update pylint (#630)

* unpin pylint, remove setuptools constraint

* bump cache

* install simpleeval in noxfile

* re-pin pylint

* fix lint

* nox uses setuptools < 58.0.0

.github/workflows/ci-tests.yml

@@ -17,7 +17,7 @@ env:
   DEFAULT_PYTHON: 3.8
   CI: "true"
   # Increase this value to reset cache if environment.yml has not changed
-  CACHE_VERSION: 4
+  CACHE_VERSION: 5
 
 jobs:
 

.pre-commit-config.yaml

@@ -41,7 +41,7 @@ repos:
         args: ["--line-length=79"]
 
   - repo: https://github.com/pycqa/pylint
-    rev: v2.10.2
+    rev: v2.11.1
     hooks:
       - id: pylint
         args: ["--disable=import-error"]

environment.yml

@@ -26,13 +26,12 @@ dependencies:
   - isort >= 5.7.0
   - codecov
   - mypy >= 0.902 # mypy no longer bundle stubs for third-party libraries
-  - pylint = 2.10.2
+  - pylint = v2.11.1
   - pytest
   - pytest-cov
   - pytest-xdist
   - pytest-asyncio
   - xdoctest
-  - setuptools < 58.0.0
   - nox = 2020.12.31 # pinning due to UnicodeDecodeError, see https://github.com/pandera-dev/pandera/pull/504/checks?check_run_id=2841360122
   - importlib_metadata # required if python < 3.8
 

pandera/hypotheses.py

@@ -210,8 +210,8 @@ def _relationships(self, relationship: Union[str, Callable]):
             relationship = self.RELATIONSHIPS[relationship]
         elif not callable(relationship):
             raise ValueError(
-                "expected relationship to be str or callable, found %s"
-                % type(relationship)
+                "expected relationship to be str or callable, found "
+                f"{type(relationship)}"
             )
         return relationship
 

pandera/io.py

@@ -270,7 +270,7 @@ def from_yaml(yaml_schema):
     :returns: dataframe schema.
     """
     try:
-        with Path(yaml_schema).open("r") as f:
+        with Path(yaml_schema).open("r", encoding="utf-8") as f:
             serialized_schema = yaml.safe_load(f)
     except (TypeError, OSError):
         serialized_schema = yaml.safe_load(yaml_schema)
@@ -290,7 +290,7 @@ def _write_yaml(obj, stream):
         return yaml.safe_dump(obj, stream=stream, sort_keys=False)
 
     try:
-        with Path(stream).open("w") as f:
+        with Path(stream).open("w", encoding="utf-8") as f:
             _write_yaml(statistics, f)
     except (TypeError, OSError):
         return _write_yaml(statistics, stream)
@@ -437,7 +437,7 @@ def to_script(dataframe_schema, path_or_buf=None):
     if path_or_buf is None:
         return formatted_script
 
-    with Path(path_or_buf).open("w") as f:
+    with Path(path_or_buf).open("w", encoding="utf-8") as f:
         f.write(formatted_script)
 
 

pandera/schemas.py

@@ -13,7 +13,7 @@
 import numpy as np
 import pandas as pd
 
-from . import constants, errors
+from . import errors
 from . import strategies as st
 from .checks import Check
 from .deprecations import deprecate_pandas_dtype
@@ -278,9 +278,9 @@ def dtypes(self) -> Dict[str, DataType]:
         ]
         if regex_columns:
             warnings.warn(
-                "Schema has columns specified as regex column names: %s "
-                "Use the `get_dtypes` to get the datatypes for these "
-                "columns." % regex_columns,
+                "Schema has columns specified as regex column names: "
+                f"{regex_columns}. Use the `get_dtypes` to get the datatypes "
+                "for these columns.",
                 UserWarning,
             )
         return {n: c.dtype for n, c in self.columns.items() if not c.regex}
@@ -460,11 +460,10 @@ def validate(
 
         if self._is_inferred:
             warnings.warn(
-                "This %s is an inferred schema that hasn't been "
+                f"This {type(self)} is an inferred schema that hasn't been "
                 "modified. It's recommended that you refine the schema "
                 "by calling `add_columns`, `remove_columns`, or "
-                "`update_columns` before using it to validate data."
-                % type(self),
+                "`update_columns` before using it to validate data.",
                 UserWarning,
             )
 
