Replace assert statements with argument validation

Author: rgkimball

pandas/core/apply.py

@@ -513,7 +513,8 @@ def normalize_dictlike_arg(
         that a nested renamer is not passed. Also normalizes to all lists
         when values consists of a mix of list and non-lists.
         """
-        assert how in ("apply", "agg", "transform")
+        if how not in ("apply", "agg", "transform"):
+            raise ValueError('Value for how argument must be one of : apply, agg, transform')
 
         # Can't use func.values(); wouldn't work for a Series
         if (

pandas/core/arrays/_ranges.py

@@ -102,7 +102,8 @@ def _generate_range_overflow_safe(
     OutOfBoundsDatetime
     """
     # GH#14187 raise instead of incorrectly wrapping around
-    assert side in ["start", "end"]
+    if side not in ["start", "end"]:
+        raise ValueError('Value for side argument must be one of: start, end')
 
     i64max = np.uint64(np.iinfo(np.int64).max)
     msg = f"Cannot generate range with {side}={endpoint} and periods={periods}"
@@ -148,8 +149,9 @@ def _generate_range_overflow_safe_signed(
     A special case for _generate_range_overflow_safe where `periods * stride`
     can be calculated without overflowing int64 bounds.
     """
-    assert side in ["start", "end"]
-    if side == "end":
+    if side not in ['start', 'end']:
+        raise ValueError('Value for side argument must be one of: start, end')
+    if side == 'end':
         stride *= -1
 
     with np.errstate(over="raise"):

pandas/core/arrays/datetimes.py

@@ -2155,7 +2155,8 @@ def objects_to_datetime64ns(
     ------
     ValueError : if data cannot be converted to datetimes
     """
-    assert errors in ["raise", "ignore", "coerce"]
+    if errors not in ["raise", "ignore", "coerce"]:
+        raise ValueError('Value for errors argument must be one of: raise, coerce, ignore')
 
     # if str-dtype, convert
     data = np.array(data, copy=False, dtype=np.object_)

pandas/core/groupby/ops.py

@@ -967,7 +967,8 @@ def _cython_operation(
         """
         Returns the values of a cython operation.
         """
-        assert kind in ["transform", "aggregate"]
+        if kind not in ["transform", "aggregate"]:
+            raise ValueError('Value for kind argument must be one of: transform, aggregate')
 
         cy_op = WrappedCythonOp(kind=kind, how=how)
 

pandas/core/indexes/base.py

@@ -3640,7 +3640,8 @@ def _convert_slice_indexer(self, key: slice, kind: str_t):
         key : label of the slice bound
         kind : {'loc', 'getitem'}
         """
-        assert kind in ["loc", "getitem"], kind
+        if kind not in ["loc", "getitem"]:
+            raise ValueError('Value for kind argument must be one of: loc, getitem')
 
         # potentially cast the bounds to integers
         start, stop, step = key.start, key.stop, key.step
@@ -5687,7 +5688,8 @@ def _validate_indexer(self, form: str_t, key, kind: str_t):
         If we are positional indexer, validate that we have appropriate
         typed bounds must be an integer.
         """
-        assert kind in ["getitem", "iloc"]
+        if kind not in ["getitem", "iloc"]:
+            raise ValueError('Value for kind argument must be one of: getitem, iloc')
 
         if key is not None and not is_integer(key):
             raise self._invalid_indexer(form, key)
@@ -5712,7 +5714,8 @@ def _maybe_cast_slice_bound(self, label, side: str_t, kind=no_default):
         -----
         Value of `side` parameter should be validated in caller.
         """
-        assert kind in ["loc", "getitem", None, no_default]
+        if kind not in ["loc", "getitem", None, no_default]:
+            raise ValueError('Value for kind argument must be one of: loc, getitem or None')
         self._deprecated_arg(kind, "kind", "_maybe_cast_slice_bound")
 
