arb: 2.17.0 -> 2.18.0

[r-ryantm]

Semi-automatic update generated by nixpkgs-update tools. This update was made based on information from https://repology.org/metapackage/arb/versions.

meta.description for arb is: "A library for arbitrary-precision interval arithmetic"

meta.homepage for arb is: "http://arblib.org/"

meta.changelog for arb is: ""

Updates performed

• Version update

To inspect upstream changes

• Compare changes on GitHub

Impact

Checks done (click to expand)

---

• built on NixOS

• The tests defined in passthru.tests, if any, passed

• 0 of 0 passed binary check by having a zero exit code.

• 0 of 0 passed binary check by having the new version present in output.

• found 2.18.0 with grep in /nix/store/158kc1a36c2jwi8859155wgmlab58mmj-arb-2.18.0

• directory tree listing: https://gist.github.com/1085b667bcc61cdb6e38e2c56600fdc6

• du listing: https://gist.github.com/c76ee51c86f9fb428bb819ee480a88ef

---

Rebuild report (if merged into master) (click to expand)

6 total rebuild path(s)

3 package rebuild(s)

3 x86_64-linux rebuild(s)
1 i686-linux rebuild(s)
1 x86_64-darwin rebuild(s)
1 aarch64-linux rebuild(s)


First fifty rebuilds by attrpath
arb
sage
sageWithDoc

Instructions to test this update (click to expand)

---

Either download from Cachix:

nix-store -r /nix/store/158kc1a36c2jwi8859155wgmlab58mmj-arb-2.18.0 \
  --option binary-caches 'https://cache.nixos.org/ https://r-ryantm.cachix.org/' \
  --option trusted-public-keys '
  r-ryantm.cachix.org-1:gkUbLkouDAyvBdpBX0JOdIiD2/DP1ldF3Z3Y6Gqcc4c=
  cache.nixos.org-1:6NCHdD59X431o0gWypbMrAURkbJ16ZPMQFGspcDShjY=
  '

(r-ryantm's Cachix cache is only trusted for this store-path realization.)
For the Cachix download to work, your user must be in the trusted-users list or you can use sudo since root is effectively trusted.

Or, build yourself:

nix-build -A arb https://github.com/r-ryantm/nixpkgs/archive/577e63091db71d7edda1b34c84d03c2afdefffbc.tar.gz

After you've downloaded or built it, look at the files and if there are any, run the binaries:

ls -la /nix/store/158kc1a36c2jwi8859155wgmlab58mmj-arb-2.18.0
ls -la /nix/store/158kc1a36c2jwi8859155wgmlab58mmj-arb-2.18.0/bin

---

Pre-merge build results

We have automatically built all packages that will get rebuilt due to
this change.

This gives evidence on whether the upgrade will break dependent packages.
Note sometimes packages show up as failed to build independent of the
change, simply because they are already broken on the target branch.

Result of nixpkgs-review 1

2 packages failed to build:

- sage (sagemath)
- sageWithDoc

1 package built:

- arb

---

Maintainer pings

cc @7c6f434c @timokau for testing.

[timokau]

Some adjustments to the sage testsuite necessary.

[timokau]

https://trac.sagemath.org/ticket/28623#comment:9

[stale]

I marked this as stale due to inactivity. → More info

[timokau]

The latest sage master now supports/requires arb 2.18.1. I'm taking a step back from sage maintenance, so I will not work on the update.

CC @omasanori and @collares in case either of you plans to work on the next sage update.

[omasanori]

@timokau Ack. Thank you for letting us know.
Now SageMath 9.3 is in beta stage and I expect rc1 would be suitable for starting actual upgrade effort.

[timokau]

Related: #109890

The way I have always kept it is to

1. update sage dependencies when upstream support is available, backporting patches to sage if necessary.
2. keep a WIP PR in which I track the intermediate sage releases.

The reason for (1) is that it simplifies the actual update (ideally we only need to update the sage source and remove most of the patches) and does not block the dependencies. The downside is that sage in nixpkgs might deviate slightly from upstream sage, which could introduce bugs or be confusing for a user.

The reason for (2) is again that it simplifies the update process. Instead of having to fix all the breakage that the update introduces at once, the work can be split over a longer time and multiple commits. Nix specific regressions can be reported/fixed during the beta phase. By the time the new release is official, the update is already almost finished. Having intermediate commits for the beta releases can also help when trying to pin down regressions with git-bisect.

That's just my process though. It has its upsides and its downsides. If you want to tackle the next update, whatever works best for you is good :)

[omasanori]

Thank you for your explanation, @timokau! That makes sense. I will try 2.18.1 first then, as 2.19 (latest) is not known to work with SageMath.

[timokau]

Okay, feel free to ping me if you need a review. If you end up trying 2.19 you could open an upstream trac ticket and report your results there. If you encounter packaging specific issues, the sage-packaging mailing list may also be able to help.