sqlframe 3.37.0__tar.gz → 3.38.1__tar.gz

{sqlframe-3.37.0 → sqlframe-3.38.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 3.37.0
+Version: 3.38.1
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman

{sqlframe-3.37.0 → sqlframe-3.38.1}/docs/bigquery.md

@@ -365,6 +365,10 @@ See something that you would like to see supported? [Open an issue](https://gith
 * [cosh](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.cosh.html)
 * [cot](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.cot.html)
 * [count](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.count.html)
+* [countDistinct](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.countDistinct.html)
+  * Can only provide a single column 
+* [count_distinct](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.count_distinct.html)
+  * Can only provide a single column 
 * [count_if](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.count_if.html)
 * [covar_pop](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.covar_pop.html)
 * [covar_samp](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.covar_samp.html)

{sqlframe-3.37.0 → sqlframe-3.38.1}/setup.py

@@ -20,7 +20,7 @@ setup(
     python_requires=">=3.9",
     install_requires=[
         "prettytable<4",
-        "sqlglot>=24.0.0,<27.2",
+        "sqlglot>=24.0.0,<27.7",
         "typing_extensions",
     ],
     extras_require={

{sqlframe-3.37.0 → sqlframe-3.38.1}/sqlframe/_version.py

@@ -17,5 +17,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '3.37.0'
-__version_tuple__ = version_tuple = (3, 37, 0)
+__version__ = version = '3.38.1'
+__version_tuple__ = version_tuple = (3, 38, 1)

{sqlframe-3.37.0 → sqlframe-3.38.1}/sqlframe/base/column.py

@@ -222,6 +222,12 @@ class Column:
             else:
                 value = value.astimezone(datetime.timezone.utc).isoformat(sep=" ")
                 return cls(exp.cast(exp.Literal.string(value), exp.DataType.Type.TIMESTAMPTZ))
+        elif isinstance(value, datetime.timedelta):
+            return cls(
+                exp.Interval(
+                    this=exp.Literal.string(int(value.total_seconds())), unit=exp.Var(this="SECOND")
+                )
+            )
         return cls(exp.convert(value))
 
     @classmethod
@@ -413,8 +419,9 @@ class Column:
         return self.invoke_expression_over_column(self, exp.StartsWith, expression=value.expression)
 
     def endswith(self, value: t.Union[str, Column]) -> Column:
+        ends_with_func = get_func_from_session("endswith")
         value = self._lit(value) if not isinstance(value, Column) else value
-        return self.invoke_anonymous_function(self, "ENDSWITH", value)
+        return ends_with_func(self, value)
 
     def rlike(self, regexp: str) -> Column:
         return self.invoke_expression_over_column(

{sqlframe-3.37.0 → sqlframe-3.38.1}/sqlframe/base/functions.py

@@ -14,6 +14,7 @@ from sqlglot.helper import flatten as _flatten
 
 from sqlframe.base.column import Column
 from sqlframe.base.decorators import func_metadata as meta
+from sqlframe.base.exceptions import UnsupportedOperationError
 from sqlframe.base.util import (
     get_func_from_session,
 )
@@ -81,9 +82,13 @@ def least(*cols: ColumnOrName) -> Column:
     return Column.invoke_expression_over_column(cols[0], expression.Least)
 
 
-@meta(unsupported_engines="bigquery")
+@meta()
 def count_distinct(col: ColumnOrName, *cols: ColumnOrName) -> Column:
     columns = [Column.ensure_col(x) for x in [col] + list(cols)]
+    if len(columns) > 1 and _get_session()._is_bigquery:
+        raise UnsupportedOperationError(
+            "BigQuery does not support multiple columns in countDistinct"
+        )
     return Column(
         expression.Count(
             this=expression.Distinct(expressions=[x.column_expression for x in columns])

{sqlframe-3.37.0 → sqlframe-3.38.1}/sqlframe/base/session.py

@@ -12,6 +12,7 @@ from collections import defaultdict
 from functools import cached_property
 
 import sqlglot
+from dateutil.relativedelta import relativedelta
 from sqlglot import Dialect, exp
 from sqlglot.dialects.dialect import DialectType, NormalizationStrategy
 from sqlglot.expressions import parse_identifier
@@ -613,6 +614,11 @@ class _BaseSession(t.Generic[CATALOG, READER, WRITER, DF, TABLE, CONN, UDF_REGIS
             return [cls._to_value(x) for x in value]
         elif isinstance(value, datetime.datetime):
             return value.replace(tzinfo=None)
+        elif isinstance(value, relativedelta):
+            return datetime.timedelta(
+                days=value.days, hours=value.hours, minutes=value.minutes, seconds=value.seconds
+            )
+
         return value
 
