sqlframe 3.20.0__py3-none-any.whl → 3.21.1__py3-none-any.whl

sqlframe/_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '3.20.0'
-__version_tuple__ = version_tuple = (3, 20, 0)
+__version__ = version = '3.21.1'
+__version_tuple__ = version_tuple = (3, 21, 1)

sqlframe/base/dataframe.py

@@ -296,6 +296,12 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
 
     @property
     def columns(self) -> t.List[str]:
+        expression_display_names = self.expression.copy()
+        self._set_display_names(expression_display_names)
+        return expression_display_names.named_selects
+
+    @property
+    def _columns(self) -> t.List[str]:
         return self.expression.named_selects
 
     @property
@@ -611,6 +617,18 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
         }
         self.display_name_mapping.update(zipped)
 
+    def _set_display_names(self, select_expression: exp.Select) -> None:
+        for index, column in enumerate(select_expression.expressions):
+            column_name = quote_preserving_alias_or_name(column)
+            if column_name in self.display_name_mapping:
+                display_name_identifier = exp.to_identifier(
+                    self.display_name_mapping[column_name], quoted=True
+                )
+                display_name_identifier._meta = {"case_sensitive": True, **(column._meta or {})}
+                select_expression.expressions[index] = exp.alias_(
+                    column.unalias(), display_name_identifier, quoted=True
+                )
+
     def _get_expressions(
         self,
         optimize: bool = True,
@@ -631,16 +649,7 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
             select_expression = select_expression.transform(
                 replace_id_value, replacement_mapping
             ).assert_is(exp.Select)
-            for index, column in enumerate(select_expression.expressions):
-                column_name = quote_preserving_alias_or_name(column)
-                if column_name in self.display_name_mapping:
-                    display_name_identifier = exp.to_identifier(
-                        self.display_name_mapping[column_name], quoted=True
-                    )
-                    display_name_identifier._meta = {"case_sensitive": True, **(column._meta or {})}
-                    select_expression.expressions[index] = exp.alias_(
-                        column.unalias(), display_name_identifier, quoted=True
-                    )
+            self._set_display_names(select_expression)
             if optimize:
                 select_expression = t.cast(
                     exp.Select,
@@ -1158,8 +1167,8 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
 
     @operation(Operation.FROM)
     def unionByName(self, other: Self, allowMissingColumns: bool = False) -> Self:
-        l_columns = self.columns
-        r_columns = other.columns
+        l_columns = self._columns
+        r_columns = other._columns
         if not allowMissingColumns:
             l_expressions = l_columns
             r_expressions = l_columns
@@ -1619,9 +1628,9 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
         | 16|  Bob|
         +---+-----+
         """
-        if len(cols) != len(self.columns):
+        if len(cols) != len(self._columns):
             raise ValueError(
-                f"Number of column names does not match number of columns: {len(cols)} != {len(self.columns)}"
+                f"Number of column names does not match number of columns: {len(cols)} != {len(self._columns)}"
             )
         expression = self.expression.copy()
         expression = expression.select(
@@ -1718,6 +1727,114 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
             grouping_columns.extend([list(x) for x in itertools.combinations(columns, i)])
         return self._group_data(self, grouping_columns, self.last_op)
 
