Yeah, agreed. I'll create rc2 fixing the urllib limit On Tue, 20 May 2025 at 07:19, Jarek Potiuk <ja...@potiuk.com> wrote:
> +1 binding - but with a few caveats, and some bad consequences if we > release it (up to Kaxil to decide) > > Tested reproducibility, checksums signatures, licences are ok since all the > code is generated from "airflow", the package is **almost** reproducible - > the one difference is EOL in version,txt - which I attribute to building > the package before running pre-commit. Not a blocker since the diff is > easily verifiable. I ran a few runs - the client works in breeze in Python > 3.10, but it fails with 3.9 because urllib3 in Python 3.9 is 1.26.8. > > The current rc has "urllib3>=1.25.3" but in fact it will not work with it - > we already have a fix for that in main > https://github.com/apache/airflow/pull/50770 (because it failed the tests) > - and I think we should fix it. Again - not a blocker, but it would likely > be better to fix it. > > I tested that it does not work with urllib <2.1. (bad key passed to > lambda).. That would not be a big problem on its own if not for a Python > 3.9 problem: > > The client seems to work with 3.9 + urllib3 > 2.1. But several of our > packages limit urllib3 to < 2 for Python 3.9 (botocore, snowflake, tableau, > opensearch) - all of them with the samelimit (if python < 3.10 -> urllib3 < > 2). Probably they have a good reason for that - but our client seems to > work fine on Python 3.9 with urllib3. I can imagine a number of people will > try to install the client inside the airflow image, so that they can use it > to interact with Airflow. And as it is now, without the `urllib > 2.1` > limit, the client can be installed on Airflow 3, (reference image) with > python 3.9 - but it will not work when at the same time botocore, > snowflake providers are installed. > > I think it would be better to see "conflict" rather than cryptic 'bad key > passed to lambda'. Possibly rc2 with the urllib limit applied is a good > idea. > > BTW. It is yet another reason why we should consider dropping Python 3.9 > faster. > > J. > > > On Mon, May 19, 2025 at 8:35 AM Kaxil Naik <kaxiln...@gmail.com> wrote: > > > Hey fellow Airflowers, > > > > I have cut the first release candidate for the Apache Airflow Python > Client > > 3.0.0. > > This email is calling for a vote on the release, > > which will last for 72 hours. Consider this my (binding) +1. > > > > Airflow Client 3.0.0rc1 is available at: > > https://dist.apache.org/repos/dist/dev/airflow/clients/python/3.0.0rc1/ > > > > The apache_airflow_client-3.0.0.tar.gz is an sdist release that contains > > INSTALL instructions, and also > > is the official source release. > > > > The apache_airflow_client-3.0.0-py3-none-any.whl is a binary wheel > release > > that pip can install. > > > > Those packages do not contain .rc* version as, when approved, they will > be > > released as the final version. > > > > The rc packages are also available at PyPI (with rc suffix) and you can > > install it with pip as usual: > > https://pypi.org/project/apache-airflow-client/3.0.0rc1/ > > > > Public keys are available at: > > https://dist.apache.org/repos/dist/release/airflow/KEYS > > > > Only votes from PMC members are binding, but all members of the community > > are encouraged to test the release and vote with "(non-binding)". > > > > The test procedure for PMC members is described in: > > > > > https://github.com/apache/airflow/blob/main/dev/README_RELEASE_PYTHON_CLIENT.md#verify-the-release-candidate-by-pmc-members > > > > The test procedure for contributors and members of the community who > would > > like to test this RC is described in: > > > > > https://github.com/apache/airflow/blob/main/dev/README_RELEASE_PYTHON_CLIENT.md#verify-the-release-candidate-by-contributors > > > > *Changelog*: > > > > > https://github.com/apache/airflow/blob/main/clients/python/CHANGELOG.md#v300 > > > > Regards, > > Kaxil > > >
