sqlframe 3.43.6__tar.gz → 3.43.8__tar.gz

{sqlframe-3.43.6 → sqlframe-3.43.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 3.43.6
+Version: 3.43.8
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman

{sqlframe-3.43.6 → sqlframe-3.43.8}/docs/duckdb.md

@@ -397,6 +397,7 @@ See something that you would like to see supported? [Open an issue](https://gith
 * [left](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.left.html)
 * [length](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.length.html)
 * [levenshtein](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.levenshtein.html)
+* [like](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.like.html)
 * [lit](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.lit.html)
 * [ln](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.ln.html)
 * [locate](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.locate.html)

{sqlframe-3.43.6 → sqlframe-3.43.8}/docs/postgres.md

@@ -392,6 +392,7 @@ See something that you would like to see supported? [Open an issue](https://gith
 * [left](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.left.html)
 * [length](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.length.html)
 * [levenshtein](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.levenshtein.html)
+* [like](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.like.html)
 * [lit](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.lit.html)
 * [ln](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.ln.html)
 * [locate](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.locate.html)

{sqlframe-3.43.6 → sqlframe-3.43.8}/docs/snowflake.md

@@ -389,6 +389,7 @@ See something that you would like to see supported? [Open an issue](https://gith
 * [e](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.e.html)
 * [element_at](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.element_at.html)
 * [endswith](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.Column.endswith.html)
+* [equal_null](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.equal_null.html)
 * [exp](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.exp.html)
 * [explode](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.explode.html)
 * [expm1](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.expm1.html)
@@ -426,6 +427,7 @@ See something that you would like to see supported? [Open an issue](https://gith
 * [length](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.length.html)
 * [levenshtein](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.levenshtein.html)
   * threshold is not supported
+* [like](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.like.html)
 * [lit](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.lit.html)
 * [ln](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.ln.html)
 * [locate](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.locate.html)

{sqlframe-3.43.6 → sqlframe-3.43.8}/setup.py

@@ -20,7 +20,7 @@ setup(
     python_requires=">=3.9",
     install_requires=[
         "prettytable<4",
-        "sqlglot>=24.0.0,<27.27",
+        "sqlglot>=24.0.0,<27.30",
         "typing_extensions",
         "more-itertools",
     ],
@@ -32,12 +32,14 @@ setup(
         "dev": [
             "duckdb>=1.2,<1.5",
             "findspark>=2,<3",
+            # 3.0.0 dropped support for Python 3.9 and this is needed by pytest-postgresql
+            "mirakuru<3.0.0",
             "mypy>=1.10.0,<1.19",
             "openai>=1.30,<3",
             "pandas>=2,<3",
             "pandas-stubs>=2,<3",
             "psycopg>=3.1,<4",
-            "pyarrow>=10,<22",
+            "pyarrow>=10,<23",
             "pyspark>=2,<3.6",
             "pytest>=8.2.0,<8.5",
             "pytest-forked",

{sqlframe-3.43.6 → sqlframe-3.43.8}/sqlframe/__init__.py

@@ -6,6 +6,10 @@ import typing as t
 from contextlib import contextmanager
 from unittest.mock import MagicMock
 
+from sqlglot.dialects import spark
+
+spark.Spark.Generator.readparquet_sql = lambda self, x: x.sql(dialect="duckdb")  # type: ignore
+
 if t.TYPE_CHECKING:
     from sqlframe.base.session import CONN
 

{sqlframe-3.43.6 → sqlframe-3.43.8}/sqlframe/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '3.43.6'
-__version_tuple__ = version_tuple = (3, 43, 6)
+__version__ = version = '3.43.8'
+__version_tuple__ = version_tuple = (3, 43, 8)
 
-__commit_id__ = commit_id = 'g7c6cd19ef'
+__commit_id__ = commit_id = 'g371f414f3'

