fakesnow 0.9.10__tar.gz → 0.9.12__tar.gz

{fakesnow-0.9.10 → fakesnow-0.9.12}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: fakesnow
-Version: 0.9.10
+Version: 0.9.12
 Summary: Fake Snowflake Connector for Python. Run, mock and test Snowflake DB locally.
 License:                                  Apache License
                                    Version 2.0, January 2004
@@ -357,7 +357,7 @@ For more detail see [tests/test_fakes.py](tests/test_fakes.py)
 ## Caveats
 
 - The order of rows is non deterministic and may not match Snowflake unless ORDER BY is fully specified.
-- A more liberal Snowflake SQL dialect than a real Snowflake instance is supported, ie: some queries might pass using fakesnow that a real Snowflake instance would reject.
+- A more liberal Snowflake SQL dialect than used by a real Snowflake instance is supported, ie: some queries might pass using fakesnow that a real Snowflake instance would reject.
 
 ## Contributing
 

{fakesnow-0.9.10 → fakesnow-0.9.12}/README.md

@@ -127,7 +127,7 @@ For more detail see [tests/test_fakes.py](tests/test_fakes.py)
 ## Caveats
 
 - The order of rows is non deterministic and may not match Snowflake unless ORDER BY is fully specified.
-- A more liberal Snowflake SQL dialect than a real Snowflake instance is supported, ie: some queries might pass using fakesnow that a real Snowflake instance would reject.
+- A more liberal Snowflake SQL dialect than used by a real Snowflake instance is supported, ie: some queries might pass using fakesnow that a real Snowflake instance would reject.
 
 ## Contributing
 

{fakesnow-0.9.10 → fakesnow-0.9.12}/fakesnow/fakes.py

@@ -16,6 +16,7 @@ from sqlglot import exp
 if TYPE_CHECKING:
     import pandas as pd
     import pyarrow.lib
+import numpy as np
 import pyarrow
 import snowflake.connector.converter
 import snowflake.connector.errors
@@ -606,9 +607,7 @@ class FakeSnowflakeConnection:
     def rollback(self) -> None:
         self.cursor().execute("ROLLBACK")
 
-    def _insert_df(
-        self, df: pd.DataFrame, table_name: str, database: str | None = None, schema: str | None = None
-    ) -> int:
+    def _insert_df(self, df: pd.DataFrame, table_name: str) -> int:
         # Objects in dataframes are written as parquet structs, and snowflake loads parquet structs as json strings.
         # Whereas duckdb analyses a dataframe see https://duckdb.org/docs/api/python/data_ingestion.html#pandas-dataframes--object-columns
         # and converts a object to the most specific type possible, eg: dict -> STRUCT, MAP or varchar, and list -> LIST
@@ -680,6 +679,15 @@ WritePandasResult = tuple[
 ]
 
 
+def sql_type(dtype: np.dtype) -> str:
+    if str(dtype) == "int64":
+        return "NUMBER"
+    elif str(dtype) == "object":
+        return "VARCHAR"
+    else:
+        raise NotImplementedError(f"sql_type {dtype=}")
+
+
 def write_pandas(
     conn: FakeSnowflakeConnection,
     df: pd.DataFrame,
@@ -697,7 +705,18 @@ def write_pandas(
     table_type: Literal["", "temp", "temporary", "transient"] = "",
     **kwargs: Any,
 ) -> WritePandasResult:
-    count = conn._insert_df(df, table_name, database, schema)  # noqa: SLF001
+    name = table_name
+    if schema:
+        name = f"{schema}.{name}"
+    if database:
+        name = f"{database}.{name}"
+
+    if auto_create_table:
+        cols = [f"{c} {sql_type(t)}" for c, t in df.dtypes.to_dict().items()]
+
+        conn.cursor().execute(f"CREATE TABLE IF NOT EXISTS {name} ({','.join(cols)})")
+
+    count = conn._insert_df(df, name)  # noqa: SLF001
 
     # mocks https://docs.snowflake.com/en/sql-reference/sql/copy-into-table.html#output
     mock_copy_results = [("fakesnow/file0.txt", "LOADED", count, count, 1, 0, None, None, None, None)]

