Extras should be sorted in lock file

[baryluk]

• I am on the latest Poetry version.

• I have searched the issues of this repo and believe that this is not a duplicate.

• OS version and name: Linux

• Poetry version: 1.1.7

Issue

poetry.lock does not use stable order:

Example:

-webhdfs = ["hdfs[kerberos,dataframe,avro] (>=2.0.4)"]
+webhdfs = ["hdfs[dataframe,kerberos,avro] (>=2.0.4)"]

-devel_hadoop = ["aws-xray-sdk", "bcrypt (>=2.0.0)", "beautifulsoup4 (>=4.7.1,<4.8.0)", "black", "blinker", "bowler", "cgroupspy (>=0.1.4)", "click (>=7.1,<8.0)", "coverage", "cryptography (>=2.0.0)", "filelock", "flake8-colors", "flake8 (>=3.6.0)", "flaky", "flask-bcrypt (>=0.7.1)", "freezegun", "github3.py", "gitpython", "hdfs[kerberos,dataframe,avro] (>=2.0.4)", "hmsclient (>=0.1.0)", "importlib-resources (>=1.4,<2.0)", "ipdb", "jira", "jsondiff", "kubernetes (>=3.0.0,<12.0.0)", "mongomock", "moto (>=2.0,<3.0)", "mypy (==0.770)", "mysql-connector-python (>=8.0.11,<=8.0.22)", "mysqlclient (>=1.3.6,<3)", "parameterized", "paramiko", "pipdeptree", "pre-commit", "presto-python-client (>=0.7.0,<0.8)", "pygithub", "pykerberos (>=1.1.13)", "pysftp", "pytest-cov", "pytest-httpx", "pytest-instafail", "pytest-rerunfailures (>=9.1,<10.0)", "pytest-timeouts", "pytest-xdist", "pytest (>=6.0,<7.0)", "python-jose", "pywinrm", "qds-sdk (>=1.9.6)", "requests-kerberos (>=0.10.0)", "requests-mock", "snakebite-py3", "sphinx-airflow-theme", "sphinx-argparse (>=0.1.13)", "sphinx-autoapi (==1.0.0)", "sphinx-copybutton", "sphinx-jinja (>=1.1,<2.0)", "sphinx-rtd-theme (>=0.1.6)", "sphinx (>=2.1.2,<3.5.0)", "sphinxcontrib-httpdomain (>=1.7.0)", "sphinxcontrib-redoc (>=1.6.0)", "sphinxcontrib-spelling (==5.2.1)", "thrift (>=0.9.2)", "thrift-sasl (>=0.2.0)", "wheel", "yamllint", "apache-airflow-providers-apache-hdfs", "apache-airflow-providers-apache-hive", "apache-airflow-providers-presto", "apache-airflow-providers-trino", "pyhive[hive] (>=0.6.0)"]
+devel_hadoop = ["aws-xray-sdk", "bcrypt (>=2.0.0)", "beautifulsoup4 (>=4.7.1,<4.8.0)", "black", "blinker", "bowler", "cgroupspy (>=0.1.4)", "click (>=7.1,<8.0)", "coverage", "cryptography (>=2.0.0)", "filelock", "flake8-colors", "flake8 (>=3.6.0)", "flaky", "flask-bcrypt (>=0.7.1)", "freezegun", "github3.py", "gitpython", "hdfs[dataframe,kerberos,avro] (>=2.0.4)", "hmsclient (>=0.1.0)", "importlib-resources (>=1.4,<2.0)", "ipdb", "jira", "jsondiff", "kubernetes (>=3.0.0,<12.0.0)", "mongomock", "moto (>=2.0,<3.0)", "mypy (==0.770)", "mysql-connector-python (>=8.0.11,<=8.0.22)", "mysqlclient (>=1.3.6,<3)", "parameterized", "paramiko", "pipdeptree", "pre-commit", "presto-python-client (>=0.7.0,<0.8)", "pygithub", "pykerberos (>=1.1.13)", "pysftp", "pytest-cov", "pytest-httpx", "pytest-instafail", "pytest-rerunfailures (>=9.1,<10.0)", "pytest-timeouts", "pytest-xdist", "pytest (>=6.0,<7.0)", "python-jose", "pywinrm", "qds-sdk (>=1.9.6)", "requests-kerberos (>=0.10.0)", "requests-mock", "snakebite-py3", "sphinx-airflow-theme", "sphinx-argparse (>=0.1.13)", "sphinx-autoapi (==1.0.0)", "sphinx-copybutton", "sphinx-jinja (>=1.1,<2.0)", "sphinx-rtd-theme (>=0.1.6)", "sphinx (>=2.1.2,<3.5.0)", "sphinxcontrib-httpdomain (>=1.7.0)", "sphinxcontrib-redoc (>=1.6.0)", "sphinxcontrib-spelling (==5.2.1)", "thrift (>=0.9.2)", "thrift-sasl (>=0.2.0)", "wheel", "yamllint", "apache-airflow-providers-apache-hdfs", "apache-airflow-providers-apache-hive", "apache-airflow-providers-presto", "apache-airflow-providers-trino", "pyhive[hive] (>=0.6.0)"]

[philippefutureboy]

+1 - This is problematic for PR merges - we are using Apache Airflow and about every dependency bump results in merge conflict.

[mgdelmonte]

+1 - issue still persists as of Poetry version 1.1.13

This is a trivially solvable problem which, until solved, causes no end of headaches on teams with multiple contributors whose machines may list extras in different orders. Sorting is an obvious solution.

[radoering]

Should be fixed in poetry 1.2.0b1.

Duplicate: #5125
Resolved by: python-poetry/poetry-core#280

[github-actions]

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.