{sqlframe-3.43.6 → sqlframe-3.43.8}/sqlframe/base/functions.py

@@ -60,7 +60,11 @@ def lit(value: t.Optional[t.Any] = None) -> Column:
     if isinstance(value, str):
         return Column(expression.Literal.string(value))
     if isinstance(value, float) and value in {float("inf"), float("-inf")}:
-        return Column(expression.Literal.string(str(value)))
+        return Column(
+            expression.cast(
+                expression.Literal.string(str(value)), to=expression.DataType.Type.FLOAT
+            )
+        )
     return Column(value)
 
 
@@ -414,7 +418,7 @@ def radians(col: ColumnOrName) -> Column:
     if session._is_bigquery:
         return radians_bgutil(col)
 
-    return Column.invoke_anonymous_function(col, "RADIANS")
+    return Column.invoke_expression_over_column(col, expression.Radians)
 
 
 toRadians = radians
@@ -1078,7 +1082,7 @@ def hour(col: ColumnOrName) -> Column:
     if session._is_bigquery or session._is_postgres:
         return hour_from_extract(col)
 
-    return Column.invoke_anonymous_function(col, "HOUR")
+    return Column.invoke_expression_over_column(col, expression.Hour)
 
 
 @meta()
@@ -1090,7 +1094,7 @@ def minute(col: ColumnOrName) -> Column:
     if session._is_bigquery or session._is_postgres:
         return minute_from_extract(col)
 
-    return Column.invoke_anonymous_function(col, "MINUTE")
+    return Column.invoke_expression_over_column(col, expression.Minute)
 
 
 @meta()
@@ -1102,7 +1106,7 @@ def second(col: ColumnOrName) -> Column:
     if session._is_bigquery or session._is_postgres:
         return second_from_extract(col)
 
-    return Column.invoke_anonymous_function(col, "SECOND")
+    return Column.invoke_expression_over_column(col, expression.Second)
 
 
 @meta()
@@ -1398,7 +1402,7 @@ def next_day(col: ColumnOrName, dayOfWeek: str) -> Column:
     if session._is_bigquery:
         return next_day_bgutil(col, dayOfWeek)
 
-    return Column.invoke_anonymous_function(col, "NEXT_DAY", lit(dayOfWeek))
+    return Column.invoke_expression_over_column(col, expression.NextDay, expression=lit(dayOfWeek))
 
 
 @meta()
@@ -2160,7 +2164,7 @@ def bit_count(col: ColumnOrName) -> Column:
     if session._is_duckdb:
         return Column.invoke_anonymous_function(col, "BIT_COUNT")
 
-    return Column.invoke_expression_over_column(col, expression.BitwiseCountAgg)
+    return Column.invoke_expression_over_column(col, expression.BitwiseCount)
 
 
 @meta(unsupported_engines="*")
@@ -3308,9 +3312,9 @@ def endswith(str: ColumnOrName, suffix: ColumnOrName) -> Column:
     )
 
 
-@meta(unsupported_engines="*")
+@meta(unsupported_engines=["bigquery", "duckdb", "postgres"])
 def equal_null(col1: ColumnOrName, col2: ColumnOrName) -> Column:
-    return Column.invoke_anonymous_function(col1, "equal_null", col2)
+    return Column.invoke_expression_over_column(col1, expression.EqualNull, expression=col2)
 
 
 @meta(unsupported_engines="*")
@@ -4108,7 +4112,7 @@ def left(str: ColumnOrName, len: ColumnOrName) -> Column:
     return Column.invoke_expression_over_column(str, expression.Left, expression=len)
 
 
-@meta(unsupported_engines="*")
+@meta(unsupported_engines="bigquery")
 def like(
     str: ColumnOrName, pattern: ColumnOrName, escapeChar: t.Optional["Column"] = None
 ) -> Column:
@@ -6805,7 +6809,7 @@ def weekday(col: ColumnOrName) -> Column:
     return Column.invoke_anonymous_function(col, "weekday")
 
