-
-
Notifications
You must be signed in to change notification settings - Fork 14.1k
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
arrow-cpp: 17.0.0 -> 18.0.0 #351868
base: master
Are you sure you want to change the base?
arrow-cpp: 17.0.0 -> 18.0.0 #351868
Conversation
Running |
Btw, I'd like to enable support for the Azure Filesystem Integration. But someone needs to take a look at #348202. |
|
Ceph is using an api function that was deprecated and has now been removed. The upstream repo has already been fixed, but that was not integrated into a release yet: ceph/s3select#156. |
diff --git a/pkgs/tools/filesystems/ceph/default.nix b/pkgs/tools/filesystems/ceph/default.nix
index 25075f612df8..004d0bbca94b 100644
--- a/pkgs/tools/filesystems/ceph/default.nix
+++ b/pkgs/tools/filesystems/ceph/default.nix
@@ -4,7 +4,7 @@
, fetchurl
, fetchFromGitHub
, fetchPypi
-, fetchpatch
+, fetchpatch2
# Build time
, autoconf
@@ -314,6 +314,16 @@ in rec {
pname = "ceph";
inherit src version;
+ patches = [
+ (fetchpatch2 {
+ name = "ceph-s3select-arrow-18-compat.patch";
+ url = "https://github.com/ceph/s3select/commit/f333ec82e6e8a3f7eb9ba1041d1442b2c7cd0f05.patch";
+ hash = "sha256-21fi5tMIs/JmuhwPYMWtampv/aqAe+EoPAXZLJlOvgo=";
+ stripLen = 1;
+ extraPrefix = "src/s3select/";
+ })
+ ];
+
nativeBuildInputs = [
autoconf # `autoreconf` is called, e.g. for `qatlib_ext`
automake # `aclocal` is called, e.g. for `qatlib_ext` Fixes ceph (cc @nh2). |
@tobim Thanks for poking around and figuring that out, I applied the patch you suggested. |
I opened OSGeo/gdal#11163 for the gdal test failure. |
Thanks, looks like that's already been fixed. Should I pull in a patch of those changes if possible? |
Seems like the patch applies cleanly directly from upstream. |
I can't seem to run nixpkgs-review in a reasonable amount of time without running out of memory 😞. |
You could try
or some value that makes sense to you. |
Yep, I'm tinkering with |
|
Datafusion can be fixed with diff --git a/pkgs/development/python-modules/datafusion/default.nix b/pkgs/development/python-modules/datafusion/default.nix
index 96ac76d17530..9badd981aed4 100644
--- a/pkgs/development/python-modules/datafusion/default.nix
+++ b/pkgs/development/python-modules/datafusion/default.nix
@@ -8,6 +8,7 @@
libiconv,
numpy,
protobuf,
+ protoc,
pyarrow,
Security,
SystemConfiguration,
@@ -55,6 +56,7 @@ buildPythonPackage rec {
nativeBuildInputs = with rustPlatform; [
cargoSetupHook
maturinBuildHook
+ protoc
];
buildInputs =
diff --git a/pkgs/top-level/python-packages.nix b/pkgs/top-level/python-packages.nix
index e386f2980ce1..bbf4e57da195 100644
--- a/pkgs/top-level/python-packages.nix
+++ b/pkgs/top-level/python-packages.nix
@@ -2908,6 +2908,7 @@ self: super: with self; {
datafusion = callPackage ../development/python-modules/datafusion {
inherit (pkgs.darwin.apple_sdk.frameworks) Security SystemConfiguration;
+ protoc = pkgs.protobuf;
};
datalad = callPackage ../development/python-modules/datalad { }; I'd prefer that to trying to update in this PR. |
|
|
Result of 117 packages marked as broken and skipped:
169 packages failed to build:
412 packages built:
|
|
I'm now having to build chromium from source. Is there some way we can move this along without having to clear an ever shifting goal post? |
Fixing a bunch of already broken packages that happen to have arrow as a dependency and break because of an unrelated reason is not something that I've ever had to do when bumping arrow. |
From my perspective the changes here are okay, but we can't be sure we're not breaking stuff that can't be built because of unrelated breakages at the moment. In normal times this would not be necessary, but we are in feature freeze now, and if we can't make a good case that this isn't introducing new breakage then it is likely this PR will be blocked until 24.11 is done. In any case, I agree that fixing chromium for this bump is unreasonable. Summoning @RossComputerGuy to decide whether this can be merged now. (See #351868 (comment) and following for context). |
Result of 70 packages marked as broken and skipped:
669 packages failed to build:
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
arrow-cpp
fails to build due to a test:
arrow-cpp> 99% tests passed, 1 tests failed out of 90
arrow-cpp>
arrow-cpp> Label Time Summary:
arrow-cpp> arrow-compute-tests = 12.43 sec*proc (13 tests)
arrow-cpp> arrow-tests = 41.25 sec*proc (39 tests)
arrow-cpp> arrow_acero = 46.71 sec*proc (13 tests)
arrow-cpp> arrow_dataset = 24.29 sec*proc (13 tests)
arrow-cpp> arrow_flight = 0.51 sec*proc (1 test)
arrow-cpp> arrow_flight_sql = 0.85 sec*proc (1 test)
arrow-cpp> arrow_substrait = 0.52 sec*proc (1 test)
arrow-cpp> filesystem = 2.04 sec*proc (3 tests)
arrow-cpp> parquet-tests = 14.99 sec*proc (9 tests)
arrow-cpp> unittest = 141.57 sec*proc (90 tests)
arrow-cpp>
arrow-cpp> Total Test time (real) = 24.70 sec
arrow-cpp>
arrow-cpp> The following tests FAILED:
arrow-cpp> 4 - arrow-misc-test (Failed)
arrow-cpp> Errors while running CTest
Seems to be something with the jemalloc tests in arrow-cpp?
|
Thanks! @cpcloud can you set the default for |
@tobim Done! |
Bump
arrow-cpp
to the latest release, 18.0.0.Things done
nix.conf
? (See Nix manual)sandbox = relaxed
sandbox = true
nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD"
. Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/
)Add a 👍 reaction to pull requests you find important.