@@ -1744,10 +1743,9 @@ def validate(
 
         if self._is_inferred:
             warnings.warn(
-                "This %s is an inferred schema that hasn't been "
+                f"This {type(self)} is an inferred schema that hasn't been "
                 "modified. It's recommended that you refine the schema "
-                "by calling `set_checks` before using it to validate data."
-                % type(self),
+                "by calling `set_checks` before using it to validate data.",
                 UserWarning,
             )
 
@@ -1771,10 +1769,9 @@ def validate(
         )
 
         if self.name is not None and series.name != self._name:
-            msg = "Expected %s to have name '%s', found '%s'" % (
-                type(self),
-                self._name,
-                series.name,
+            msg = (
+                f"Expected {type(self)} to have name '{self._name}', found "
+                f"'{series.name}'"
             )
             error_handler.collect_error(
                 "wrong_field_name",
@@ -1790,9 +1787,10 @@ def validate(
         if not self._nullable:
             nulls = series.isna()
             if sum(nulls) > 0:
-                msg = "non-nullable series '%s' contains null values: %s" % (
-                    series.name,
-                    series[nulls].head(constants.N_FAILURE_CASES).to_dict(),
+                failed = series[nulls]
+                msg = (
+                    f"non-nullable series '{series.name}' contains null "
+                    f"values:\n{failed}"
                 )
                 error_handler.collect_error(
                     "series_contains_nulls",
@@ -1811,11 +1809,10 @@ def validate(
         if self._unique:
             duplicates = series.duplicated()
             if any(duplicates):
-                msg = "series '%s' contains duplicate values: %s" % (
-                    series.name,
-                    series[duplicates]
-                    .head(constants.N_FAILURE_CASES)
-                    .to_dict(),
+                failed = series[duplicates]
+                msg = (
+                    f"series '{series.name}' contains duplicate values:\n"
+                    f"{series[duplicates]}"
                 )
                 error_handler.collect_error(
                     "series_contains_duplicates",

requirements-dev.txt

@@ -17,13 +17,12 @@ black >= 20.8b1
 isort >= 5.7.0
 codecov
 mypy >= 0.902
-pylint == 2.10.2
+pylint == v2.11.1
 pytest
 pytest-cov
 pytest-xdist
 pytest-asyncio
 xdoctest
-setuptools < 58.0.0
 nox == 2020.12.31
 importlib_metadata
 sphinx

setup.py

@@ -44,7 +44,6 @@
         "typing_inspect >= 0.6.0",
         "wrapt",
         "pyarrow",
-        "setuptools < 58.0.0",
     ],
     extras_require=extras_require,
     python_requires=">=3.7",

tests/core/test_decorators.py

@@ -819,7 +819,7 @@ class Meta(type):
         @check_output(Schema.to_schema())
         @check_input(Schema.to_schema(), "df1")
         @check_io(df1=Schema.to_schema(), out=Schema.to_schema())
-        async def regular_meta_coroutine(
+        async def regular_meta_coroutine(  # pylint: disable=no-self-use
             cls,
             df1: DataFrame[Schema],
         ) -> DataFrame[Schema]:

tests/core/test_schemas.py

@@ -160,15 +160,15 @@ def test_dataframe_schema_strict_regex() -> None:
         {"foo_*": Column(int, regex=True)},
         strict=True,
     )
-    df = pd.DataFrame({"foo_%d" % i: range(10) for i in range(5)})
+    df = pd.DataFrame({f"foo_{i}": range(10) for i in range(5)})
 
     assert isinstance(schema.validate(df), pd.DataFrame)
 
     # Raise a SchemaError if schema is strict and a regex pattern yields
     # no matches
     with pytest.raises(errors.SchemaError):
         schema.validate(
-            pd.DataFrame({"bar_%d" % i: range(10) for i in range(5)})
+            pd.DataFrame({f"bar_{i}": range(10) for i in range(5)})
         )