sqlframe 2.3.0__tar.gz → 2.4.0__tar.gz

{sqlframe-2.3.0 → sqlframe-2.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 2.3.0
+Version: 2.4.0
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman

{sqlframe-2.3.0 → sqlframe-2.4.0}/docs/duckdb.md

@@ -369,7 +369,6 @@ See something that you would like to see supported? [Open an issue](https://gith
 * [regexp_extract](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.regexp_extract.html)
 * [regexp_like](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.regexp_like.html)
 * [regexp_replace](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.regexp_replace.html)
-    * Replaces first match only
 * [repeat](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.repeat.html)
 * [reverse](https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.functions.reverse.html)
     * Only works on strings (does not work on arrays)

{sqlframe-2.3.0 → sqlframe-2.4.0}/setup.py

@@ -20,7 +20,7 @@ setup(
     python_requires=">=3.8",
     install_requires=[
         "prettytable<3.11.1",
-        "sqlglot>=24.0.0,<25.16",
+        "sqlglot>=24.0.0,<25.17",
         "typing_extensions>=4.8,<5",
     ],
     extras_require={

{sqlframe-2.3.0 → sqlframe-2.4.0}/sqlframe/_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '2.3.0'
-__version_tuple__ = version_tuple = (2, 3, 0)
+__version__ = version = '2.4.0'
+__version_tuple__ = version_tuple = (2, 4, 0)

{sqlframe-2.3.0 → sqlframe-2.4.0}/sqlframe/base/column.py

@@ -339,7 +339,9 @@ class Column:
         return Column(new_expression)
 
     def cast(
-        self, dataType: t.Union[str, DataType], dialect: t.Optional[t.Union[str, Dialect]] = None
+        self,
+        dataType: t.Union[str, DataType, exp.DataType, exp.DataType.Type],
+        dialect: t.Optional[t.Union[str, Dialect]] = None,
     ) -> Column:
         from sqlframe.base.session import _BaseSession
 

{sqlframe-2.3.0 → sqlframe-2.4.0}/sqlframe/base/function_alternatives.py

@@ -1593,7 +1593,7 @@ def try_to_timestamp_pgtemp(col: ColumnOrName, format: t.Optional[ColumnOrName]
 def typeof_pg_typeof(col: ColumnOrName) -> Column:
     return (
         Column.invoke_anonymous_function(col, "pg_typeof")
-        .cast("regtype", dialect="postgres")
+        .cast(expression.DataType.build("regtype", dialect="postgres"))
         .cast("text")
     )
 
@@ -1603,6 +1603,29 @@ def typeof_from_variant(col: ColumnOrName) -> Column:
     return Column.invoke_anonymous_function(col, "TYPEOF")
 
 
+def regexp_replace_global_option(
+    str: ColumnOrName, pattern: str, replacement: str, position: t.Optional[int] = None
+) -> Column:
+    lit = get_func_from_session("lit")
+
+    if position is not None:
+        return Column.invoke_expression_over_column(
+            str,
+            expression.RegexpReplace,
+            expression=lit(pattern),
+            replacement=lit(replacement),
+            position=lit(position),
+            modifiers=lit("g"),
+        )
+    return Column.invoke_expression_over_column(
+        str,
+        expression.RegexpReplace,
+        expression=lit(pattern),
+        replacement=lit(replacement),
+        modifiers=lit("g"),
+    )
+
+
 def _is_string_using_typeof_varchar(col: ColumnOrName) -> Column:
     typeof = get_func_from_session("typeof")
     lit = get_func_from_session("lit")

{sqlframe-2.3.0 → sqlframe-2.4.0}/sqlframe/duckdb/functions.py

@@ -52,4 +52,5 @@ from sqlframe.base.function_alternatives import (  # noqa
     array_append_list_append as array_append,
     endswith_with_underscore as endswith,
     last_day_with_cast as last_day,
+    regexp_replace_global_option as regexp_replace,
 )

