ENH: Move UndefinedVariableError to error/__init__.py per GH27656 (#47338)

Author: dataxerik

doc/source/reference/testing.rst

@@ -45,6 +45,7 @@ Exceptions and warnings
    errors.SettingWithCopyError
    errors.SettingWithCopyWarning
    errors.SpecificationError
+   errors.UndefinedVariableError
    errors.UnsortedIndexError
    errors.UnsupportedFunctionCall
 

doc/source/whatsnew/v1.5.0.rst

@@ -152,7 +152,7 @@ Other enhancements
 - A :class:`errors.PerformanceWarning` is now thrown when using ``string[pyarrow]`` dtype with methods that don't dispatch to ``pyarrow.compute`` methods (:issue:`42613`)
 - Added ``numeric_only`` argument to :meth:`Resampler.sum`, :meth:`Resampler.prod`, :meth:`Resampler.min`, :meth:`Resampler.max`, :meth:`Resampler.first`, and :meth:`Resampler.last` (:issue:`46442`)
 - ``times`` argument in :class:`.ExponentialMovingWindow` now accepts ``np.timedelta64`` (:issue:`47003`)
-- :class:`DataError`, :class:`SpecificationError`, :class:`SettingWithCopyError`, :class:`SettingWithCopyWarning`, and :class:`NumExprClobberingError` are now exposed in ``pandas.errors`` (:issue:`27656`)
+- :class:`DataError`, :class:`SpecificationError`, :class:`SettingWithCopyError`, :class:`SettingWithCopyWarning`, :class:`NumExprClobberingError`, :class:`UndefinedVariableError` are now exposed in ``pandas.errors`` (:issue:`27656`)
 - Added ``check_like`` argument to :func:`testing.assert_series_equal` (:issue:`47247`)
 
 .. ---------------------------------------------------------------------------

pandas/core/computation/expr.py

@@ -18,6 +18,7 @@
 import numpy as np
 
 from pandas.compat import PY39
+from pandas.errors import UndefinedVariableError
 
 import pandas.core.common as com
 from pandas.core.computation.ops import (
@@ -35,7 +36,6 @@
     Op,
     Term,
     UnaryOp,
-    UndefinedVariableError,
     is_term,
 )
 from pandas.core.computation.parsing import (

pandas/core/computation/ops.py

@@ -65,20 +65,6 @@
 LOCAL_TAG = "__pd_eval_local_"
 
 
-class UndefinedVariableError(NameError):
-    """
-    NameError subclass for local variables.
-    """
-
-    def __init__(self, name: str, is_local: bool | None = None) -> None:
-        base_msg = f"{repr(name)} is not defined"
-        if is_local:
-            msg = f"local variable {base_msg}"
-        else:
-            msg = f"name {base_msg}"
-        super().__init__(msg)
-
-
 class Term:
     def __new__(cls, name, env, side=None, encoding=None):
         klass = Constant if not isinstance(name, str) else cls

pandas/core/computation/pytables.py

@@ -13,6 +13,7 @@
 )
 from pandas._typing import npt
 from pandas.compat.chainmap import DeepChainMap
+from pandas.errors import UndefinedVariableError
 
 from pandas.core.dtypes.common import is_list_like
 
@@ -24,10 +25,7 @@
 )
 from pandas.core.computation.common import ensure_decoded
 from pandas.core.computation.expr import BaseExprVisitor
-from pandas.core.computation.ops import (
-    UndefinedVariableError,
-    is_term,
-)
+from pandas.core.computation.ops import is_term
 from pandas.core.construction import extract_array
 from pandas.core.indexes.base import Index
 

pandas/core/computation/scope.py

@@ -15,6 +15,7 @@
 
 from pandas._libs.tslibs import Timestamp
 from pandas.compat.chainmap import DeepChainMap
+from pandas.errors import UndefinedVariableError
 
 
 def ensure_scope(
@@ -207,9 +208,6 @@ def resolve(self, key: str, is_local: bool):
                 # e.g., df[df > 0]
                 return self.temps[key]
             except KeyError as err:
-                # runtime import because ops imports from scope
-                from pandas.core.computation.ops import UndefinedVariableError
-
                 raise UndefinedVariableError(key, is_local) from err
 
     def swapkey(self, old_key: str, new_key: str, new_value=None) -> None:

pandas/errors/__init__.py

@@ -1,6 +1,7 @@
 """
 Expose public exceptions & warnings
 """
+from __future__ import annotations
 
