Remove old flake8 validation

mroeschke · mroeschke · commit d0043987a6ca · 2024-01-13T18:51:44.000-08:00
diff --git a/ci/code_checks.sh b/ci/code_checks.sh
@@ -69,71 +69,6 @@ if [[ -z "$CHECK" || "$CHECK" == "docstrings" ]]; then
     $BASE_DIR/scripts/validate_docstrings.py --format=actions --errors=EX01,EX02,EX04,GL01,GL02,GL03,GL04,GL05,GL06,GL07,GL09,GL10,PR03,PR04,PR05,PR06,PR08,PR09,PR10,RT01,RT02,RT04,RT05,SA02,SA03,SA04,SS01,SS02,SS03,SS04,SS05,SS06
     RET=$(($RET + $?)) ; echo $MSG "DONE"
 
-    MSG='Partially validate docstrings (EX03)' ;  echo $MSG
-    $BASE_DIR/scripts/validate_docstrings.py --format=actions --errors=EX03 --ignore_functions \
-        pandas.Series.plot.line \
-        pandas.Series.to_sql \
-        pandas.Series.to_latex \
-        pandas.errors.DatabaseError \
-        pandas.errors.IndexingError \
-        pandas.errors.InvalidColumnName \
-        pandas.errors.PossibleDataLossError \
-        pandas.errors.PossiblePrecisionLoss \
-        pandas.errors.SettingWithCopyError \
-        pandas.errors.SettingWithCopyWarning \
-        pandas.errors.SpecificationError \
-        pandas.errors.UndefinedVariableError \
-        pandas.errors.ValueLabelTypeMismatch \
-        pandas.Timestamp.ceil \
-        pandas.Timestamp.floor \
-        pandas.Timestamp.round \
-        pandas.read_pickle \
-        pandas.ExcelWriter \
-        pandas.read_json \
-        pandas.io.json.build_table_schema \
-        pandas.DataFrame.to_latex \
-        pandas.io.formats.style.Styler.to_latex \
-        pandas.read_parquet \
-        pandas.DataFrame.to_sql \
-        pandas.read_stata \
-        pandas.core.resample.Resampler.pipe \
-        pandas.core.resample.Resampler.interpolate \
-        pandas.plotting.scatter_matrix \
-        pandas.pivot \
-        pandas.merge_asof \
-        pandas.wide_to_long \
-        pandas.Index.rename \
-        pandas.Index.droplevel \
-        pandas.Index.isin \
-        pandas.MultiIndex.names \
-        pandas.MultiIndex.droplevel \
-        pandas.IndexSlice \
-        pandas.Grouper \
-        pandas.io.formats.style.Styler.map \
-        pandas.io.formats.style.Styler.apply_index \
-        pandas.io.formats.style.Styler.map_index \
-        pandas.io.formats.style.Styler.format \
-        pandas.io.formats.style.Styler.format_index \
-        pandas.io.formats.style.Styler.relabel_index \
-        pandas.io.formats.style.Styler.hide \
-        pandas.io.formats.style.Styler.set_td_classes \
-        pandas.io.formats.style.Styler.set_tooltips \
-        pandas.io.formats.style.Styler.set_uuid \
-        pandas.io.formats.style.Styler.pipe \
-        pandas.io.formats.style.Styler.highlight_between \
-        pandas.io.formats.style.Styler.highlight_quantile \
-        pandas.io.formats.style.Styler.background_gradient \
-        pandas.io.formats.style.Styler.text_gradient \
-        pandas.DataFrame.values \
-        pandas.DataFrame.groupby \
-        pandas.DataFrame.idxmax \
-        pandas.DataFrame.idxmin \
-        pandas.DataFrame.pivot \
-        pandas.DataFrame.sort_values \
-        pandas.DataFrame.plot.hexbin \
-        pandas.DataFrame.plot.line \
-    RET=$(($RET + $?)) ; echo $MSG "DONE"
-
 fi
 
 ### DOCUMENTATION NOTEBOOKS ###
diff --git a/environment.yml b/environment.yml
@@ -75,7 +75,6 @@ dependencies:
   - cxx-compiler
 
   # code checks
-  - flake8=6.1.0  # run in subprocess over docstring examples
   - mypy=1.8.0  # pre-commit uses locally installed mypy
   - tokenize-rt  # scripts/check_for_inconsistent_pandas_namespace.py
   - pre-commit>=3.6.0
diff --git a/requirements-dev.txt b/requirements-dev.txt
@@ -52,7 +52,6 @@ seaborn
 moto
 flask
 asv>=0.6.1
-flake8==6.1.0
 mypy==1.8.0
 tokenize-rt
 pre-commit>=3.6.0
diff --git a/scripts/tests/test_validate_docstrings.py b/scripts/tests/test_validate_docstrings.py
@@ -44,37 +44,6 @@ def redundant_import(self, paramx=None, paramy=None) -> None:
         Series([], dtype: bool)
         """
 
