New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add GAP's Semigroups package to gap_packages #27295
Comments
Dependencies: #26930 |
This comment has been minimized.
This comment has been minimized.
comment:2
of these, only Semigroups doesn't want to get installed.
And indeed:
|
comment:3
This is after applying diff --git a/build/pkgs/gap_packages/spkg-install b/build/pkgs/gap_packages/spkg-install
index d8260b9aa0..4592675553 100644
--- a/build/pkgs/gap_packages/spkg-install
+++ b/build/pkgs/gap_packages/spkg-install
@@ -19,6 +19,7 @@ sdh_install \
crystcat \
design \
gbnp \
+ genss-* \
Hap* \
HAPcryst \
hecke-* \
@@ -76,7 +77,7 @@ done
# These packages have a new-style autoconf ./configure
# that takes --with-gaproot
-for pkg in nq-* io-*
+for pkg in nq-* io-* orb-* digraphs-* semigroups-*
do
cd "$PKG_SRC_DIR/$pkg"
sdh_configure --with-gaproot="$GAP_ROOT" to the branch of #26930. |
comment:4
This is a sort of upstream issue - they install a libsemigroup, a C++ library, in the tree, etc etc. I've opened semigroups/Semigroups#580 |
Upstream: Reported upstream. No feedback yet. |
Changed upstream from Reported upstream. No feedback yet. to Reported upstream. Developers acknowledge bug. |
comment:5
upstream welcomes PRs, so I'll try doing this in a proper way. |
Changed keywords from none to GAP packages, semigroups |
comment:8
|
comment:9
I don't seem to have a problem with digraphs, it installs just fine, and uses a static copy of bliss, it seems. Do you mean to say one has to watch out for something, still? |
comment:10
Replying to @dimpase:
No, just don't like it I guess. |
comment:11
At least with semigroups, hopefully after my semigroups/Semigroups#584 is in, it can be done in a clean way. |
comment:12
Replying to @dimpase:
Looks good. It looks like that will have to wait gap-4.10.2 for it. I just finished updating all my sage-on-gentoo gap packages to offer 4.10.1 and I must say I was shocked when |
comment:13
Upstream has https://github.com/gap-packages/Semigroups/tree/v3.1.2 which has all the needed changes. Should we make a separate optional package for this? |
comment:14
Ticket retargeted after milestone closed (if you don't believe this ticket is appropriate for the Sage 8.8 release please retarget manually) |
comment:15
As the Sage-8.8 release milestone is pending, we should delete the sage-8.8 milestone for tickets that are not actively being worked on or that still require significant work to move forward. If you feel that this ticket should be included in the next Sage release at the soonest please set its milestone to the next release milestone (sage-8.9). |
Branch: u/dimpase/packages/semigroupsgap |
Commit: |
comment:24
Only patch needed by planarity is extern patch that sage already has.
Probably. Digraphs adds planarity as a subdirectory just like semigroups does with libsemigroups and therefore planarity inside digraphs is not installed. Since planarity is installed in sage and the two versions of planarity are ABI compatible, you might not see a missing planarity issue. This would break if planarity versions differ in Digraphs and Sage which is pretty unlikely since planarity changes rarely. |
comment:25
Replying to @isuruf:
You're right that
So in this sense the only thing to fix is to add |
comment:26
Would still be best to use |
comment:27
Since there hasn't been a release of Digraphs, it wouldn't work, but it's still okay to add |
Branch pushed to git repo; I updated commit sha1. New commits:
|
Branch pushed to git repo; I updated commit sha1. New commits:
|
Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:
|
Branch pushed to git repo; I updated commit sha1. New commits:
|
comment:32
Replying to @isuruf:
I've added this, and missing deps, and the correct tarball checksum for libsemigroups from #27396. I didn't try to remove dead wood copy of planarity installed by |
comment:33
Dear me, I forgot those horrors so fast. No only does Digraphs use plain "planarity" it also include its own vendored (and prefixed so it doesn't clash) version of bliss-0.73. I had some fun patching that in 0.13.0 and 0.15.0. No I may have to reproduce it for 0.15.2. |
comment:35
This package vendors bliss. Is that okay? |
comment:36
We know about vendored bliss. It builds it statically, so it should not be a show-stopper. It should be unvendored upstream. Some of us actually do research in semigroup algorithms, so we want it in... |
Reviewer: Isuru Fernando |
comment:37
Do you mind creating a new trac ticket for unvendoring bliss? |
comment:38
To be more precise, it's Digraphs, a dependency of Semigroups, that vendors bliss. So this is in fact digraphs/Digraphs#205 |
comment:39
I have an unvendoring patch in sage-on-gentoo, but it is minimal. I pass |
comment:40
Improved unvendoring patch: https://github.com/cschwan/sage-on-gentoo/blob/master/dev-gap/digraphs/files/digraphs-0.15.2-system_pkg-r1.patch
|
Changed branch from u/dimpase/packages/semigroupsgap to |
comment:42
Seems to work on Cygwin too, FWIW. |
Changed commit from |
Semigroup depends on a number of GAP packages; io (cf. #26930), Digraphs (that uses bliss), orb, and genss.
So they all need to be added.
Depends on #27396
Depends on #28088
Upstream: Reported upstream. Developers acknowledge bug.
CC: @dimpase @isuruf
Component: interfaces
Keywords: GAP packages, semigroups
Author: Dima Pasechnik
Branch:
426edfe
Reviewer: Isuru Fernando
Issue created by migration from https://trac.sagemath.org/ticket/27295
The text was updated successfully, but these errors were encountered: