Move search related config under SearchConfig

CHANGELOG.md

@@ -52,7 +52,11 @@
   - `content` and `docling_document` are mutually exclusive
 - **BREAKING: Chunker Interface**: `DocumentChunker.chunk()` now returns `list[Chunk]` instead of `list[str]`
   - Chunks include structured metadata (doc_item_refs, labels, headings, page_numbers)
-- **BREAKING: Config Renamed**: `context_chunk_radius` renamed to `text_context_radius`
+- **Search Config**: New settings in `search` section for search behavior and context expansion
+  - `search.limit` - Default number of search results (default: 5). Used by CLI, MCP server, and API when no limit specified
+  - `search.context_radius` - DocItems before/after to include for text content expansion (default: 0)
+  - `search.max_context_items` - Maximum items in expanded context (default: 25)
+  - `search.max_context_chars` - Maximum characters in expanded context (default: 10000)
 - **Rebuild Performance**: Batched database writes during `rebuild` command reduce LanceDB versions by ~98%
   - All rebuild modes (FULL, RECHUNK, EMBED_ONLY) now batch writes across documents
   - Eliminates redundant per-document chunk deletions and vacuum calls

docs/configuration/index.md

@@ -100,6 +100,9 @@ research:
 
 search:
   limit: 5                     # Default number of results to return
+  context_radius: 0            # DocItems before/after to include for text content
+  max_context_items: 25        # Maximum items in expanded context
+  max_context_chars: 10000     # Maximum characters in expanded context
   vector_index_metric: cosine  # cosine, l2, or dot
   vector_refine_factor: 30
 
@@ -116,9 +119,6 @@ processing:
   chunker: docling-local    # docling-local or docling-serve
   chunker_type: hybrid      # hybrid or hierarchical
   chunk_size: 256
-  text_context_radius: 0
-  max_context_items: 25
-  max_context_chars: 10000
   chunking_tokenizer: "Qwen/Qwen3-Embedding-0.6B"
   chunking_merge_peers: true
   chunking_use_markdown_tables: false

docs/configuration/processing.md

@@ -11,11 +11,6 @@ processing:
   # Chunking configuration
   chunk_size: 256                            # Maximum tokens per chunk
 
-  # Context expansion for search results
-  text_context_radius: 0                     # Radius for text chunk expansion
-  max_context_items: 25                      # Maximum items in expanded context
-  max_context_chars: 10000                   # Maximum characters in expanded context
-
   # Converter selection
   converter: docling-local                   # docling-local or docling-serve
 
@@ -137,27 +132,14 @@ processing:
 - `false`: Tables as narrative text ("Value A, Column 2 = Value B")
 - `true`: Tables as markdown (preserves table structure)
 
-### Chunk Size and Context Expansion
+### Chunk Size
 
 ```yaml
 processing:
-  # Chunk size for document processing
+  chunk_size: 256  # Maximum tokens per chunk
-  chunk_size: 256
-
-  # Context expansion settings
-  # Controls how search results are expanded with surrounding content
-  text_context_radius: 0     # Chunks before/after to include for text content
-  max_context_items: 25      # Maximum doc items to include in expansion
-  max_context_chars: 10000   # Maximum characters in expanded content
 ```
 
-Context expansion enriches search results with surrounding content from the source document:
+Context expansion settings (for enriching search results with surrounding content) are configured in the `search` section. See [Search Settings](storage.md#search-settings).
-
-- **text_context_radius**: For text content (paragraphs), includes N chunks before and after. Set to 0 to disable expansion (default).
-- **max_context_items**: Limits how many document items (paragraphs, list items, etc.) can be included in expanded context.
-- **max_context_chars**: Hard limit on total characters in expanded content.
-
-Structural content (tables, code blocks, lists) uses type-aware expansion that automatically includes the complete structure regardless of how it was chunked. For example, if a table was split across multiple chunks, expansion retrieves the complete table.
 
 ## File Monitoring
 

docs/configuration/storage.md

@@ -59,16 +59,32 @@ This prevents the common mistake where a search query accidentally creates an em
 
 ## Search Settings
 
-Configure search behavior:
+Configure search behavior and context expansion:
 
 ```yaml
 search:
   limit: 5                     # Default number of results to return
+  context_radius: 0            # DocItems before/after to include for text content
+  max_context_items: 25        # Maximum items in expanded context
+  max_context_chars: 10000     # Maximum characters in expanded context
   vector_index_metric: cosine  # cosine, l2, or dot
   vector_refine_factor: 30     # Re-ranking factor for accuracy
 ```
 
 - **limit**: Default number of search results to return when no limit is specified. Used by CLI, MCP server, and API. Default: 5
+
+### Context Expansion
+
+Context expansion enriches search results with surrounding content from the source document:
+
+- **context_radius**: For text content (paragraphs), includes N DocItems before and after. Set to 0 to disable expansion (default).
+- **max_context_items**: Limits how many document items (paragraphs, list items, etc.) can be included in expanded context.
+- **max_context_chars**: Hard limit on total characters in expanded content.
+
+Structural content (tables, code blocks, lists) uses type-aware expansion that automatically includes the complete structure regardless of how it was chunked. For example, if a table was split across multiple chunks, expansion retrieves the complete table.
+
+### Vector Indexing
+
 - **vector_index_metric**: Distance metric for vector similarity:
   - `cosine`: Cosine similarity (default, best for most embeddings)
   - `l2`: Euclidean distance

evaluations/evaluations/benchmark.py

@@ -45,9 +45,10 @@ def build_experiment_metadata(
         "embedder_model": config.embeddings.model.name,
         "embedder_dim": config.embeddings.model.vector_dim,
         "chunk_size": config.processing.chunk_size,
-        "text_context_radius": config.processing.text_context_radius,
+        "search_limit": config.search.limit,
-        "max_context_items": config.processing.max_context_items,
+        "context_radius": config.search.context_radius,
-        "max_context_chars": config.processing.max_context_chars,
+        "max_context_items": config.search.max_context_items,
+        "max_context_chars": config.search.max_context_chars,
         "rerank_provider": config.reranking.model.provider
         if config.reranking.model
         else None,

haiku_rag_slim/haiku/rag/client.py

@@ -880,8 +880,8 @@ class HaikuRAG:
 
