[Bug]: Could not find a version that matches protobuf

cypreess commented 2 years ago

What happened?

Cannot install latest apache-beam[gcp]==2.42.0 because there is a conflict of protobuf version dependencies.

Could not find a version that matches protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<4,<4.0.0dev,<5.0.0dev,>=3.12.0,>=3.12.2,>=3.15.0,>=3.19.0,>=3.19.5,>=4.21.6 (from proto-plus==1.22.1->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
Tried: 2.0.3, 2.3.0, 2.4.1, 2.5.0, 2.6.0, 2.6.1, 3.0.0, 3.0.0, 3.1.0, 3.1.0.post1, 3.1.0.post1, 3.2.0, 3.2.0, 3.3.0, 3.4.0, 3.4.0, 3.5.0.post1, 3.5.0.post1, 3.5.1, 3.5.1, 3.5.2, 3.5.2, 3.5.2.post1, 3.5.2.post1, 3.6.0, 3.6.0, 3.6.1, 3.6.1, 3.7.0, 3.7.0, 3.7.1, 3.7.1, 3.8.0, 3.8.0, 3.9.0, 3.9.0, 3.9.1, 3.9.1, 3.9.2, 3.9.2, 3.10.0, 3.10.0, 3.11.0, 3.11.0, 3.11.1, 3.11.1, 3.11.2, 3.11.2, 3.11.3, 3.11.3, 3.12.2, 3.12.2, 3.12.4, 3.12.4, 3.13.0, 3.13.0, 3.14.0, 3.14.0, 3.15.0, 3.15.0, 3.15.1, 3.15.1, 3.15.2, 3.15.2, 3.15.3, 3.15.3, 3.15.4, 3.15.4, 3.15.5, 3.15.5, 3.15.6, 3.15.6, 3.15.7, 3.15.7, 3.15.8, 3.15.8, 3.16.0, 3.16.0, 3.17.0, 3.17.0, 3.17.0, 3.17.1, 3.17.1, 3.17.1, 3.17.2, 3.17.2, 3.17.2, 3.17.3, 3.17.3, 3.17.3, 3.18.0, 3.18.0, 3.18.0, 3.18.1, 3.18.1, 3.18.1, 3.18.3, 3.18.3, 3.18.3, 3.19.0, 3.19.0, 3.19.0, 3.19.1, 3.19.1, 3.19.1, 3.19.2, 3.19.2, 3.19.2, 3.19.3, 3.19.3, 3.19.3, 3.19.4, 3.19.4, 3.19.4, 3.19.5, 3.19.5, 3.19.5, 3.19.6, 3.19.6, 3.19.6, 3.20.0, 3.20.0, 3.20.0, 3.20.1, 3.20.1, 3.20.1, 3.20.2, 3.20.2, 3.20.2, 3.20.3, 3.20.3, 3.20.3, 4.21.0, 4.21.0, 4.21.0, 4.21.0, 4.21.1, 4.21.1, 4.21.1, 4.21.1, 4.21.2, 4.21.2, 4.21.2, 4.21.2, 4.21.3, 4.21.3, 4.21.3, 4.21.3, 4.21.4, 4.21.4, 4.21.4, 4.21.4, 4.21.5, 4.21.5, 4.21.5, 4.21.5, 4.21.6, 4.21.6, 4.21.6, 4.21.6, 4.21.7, 4.21.7, 4.21.7, 4.21.7, 4.21.8, 4.21.8, 4.21.8, 4.21.8, 4.21.9, 4.21.9, 4.21.9, 4.21.9
Skipped pre-versions: 2.0.0b0, 3.0.0a2, 3.0.0a3, 3.0.0b1, 3.0.0b1.post1, 3.0.0b1.post2, 3.0.0b2, 3.0.0b2, 3.0.0b2.post1, 3.0.0b2.post1, 3.0.0b2.post2, 3.0.0b2.post2, 3.0.0b3, 3.0.0b4, 3.0.0b4, 3.2.0rc1, 3.2.0rc1, 3.2.0rc1.post1, 3.2.0rc1.post1, 3.2.0rc2, 3.2.0rc2, 3.7.0rc2, 3.7.0rc2, 3.7.0rc3, 3.7.0rc3, 3.8.0rc1, 3.8.0rc1, 3.9.0rc1, 3.9.0rc1, 3.10.0rc1, 3.10.0rc1, 3.11.0rc1, 3.11.0rc1, 3.11.0rc2, 3.11.0rc2, 3.13.0rc3, 3.13.0rc3, 3.14.0rc1, 3.14.0rc1, 3.14.0rc2, 3.14.0rc2, 3.14.0rc3, 3.14.0rc3, 3.15.0rc1, 3.15.0rc1, 3.15.0rc2, 3.15.0rc2, 3.16.0rc1, 3.16.0rc1, 3.16.0rc2, 3.16.0rc2, 3.17.0rc1, 3.17.0rc1, 3.17.0rc1, 3.17.0rc2, 3.17.0rc2, 3.17.0rc2, 3.18.0rc1, 3.18.0rc1, 3.18.0rc1, 3.18.0rc2, 3.18.0rc2, 3.18.0rc2, 3.19.0rc1, 3.19.0rc1, 3.19.0rc1, 3.19.0rc2, 3.19.0rc2, 3.19.0rc2, 3.20.0rc1, 3.20.0rc1, 3.20.0rc1, 3.20.0rc2, 3.20.0rc2, 3.20.0rc2, 3.20.1rc1, 3.20.1rc1, 3.20.1rc1, 4.0.0rc1, 4.0.0rc1, 4.0.0rc2, 4.0.0rc2, 4.21.0rc1, 4.21.0rc1, 4.21.0rc1, 4.21.0rc1, 4.21.0rc2, 4.21.0rc2, 4.21.0rc2, 4.21.0rc2
There are incompatible versions in the resolved dependencies:
  protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5 (from google-api-core[grpc,grpcgcp]==2.10.2->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<5.0.0dev,>=3.19.0 (from google-cloud-recommendations-ai==0.7.1->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5 (from google-cloud-dlp==3.9.2->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf>=4.21.6 (from grpcio-status==1.50.0->google-cloud-pubsub==2.13.10->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<4.0.0dev (from google-cloud-bigtable==1.7.2->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<4.0.0dev,>=3.12.0 (from google-cloud-bigquery==2.34.4->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<5.0.0dev,>=3.15.0 (from googleapis-common-protos==1.56.4->google-api-core==2.10.2->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<4.0.0dev (from google-cloud-datastore==1.15.5->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<4.0.0dev,>=3.19.0 (from google-cloud-bigquery-storage==2.13.2->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<4.0.0dev (from google-cloud-language==1.3.2->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<4.0.0dev (from google-cloud-videointelligence==1.16.3->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5 (from google-cloud-pubsub==2.13.10->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<4.0.0dev (from google-cloud-spanner==1.19.3->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<5.0.0dev,>=3.19.0 (from proto-plus==1.22.1->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<4.0.0dev (from google-cloud-vision==1.0.2->apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))
  protobuf<4,>=3.12.2 (from apache-beam[gcp]==2.42.0->uni-bi-export (setup.py))

Issue Priority

Priority: 1

Issue Component

Component: io-py-gcp

kennknowles commented 1 year ago

This appears to be caused by other dependencies in your project, perhaps proto-plus. Or if there is an incompatibility in the GCP dependencies, we would certainly like to have a set of dependencies that works together. And of course it would be useful for Beam to be compatible with as many things as possible.

CC @tvalentyn

tvalentyn commented 1 year ago

@cypreess I don't reproduce the issue in a clean virutal environment. Please add relevant details if you have and reopen the issue if still present.

LaurenzReitsam commented 1 year ago

I have a similar issue. Seems like apache-beam[gcp] dependencies (protobuf==3.*) are not up to date with other google packages (protobuf==4.*).

python -m venv .venv
source .venv/bin/activate
pip install pip-tools
echo -e "apache-beam[gcp]\ngoogle-cloud-secret-manager" > requirements.in
pip-compile

Out:


There are incompatible versions in the resolved dependencies:
  protobuf<5.0.0dev,>=3.19.0 (from google-cloud-recommendations-ai==0.7.1->apache-beam[gcp]==2.43.0->-r requirements.in (line 5))
  protobuf<4.0.0dev (from google-cloud-bigtable==1.7.3->apache-beam[gcp]==2.43.0->-r requirements.in (line 5))
  protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5 (from google-cloud-dlp==3.9.2->apache-beam[gcp]==2.43.0->-r requirements.in (line 5))
  protobuf<4.0.0dev (from google-cloud-language==1.3.2->apache-beam[gcp]==2.43.0->-r requirements.in (line 5))
  protobuf<4,>3.12.2 (from apache-beam[gcp]==2.43.0->-r requirements.in (line 5))
  protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5 (from google-api-core[grpc]==2.11.0->google-cloud-secret-manager==2.12.6->-r requirements.in (line 6))
  protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5 (from google-cloud-secret-manager==2.12.6->-r requirements.in (line 6))
  protobuf!=3.20.0,!=3.20.1,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5 (from googleapis-common-protos[grpc]==1.57.0->grpc-google-iam-v1==0.12.4->google-cloud-secret-manager==2.12.6->-r requirements.in (line 6))
  protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5 (from google-cloud-spanner==3.24.0->apache-beam[gcp]==2.43.0->-r requirements.in (line 5))
  protobuf<4.0.0dev (from google-cloud-vision==1.0.2->apache-beam[gcp]==2.43.0->-r requirements.in (line 5))
  protobuf>=4.21.6 (from grpcio-status==1.51.1->google-cloud-pubsub==2.13.11->apache-beam[gcp]==2.43.0->-r requirements.in (line 5))
  protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5 (from google-cloud-pubsub==2.13.11->apache-beam[gcp]==2.43.0->-r requirements.in (line 5))

Can we reopen this?

tvalentyn commented 1 year ago

It seems like a bug in dependency resolution in piptools. But it may be also time to try to upgrade our protobuf depdendency. Opened https://github.com/apache/beam/issues/24569

LaurenzReitsam commented 1 year ago

@tvalentyn yes it is a problem with pip-tools. Problem can be fixed with adding grpcio-status==1.48.2 as dependency... 🙄

tvalentyn commented 1 year ago

We prefer not to add indirect dependencies, let alone pin versions thereof. Can this be worked around by preinstalling protobuf v3, or explicitly adding it in the requirements.in ?

LaurenzReitsam commented 1 year ago

grpcio-status is not an apache-beam dependency. This is just a workaround for me to fix issues with pipenv.

Just pinning protobuf<4.0.0 does not solve the issue.

Just tried the same procedure with poetry as an alternative dependency solver. Seems like the same error:

pip install poetry
mkdir foo && cd ./foo
poetry init
poetry add apache-beam[gcp] google-cloud-secret-manager

Because no versions of google-cloud-secret-manager match >2.12.6,<3.0.0
 and google-cloud-secret-manager (2.12.6) depends on protobuf (>=3.19.5,<3.20.0 || >3.20.0,<3.20.1 || >3.20.1,<4.21.0 || >4.21.0,<4.21.1 || >4.21.1,<4.21.2 || >4.21.2,<4.21.3 || >4.21.3,<4.21.4 || >4.21.4,<4.21.5 || >4.21.5,<5.0.0dev), google-cloud-secret-manager (>=2.12.6,<3.0.0) requires protobuf (>=3.19.5,<3.20.0 || >3.20.0,<3.20.1 || >3.20.1,<4.21.0 || >4.21.0,<4.21.1 || >4.21.1,<4.21.2 || >4.21.2,<4.21.3 || >4.21.3,<4.21.4 || >4.21.4,<4.21.5 || >4.21.5,<5.0.0dev).
And because apache-beam (2.43.0) depends on protobuf (>3.12.2,<3.19.5)
 and no versions of apache-beam match >2.43.0,<3.0.0, google-cloud-secret-manager (>=2.12.6,<3.0.0) is incompatible with apache-beam (>=2.43.0,<3.0.0).
So, because foo depends on both apache-beam (^2.43.0) and google-cloud-secret-manager (^2.12.6), version solving failed.

Adding protobuf<4.0.0 does not change result.

tvalentyn commented 1 year ago

@AnandInguva do we still need this mac-os only restriction? https://github.com/apache/beam/blob/da100f967a6fbc86a2cc4ae6c9fe69180c70e74f/sdks/python/setup.py#L189

IIRC we added b/c MacOS github action tests were failing. but if that not an issue anymore, it would be good to relax that.

AnandInguva commented 1 year ago

echo -e "apache-beam[gcp]\ngoogle-cloud-secret-manager" > requirements.in
pip-compile

It is due to pip-compile resolver.

You can try using

pip-compile --resolver backtracking with latest version of pip-tools. You can find more details at https://github.com/jazzband/pip-tools/issues/1190

AnandInguva commented 1 year ago

#
# This file is autogenerated by pip-compile with Python 3.8
# by the following command:
#
#    pip-compile --resolver=backtracking
#
--extra-index-url https://pypi.ngc.nvidia.com
--trusted-host pypi.ngc.nvidia.com

apache-beam[gcp]==2.43.0
    # via -r requirements.in
cachetools==4.2.4
    # via
    #   apache-beam
    #   google-auth
certifi==2022.9.24
    # via requests
charset-normalizer==2.1.1
    # via requests
cloudpickle==2.2.0
    # via apache-beam
crcmod==1.7
    # via apache-beam
dill==0.3.1.1
    # via apache-beam
docopt==0.6.2
    # via hdfs
fastavro==1.7.0
    # via apache-beam
fasteners==0.18
    # via
    #   apache-beam
    #   google-apitools
google-api-core[grpc]==2.11.0
    # via
    #   google-cloud-bigquery
    #   google-cloud-bigquery-storage
    #   google-cloud-bigtable
    #   google-cloud-core
    #   google-cloud-datastore
    #   google-cloud-dlp
    #   google-cloud-language
    #   google-cloud-pubsub
    #   google-cloud-pubsublite
    #   google-cloud-recommendations-ai
    #   google-cloud-secret-manager
    #   google-cloud-spanner
    #   google-cloud-videointelligence
    #   google-cloud-vision
google-apitools==0.5.31
    # via apache-beam
google-auth==2.15.0
    # via
    #   apache-beam
    #   google-api-core
    #   google-auth-httplib2
    #   google-cloud-core
google-auth-httplib2==0.1.0
    # via apache-beam
google-cloud-bigquery==3.4.0
    # via apache-beam
google-cloud-bigquery-storage==2.13.2
    # via
    #   apache-beam
    #   google-cloud-bigquery
google-cloud-bigtable==1.7.3
    # via apache-beam
google-cloud-core==2.3.2
    # via
    #   apache-beam
    #   google-cloud-bigquery
    #   google-cloud-bigtable
    #   google-cloud-datastore
    #   google-cloud-spanner
google-cloud-datastore==1.15.5
    # via apache-beam
google-cloud-dlp==3.9.2
    # via apache-beam
google-cloud-language==1.3.2
    # via apache-beam
google-cloud-pubsub==2.13.11
    # via
    #   apache-beam
    #   google-cloud-pubsublite
google-cloud-pubsublite==1.6.0
    # via apache-beam
google-cloud-recommendations-ai==0.7.1
    # via apache-beam
google-cloud-secret-manager==2.12.6
    # via -r requirements.in
google-cloud-spanner==3.24.0
    # via apache-beam
google-cloud-videointelligence==1.16.3
    # via apache-beam
google-cloud-vision==1.0.2
    # via apache-beam
google-crc32c==1.5.0
    # via google-resumable-media
google-resumable-media==2.4.0
    # via google-cloud-bigquery
googleapis-common-protos[grpc]==1.57.0
    # via
    #   google-api-core
    #   grpc-google-iam-v1
    #   grpcio-status
grpc-google-iam-v1==0.12.4
    # via
    #   google-cloud-bigtable
    #   google-cloud-pubsub
    #   google-cloud-secret-manager
    #   google-cloud-spanner
grpcio==1.51.1
    # via
    #   apache-beam
    #   google-api-core
    #   google-cloud-bigquery
    #   google-cloud-pubsub
    #   google-cloud-pubsublite
    #   googleapis-common-protos
    #   grpc-google-iam-v1
    #   grpcio-status
grpcio-status==1.48.2
    # via
    #   google-api-core
    #   google-cloud-pubsub
    #   google-cloud-pubsublite
hdfs==2.7.0
    # via apache-beam
httplib2==0.20.4
    # via
    #   apache-beam
    #   google-apitools
    #   google-auth-httplib2
    #   oauth2client
idna==3.4
    # via requests
numpy==1.22.4
    # via
    #   apache-beam
    #   pyarrow
oauth2client==4.1.3
    # via google-apitools
objsize==0.5.2
    # via apache-beam
orjson==3.8.3
    # via apache-beam
overrides==6.5.0
    # via google-cloud-pubsublite
packaging==21.3
    # via
    #   google-cloud-bigquery
    #   google-cloud-spanner
proto-plus==1.22.1
    # via
    #   apache-beam
    #   google-cloud-bigquery
    #   google-cloud-bigquery-storage
    #   google-cloud-dlp
    #   google-cloud-pubsub
    #   google-cloud-recommendations-ai
    #   google-cloud-secret-manager
    #   google-cloud-spanner
protobuf==3.20.3
    # via
    #   apache-beam
    #   google-api-core
    #   google-cloud-bigquery
    #   google-cloud-bigquery-storage
    #   google-cloud-bigtable
    #   google-cloud-datastore
    #   google-cloud-dlp
    #   google-cloud-language
    #   google-cloud-pubsub
    #   google-cloud-recommendations-ai
    #   google-cloud-secret-manager
    #   google-cloud-spanner
    #   google-cloud-videointelligence
    #   google-cloud-vision
    #   googleapis-common-protos
    #   grpcio-status
    #   proto-plus
pyarrow==9.0.0
    # via
    #   apache-beam
    #   google-cloud-bigquery
pyasn1==0.4.8
    # via
    #   oauth2client
    #   pyasn1-modules
    #   rsa
pyasn1-modules==0.2.8
    # via
    #   google-auth
    #   oauth2client
pydot==1.4.2
    # via apache-beam
pymongo==3.13.0
    # via apache-beam
pyparsing==3.0.9
    # via
    #   httplib2
    #   packaging
    #   pydot
python-dateutil==2.8.2
    # via
    #   apache-beam
    #   google-cloud-bigquery
pytz==2022.6
    # via apache-beam
regex==2022.10.31
    # via apache-beam
requests==2.28.1
    # via
    #   apache-beam
    #   google-api-core
    #   google-cloud-bigquery
    #   hdfs
rsa==4.9
    # via
    #   google-auth
    #   oauth2client
six==1.16.0
    # via
    #   google-apitools
    #   google-auth
    #   google-auth-httplib2
    #   hdfs
    #   oauth2client
    #   python-dateutil
sqlparse==0.4.3
    # via google-cloud-spanner
typing-extensions==4.4.0
    # via apache-beam
urllib3==1.26.13
    # via requests
zstandard==0.19.0
    # via apache-beam

