sqlframe 3.15.1__py3-none-any.whl → 3.16.0__py3-none-any.whl

sqlframe/_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '3.15.1'
-__version_tuple__ = version_tuple = (3, 15, 1)
+__version__ = version = '3.16.0'
+__version_tuple__ = version_tuple = (3, 16, 0)

sqlframe/base/dataframe.py

@@ -391,7 +391,9 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
 
         cols = self._ensure_list_of_columns(cols)
         normalize(self.session, expression or self.expression, cols)
-        return list(flatten([self._expand_star(col) for col in cols]))
+        cols = list(flatten([self._expand_star(col) for col in cols]))
+        self._resolve_ambiguous_columns(cols)
+        return cols
 
     def _ensure_and_normalize_col(self, col):
         from sqlframe.base.column import Column
@@ -399,6 +401,7 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
 
         col = Column.ensure_col(col)
         normalize(self.session, self.expression, col)
+        self._resolve_ambiguous_columns(col)
         return col
 
     def _convert_leaf_to_cte(
@@ -745,10 +748,55 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
         kwargs["join_on_uuid"] = str(uuid4())
         return self.__class__(**object_to_dict(self, **kwargs))
 
+    def _resolve_ambiguous_columns(self, columns: t.Union[Column, t.List[Column]]) -> None:
+        if "joins" not in self.expression.args:
+            return
+
+        columns = ensure_list(columns)
+        ambiguous_cols: t.List[exp.Column] = list(
+            flatten(
+                [
+                    sub_col
+                    for col in columns
+                    for sub_col in col.expression.find_all(exp.Column)
+                    if not sub_col.table
+                ]
+            )
+        )
+        if ambiguous_cols:
+            join_table_identifiers = [
+                x.this for x in get_tables_from_expression_with_join(self.expression)
+            ]
+            cte_names_in_join = [x.this for x in join_table_identifiers]
+            # If we have columns that resolve to multiple CTE expressions then we want to use each CTE left-to-right
+            # (or right to left if a right join) and therefore we allow multiple columns with the same
+            # name in the result. This matches the behavior of Spark.
+            resolved_column_position: t.Dict[exp.Column, int] = {
+                col.copy(): -1 for col in ambiguous_cols
+            }
+            for ambiguous_col in ambiguous_cols:
+                ctes = (
+                    list(reversed(self.expression.ctes))
+                    if self.expression.args["joins"][0].args.get("side", "") == "right"
+                    else self.expression.ctes
+                )
+                ctes_with_column = [
+                    cte
+                    for cte in ctes
+                    if cte.alias_or_name in cte_names_in_join
+                    and ambiguous_col.alias_or_name in cte.this.named_selects
+                ]
+                # Check if there is a CTE with this column that we haven't used before. If so, use it. Otherwise,
+                # use the same CTE we used before
+                cte = seq_get(ctes_with_column, resolved_column_position[ambiguous_col] + 1)
+                if cte:
+                    resolved_column_position[ambiguous_col] += 1
+                else:
+                    cte = ctes_with_column[resolved_column_position[ambiguous_col]]
+                ambiguous_col.set("table", exp.to_identifier(cte.alias_or_name))
+
     @operation(Operation.SELECT)
     def select(self, *cols, **kwargs) -> Self:
-        from sqlframe.base.column import Column
-
         if not cols:
             return self
 
