sqlframe 3.39.2__py3-none-any.whl → 3.39.4__py3-none-any.whl

sqlframe/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '3.39.2'
-__version_tuple__ = version_tuple = (3, 39, 2)
+__version__ = version = '3.39.4'
+__version_tuple__ = version_tuple = (3, 39, 4)
 
-__commit_id__ = commit_id = 'g772b3a6bf'
+__commit_id__ = commit_id = 'g7103a1e73'

sqlframe/base/dataframe.py

@@ -31,6 +31,7 @@ from sqlframe.base.util import (
     get_func_from_session,
     get_tables_from_expression_with_join,
     normalize_string,
+    partition_to,
     quote_preserving_alias_or_name,
     sqlglot_to_spark,
     verify_openai_installed,
@@ -1633,14 +1634,30 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
 
     @operation(Operation.SELECT)
     def drop(self, *cols: t.Union[str, Column]) -> Self:
-        all_columns = self._get_outer_select_columns(self.expression)
-        drop_cols = self._ensure_and_normalize_cols(cols)
-        new_columns = [
-            col
-            for col in all_columns
-            if col.alias_or_name not in [drop_column.alias_or_name for drop_column in drop_cols]
-        ]
-        return self.copy().select(*new_columns, append=False)
+        # Separate string column names from Column objects for different handling
+        column_objs, column_names = partition_to(lambda x: isinstance(x, str), cols, list, set)
+
+        # Normalize only the Column objects (strings will be handled as unqualified)
+        drop_cols = self._ensure_and_normalize_cols(column_objs) if column_objs else []
+
+        # Work directly with the expression's select columns to preserve table qualifiers
+        current_expressions = self.expression.expressions
+        drop_sql = {drop_col.expression.sql() for drop_col in drop_cols}
+
+        # Create a more sophisticated matching function that considers table qualifiers
+        def should_drop_expression(expr: exp.Expression) -> bool:
+            # Check against fully qualified Column objects and
+            # Check against unqualified string column names (drop ALL columns with this name)
+            if expr.sql() in drop_sql or (
+                isinstance(expr, exp.Column) and expr.alias_or_name in column_names
+            ):
+                return True
+            return False
+
+        new_expressions = [expr for expr in current_expressions if not should_drop_expression(expr)]
+        return self.select.__wrapped__(  # type: ignore
+            self, *new_expressions, skip_update_display_name_mapping=True
+        )
 
     @operation(Operation.LIMIT)
     def limit(self, num: int) -> Self:

sqlframe/base/functions.py

@@ -37,9 +37,7 @@ def _get_session() -> _BaseSession:
 
 @meta()
 def col(column_name: t.Union[ColumnOrName, t.Any]) -> Column:
-    from sqlframe.base.session import _BaseSession
-
-    dialect = _BaseSession().input_dialect
+    dialect = _get_session().input_dialect
     if isinstance(column_name, str):
         col_expression = expression.to_column(column_name, dialect=dialect).transform(
             dialect.normalize_identifier
@@ -662,9 +660,7 @@ def grouping_id(*cols: ColumnOrName) -> Column:
 
 @meta()
 def input_file_name() -> Column:
-    from sqlframe.base.session import _BaseSession
-
-    return Column(expression.Literal.string(_BaseSession()._last_loaded_file or ""))
+    return Column(expression.Literal.string(_get_session()._last_loaded_file or ""))
 
 
 @meta()
@@ -959,12 +955,10 @@ def current_timestamp() -> Column:
 
 @meta()
 def date_format(col: ColumnOrName, format: str) -> Column:
-    from sqlframe.base.session import _BaseSession
-
     return Column.invoke_expression_over_column(
         Column(expression.TimeStrToTime(this=Column.ensure_col(col).column_expression)),
         expression.TimeToStr,
-        format=_BaseSession().format_time(format),
+        format=_get_session().format_time(format),
     )
 
 
@@ -1450,6 +1444,9 @@ def unix_timestamp(
 
     session = _get_session()
 
+    if session._is_duckdb or session._is_postgres or session._is_snowflake or session._is_bigquery:
+        timestamp = Column.ensure_col(timestamp).cast("string")
+
     if session._is_bigquery:
         return unix_timestamp_bgutil(timestamp, format)
 
@@ -3375,10 +3372,9 @@ def get(col: ColumnOrName, index: t.Union[ColumnOrName, int]) -> Column:
 def get_active_spark_context() -> SparkContext:
     """Raise RuntimeError if SparkContext is not initialized,
     otherwise, returns the active SparkContext."""
-    from sqlframe.base.session import _BaseSession
     from sqlframe.spark.session import SparkSession
 
