Chunkers now set chunk.order directly

CHANGELOG.md

@@ -14,6 +14,7 @@
 
 ### Changed
 
+- **Chunker Sets Order**: Chunkers now set `chunk.order` directly
 - **Evaluations Vacuum Strategy**: `populate_db` now uses periodic vacuum to prevent disk exhaustion with large datasets
   - Disables auto_vacuum during population, vacuums every N documents with retention=0
   - New `--vacuum-interval` CLI option (default: 100) to control vacuum frequency

haiku_rag_slim/haiku/rag/chunkers/docling_local.py

@@ -155,6 +155,12 @@ class DoclingLocalChunker(DocumentChunker):
                 labels=labels,
                 page_numbers=sorted(page_numbers),
             )
-            result.append(Chunk(content=text, metadata=chunk_metadata.model_dump()))
+            result.append(
+                Chunk(
+                    content=text,
+                    metadata=chunk_metadata.model_dump(),
+                    order=len(result),
+                )
+            )
 
         return result

haiku_rag_slim/haiku/rag/chunkers/docling_serve.py

@@ -192,6 +192,12 @@ class DoclingServeChunker(DocumentChunker):
                 labels=labels,
                 page_numbers=sorted(page_numbers) if page_numbers else [],
             )
-            result.append(Chunk(content=text, metadata=chunk_metadata.model_dump()))
+            result.append(
+                Chunk(
+                    content=text,
+                    metadata=chunk_metadata.model_dump(),
+                    order=len(result),
+                )
+            )
 
         return result

haiku_rag_slim/haiku/rag/client.py

@@ -194,13 +194,7 @@ class HaikuRAG:
         from haiku.rag.chunkers import get_chunker
 
         chunker = get_chunker(self._config)
-        chunks = await chunker.chunk(docling_document)
+        return await chunker.chunk(docling_document)
-
-        # Set order for each chunk
-        for i, chunk in enumerate(chunks):
-            chunk.order = i
-
-        return chunks
 
     async def _ensure_chunks_embedded(self, chunks: list[Chunk]) -> list[Chunk]:
         """Ensure all chunks have embeddings, embedding any that don't.

tests/test_chunker.py

@@ -144,6 +144,33 @@ async def test_local_chunker_markdown_tables():
     assert "," in table_content and "|" not in table_content
 
 
+@pytest.mark.asyncio
+async def test_local_chunker_sets_order():
+    """Test that DoclingLocalChunker sets sequential order on chunks."""
+    sample_md = """# Introduction
+
+First paragraph with some content.
+
+## Section One
+
+Second paragraph.
+
+## Section Two
+
+Third paragraph.
+"""
+    converter = get_converter(Config)
+    doc = await converter.convert_text(sample_md, name="test.md")
+
+    chunker = DoclingLocalChunker()
+    chunks = await chunker.chunk(doc)
+
+    assert len(chunks) > 0
+    # Verify order is set sequentially starting from 0
+    for i, chunk in enumerate(chunks):
+        assert chunk.order == i, f"Chunk {i} has order {chunk.order}, expected {i}"
+
+
 @pytest.mark.asyncio
 async def test_local_chunker_metadata_extraction():
     """Test that DoclingLocalChunker extracts metadata correctly."""