{fakesnow-0.9.10 → fakesnow-0.9.12}/fakesnow.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: fakesnow
-Version: 0.9.10
+Version: 0.9.12
 Summary: Fake Snowflake Connector for Python. Run, mock and test Snowflake DB locally.
 License:                                  Apache License
                                    Version 2.0, January 2004
@@ -357,7 +357,7 @@ For more detail see [tests/test_fakes.py](tests/test_fakes.py)
 ## Caveats
 
 - The order of rows is non deterministic and may not match Snowflake unless ORDER BY is fully specified.
-- A more liberal Snowflake SQL dialect than a real Snowflake instance is supported, ie: some queries might pass using fakesnow that a real Snowflake instance would reject.
+- A more liberal Snowflake SQL dialect than used by a real Snowflake instance is supported, ie: some queries might pass using fakesnow that a real Snowflake instance would reject.
 
 ## Contributing
 

{fakesnow-0.9.10 → fakesnow-0.9.12}/pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "fakesnow"
 description = "Fake Snowflake Connector for Python. Run, mock and test Snowflake DB locally."
-version = "0.9.10"
+version = "0.9.12"
 readme = "README.md"
 license = { file = "LICENSE" }
 classifiers = ["License :: OSI Approved :: MIT License"]

{fakesnow-0.9.10 → fakesnow-0.9.12}/tests/test_fakes.py

@@ -1502,6 +1502,21 @@ def test_json_extract_cast_as_varchar(dcur: snowflake.connector.cursor.DictCurso
     assert dcur.fetchall() == [{"C_STR_NUMBER": 100, "C_NUM_NUMBER": 100}]
 
 
+def test_write_pandas_auto_create(conn: snowflake.connector.SnowflakeConnection):
+    with conn.cursor() as cur:
+        df = pd.DataFrame.from_records(
+            [
+                {"ID": 1, "FIRST_NAME": "Jenny"},
+                {"ID": 2, "FIRST_NAME": "Jasper"},
+            ]
+        )
+        snowflake.connector.pandas_tools.write_pandas(conn, df, "CUSTOMERS", auto_create_table=True)
+
+        cur.execute("select id, first_name from customers")
+
+        assert cur.fetchall() == [(1, "Jenny"), (2, "Jasper")]
+
+
 def test_write_pandas_quoted_column_names(conn: snowflake.connector.SnowflakeConnection):
     with conn.cursor(snowflake.connector.cursor.DictCursor) as dcur:
         # colunmn names with spaces
@@ -1611,6 +1626,26 @@ def test_write_pandas_dict_different_keys(conn: snowflake.connector.SnowflakeCon
         assert indent(cur.fetchall()) == [('{\n  "k": "v1"\n}',), ('{\n  "k2": [\n    "v\'2",\n    "v\\"3"\n  ]\n}',)]
 
 
+def test_write_pandas_db_schema(conn: snowflake.connector.SnowflakeConnection):
+    with conn.cursor() as cur:
+        cur.execute("create database db2")
+        cur.execute("create schema db2.schema2")
+        cur.execute("create or replace table db2.schema2.customers (ID int, FIRST_NAME varchar, LAST_NAME varchar)")
+
+        df = pd.DataFrame.from_records(
+            [
+                {"ID": 1, "FIRST_NAME": "Jenny"},
+                {"ID": 2, "FIRST_NAME": "Jasper"},
+            ]
+        )
+        snowflake.connector.pandas_tools.write_pandas(conn, df, "CUSTOMERS", "DB2", "SCHEMA2")
+
+        cur.execute("select id, first_name, last_name from db2.schema2.customers")
+
+        # columns not in dataframe will receive their default value
+        assert cur.fetchall() == [(1, "Jenny", None), (2, "Jasper", None)]
+
+
 def indent(rows: Sequence[tuple] | Sequence[dict]) -> list[tuple]:
     # indent duckdb json strings tuple values to match snowflake json strings
     assert isinstance(rows[0], tuple)