti(rC,
         if opts.DegreeLimit === infinity then {} else

[mahrud]

Mike, is this done? The tests have passed. GitHub is not happy with your last merge commit, which would be resolved if you rebased and force pushed, but we can also just merge this without rebasing.

[mikestillman]

Yes, I'll accept it, but use a merge pull request.