Output with pip-compile --resolver backtracking

AnandInguva commented 1 year ago

@AnandInguva do we still need this mac-os only restriction?

https://github.com/apache/beam/blob/da100f967a6fbc86a2cc4ae6c9fe69180c70e74f/sdks/python/setup.py#L189

IIRC we added b/c MacOS github action tests were failing. but if that not an issue anymore, it would be good to relax that.

Yes, I will take a look at this.

LaurenzReitsam commented 1 year ago

@AnandInguva you are right pip-compile --resolver backtracking works for me on Linux + Python3.9. Just tried this on Windows + Python3.10 until now what resulted in several errors...

...
numpy_linalg_lapack_lite:
  FOUND:
   language = c
  define_macros = [('HAVE_BLAS_ILP64', None), ('BLAS_SYMBOL_SUFFIX', '64_')]

C:\Users\laurenz.reitsam\.pyenv\pyenv-win\versions\3.10.4\lib\distutils\dist.py:274: UserWarning: Unknown distribution option: 'define_macros'
 warnings.warn(msg)
running dist_info
running build_src
build_src
building py_modules sources
creating build
creating build\src.win-amd64-3.10
creating build\src.win-amd64-3.10\numpy
creating build\src.win-amd64-3.10\numpy\distutils
building library "npymath" sources
error: Microsoft Visual C++ 14.0 is required. Get it with "Build Tools for Visual Studio": https://visualstudio.microsoft.com/downloads/
...

Concerning poetry I had the issue in both environments.

jhoareau commented 1 year ago

Hi @AnandInguva we are blocked because of the above Protobuf restriction ('protobuf>3.12.2,<3.19.5') applies to all environments (also non-MacOS) when using Poetry. The reason is that the PyPI repository has this quite restrictive requirement in place: https://pypi.org/pypi/apache-beam/2.43.0/json instead of the looser "<4" for non-macOS environments.

It'd be great if you could either remove the condition from setup.py or fix the PyPI API requirements!

AnandInguva commented 1 year ago

Yes, protobuf is tightly bounded and as of now, there are efforts going on for upgrading protobuf to 4.x.x version in Apache Beam.

We need to have a tight restriction in protobuf because protobuf is a build dependency of beam and minor changes on protobuf version needs to be validated so we couldn't have a loose bound.

jhoareau commented 1 year ago

I understand that. However, the setup.py file seems to have the loose bound (protobuf>3.12.2,<4), while the PyPI bound is tighter, like the macOS build one. Since PyPI usecases usually do not include building but simply installing from wheel, I think it would make sense to relax the PyPI bound to be the same as the loose setup.py bound. Please correct me if I'm wrong!

tvalentyn commented 1 year ago

intresting. is poetry using : the metadata from https://packaging.python.org/en/latest/specifications/core-metadata/#requires-dist-multiple-use / https://pypi.org/pypi/apache-beam/2.43.0/json as SOT?

it appears to be generated incorrectly. @AnandInguva I think we could fix it by adding a platform constraint (see https://peps.python.org/pep-0508/) instead of evaluating the platform dynamically in https://github.com/apache/beam/blob/f40057b7fb68ca556be7fbd0cc420311937dcf80/sdks/python/setup.py#L183

jhoareau commented 1 year ago

Indeed @tvalentyn @AnandInguva https://pypi.org/pypi/apache-beam/2.43.0/json is the source of truth for Poetry, and this source of truth seems to be overly restrictive. Platform constraints sound like a reasonable approach.

apache / beam