+    @operation(Operation.SELECT)
+    def unpivot(
+        self,
+        ids: t.Union[ColumnOrName, t.List[ColumnOrName], t.Tuple[ColumnOrName, ...]],
+        values: t.Optional[t.Union[ColumnOrName, t.List[ColumnOrName], t.Tuple[ColumnOrName, ...]]],
+        variableColumnName: str,
+        valueColumnName: str,
+    ) -> Self:
+        """
+        Unpivot a DataFrame from wide format to long format, optionally leaving
+        identifier columns set. This is the reverse to `groupBy(...).pivot(...).agg(...)`,
+        except for the aggregation, which cannot be reversed.
+
+        This function is useful to massage a DataFrame into a format where some
+        columns are identifier columns ("ids"), while all other columns ("values")
+        are "unpivoted" to the rows, leaving just two non-id columns, named as given
+        by `variableColumnName` and `valueColumnName`.
+
+        When no "id" columns are given, the unpivoted DataFrame consists of only the
+        "variable" and "value" columns.
+
+        The `values` columns must not be empty so at least one value must be given to be unpivoted.
+        When `values` is `None`, all non-id columns will be unpivoted.
+
+        All "value" columns must share a least common data type. Unless they are the same data type,
+        all "value" columns are cast to the nearest common data type. For instance, types
+        `IntegerType` and `LongType` are cast to `LongType`, while `IntegerType` and `StringType`
+        do not have a common data type and `unpivot` fails.
+
+        .. versionadded:: 3.4.0
+
+        Parameters
+        ----------
+        ids : str, Column, tuple, list
+            Column(s) to use as identifiers. Can be a single column or column name,
+            or a list or tuple for multiple columns.
+        values : str, Column, tuple, list, optional
+            Column(s) to unpivot. Can be a single column or column name, or a list or tuple
+            for multiple columns. If specified, must not be empty. If not specified, uses all
+            columns that are not set as `ids`.
+        variableColumnName : str
+            Name of the variable column.
+        valueColumnName : str
+            Name of the value column.
+
+        Returns
+        -------
+        :class:`DataFrame`
+            Unpivoted DataFrame.
+
+        Notes
+        -----
+        Supports Spark Connect.
+
+        Examples
+        --------
+        >>> df = spark.createDataFrame(
+        ...     [(1, 11, 1.1), (2, 12, 1.2)],
+        ...     ["id", "int", "double"],
+        ... )
+        >>> df.show()
+        +---+---+------+
+        | id|int|double|
+        +---+---+------+
+        |  1| 11|   1.1|
+        |  2| 12|   1.2|
+        +---+---+------+
+
+        >>> df.unpivot("id", ["int", "double"], "var", "val").show()
+        +---+------+----+
+        | id|   var| val|
+        +---+------+----+
+        |  1|   int|11.0|
+        |  1|double| 1.1|
+        |  2|   int|12.0|
+        |  2|double| 1.2|
+        +---+------+----+
+
+        See Also
+        --------
+        DataFrame.melt
+        """
+        from sqlframe.base import functions as F
+
+        id_columns = self._ensure_and_normalize_cols(ids)
+        if not values:
+            outer_selects = self._get_outer_select_columns(self.expression)
+            values = [
+                column
+                for column in outer_selects
+                if column.alias_or_name not in {x.alias_or_name for x in id_columns}
+            ]
+        value_columns = self._ensure_and_normalize_cols(values)
+
+        df = self._convert_leaf_to_cte()
+        selects = []
+        for value in value_columns:
+            selects.append(
+                exp.select(
+                    *[x.column_expression for x in id_columns],
+                    F.lit(value.alias_or_name).alias(variableColumnName).expression,
+                    value.alias(valueColumnName).expression,
+                ).from_(df.expression.ctes[-1].alias_or_name)
+            )
+        unioned_expression = functools.reduce(lambda x, y: x.union(y, distinct=False), selects)  # type: ignore
+        final_expression = self._add_ctes_to_expression(unioned_expression, df.expression.ctes)
+        return self.copy(expression=final_expression)._convert_leaf_to_cte()
+
     def collect(self) -> t.List[Row]:
         return self._collect()
 

sqlframe/base/function_alternatives.py

@@ -193,10 +193,6 @@ def factorial_ensure_int(col: ColumnOrName) -> Column:
     return Column.invoke_anonymous_function(col_func(col).cast("integer"), "FACTORIAL")
 
 
-def skewness_from_skew(col: ColumnOrName) -> Column:
-    return Column.invoke_anonymous_function(col, "SKEW")
-
-
 def isnan_using_equal(col: ColumnOrName) -> Column:
     lit = get_func_from_session("lit")
     return Column(

sqlframe/base/functions.py

@@ -486,14 +486,32 @@ def var_pop(col: ColumnOrName) -> Column:
 
 @meta(unsupported_engines=["bigquery", "postgres"])
 def skewness(col: ColumnOrName) -> Column:
-    from sqlframe.base.function_alternatives import skewness_from_skew
-
     session = _get_session()
 
+    func_name = "SKEWNESS"
+
     if session._is_snowflake:
-        return skewness_from_skew(col)
+        func_name = "SKEW"
+
+    if session._is_duckdb or session._is_snowflake:
+        when_func = get_func_from_session("when")
+        count_func = get_func_from_session("count")
+        count_star = count_func("*")
+        lit_func = get_func_from_session("lit")
+        sqrt_func = get_func_from_session("sqrt")
+        col = Column.ensure_col(col)
+        return (
+            when_func(count_star == lit_func(0), lit_func(None))
+            .when(count_star == lit_func(1), lit_func(float("nan")))
+            .when(count_star == lit_func(2), lit_func(0.0))
+            .otherwise(
+                Column.invoke_anonymous_function(col, func_name)
+                * (count_star - lit_func(2))
+                / (sqrt_func(count_star * (count_star - lit_func(1))))
+            )
+        )
 
-    return Column.invoke_anonymous_function(col, "SKEWNESS")
+    return Column.invoke_anonymous_function(col, func_name)
 