@@ -756,48 +804,6 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
             cols = cols[0]  # type: ignore
         columns = self._ensure_and_normalize_cols(cols)
         kwargs["append"] = kwargs.get("append", False)
-        if self.expression.args.get("joins"):
-            ambiguous_cols: t.List[exp.Column] = list(
-                flatten(
-                    [
-                        sub_col
-                        for col in columns
-                        for sub_col in col.expression.find_all(exp.Column)
-                        if not sub_col.table
-                    ]
-                )
-            )
-            if ambiguous_cols:
-                join_table_identifiers = [
-                    x.this for x in get_tables_from_expression_with_join(self.expression)
-                ]
-                cte_names_in_join = [x.this for x in join_table_identifiers]
-                # If we have columns that resolve to multiple CTE expressions then we want to use each CTE left-to-right
-                # (or right to left if a right join) and therefore we allow multiple columns with the same
-                # name in the result. This matches the behavior of Spark.
-                resolved_column_position: t.Dict[exp.Column, int] = {
-                    col.copy(): -1 for col in ambiguous_cols
-                }
-                for ambiguous_col in ambiguous_cols:
-                    ctes = (
-                        list(reversed(self.expression.ctes))
-                        if self.expression.args["joins"][0].args.get("side", "") == "right"
-                        else self.expression.ctes
-                    )
-                    ctes_with_column = [
-                        cte
-                        for cte in ctes
-                        if cte.alias_or_name in cte_names_in_join
-                        and ambiguous_col.alias_or_name in cte.this.named_selects
-                    ]
-                    # Check if there is a CTE with this column that we haven't used before. If so, use it. Otherwise,
-                    # use the same CTE we used before
-                    cte = seq_get(ctes_with_column, resolved_column_position[ambiguous_col] + 1)
-                    if cte:
-                        resolved_column_position[ambiguous_col] += 1
-                    else:
-                        cte = ctes_with_column[resolved_column_position[ambiguous_col]]
-                    ambiguous_col.set("table", exp.to_identifier(cte.alias_or_name))
         # If an expression is `CAST(x AS DATETYPE)` then we want to alias so that `x` is the result column name
         columns = [
             col.alias(col.expression.alias_or_name)

sqlframe/base/decorators.py

@@ -43,7 +43,7 @@ def func_metadata(unsupported_engines: t.Optional[t.Union[str, t.List[str]]] = N
                         col_name = col_name.this
                 alias_name = f"{func.__name__}__{col_name or ''}__"
                 # BigQuery has restrictions on alias names so we constrain it to alphanumeric characters and underscores
-                return result.alias(re.sub("\W", "_", alias_name))  # type: ignore
+                return result.alias(re.sub(r"\W", "_", alias_name))  # type: ignore
             return result
 
         wrapper.unsupported_engines = (  # type: ignore

sqlframe/base/functions.py

@@ -2851,12 +2851,14 @@ def bool_or(col: ColumnOrName) -> Column:
     return Column.invoke_expression_over_column(col, expression.LogicalOr)
 
 
-@meta(unsupported_engines="*")
+@meta()
 def btrim(str: ColumnOrName, trim: t.Optional[ColumnOrName] = None) -> Column:
     if trim is not None:
-        return Column.invoke_anonymous_function(str, "btrim", trim)
+        return Column.invoke_expression_over_column(
+            str, expression.Trim, expression=Column.ensure_col(trim).column_expression
+        )
     else:
-        return Column.invoke_anonymous_function(str, "btrim")
+        return Column.invoke_expression_over_column(str, expression.Trim)
 
 
 @meta(unsupported_engines="*")

{sqlframe-3.15.1.dist-info → sqlframe-3.16.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 3.15.1
+Version: 3.16.0
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman

