Fix write_pandas() truncate-only flow to prevent CREATE TABLE IF NOT EXISTS (#1899)

* Fix write_pandas() truncate-only flow to prevent CREATE TABLE IF NOT EXISTS

* Updated DESCRIPTION.md

Author: sfc-gh-kterada

DESCRIPTION.md

@@ -20,7 +20,8 @@ Source code is also available at: https://github.com/snowflakedb/snowflake-conne
     - Add `SNOWFLAKE_AUTH_SOCKET_MSG_DONTWAIT` flag (usage: `SNOWFLAKE_AUTH_SOCKET_MSG_DONTWAIT=true`) to make a non-blocking socket.recv call and retry on Error
       - Consider using this if running in a containerized environment and externalbrowser auth frequently hangs while waiting for callback
       - NOTE: this has not been tested extensively, but has been shown to improve the experience when using WSL
-      - Updated diagnostics to use system$allowlist instead of system$whitelist.
+  - Updated diagnostics to use system$allowlist instead of system$whitelist.
+  - Update `write_pandas` to skip TABLE IF NOT EXISTS in truncate mode
 
 - v3.7.1(February 21, 2024)
 

src/snowflake/connector/pandas_tools.py

@@ -393,13 +393,14 @@ def drop_object(name: str, object_type: str) -> None:
             quote_identifiers,
         )
 
-        create_table_sql = (
-            f"CREATE {table_type.upper()} TABLE IF NOT EXISTS {target_table_location} "
-            f"({create_table_columns})"
-            f" /* Python:snowflake.connector.pandas_tools.write_pandas() */ "
-        )
-        logger.debug(f"auto creating table with '{create_table_sql}'")
-        cursor.execute(create_table_sql, _is_internal=True)
+        if auto_create_table:
+            create_table_sql = (
+                f"CREATE {table_type.upper()} TABLE IF NOT EXISTS {target_table_location} "
+                f"({create_table_columns})"
+                f" /* Python:snowflake.connector.pandas_tools.write_pandas() */ "
+            )
+            logger.debug(f"auto creating table with '{create_table_sql}'")
+            cursor.execute(create_table_sql, _is_internal=True)
         # need explicit casting when the underlying table schema is inferred
         parquet_columns = "$1:" + ",$1:".join(
             f"{quote}{snowflake_col}{quote}::{column_type_mapping[col]}"

test/integ/pandas/test_pandas_tools.py

@@ -115,6 +115,7 @@ def test_write_pandas_with_overwrite(
         f"({col_name} STRING, {col_points} INT, {col_id} INT AUTOINCREMENT)"
     )
 
+    show_sql = f"SHOW TABLES LIKE '{random_table_name}'"  # SHOW command like is case-insensitive
     select_sql = f"SELECT * FROM {table_name}"
     select_count_sql = f"SELECT count(*) FROM {table_name}"
     drop_sql = f"DROP TABLE IF EXISTS {table_name}"
@@ -148,6 +149,35 @@ def test_write_pandas_with_overwrite(
             # Check number of rows
             assert result["COUNT(*)"] == 1
 
+            # Truncate-only mode
+            if not auto_create_table:
+                # Capture the created_on timestamp of the target table before truncation
+                result = cnx.cursor(DictCursor).execute(show_sql).fetchone()
+                create_ts_before_trunc = result["created_on"]
+
+                # This operation should only truncate the table
+                success, nchunks, nrows, _ = write_pandas(
+                    cnx,
+                    df2,
+                    random_table_name,
+                    quote_identifiers=quote_identifiers,
+                    auto_create_table=auto_create_table,
+                    overwrite=True,
+                    index=index,
+                )
+
+                # Capture the created_on timestamp of the target table after truncation
+                result = cnx.cursor(DictCursor).execute(show_sql).fetchone()
+                create_ts_after_trunc = result["created_on"]
+                assert create_ts_before_trunc == create_ts_after_trunc
+
+                # Check write_pandas output
+                assert success
+                assert nchunks == 1
+                result = cnx.cursor(DictCursor).execute(select_count_sql).fetchone()
+                # Check number of rows
+                assert result["COUNT(*)"] == 1
+
             # Write dataframe with a different schema
             if auto_create_table:
                 # Should drop table and SUCCEED because the new table will be created with new schema of df3