 
-@meta(unsupported_engines="*")
+@meta(unsupported_engines=["bigquery", "duckdb", "postgres"])
 def width_bucket(
     v: ColumnOrName,
     min: ColumnOrName,
@@ -6854,7 +6858,9 @@ def width_bucket(
     +----------------------------+
     """
     numBucket = lit(numBucket) if isinstance(numBucket, int) else numBucket
-    return Column.invoke_anonymous_function(v, "width_bucket", min, max, numBucket)
+    return Column.invoke_expression_over_column(
+        v, expression.WidthBucket, min_value=min, max_value=max, num_buckets=numBucket
+    )
 
 
 @meta(unsupported_engines=["*", "spark"])

{sqlframe-3.43.6 → sqlframe-3.43.8}/sqlframe.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 3.43.6
+Version: 3.43.8
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman

{sqlframe-3.43.6 → sqlframe-3.43.8}/sqlframe.egg-info/requires.txt

@@ -1,6 +1,6 @@
 more-itertools
 prettytable<4
-sqlglot<27.27,>=24.0.0
+sqlglot<27.30,>=24.0.0
 typing_extensions
 
 [bigquery]
@@ -13,13 +13,14 @@ databricks-sql-connector[pyarrow]<5,>=3.6
 [dev]
 duckdb<1.5,>=1.2
 findspark<3,>=2
+mirakuru<3.0.0
 mypy<1.19,>=1.10.0
 openai<3,>=1.30
 pandas-stubs<3,>=2
 pandas<3,>=2
 pre-commit<5,>=3.7
 psycopg<4,>=3.1
-pyarrow<22,>=10
+pyarrow<23,>=10
 pyspark<3.6,>=2
 pytest-forked
 pytest-postgresql<8,>=6

{sqlframe-3.43.6 → sqlframe-3.43.8}/tests/integration/engines/test_int_functions.py

@@ -122,6 +122,7 @@ 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)),
+        (float("inf"), float("inf")),
     ],
 )
 def test_lit(get_session_and_func, arg, expected):

{sqlframe-3.43.6 → sqlframe-3.43.8}/tests/integration/test_int_dataframe.py

@@ -2938,3 +2938,26 @@ def test_join_filter_join_filter(
     )
 
     assert result == dfs_result
+
+
+# https://github.com/eakmanrq/sqlframe/issues/542
+def test_float_infinity(
+    pyspark_employee: PySparkDataFrame,
+    get_df: t.Callable[[str], BaseDataFrame],
+    compare_frames: t.Callable,
+):
+    session = pyspark_employee.sparkSession
+    data = {"a": [1, 2, 3], "b": [4, 5, 6], "i": [0, 1, 2]}
+    rows = [
+        {key: value[i] for key, value in data.items()}
+        for i in range(len(data[next(iter(data.keys()))]))
+    ]
+    df = session.createDataFrame(rows)  # type: ignore
+    df = df.withColumns({"b": F.col("a") != F.lit(float("inf"))})
+
+    employee = get_df("employee")
+    sf_session = employee.session
+    dfs = sf_session.createDataFrame(rows)
+    dfs = dfs.withColumns({"b": SF.col("a") != SF.lit(float("inf"))})
+
+    compare_frames(df, dfs, compare_schema=False, sort=True)

{sqlframe-3.43.6 → sqlframe-3.43.8}/tests/unit/standalone/test_functions.py

@@ -57,8 +57,8 @@ def test_invoke_anonymous(name, func):
         ),
         (SF.lit({"cola": 1, "colb": "test"}), "MAP('cola', 1, 'colb', 'test')"),
         (SF.lit(Row(cola=1, colb="test")), "STRUCT(1 AS cola, 'test' AS colb)"),
-        (SF.lit(float("inf")), "'inf'"),
-        (SF.lit(float("-inf")), "'-inf'"),
+        (SF.lit(float("inf")), "CAST('inf' AS FLOAT)"),
+        (SF.lit(float("-inf")), "CAST('-inf' AS FLOAT)"),
     ],
 )
 def test_lit(expression, expected):