-    session: _BaseSession = _BaseSession()
+    session = _get_session()
     if not isinstance(session, SparkSession):
         raise RuntimeError("This function is only available in SparkSession.")
     return session.spark_session.sparkContext
@@ -6341,7 +6337,8 @@ def to_unix_timestamp(
     session = _get_session()
 
     if session._is_duckdb:
-        format = format or _BaseSession().default_time_format
+        format = format or session.default_time_format
+        timestamp = Column.ensure_col(timestamp).cast("string")
 
     if format is not None:
         return Column.invoke_expression_over_column(

sqlframe/base/session.py

@@ -179,7 +179,7 @@ class _BaseSession(t.Generic[CATALOG, READER, WRITER, DF, TABLE, CONN, UDF_REGIS
         return self._table(self, *args, **kwargs)
 
     def __new__(cls, *args, **kwargs):
-        if _BaseSession._instance is None:
+        if _BaseSession._instance is None or not isinstance(_BaseSession._instance, cls):
             _BaseSession._instance = super().__new__(cls)
         return _BaseSession._instance
 
@@ -194,6 +194,11 @@ class _BaseSession(t.Generic[CATALOG, READER, WRITER, DF, TABLE, CONN, UDF_REGIS
     def getActiveSession(self) -> Self:
         return self
 
+    def stop(self) -> None:
+        if connection := getattr(self, "_connection", None):
+            connection.close()
+        _BaseSession._instance = None
+
     def range(
         self,
         start: int,

sqlframe/base/util.py

@@ -6,6 +6,7 @@ import string
 import typing as t
 import unicodedata
 
+from more_itertools import partition
 from sqlglot import expressions as exp
 from sqlglot import parse_one, to_table
 from sqlglot.dialects import DuckDB
@@ -537,3 +538,17 @@ def is_relativedelta_like(value: t.Any) -> bool:
         and hasattr(value, "weeks")
         and hasattr(value, "leapdays")
     )
+
+
+T = t.TypeVar("T")
+R1 = t.TypeVar("R1")
+R2 = t.TypeVar("R2")
+
+
+def partition_to(
+    pred: t.Callable[[T], bool],
+    iterable: t.Iterable[T],
+    result1: t.Type[R1],
+    result2: t.Type[R2],
+) -> tuple[R1, R2]:
+    return (lambda x, y: (result1(x), result2(y)))(*partition(pred, iterable))  # type: ignore

{sqlframe-3.39.2.dist-info → sqlframe-3.39.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 3.39.2
+Version: 3.39.4
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman
@@ -16,8 +16,9 @@ Classifier: Programming Language :: Python :: 3 :: Only
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
+Requires-Dist: more-itertools
 Requires-Dist: prettytable <4
-Requires-Dist: sqlglot <27.9,>=24.0.0
+Requires-Dist: sqlglot <27.10,>=24.0.0
 Requires-Dist: typing-extensions
 Provides-Extra: bigquery
 Requires-Dist: google-cloud-bigquery-storage <3,>=2 ; extra == 'bigquery'