{sqlframe-3.15.1.dist-info → sqlframe-3.16.0.dist-info}/RECORD

@@ -1,14 +1,14 @@
 sqlframe/__init__.py,sha256=wfqm98eLoLid9oV_FzzpG5loKC6LxOhj2lXpfN7SARo,3138
-sqlframe/_version.py,sha256=rNfI2qI8EULJid-fGjytQ8KiqfMi0Ktaq6sNSFSM_1s,413
+sqlframe/_version.py,sha256=CtTis8a_OeN0EsLFoVgtqX-ARqHjuin2ATomgRROY1Y,413
 sqlframe/base/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sqlframe/base/_typing.py,sha256=b2clI5HI1zEZKB_3Msx3FeAJQyft44ubUifJwQRVXyQ,1298
 sqlframe/base/catalog.py,sha256=SzFQalTWdhWzxUY-4ut1f9TfOECp_JmJEgNPfrRKCe0,38457
 sqlframe/base/column.py,sha256=wRghgieYAA51aw4WuFQWOvl0TFOToZbBhBuIamEzxx4,18011
-sqlframe/base/dataframe.py,sha256=E1zWlB_a2FNOxjTcQ68MtL_A4c8fnLiHY3MeZttK4Xk,76570
-sqlframe/base/decorators.py,sha256=P56cgs8DANxGRIwVs5uOMnDy-BlXZZYMbf4fdnkpWPI,1889
+sqlframe/base/dataframe.py,sha256=KKBwtn73xNGt2gRwUB8Vri7Ee6_ivP5a_qij4Eq96zE,76622
+sqlframe/base/decorators.py,sha256=ms-CvDOIW3T8IVB9VqDmLwAiaEsqXLYRXEqVQaxktiM,1890
 sqlframe/base/exceptions.py,sha256=9Uwvqn2eAkDpqm4BrRgbL61qM-GMCbJEMAW8otxO46s,370
 sqlframe/base/function_alternatives.py,sha256=NV31IaEhVYmfUSWetAEFISAvLzs2DxQ7bp-iMNgj0hQ,53786
-sqlframe/base/functions.py,sha256=9mN54Nx6yqos1njfyW2-WRzfFUsA96P9z1ldJVtovSs,220543
+sqlframe/base/functions.py,sha256=o8zwbS8zCsyNe5arcb6dbAGBL8a1tH99rGyRimwzzUk,220614
 sqlframe/base/group.py,sha256=fsyG5990_Pd7gFPjTFrH9IEoAquL_wEkVpIlBAIkZJU,4091
 sqlframe/base/normalize.py,sha256=nXAJ5CwxVf4DV0GsH-q1w0p8gmjSMlv96k_ez1eVul8,3880
 sqlframe/base/operations.py,sha256=xSPw74e59wYvNd6U1AlwziNCTG6Aftrbl4SybN9u9VE,3450
@@ -129,8 +129,8 @@ sqlframe/standalone/udf.py,sha256=azmgtUjHNIPs0WMVNId05SHwiYn41MKVBhKXsQJ5dmY,27
 sqlframe/standalone/window.py,sha256=6GKPzuxeSapJakBaKBeT9VpED1ACdjggDv9JRILDyV0,35
 sqlframe/testing/__init__.py,sha256=VVCosQhitU74A3NnE52O4mNtGZONapuEXcc20QmSlnQ,132
 sqlframe/testing/utils.py,sha256=PFsGZpwNUE_4-g_f43_vstTqsK0AQ2lBneb5Eb6NkFo,13008
-sqlframe-3.15.1.dist-info/LICENSE,sha256=VZu79YgW780qxaFJMr0t5ZgbOYEh04xWoxaWOaqIGWk,1068
-sqlframe-3.15.1.dist-info/METADATA,sha256=-MxovSCoyQnT-6Ujd4BDA_yVpf9KWra2v1CQGN2TmG4,8970
-sqlframe-3.15.1.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
-sqlframe-3.15.1.dist-info/top_level.txt,sha256=T0_RpoygaZSF6heeWwIDQgaP0varUdSK1pzjeJZRjM8,9
-sqlframe-3.15.1.dist-info/RECORD,,
+sqlframe-3.16.0.dist-info/LICENSE,sha256=VZu79YgW780qxaFJMr0t5ZgbOYEh04xWoxaWOaqIGWk,1068
+sqlframe-3.16.0.dist-info/METADATA,sha256=SMpgyXmxbVMqeeRuByF19qKm9iLDYubcniTCYBUmyNo,8970
+sqlframe-3.16.0.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
+sqlframe-3.16.0.dist-info/top_level.txt,sha256=T0_RpoygaZSF6heeWwIDQgaP0varUdSK1pzjeJZRjM8,9
+sqlframe-3.16.0.dist-info/RECORD,,