First query and visualization test with duckdb and matplotlib

issues-with-duckdb_gh/.sqlfluff

+[sqlfluff]
+dialect = duckdb
+
+[sqlfluff:indentation]
+tab_space_size = 4
+
+[sqlfluff:rules:capitalisation.keywords]
+capitalisation_policy = upper

issues-with-duckdb_gh/duckdb-config-cli.sql

+-- How do I make duckbox output show more than 40 rows?
+-- https://github.com/duckdb/duckdb/discussions/10562
+
+-- Je pensais que c'était un paramètre de configuration :
+--     https://duckdb.org/docs/configuration/overview
+-- mais non c'est une "dot" commande du CLI
+--     https://duckdb.org/docs/api/cli/dot_commands.html
+-- qui ne serait donc pas disponible pour Python
+--     https://stackoverflow.com/a/76860941
+
+.maxrows 80

issues-with-duckdb_gh/plot_issues_stats_by_month.py

+"""
+https://matplotlib.org/2.0.2/users/pyplot_tutorial.html
+https://matplotlib.org/2.0.2/examples/api/barchart_demo.html
+"""
+
+from pathlib import Path
+
+import duckdb
+import dotenv
+
+# import pandas as pd
+import matplotlib.pyplot as plt
+
+
+def generate_date_tick(row):
+    """Generate a month-date label for horizontal axis.
+
+    ATTENTION : row will be a Series with uniform data types
+    - if a column contains float64 and other columns are int64,
+      all row elements will be float64
+    - if a column contains strings (object), all row elements
+      will be object
+    See "Why data types was changed while calling the
+    apply function in Pandas?" : https://stackoverflow.com/a/40315866
+
+    :param _type_ row: A Pandas row with all columns
+    :return _type_: string
+    """
+    return f"{row['cl_year']}-{row['cl_month']:02d}"
+
+
+if __name__ == "__main__":
+    # Get environment variables in an OrderedDict
+    # https://github.com/theskumar/python-dotenv
+    env_vars = dotenv.dotenv_values()
+
+    # Build the path to the issues DuckDB database file
+    issues_db = Path(env_vars.get("PYPI_DB_PATH")) / env_vars.get("PYPI_DB_NAME")
+
+    # Connect to the file (no failure if nothing to expand)
+    cnx = duckdb.connect(str(issues_db.expanduser()), read_only=True)
+
+    # Load query from file
+    with open(env_vars.get("QUERY_ISSUES_STATS_BY_MONTH"), "r", encoding="utf-8") as f:
+        QUERY = f.read()
+
+    # Execute query and get a Pandas DataFrame back
+    # https://duckdb.org/docs/api/python/overview#result-conversion
+    df = cnx.sql(QUERY).df()
+
+    # Create a value for x tick labelling
+    df["x_tick"] = df.apply(generate_date_tick, axis=1)  # , result_type="reduce")
+
+    # fig = plt.figure(num="PyPI Issues", figsize=(12, 8), dpi=600)
+    fig = plt.figure(num="PyPI Issues", figsize=(12, 8), layout="constrained")
+    ax = fig.add_subplot(1, 1, 1)
+
+    # https://matplotlib.org/stable/users/explain/colors/colors.html#colors-def
+    # https://matplotlib.org/stable/users/explain/colors/colormaps.html
+    ax.plot(
+        df["x_tick"],
+        df["nb_days_avg"],
+        color="b",
+        marker="o",
+        label="Number of days between opening and close",
+    )
+
+    ax.plot(
+        df["x_tick"],
+        df["closed_nb"],
+        color="g",
+        marker="o",
+        label="Issues closed by someone else than the opener",
+    )
+
+    ax.set_title(
+        "Statistics on PyPI issues, by month",
+        pad=10,
+    )
+
+    ax.legend()
+
+    # ~/Documents/DEVLOG/Evaluations/2024/Bilan-Formations/bilan-8-global-reseau.matplotlib.py
+    plt.setp(ax.get_xticklabels(), rotation=45, ha="right", rotation_mode="anchor")
+
+    plt.show()