{sqlframe-3.39.2.dist-info → sqlframe-3.39.4.dist-info}/RECORD

@@ -1,25 +1,25 @@
 sqlframe/__init__.py,sha256=SB80yLTITBXHI2GCDS6n6bN5ObHqgPjfpRPAUwxaots,3403
-sqlframe/_version.py,sha256=2ZMyDPGKBrqT_KjBcZ7ni5_lsj0fVr5EDt184buBf6w,714
+sqlframe/_version.py,sha256=RTACos9x6Q52oWKZxADQ1aU73aw2iil2MarQYuwWMsM,714
 sqlframe/py.typed,sha256=Nqnn8clbgv-5l0PgxcTOldg8mkMKrFn4TvPL-rYUUGg,1
 sqlframe/base/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sqlframe/base/_typing.py,sha256=b2clI5HI1zEZKB_3Msx3FeAJQyft44ubUifJwQRVXyQ,1298
 sqlframe/base/catalog.py,sha256=-YulM2BMK8MoWbXi05AsJIPxd4AuiZDBCZuk4HoeMlE,38900
 sqlframe/base/column.py,sha256=f6rK6-hTiNx9WwJP7t6tqL3xEC2gwERPDlhWCS5iCBw,21417
-sqlframe/base/dataframe.py,sha256=0diYONDlet8iZt49LC3vcmfXHAAZ2MovPL2pTXYHj2U,85974
+sqlframe/base/dataframe.py,sha256=-jeoqP5jS8Rk1fp_Og9ie_e2fjo19uX7JVpi9PeU5qI,86943
 sqlframe/base/decorators.py,sha256=IhE5xNQDkwJHacCvulq5WpUKyKmXm7dL2A3o5WuKGP4,2131
 sqlframe/base/exceptions.py,sha256=9Uwvqn2eAkDpqm4BrRgbL61qM-GMCbJEMAW8otxO46s,370
 sqlframe/base/function_alternatives.py,sha256=aTu3nQhIAkZoxrI1IpjpaHEAMxBNms0AnhS0EMR-TwY,51727
-sqlframe/base/functions.py,sha256=9hW5aYke5EFU4C7Epx-TlyG2ZxjYnFGskv4LwHiQ2dw,227752
+sqlframe/base/functions.py,sha256=lMwNtOAj7MbmaFtweo5N8DJp-8ent1fT6lr3J3YcQsA,227753
 sqlframe/base/group.py,sha256=fBm8EUve7W7xz11nybTXr09ih-yZxL_vvEiZVE1eb_0,12025
 sqlframe/base/normalize.py,sha256=nXAJ5CwxVf4DV0GsH-q1w0p8gmjSMlv96k_ez1eVul8,3880
 sqlframe/base/operations.py,sha256=g-YNcbvNKTOBbYm23GKfB3fmydlR7ZZDAuZUtXIHtzw,4438
 sqlframe/base/readerwriter.py,sha256=Nb2VJ_HBmLQp5mK8JhnFooZh2ydAaboCAFVPb-4MNX4,31241
-sqlframe/base/session.py,sha256=8oaEgGbyctKKEaI0GW6k7Praku7nwx3YRYgAW3mZNk0,27481
+sqlframe/base/session.py,sha256=99X-ShK9ohHCX6WdIJs0HhjfK23snaE3Gv6RYc5wqUI,27687
 sqlframe/base/table.py,sha256=rCeh1W5SWbtEVfkLAUiexzrZwNgmZeptLEmLcM1ABkE,6961
 sqlframe/base/transforms.py,sha256=y0j3SGDz3XCmNGrvassk1S-owllUWfkHyMgZlY6SFO4,467
 sqlframe/base/types.py,sha256=OktuJ5f7tEogOW0oupI0RBlHfzZMmKh7zGLke9cwllo,12305
 sqlframe/base/udf.py,sha256=O6hMhBUy9NVv-mhJRtfFhXTIa_-Z8Y_FkmmuOHu0l90,1117
-sqlframe/base/util.py,sha256=D4HAhtu4DMz5mXyxlUHRP_GrsjLJACpBYlLriyGoT0g,19435
+sqlframe/base/util.py,sha256=11rBF_GBFXGBCllSdlWWWo8EiZZATJn4me3u7OUNIFg,19782
 sqlframe/base/window.py,sha256=7NaKDTlhun-95LEghukBCjFBwq0RHrPaajWQNCsLxok,4818
 sqlframe/base/mixins/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sqlframe/base/mixins/catalog_mixins.py,sha256=9fZGWToz9xMJSzUl1vsVtj6TH3TysP3fBCKJLnGUQzE,23353
@@ -130,8 +130,8 @@ 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-3.39.2.dist-info/LICENSE,sha256=VZu79YgW780qxaFJMr0t5ZgbOYEh04xWoxaWOaqIGWk,1068
-sqlframe-3.39.2.dist-info/METADATA,sha256=M0k0V_XPUzeL9-tCwZWKMMv9DVhVstFonKVOWRc7wRk,9039
-sqlframe-3.39.2.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
-sqlframe-3.39.2.dist-info/top_level.txt,sha256=T0_RpoygaZSF6heeWwIDQgaP0varUdSK1pzjeJZRjM8,9
-sqlframe-3.39.2.dist-info/RECORD,,
+sqlframe-3.39.4.dist-info/LICENSE,sha256=VZu79YgW780qxaFJMr0t5ZgbOYEh04xWoxaWOaqIGWk,1068
+sqlframe-3.39.4.dist-info/METADATA,sha256=wv8nBkcg2ofsdZYo3qC0g8q_-QR_flvR1YPoKQ6uGeU,9070
+sqlframe-3.39.4.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
+sqlframe-3.39.4.dist-info/top_level.txt,sha256=T0_RpoygaZSF6heeWwIDQgaP0varUdSK1pzjeJZRjM8,9
+sqlframe-3.39.4.dist-info/RECORD,,