         # We are a plain index here (sub-class override this method if they
@@ -5756,7 +5759,8 @@ def get_slice_bound(self, label, side: str_t, kind=None) -> int:
         int
             Index of label.
         """
-        assert kind in ["loc", "getitem", None]
+        if kind not in ["loc", "getitem", None]:
+            raise ValueError('Value for kind argument must be one of: loc, getitem or None')
 
         if side not in ("left", "right"):
             raise ValueError(

pandas/core/indexes/datetimes.py

@@ -730,7 +730,8 @@ def _maybe_cast_slice_bound(self, label, side: str, kind=lib.no_default):
         -----
         Value of `side` parameter should be validated in caller.
         """
-        assert kind in ["loc", "getitem", None, lib.no_default]
+        if kind not in ["loc", "getitem", None, lib.no_default]:
+            raise ValueError('Value for kind argument must be one of: loc, getitem or None')
         self._deprecated_arg(kind, "kind", "_maybe_cast_slice_bound")
 
         if isinstance(label, str):

pandas/core/indexes/numeric.py

@@ -235,7 +235,8 @@ def _should_fallback_to_positional(self) -> bool:
     @doc(Index._convert_slice_indexer)
     def _convert_slice_indexer(self, key: slice, kind: str):
         if is_float_dtype(self.dtype):
-            assert kind in ["loc", "getitem"]
+            if kind not in ["loc", "getitem"]:
+                raise ValueError('Value for kind argument must be one of: loc, getitem')
 
             # We always treat __getitem__ slicing as label-based
             # translate to locations
@@ -245,7 +246,8 @@ def _convert_slice_indexer(self, key: slice, kind: str):
 
     @doc(Index._maybe_cast_slice_bound)
     def _maybe_cast_slice_bound(self, label, side: str, kind=lib.no_default):
-        assert kind in ["loc", "getitem", None, lib.no_default]
+        if kind not in ["loc", "getitem", None, lib.no_default]:
+            raise ValueError('Value for kind argument must be one of: loc, getitem or None')
         self._deprecated_arg(kind, "kind", "_maybe_cast_slice_bound")
 
         # we will try to coerce to integers

pandas/core/indexes/period.py

@@ -485,7 +485,8 @@ def _maybe_cast_slice_bound(self, label, side: str, kind=lib.no_default):
         Value of `side` parameter should be validated in caller.
 
         """
-        assert kind in ["loc", "getitem", None, lib.no_default]
+        if kind not in ["loc", "getitem", None, lib.no_default]:
+            raise ValueError('Value for kind argument must be one of: loc, getitem or None')
         self._deprecated_arg(kind, "kind", "_maybe_cast_slice_bound")
 
         if isinstance(label, datetime):

pandas/core/indexes/timedeltas.py

@@ -199,7 +199,8 @@ def _maybe_cast_slice_bound(self, label, side: str, kind=lib.no_default):
         -------
         label : object
         """
-        assert kind in ["loc", "getitem", None, lib.no_default]
+        if kind not in ["loc", "getitem", None, lib.no_default]:
+            raise ValueError('Value for kind argument must be one of: loc, getitem or None')
         self._deprecated_arg(kind, "kind", "_maybe_cast_slice_bound")
 
         if isinstance(label, str):

pandas/core/internals/blocks.py

@@ -1204,7 +1204,8 @@ def where(self, other, cond, errors="raise") -> list[Block]:
         assert cond.ndim == self.ndim
         assert not isinstance(other, (ABCIndex, ABCSeries, ABCDataFrame))
 
-        assert errors in ["raise", "ignore"]
+        if errors not in ["raise", "ignore"]:
+            raise ValueError('Value for errors argument must be one of: raise, ignore')
         transpose = self.ndim == 2
 
         values = self.values

pandas/core/missing.py

@@ -178,7 +178,8 @@ def find_valid_index(values, *, how: str) -> int | None:
     -------
     int or None
     """
-    assert how in ["first", "last"]
+    if how not in ["first", "last"]:
+        raise ValueError(f'Value for how argument must be one of : first, last')
 
     if len(values) == 0:  # early stop
         return None

pandas/io/excel/_util.py

@@ -60,7 +60,8 @@ def get_default_engine(ext, mode="reader"):
         "xls": "xlwt",
         "ods": "odf",
     }
-    assert mode in ["reader", "writer"]
+    if mode not in ["reader", "writer"]:
+        raise ValueError('File mode must be either "reader" or "writer".')
     if mode == "writer":
         # Prefer xlsxwriter over openpyxl if installed
         xlsxwriter = import_optional_dependency("xlsxwriter", errors="warn")