issues-with-duckdb_gh/requirements-dev.in

+-c requirements.txt
+ipython
+pytest

issues-with-duckdb_gh/requirements-dev.txt

+#
+# This file is autogenerated by pip-compile with Python 3.10
+# by the following command:
+#
+#    pip-compile requirements-dev.in
+#
+asttokens==2.4.1
+    # via stack-data
+decorator==5.1.1
+    # via ipython
+exceptiongroup==1.2.1
+    # via
+    #   ipython
+    #   pytest
+executing==2.0.1
+    # via stack-data
+iniconfig==2.0.0
+    # via pytest
+ipython==8.25.0
+    # via -r requirements-dev.in
+jedi==0.19.1
+    # via ipython
+matplotlib-inline==0.1.7
+    # via ipython
+packaging==24.1
+    # via
+    #   -c requirements.txt
+    #   pytest
+parso==0.8.4
+    # via jedi
+pexpect==4.9.0
+    # via ipython
+pluggy==1.5.0
+    # via pytest
+prompt-toolkit==3.0.47
+    # via ipython
+ptyprocess==0.7.0
+    # via pexpect
+pure-eval==0.2.2
+    # via stack-data
+pygments==2.18.0
+    # via ipython
+pytest==8.2.2
+    # via -r requirements-dev.in
+six==1.16.0
+    # via
+    #   -c requirements.txt
+    #   asttokens
+stack-data==0.6.3
+    # via ipython
+tomli==2.0.1
+    # via pytest
+traitlets==5.14.3
+    # via
+    #   ipython
+    #   matplotlib-inline
+typing-extensions==4.12.2
+    # via ipython
+wcwidth==0.2.13
+    # via prompt-toolkit

issues-with-duckdb_gh/requirements.in

+duckdb
+matplotlib
+pandas
+python-dotenv

issues-with-duckdb_gh/requirements.txt

+#
+# This file is autogenerated by pip-compile with Python 3.10
+# by the following command:
+#
+#    pip-compile requirements.in
+#
+contourpy==1.2.1
+    # via matplotlib
+cycler==0.12.1
+    # via matplotlib
+duckdb==1.0.0
+    # via -r requirements.in
+fonttools==4.53.0
+    # via matplotlib
+kiwisolver==1.4.5
+    # via matplotlib
+matplotlib==3.9.0
+    # via -r requirements.in
+numpy==1.26.4
+    # via
+    #   contourpy
+    #   matplotlib
+    #   pandas
+packaging==24.1
+    # via matplotlib
+pandas==2.2.2
+    # via -r requirements.in
+pillow==10.3.0
+    # via matplotlib
+pyparsing==3.1.2
+    # via matplotlib
+python-dateutil==2.9.0.post0
+    # via
+    #   matplotlib
+    #   pandas
+python-dotenv==1.0.1
+    # via -r requirements.in
+pytz==2024.1
+    # via pandas
+six==1.16.0
+    # via python-dateutil
+tzdata==2024.1
+    # via pandas

issues-with-duckdb_gh/select-issues-stats-by-month.sql

+-- Compter les issues "PEP 541" fermées par d'autres personnes que celles qui
+-- les ont ouvertes. Par mois
+
+SELECT
+    year(i.closed_at) AS cl_year,
+    month(i.closed_at) AS cl_month,
+    cast(round(avg(datediff('day', i.created_at, i.closed_at)), 0) AS BIGINT) AS nb_days_avg,
+    count(*) AS closed_nb
+FROM
+    pypi_issues AS i
+INNER JOIN pypi_issues_events AS e
+    ON i.number = e.issue_number
+INNER JOIN pypi_issues_labels AS l
+    ON l.issue_id = i.id
+WHERE
+    i.state = 'closed'
+    AND i.user_login != e.actor_login
+GROUP BY
+    cl_year, cl_month
+ORDER BY
+    cl_year, cl_month;