{sqlframe-2.3.0 → sqlframe-2.4.0}/sqlframe/duckdb/functions.pyi

@@ -28,6 +28,7 @@ from sqlframe.base.function_alternatives import (  # noqa
     sequence_from_generate_series as sequence,
     try_element_at_zero_based as try_element_at,
     to_unix_timestamp_include_default_format as to_unix_timestamp,
+    regexp_replace_global_option as regexp_replace,
 )
 from sqlframe.base.functions import (
     abs as abs,
@@ -156,7 +157,6 @@ from sqlframe.base.functions import (
     rank as rank,
     regexp_extract as regexp_extract,
     regexp_like as regexp_like,
-    regexp_replace as regexp_replace,
     repeat as repeat,
     reverse as reverse,
     right as right,

{sqlframe-2.3.0 → sqlframe-2.4.0}/sqlframe/postgres/functions.py

@@ -69,4 +69,5 @@ from sqlframe.base.function_alternatives import (  # noqa
     _is_string_using_typeof_char_varying as _is_string,
     endswith_using_like as endswith,
     last_day_with_cast as last_day,
+    regexp_replace_global_option as regexp_replace,
 )

{sqlframe-2.3.0 → sqlframe-2.4.0}/sqlframe/postgres/functions.pyi

@@ -47,6 +47,7 @@ from sqlframe.base.function_alternatives import (  # noqa
     right_cast_len as right,
     position_cast_start as position,
     try_element_at_zero_based as try_element_at,
+    regexp_replace_global_option as regexp_replace,
 )
 from sqlframe.base.functions import (
     abs as abs,
@@ -148,7 +149,6 @@ from sqlframe.base.functions import (
     radians as radians,
     rank as rank,
     regexp_like as regexp_like,
-    regexp_replace as regexp_replace,
     repeat as repeat,
     reverse as reverse,
     rlike as rlike,

{sqlframe-2.3.0 → sqlframe-2.4.0}/sqlframe.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 2.3.0
+Version: 2.4.0
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman

{sqlframe-2.3.0 → sqlframe-2.4.0}/sqlframe.egg-info/requires.txt

@@ -1,5 +1,5 @@
 prettytable<3.11.1
-sqlglot<25.16,>=24.0.0
+sqlglot<25.17,>=24.0.0
 typing_extensions<5,>=4.8
 
 [bigquery]

{sqlframe-2.3.0 → sqlframe-2.4.0}/tests/integration/engines/test_int_functions.py

@@ -1944,23 +1944,11 @@ def test_regexp_replace(get_session_and_func, get_func):
     session, regexp_replace = get_session_and_func("regexp_replace")
     col = get_func("col", session)
     df = session.createDataFrame([("100-200", r"(\d+)", "--")], ["str", "pattern", "replacement"])
-    # Spark replaces all matches while most just replace the first
-    if isinstance(session, (BigQuerySession, PySparkSession, SparkSession, SnowflakeSession)):
-        assert df.select(regexp_replace("str", r"(\d+)", "--").alias("d")).first()[0] == "-----"
-        assert (
-            df.select(regexp_replace("str", col("pattern"), col("replacement")).alias("d")).first()[
-                0
-            ]
-            == "-----"
-        )
-    else:
-        assert df.select(regexp_replace("str", r"(\d+)", "--").alias("d")).first()[0] == "---200"
-        assert (
-            df.select(regexp_replace("str", col("pattern"), col("replacement")).alias("d")).first()[
-                0
-            ]
-            == "---200"
-        )
+    assert df.select(regexp_replace("str", r"(\d+)", "--").alias("d")).first()[0] == "-----"
+    assert (
+        df.select(regexp_replace("str", col("pattern"), col("replacement")).alias("d")).first()[0]
+        == "-----"
+    )
 
 
 def test_initcap(get_session_and_func):