-    def unused_import(self) -> None:
-        """
-        Examples
-        --------
-        >>> import pandas as pdf
-        >>> df = pd.DataFrame(np.ones((3, 3)), columns=("a", "b", "c"))
-        """
-
-    def missing_whitespace_around_arithmetic_operator(self) -> None:
-        """
-        Examples
-        --------
-        >>> 2 + 5
-        7
-        """
-
-    def indentation_is_not_a_multiple_of_four(self) -> None:
-        """
-        Examples
-        --------
-        >>> if 2 + 5:
-        ...     pass
-        """
-
-    def missing_whitespace_after_comma(self) -> None:
-        """
-        Examples
-        --------
-        >>> df = pd.DataFrame(np.ones((3, 3)), columns=("a", "b", "c"))
-        """
-
     def write_array_like_with_hyphen_not_underscore(self) -> None:
         """
         In docstrings, use array-like over array_like
@@ -153,37 +122,6 @@ def test_bad_class(self, capsys) -> None:
                 "redundant_import",
                 ("Do not import pandas, as it is imported automatically",),
             ),
-            (
-                "BadDocstrings",
-                "unused_import",
-                (
-                    "flake8 error: line 1, col 1: F401 'pandas as pdf' "
-                    "imported but unused",
-                ),
-            ),
-            (
-                "BadDocstrings",
-                "missing_whitespace_around_arithmetic_operator",
-                (
-                    "flake8 error: line 1, col 2: "
-                    "E226 missing whitespace around arithmetic operator",
-                ),
-            ),
-            (
-                "BadDocstrings",
-                "indentation_is_not_a_multiple_of_four",
-                # with flake8 3.9.0, the message ends with four spaces,
-                #  whereas in earlier versions, it ended with "four"
-                (
-                    "flake8 error: line 2, col 3: E111 indentation is not a "
-                    "multiple of 4",
-                ),
-            ),
-            (
-                "BadDocstrings",
-                "missing_whitespace_after_comma",
-                ("flake8 error: line 1, col 33: E231 missing whitespace after ','",),
-            ),
             (
                 "BadDocstrings",
                 "write_array_like_with_hyphen_not_underscore",
diff --git a/scripts/validate_docstrings.py b/scripts/validate_docstrings.py
@@ -22,9 +22,7 @@
 import json
 import os
 import pathlib
-import subprocess
 import sys
-import tempfile
 
 import matplotlib
 import matplotlib.pyplot as plt
@@ -64,8 +62,6 @@
     "SA05": "{reference_name} in `See Also` section does not need `pandas` "
     "prefix, use {right_reference} instead.",
     "EX02": "Examples do not pass tests:\n{doctest_log}",
-    "EX03": "flake8 error: line {line_number}, col {col_number}: {error_code} "
-    "{error_message}",
     "EX04": "Do not import {imported_library}, as it is imported "
     "automatically for the examples (numpy as np, pandas as pd)",
 }
@@ -198,55 +194,6 @@ def examples_source_code(self):
         lines = doctest.DocTestParser().get_examples(self.raw_doc)
         return [line.source for line in lines]
 
-    def validate_pep8(self):
-        if not self.examples:
-            return
-
-        # F401 is needed to not generate flake8 errors in examples
-        # that do not user numpy or pandas
-        content = "".join(
-            (
-                "import numpy as np  # noqa: F401\n",
-                "import pandas as pd  # noqa: F401\n",
-                *self.examples_source_code,
-            )
-        )
-
-        error_messages = []
-
-        file = tempfile.NamedTemporaryFile(mode="w", encoding="utf-8", delete=False)
-        try:
-            file.write(content)
-            file.flush()
-            cmd = [
-                "python",
-                "-m",
-                "flake8",
-                "--format=%(row)d\t%(col)d\t%(code)s\t%(text)s",
-                "--max-line-length=88",
-                "--ignore=E203,E3,W503,W504,E402,E731",
-                file.name,
-            ]
-            response = subprocess.run(cmd, capture_output=True, check=False, text=True)
-            for output in ("stdout", "stderr"):
-                out = getattr(response, output)
-                out = out.replace(file.name, "")
-                messages = out.strip("\n").splitlines()
-                if messages:
-                    error_messages.extend(messages)
-        finally:
-            file.close()
-            os.unlink(file.name)
-
-        for error_message in error_messages:
-            line_number, col_number, error_code, message = error_message.split(
-                "\t", maxsplit=3
-            )
-            # Note: we subtract 2 from the line number because
-            # 'import numpy as np\nimport pandas as pd\n'
-            # is prepended to the docstrings.
-            yield error_code, message, int(line_number) - 2, int(col_number)
-
     def non_hyphenated_array_like(self):
         return "array_like" in self.raw_doc
 
@@ -296,16 +243,6 @@ def pandas_validate(func_name: str):
                 pandas_error("EX02", doctest_log=result["examples_errs"])
             )
 
-        for error_code, error_message, line_number, col_number in doc.validate_pep8():
-            result["errors"].append(
-                pandas_error(
-                    "EX03",
-                    error_code=error_code,
-                    error_message=error_message,
-                    line_number=line_number,
-                    col_number=col_number,
-                )
-            )
         examples_source_code = "".join(doc.examples_source_code)
         result["errors"].extend(
             pandas_error("EX04", imported_library=wrong_import)
