[1/3] sdks/python: enrich data with Milvus Search [Vector, Keyword, Hybrid]

CHANGES.md

@@ -75,6 +75,9 @@
 
 * X feature added (Java/Python) ([#X](https://github.com/apache/beam/issues/X)).
 * Add pip-based install support for JupyterLab Sidepanel extension ([#35397](https://github.com/apache/beam/issues/#35397)).
+* Milvus enrichment handler added (Python) ([#35216](https://github.com/apache/beam/pull/35216)).
+  Beam now supports Milvus enrichment handler capabilities for vector, keyword,
+  and hybrid search operations.
 
 ## Breaking Changes
 

sdks/python/apache_beam/ml/rag/enrichment/bigquery_vector_search.py

@@ -220,11 +220,11 @@ def format_query(self, chunks: List[Chunk]) -> str:
       # Create embeddings subquery for this group
       embedding_unions = []
       for chunk in group_chunks:
-        if chunk.embedding is None or chunk.embedding.dense_embedding is None:
+        if not chunk.dense_embedding:
           raise ValueError(f"Chunk {chunk.id} missing embedding")
         embedding_str = (
             f"SELECT '{chunk.id}' as id, "
-            f"{[float(x) for x in chunk.embedding.dense_embedding]} "
+            f"{[float(x) for x in chunk.dense_embedding]} "
             f"as embedding")
         embedding_unions.append(embedding_str)
       group_embeddings = " UNION ALL ".join(embedding_unions)