Replace

Author: phofl

asv_bench/benchmarks/join_merge.py

@@ -273,7 +273,7 @@ def time_merge_dataframes_cross(self, sort):
         merge(self.left.loc[:2000], self.right.loc[:2000], how="cross", sort=sort)
 
     def time_merge_semi(self, sort):
-        merge(self.df, self.df2, on="key1", how="leftsemi")
+        merge(self.df, self.df2, on="key1", how="left_semi")
 
 
 class MergeEA:
@@ -384,7 +384,7 @@ def time_merge(self, units, tz, monotonic):
         merge(self.left, self.right)
 
     def time_merge_semi(self, units, tz, monotonic):
-        merge(self.left, self.right, how="leftsemi")
+        merge(self.left, self.right, how="left_semi")
 
 
 class MergeCategoricals:

doc/source/user_guide/merging.rst

@@ -407,7 +407,7 @@ either the left or right tables, the values in the joined table will be
     ``right``, ``RIGHT OUTER JOIN``, Use keys from right frame only
     ``outer``, ``FULL OUTER JOIN``, Use union of keys from both frames
     ``inner``, ``INNER JOIN``, Use intersection of keys from both frames
-    ``leftsemi``, ``LEFT SEMI JOIN``, Filter rows on left based on occurrences in right.
+    ``left_semi``, ``LEFT SEMI JOIN``, Filter rows on left based on occurrences in right.
     ``cross``, ``CROSS JOIN``, Create the cartesian product of rows of both frames
 
 .. ipython:: python

pandas/core/frame.py

@@ -338,7 +338,7 @@
       join; sort keys lexicographically.
     * inner: use intersection of keys from both frames, similar to a SQL inner
       join; preserve the order of the left keys.
-    * leftsemi: Filter for rows in the left that have a match on the right;
+    * left_semi: Filter for rows in the left that have a match on the right;
       preserve the order of the left keys. Doesn't support `left_index`, `right_index`,
       `indicator` or `validate`.
 

pandas/core/reshape/merge.py

@@ -385,7 +385,7 @@ def merge(
             validate=validate,
         )
     else:
-        klass = _MergeOperation if how != "leftsemi" else _SemiMergeOperation
+        klass = _MergeOperation if how != "left_semi" else _SemiMergeOperation
         op = klass(
             left_df,
             right_df,
@@ -2038,7 +2038,7 @@ def get_join_indexers(
     left_keys: list[ArrayLike],
     right_keys: list[ArrayLike],
     sort: bool = False,
-    how: JoinHow + Literal["leftsemi"] = "inner",
+    how: JoinHow + Literal["left_semi"] = "inner",
 ) -> tuple[npt.NDArray[np.intp] | None, npt.NDArray[np.intp] | None]:
     """
 
@@ -2095,7 +2095,7 @@ def get_join_indexers(
     right = Index(rkey)
 
     if (
-        how != "leftsemi"
+        how != "left_semi"
         and left.is_monotonic_increasing
         and right.is_monotonic_increasing
         and (left.is_unique or right.is_unique)
@@ -2868,7 +2868,7 @@ def _factorize_keys(
         lk = ensure_int64(lk.codes)
         rk = ensure_int64(rk.codes)
 
-    elif how != "leftsemi" and isinstance(lk, ExtensionArray) and lk.dtype == rk.dtype:
+    elif how != "left_semi" and isinstance(lk, ExtensionArray) and lk.dtype == rk.dtype:
         if (isinstance(lk.dtype, ArrowDtype) and is_string_dtype(lk.dtype)) or (
             isinstance(lk.dtype, StringDtype) and lk.dtype.storage == "pyarrow"
         ):
@@ -2964,7 +2964,7 @@ def _factorize_keys(
             return lidx, ridx, -1
         else:
             llab = rizer.factorize(lk_data, mask=lk_mask)
-    elif how == "leftsemi":
+    elif how == "left_semi":
         # populate hashtable for right and then do a hash join
         rizer.factorize(rk_data, mask=rk_mask)
         return rizer.hash_inner_join(lk_data, lk_mask)[1], None, -1  # type: ignore[return-value]

pandas/tests/reshape/merge/test_semi.py

@@ -19,42 +19,42 @@
         ),
     ],
 )
-def test_leftsemi(vals_left, vals_right, dtype):
+def test_left_semi(vals_left, vals_right, dtype):
     vals_left = pd.Series(vals_left, dtype=dtype)
     vals_right = pd.Series(vals_right, dtype=dtype)
     left = pd.DataFrame({"a": vals_left, "b": [1, 2, 3]})
     right = pd.DataFrame({"a": vals_right, "c": 1})
     expected = pd.DataFrame({"a": vals_right, "b": [1, 2]})
-    result = left.merge(right, how="leftsemi")
+    result = left.merge(right, how="left_semi")
     tm.assert_frame_equal(result, expected)
 
     right = pd.DataFrame({"d": vals_right, "c": 1})
-    result = left.merge(right, how="leftsemi", left_on="a", right_on="d")
+    result = left.merge(right, how="left_semi", left_on="a", right_on="d")
     tm.assert_frame_equal(result, expected)
 
     right = pd.DataFrame({"d": vals_right, "c": 1})
-    result = left.merge(right, how="leftsemi", left_on=["a", "b"], right_on=["d", "c"])
+    result = left.merge(right, how="left_semi", left_on=["a", "b"], right_on=["d", "c"])
     tm.assert_frame_equal(result, expected.head(1))
 
 
-def test_leftsemi_invalid():
+def test_left_semi_invalid():
     left = pd.DataFrame({"a": [1, 2, 3], "b": [1, 2, 3]})
     right = pd.DataFrame({"a": [1, 2], "c": 1})
 
     msg = "left_index or right_index are not supported for semi-join."
     with pytest.raises(NotImplementedError, match=msg):
-        left.merge(right, how="leftsemi", left_index=True, right_on="a")
+        left.merge(right, how="left_semi", left_index=True, right_on="a")
     with pytest.raises(NotImplementedError, match=msg):
-        left.merge(right, how="leftsemi", right_index=True, left_on="a")
+        left.merge(right, how="left_semi", right_index=True, left_on="a")
 
     msg = "validate is not supported for semi-join."
     with pytest.raises(NotImplementedError, match=msg):
-        left.merge(right, how="leftsemi", validate="one_to_one")
+        left.merge(right, how="left_semi", validate="one_to_one")
 
     msg = "indicator is not supported for semi-join."
     with pytest.raises(NotImplementedError, match=msg):
-        left.merge(right, how="leftsemi", indicator=True)
+        left.merge(right, how="left_semi", indicator=True)
 
     msg = "sort is not supported for semi-join. Sort your DataFrame afterwards."
     with pytest.raises(NotImplementedError, match=msg):
-        left.merge(right, how="leftsemi", sort=True)
+        left.merge(right, how="left_semi", sort=True)