 
 @meta(unsupported_engines=["bigquery", "postgres"])

sqlframe/base/util.py

@@ -97,12 +97,8 @@ def get_column_mapping_from_schema_input(
     else:
         value = {x.strip(): None for x in schema}
     return {
-        exp.to_column(k).sql(dialect=dialect): exp.DataType.build(v, dialect=dialect)
-        if v is not None
-        else v
-        for k, v in value.items()
+        k: exp.DataType.build(v, dialect=dialect) if v is not None else v for k, v in value.items()
     }
-    # return {x.strip(): None for x in schema}  # type: ignore
 
 
 def get_tables_from_expression_with_join(expression: exp.Select) -> t.List[exp.Table]:

{sqlframe-3.20.0.dist-info → sqlframe-3.21.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 3.20.0
+Version: 3.21.1
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman

{sqlframe-3.20.0.dist-info → sqlframe-3.21.1.dist-info}/RECORD

@@ -1,14 +1,14 @@
 sqlframe/__init__.py,sha256=wfqm98eLoLid9oV_FzzpG5loKC6LxOhj2lXpfN7SARo,3138
-sqlframe/_version.py,sha256=nzt1OjXbH5tyyHQvLpmIr9I_E9sBcud1ZUXFSGz-12c,413
+sqlframe/_version.py,sha256=fmhKf9XPZdwZdKpQ-ESJ_LGssm7Q8K_NJEGVKwXLGQM,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=oHVwkSWABO3ZlAbgBShsxSSlgbI06BOup5XJrRhgqJI,18097
-sqlframe/base/dataframe.py,sha256=mKXbIKYiKH5mh6qj0Dg7L_znmCL85q9kHlmHtCW4kJ4,79352
+sqlframe/base/dataframe.py,sha256=FOgLdCpscLsBntkRvutcgSVqXqMgXo9DYa892mXu00E,83907
 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=nfDf2oKoBq2hrutTfuVHKmGvkm_X_ZvhfnFPv1rn0oU,222350
+sqlframe/base/function_alternatives.py,sha256=KFkEm0aIHzajvQmiPZnzTLh-Ud9wjeg4lJ4Rk0vk-YU,53674
+sqlframe/base/functions.py,sha256=jfLgboldiTB9CPkoZMtKUAwx6XSvFnEOIpCZQfoEJJU,223060
 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
@@ -18,7 +18,7 @@ sqlframe/base/table.py,sha256=rCeh1W5SWbtEVfkLAUiexzrZwNgmZeptLEmLcM1ABkE,6961
 sqlframe/base/transforms.py,sha256=y0j3SGDz3XCmNGrvassk1S-owllUWfkHyMgZlY6SFO4,467
 sqlframe/base/types.py,sha256=iBNk9bpFtb2NBIogYS8i7OlQZMRvpR6XxqzBebsjQDU,12280
 sqlframe/base/udf.py,sha256=O6hMhBUy9NVv-mhJRtfFhXTIa_-Z8Y_FkmmuOHu0l90,1117
-sqlframe/base/util.py,sha256=rdnH3Kg6gZVT3DehU_ZHjfum79vc-I5W_Il6OiCtWF4,15284
+sqlframe/base/util.py,sha256=_s2M-qHzTLgyGu1v8laRHJorUpUO6-fr3kk7CsvcuXw,15161
 sqlframe/base/window.py,sha256=8hOv-ignPPIsZA9FzvYzcLE9J_glalVaYjIAUdRUX3o,4943
 sqlframe/base/mixins/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sqlframe/base/mixins/catalog_mixins.py,sha256=9tn0mK8oPoqIIjNItystD5tdBMdK9YpkxTG7G9KQl8k,18619
@@ -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-3.20.0.dist-info/LICENSE,sha256=VZu79YgW780qxaFJMr0t5ZgbOYEh04xWoxaWOaqIGWk,1068
-sqlframe-3.20.0.dist-info/METADATA,sha256=vEauG8vJY6ak5FN5oJpsaGRKgzD7uaodpdlFFu3uN04,8970
-sqlframe-3.20.0.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
-sqlframe-3.20.0.dist-info/top_level.txt,sha256=T0_RpoygaZSF6heeWwIDQgaP0varUdSK1pzjeJZRjM8,9
-sqlframe-3.20.0.dist-info/RECORD,,
+sqlframe-3.21.1.dist-info/LICENSE,sha256=VZu79YgW780qxaFJMr0t5ZgbOYEh04xWoxaWOaqIGWk,1068
+sqlframe-3.21.1.dist-info/METADATA,sha256=AauznGD-zSbh2cqT63w2MIrg_-0SlewyyRMNElL5O2I,8970
+sqlframe-3.21.1.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
+sqlframe-3.21.1.dist-info/top_level.txt,sha256=T0_RpoygaZSF6heeWwIDQgaP0varUdSK1pzjeJZRjM8,9
+sqlframe-3.21.1.dist-info/RECORD,,