     @classmethod

{sqlframe-3.37.0 → sqlframe-3.38.1}/sqlframe.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 3.37.0
+Version: 3.38.1
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman

{sqlframe-3.37.0 → sqlframe-3.38.1}/sqlframe.egg-info/requires.txt

@@ -1,5 +1,5 @@
 prettytable<4
-sqlglot<27.2,>=24.0.0
+sqlglot<27.7,>=24.0.0
 typing_extensions
 
 [bigquery]

{sqlframe-3.37.0 → sqlframe-3.38.1}/tests/integration/engines/test_engine_column.py

@@ -90,3 +90,21 @@ def test_contains(get_session: t.Callable[[], _BaseSession], get_func):
     df_bar_lit = df.select(df.a.contains(lit("bar"))).collect()
     assert df_bar_lit[0][0] is False
     assert df_bar_lit[1][0] is True
+
+
+def test_endswith(get_session: t.Callable[[], _BaseSession], get_func):
+    session = get_session()
+    lit = get_func("lit", session)
+    df = session.createDataFrame([Row(a="abc"), Row(a="def")])
+    df_endswith_c = df.select(df.a.endswith("c")).collect()
+    assert df_endswith_c[0][0] is True
+    assert df_endswith_c[1][0] is False
+    df_endswith_c_lit = df.select(df.a.endswith(lit("c"))).collect()
+    assert df_endswith_c_lit[0][0] is True
+    assert df_endswith_c_lit[1][0] is False
+    df_endswith_f = df.select(df.a.endswith("f")).collect()
+    assert df_endswith_f[0][0] is False
+    assert df_endswith_f[1][0] is True
+    df_endswith_f_lit = df.select(df.a.endswith(lit("f"))).collect()
+    assert df_endswith_f_lit[0][0] is False
+    assert df_endswith_f_lit[1][0] is True

{sqlframe-3.37.0 → sqlframe-3.38.1}/tests/integration/engines/test_int_functions.py

@@ -121,6 +121,7 @@ def get_types() -> t.Callable:
         ),
         ({"cola": 1}, {"cola": 1}),
         (Row(**{"cola": 1, "colb": "test"}), Row(**{"cola": 1, "colb": "test"})),
+        (datetime.timedelta(1), datetime.timedelta(1)),
     ],
 )
 def test_lit(get_session_and_func, arg, expected):
@@ -141,6 +142,11 @@ def test_lit(get_session_and_func, arg, expected):
     if isinstance(session, SnowflakeSession):
         if isinstance(arg, Row):
             pytest.skip("Snowflake doesn't support literal row types")
+        if isinstance(arg, datetime.timedelta):
+            pytest.skip("Snowflake doesn't support literal timedelta types")
+    if isinstance(session, DatabricksSession):
+        if isinstance(arg, datetime.timedelta):
+            pytest.skip("Databricks doesn't support literal timedelta types")
     if isinstance(session, DuckDBSession):
         if isinstance(arg, dict):
             expected = Row(**expected)
@@ -873,7 +879,11 @@ def test_count_distinct(get_session_and_func):
     df1 = session.createDataFrame([1, 1, 3], "int")
     df2 = session.createDataFrame([1, 2], "int")
     df_joined = df1.join(df2)
-    assert df_joined.select(count_distinct(df1.value, df2.value)).collect() == [Row(value=4)]
+    if isinstance(session, BigQuerySession):
+        # BigQuery does not support count_distinct with multiple columns
+        assert df_joined.select(count_distinct(df1.value)).collect() == [Row(value=2)]
+    else:
+        assert df_joined.select(count_distinct(df1.value, df2.value)).collect() == [Row(value=4)]
 
 
 def test_first(get_session_and_func):
@@ -3748,7 +3758,6 @@ def test_elt(get_session_and_func, get_func):
 
 def test_endswith(get_session_and_func, get_func):
     session, endswith = get_session_and_func("endswith")
-    to_binary = get_func("to_binary", session)
     df = session.createDataFrame(
         [
             (
@@ -3768,6 +3777,7 @@ def test_endswith(get_session_and_func, get_func):
         ],
         ["e", "f"],
     )
+    to_binary = get_func("to_binary", session)
     df = df.select(to_binary("e").alias("e"), to_binary("f").alias("f"))
     assert df.select(endswith("e", "f"), endswith("f", "e")).collect() == [
         Row(value1=True, value2=False)

{sqlframe-3.37.0 → sqlframe-3.38.1}/tests/unit/standalone/test_column.py

@@ -105,7 +105,7 @@ def test_startswith():
 
 
 def test_endswith():
-    assert (F.col("cola").endswith("test")).sql() == "ENDSWITH(cola, 'test')"
+    assert (F.col("cola").endswith("test")).sql() == "ENDSWITH(cola, 'test') AS endswith__cola__"
 
 
 def test_rlike():