I have two updates for F37 that will not be stable by final freeze based on time alone, but that I think should be stable to avoid disruptions to users. I could wait and ask for freeze exceptions, but I would first like to ask the community to give karma to these updates. I am willing to test and give karma to your updates, too. Let me know what updates to test, and how to test them. The two updates are: https://bodhi.fedoraproject.org/updates/FEDORA-2022-a6d8261b07 - sympy https://bodhi.fedoraproject.org/updates/FEDORA-2022-21f6ca9264 - gap, pari, & friends The sympy update deals with the disappearance of the JDK on i386. Sympy uses the antlr4 package, a Java package, to generate some grammar files. Since there is no JDK on i386, there is no antlr4 package either. This makes one of the BuildRequires for the package unresolvable. The ideal solution would be to stop building sympy on i386. However, sympy is consumed by a number of packages that still build for i386. I think that is the right long-term goal, and will work toward that in the coming months. This update is a stop-gap measure. It switches sympy from a noarch package to an archful package so that we can omit the problematic BuildRequires on i386. This will leave sympy partially broken on i386 (it is still mostly functional), but that is unavoidable at this point. I think this should go stable so that the noarch -> archful change happens at a release boundary, rather than after F37 is released. (And hopefully we can change back for F38!) For testing purposes, this build of sympy should function identically to the previous build. Install the python3-sympy and python3-ipython packages, then verify that it starts and runs at all, perhaps like this: ``` $ isympy [snip sympy banner] In [1]: import math In [2]: math.sqrt(9) Out[2]: 3.0 In [3]: from sympy import symbols In [4]: x, y = symbols('x y') In [5]: expr = x + 2*y In [6]: expr Out[6]: x + 2⋅y ``` The second update is large, containing 100 package builds. It contains an update of gap to version 4.12.0, and an update of pari to version 2.15.0. Both new versions come with significant performance improvements. The gap update in particular drops i386 builds for all packages named gap-pkg-*, as well as gap, GAPDoc, and xgap. I think this should go stable before release so that the i386 disappearing act doesn't happen post-release, and also because the gap and pari updates both contain a few backwards incompatibilities, which should be introduced at release boundaries. Almost every package in this update has a test suite that is run at build time, so they have been tested quite well already, and I have been testing them also. To test the gap part, install the gap and gap-pkg-xmod packages. (The gap-pkg-xmod package pulls in several other gap-pkg-* packages, so we can show they all work together.) Then do a few things with gap-pkg-xmod, like this: ``` $ gap [snip gap banner] gap> LoadPackage("xmod"); [snip package loading messages] gap> c5 := Group( (5,6,7,8,9) ); Group([ (5,6,7,8,9) ]) gap> SetName( c5, "c5" ); gap> id5 := IdentityMapping( c5 ); IdentityMapping( c5 ) gap> ac5 := AutomorphismGroup( c5 ); <group with 1 generator> gap> act := MappingToOne( c5, ac5 ); [ (5,6,7,8,9) ] -> [ IdentityMapping( c5 ) ] gap> XMod( id5, act ) = XModByBoundaryAndAction( id5, act ); true gap> H12 := SmallCatOneGroup( 12, 4, 3 ); Cat-1-group with underlying group Group( [ f1, f2, f3 ] ) . gap> C12 := SmallCat1Group( 12, 4, 3 ); [Group( [ f1, f2, f3 ] ) => Group( [ f1, f2, <identity> of ... ] )] gap> C12 := CatOneGroupToXMod( H12 ); [Group( [ f1, f2, f3 ] ) => Group( [ f1, f2, <identity> of ... ] )] gap> C18 := Cat1Select( 18, 4, 3 ); [(C3 x C3) : C2 => Group( [ f1, <identity> of ..., <identity> of ... ] )] gap> H18 := Cat1GroupToHAP( C18 ); Cat-1-group with underlying group (C3 x C3) : C2 . gap> IdCatOneGroup( H18 ); [ 18, 4, 3 ] gap> IdCat1Group( C18 ); [ 18, 4, 3 ] ``` To test the pari part, install the pari-gp package and try a few calculations, like this: ``` $ gp [snip pari banner] ? 1 + 1 %1 = 2 ? (x+1)^(-2) %2 = 1/(x^2 + 2*x + 1) ? Mod(x, x^2+x*y+y^2)^3 %3 = Mod(y^3, x^2 + y*x + y^2) ? T = x^2 + 1; ? factor(T*(1 + O(5^3))) %5 = [ (1 + O(5^3))*x + (2 + 5 + 2*5^2 + O(5^3)) 1] [(1 + O(5^3))*x + (3 + 3*5 + 2*5^2 + O(5^3)) 1] ? R(x) = { my(s); s = zeta(2) * sum(a=1, sqrt(x), moebius(a)*(x\a^2)); (s - x) / x^0.4; } ? R(10^7) %7 = 0.052092560787004188970344062406837112269 ? R(10^12) %8 = 0.010948893958117048274619354741656931605 ? R(10^15) %9 = 0.00012562581350676279794919752938422775380 ``` Warning: you'll notice a slight pause when computing R(10^12), and a significant pause when computing R(10^15). Pari hasn't hung. Wait a minute and it will be back with the answer. Sorry for the epistle. I will be very grateful for any testing and karma help you can give me. Let me know what I can test and give karma to for you. -- Jerry James http://www.jamezone.org/ _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
