Change gzip file mode to text with UTF-8 encoding for JSON output (#616)

Author: AsafMah

CHANGELOG.md

@@ -9,6 +9,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Fixed
 - Changed extra's name back to `aio`
+- Fixed encoding error in `ingest_from_dataframe` when using csv data format.
 
 ## [6.0.0] - 2025-11-26
 

azure-kusto-ingest/azure/kusto/ingest/base_ingest_client.py

@@ -133,7 +133,7 @@ def ingest_from_dataframe(
 
         file_name = "df_{id}_{timestamp}_{uid}.{ext}.gz".format(id=id(df), timestamp=int(time.time()), uid=uuid.uuid4(), ext="json" if is_json else "csv")
         temp_file_path = os.path.join(tempfile.gettempdir(), file_name)
-        with gzip.open(temp_file_path, "wb") as temp_file:
+        with gzip.open(temp_file_path, "wt", encoding="utf-8") as temp_file:
             if is_json:
                 df.to_json(temp_file, orient="records", date_format="iso", lines=True)
                 ingestion_properties.format = DataFormat.JSON