 from pandas._config.config import OptionError  # noqa:F401
 
@@ -326,3 +327,29 @@ class NumExprClobberingError(NameError):
     >>> pd.eval("sin + a", engine='numexpr') # doctest: +SKIP
     ... # NumExprClobberingError: Variables in expression "(sin) + (a)" overlap...
     """
+
+
+class UndefinedVariableError(NameError):
+    """
+    Exception is raised when trying to use an undefined variable name in a method
+    like query or eval. It will also specific whether the undefined variable is
+    local or not.
+
+    Examples
+    --------
+    >>> df = pd.DataFrame({'A': [1, 1, 1]})
+    >>> df.query("A > x") # doctest: +SKIP
+    ... # UndefinedVariableError: name 'x' is not defined
+    >>> df.query("A > @y") # doctest: +SKIP
+    ... # UndefinedVariableError: local variable 'y' is not defined
+    >>> pd.eval('x + 1') # doctest: +SKIP
+    ... # UndefinedVariableError: name 'x' is not defined
+    """
+
+    def __init__(self, name: str, is_local: bool | None = None) -> None:
+        base_msg = f"{repr(name)} is not defined"
+        if is_local:
+            msg = f"local variable {base_msg}"
+        else:
+            msg = f"name {base_msg}"
+        super().__init__(msg)

pandas/tests/computation/test_eval.py

@@ -12,6 +12,7 @@
 from pandas.errors import (
     NumExprClobberingError,
     PerformanceWarning,
+    UndefinedVariableError,
 )
 import pandas.util._test_decorators as td
 
@@ -44,7 +45,6 @@
 from pandas.core.computation.ops import (
     ARITH_OPS_SYMS,
     SPECIAL_CASE_ARITH_OPS_SYMS,
-    UndefinedVariableError,
     _binary_math_ops,
     _binary_ops_dict,
     _unary_math_ops,

pandas/tests/frame/test_query_eval.py

@@ -495,7 +495,7 @@ def test_query_syntax_error(self):
             df.query("i - +", engine=engine, parser=parser)
 
     def test_query_scope(self):
-        from pandas.core.computation.ops import UndefinedVariableError
+        from pandas.errors import UndefinedVariableError
 
         engine, parser = self.engine, self.parser
         skip_if_no_pandas_parser(parser)
@@ -522,7 +522,7 @@ def test_query_scope(self):
             df.query("@a > b > c", engine=engine, parser=parser)
 
     def test_query_doesnt_pickup_local(self):
-        from pandas.core.computation.ops import UndefinedVariableError
+        from pandas.errors import UndefinedVariableError
 
         engine, parser = self.engine, self.parser
         n = m = 10
@@ -618,7 +618,7 @@ def test_nested_scope(self):
         tm.assert_frame_equal(result, expected)
 
     def test_nested_raises_on_local_self_reference(self):
-        from pandas.core.computation.ops import UndefinedVariableError
+        from pandas.errors import UndefinedVariableError
 
         df = DataFrame(np.random.randn(5, 3))
 
@@ -678,7 +678,7 @@ def test_at_inside_string(self):
         tm.assert_frame_equal(result, expected)
 
     def test_query_undefined_local(self):
-        from pandas.core.computation.ops import UndefinedVariableError
+        from pandas.errors import UndefinedVariableError
 
         engine, parser = self.engine, self.parser
         skip_if_no_pandas_parser(parser)
@@ -838,7 +838,7 @@ def test_date_index_query_with_NaT_duplicates(self):
             df.query("index < 20130101 < dates3", engine=engine, parser=parser)
 
     def test_nested_scope(self):
-        from pandas.core.computation.ops import UndefinedVariableError
+        from pandas.errors import UndefinedVariableError
 
         engine = self.engine
         parser = self.parser

pandas/tests/test_errors.py

@@ -1,6 +1,9 @@
 import pytest
 
-from pandas.errors import AbstractMethodError
+from pandas.errors import (
+    AbstractMethodError,
+    UndefinedVariableError,
+)
 
 import pandas as pd
 
@@ -48,6 +51,24 @@ def test_catch_oob():
         pd.Timestamp("15000101")
 
 
+@pytest.mark.parametrize(
+    "is_local",
+    [
+        True,
+        False,
+    ],
+)
+def test_catch_undefined_variable_error(is_local):
+    variable_name = "x"
+    if is_local:
+        msg = f"local variable '{variable_name}' is not defined"
+    else:
+        msg = f"name '{variable_name}' is not defined"
+
+    with pytest.raises(UndefinedVariableError, match=msg):
+        raise UndefinedVariableError(variable_name, is_local)
+
+
 class Foo:
     @classmethod
     def classmethod(cls):

scripts/pandas_errors_documented.py

@@ -22,7 +22,7 @@ def get_defined_errors(content: str) -> set[str]:
     for node in ast.walk(ast.parse(content)):
         if isinstance(node, ast.ClassDef):
             errors.add(node.name)
-        elif isinstance(node, ast.ImportFrom):
+        elif isinstance(node, ast.ImportFrom) and node.module != "__future__":
             for alias in node.names:
                 errors.add(alias.name)
     return errors
@@ -41,7 +41,7 @@ def main(argv: Sequence[str] | None = None) -> None:
     missing = file_errors.difference(doc_errors)
     if missing:
         sys.stdout.write(
-            f"The follow exceptions and/or warnings are not documented "
+            f"The following exceptions and/or warnings are not documented "
             f"in {API_PATH}: {missing}"
         )
         sys.exit(1)