         Expansion is type-aware based on content:
         - Tables, code blocks, and lists expand to include complete structures
-        - Text content uses the configured radius (text_context_radius)
+        - Text content uses the configured radius (search.context_radius)
-        - Expansion is limited by max_context_items and max_context_chars
+        - Expansion is limited by search.max_context_items and search.max_context_chars
 
         Args:
             search_results: List of SearchResult objects from search.
@@ -889,9 +889,9 @@ class HaikuRAG:
         Returns:
             List of SearchResult objects with expanded content and resolved provenance.
         """
-        radius = self._config.processing.text_context_radius
+        radius = self._config.search.context_radius
-        max_items = self._config.processing.max_context_items
+        max_items = self._config.search.max_context_items
-        max_chars = self._config.processing.max_context_chars
+        max_chars = self._config.search.max_context_chars
 
         # Group by document_id for efficient processing
         document_groups: dict[str | None, list[SearchResult]] = {}

haiku_rag_slim/haiku/rag/config/models.py

@@ -110,9 +110,6 @@ class ConversionOptions(BaseModel):
 
 class ProcessingConfig(BaseModel):
     chunk_size: int = 256
-    text_context_radius: int = 0
-    max_context_items: int = 25
-    max_context_chars: int = 10000
     converter: str = "docling-local"
     chunker: str = "docling-local"
     chunker_type: str = "hybrid"
@@ -124,6 +121,9 @@ class ProcessingConfig(BaseModel):
 
 class SearchConfig(BaseModel):
     limit: int = 5
+    context_radius: int = 0
+    max_context_items: int = 25
+    max_context_chars: int = 10000
     vector_index_metric: Literal["cosine", "l2", "dot"] = "cosine"
     vector_refine_factor: int = 30
 

tests/test_context_enhancement.py

@@ -81,8 +81,8 @@ def small_chunk_config() -> AppConfig:
     """Config with small chunk size to force splitting."""
     config = AppConfig()
     config.processing.chunk_size = 32
-    config.processing.max_context_items = 25
+    config.search.max_context_items = 25
-    config.processing.max_context_chars = 10000
+    config.search.max_context_chars = 10000
     return config
 
 
@@ -228,7 +228,7 @@ async def test_text_expansion_uses_radius(temp_db_path):
     """Text content expansion should use radius, not structural boundaries."""
     config = AppConfig()
     config.processing.chunk_size = 32
-    config.processing.text_context_radius = 1  # Small radius
+    config.search.context_radius = 1  # Small radius
 
     # Create a document with longer paragraphs that will split
     doc = DoclingDocument(name="text_test")
@@ -312,7 +312,7 @@ async def test_max_items_limit_caps_expansion(temp_db_path):
     """Expansion should respect max_context_items limit."""
     config = AppConfig()
     config.processing.chunk_size = 32
-    config.processing.max_context_items = 2  # Very restrictive
+    config.search.max_context_items = 2  # Very restrictive
 
     docling_doc = create_list_document()
 
@@ -410,7 +410,7 @@ async def test_expand_context_radius_zero(temp_db_path):
 async def test_expand_context_multiple_documents(temp_db_path):
     """Test expand_context with results from multiple documents."""
     config = AppConfig()
-    config.processing.text_context_radius = 1
+    config.search.context_radius = 1
 
     async with HaikuRAG(temp_db_path, config=config, create=True) as client:
         # Create first document with manual chunks
@@ -471,7 +471,7 @@ async def test_expand_context_multiple_documents(temp_db_path):
 async def test_expand_context_merges_overlapping_chunks(temp_db_path):
     """Test that overlapping expanded chunks are merged into one."""
     config = AppConfig()
-    config.processing.text_context_radius = 1
+    config.search.context_radius = 1
 
     async with HaikuRAG(temp_db_path, config=config, create=True) as client:
         # Create document with 5 chunks
@@ -526,7 +526,7 @@ async def test_expand_context_merges_overlapping_chunks(temp_db_path):
 async def test_expand_context_keeps_separate_non_overlapping(temp_db_path):
     """Test that non-overlapping expanded chunks remain separate."""
     config = AppConfig()
-    config.processing.text_context_radius = 1
+    config.search.context_radius = 1
 
     async with HaikuRAG(temp_db_path, config=config, create=True) as client:
         # Create document with chunks far apart
@@ -591,7 +591,7 @@ async def test_expand_context_keeps_separate_non_overlapping(temp_db_path):
 async def test_expand_context_with_docling_merges_overlapping(temp_db_path):
     """Test that expand_context with DoclingDocument merges overlapping results."""
     config = AppConfig()
-    config.processing.text_context_radius = 3
+    config.search.context_radius = 3
 
     markdown_content = """# Chapter 1
 
@@ -646,7 +646,7 @@ This is paragraph four about topic C.
 async def test_expand_context_docling_merges_metadata(temp_db_path):
     """Test that expand_context properly merges metadata from multiple results."""
     config = AppConfig()
-    config.processing.text_context_radius = 10
+    config.search.context_radius = 10
 
     markdown_content = """# Introduction