morphik 0.2.4__tar.gz → 0.2.6__tar.gz

{morphik-0.2.4 → morphik-0.2.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: morphik
-Version: 0.2.4
+Version: 0.2.6
 Summary: Morphik Python Client
 Author-email: Morphik <founders@morphik.ai>
 Requires-Python: >=3.8

{morphik-0.2.4 → morphik-0.2.6}/morphik/__init__.py

@@ -12,4 +12,4 @@ __all__ = [
     "Document",
 ]
 
-__version__ = "0.2.4"
+__version__ = "0.2.6"

{morphik-0.2.4 → morphik-0.2.6}/morphik/_internal.py

@@ -254,6 +254,7 @@ class _MorphikClientLogic:
         chat_id: Optional[str] = None,
         schema: Optional[Union[Type[BaseModel], Dict[str, Any]]] = None,
         llm_config: Optional[Dict[str, Any]] = None,
+        padding: int = 0,
     ) -> Dict[str, Any]:
         """Prepare request for query endpoint"""
         payload = {
@@ -277,6 +278,8 @@ class _MorphikClientLogic:
             payload["chat_id"] = chat_id
         if llm_config:
             payload["llm_config"] = llm_config
+        if padding > 0:
+            payload["padding"] = padding
 
         # Add schema to payload if provided
         if schema:
@@ -303,6 +306,7 @@ class _MorphikClientLogic:
         use_colpali: bool,
         folder_name: Optional[Union[str, List[str]]],
         end_user_id: Optional[str],
+        padding: int = 0,
     ) -> Dict[str, Any]:
         """Prepare request for retrieve_chunks endpoint"""
         request = {
@@ -316,6 +320,8 @@ class _MorphikClientLogic:
             request["folder_name"] = folder_name
         if end_user_id:
             request["end_user_id"] = end_user_id
+        if padding > 0:
+            request["padding"] = padding
         return request
 
     def _prepare_retrieve_docs_request(

{morphik-0.2.4 → morphik-0.2.6}/morphik/async_.py

@@ -288,6 +288,7 @@ class AsyncFolder:
         min_score: float = 0.0,
         use_colpali: bool = True,
         additional_folders: Optional[List[str]] = None,
+        padding: int = 0,
     ) -> List[FinalChunkResult]:
         """
         Retrieve relevant chunks within this folder.
@@ -299,13 +300,14 @@ class AsyncFolder:
             min_score: Minimum similarity threshold (default: 0.0)
             use_colpali: Whether to use ColPali-style embedding model
             additional_folders: Optional list of additional folder names to further scope operations
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
 
         Returns:
             List[FinalChunkResult]: List of relevant chunks
         """
         effective_folder = self._merge_folders(additional_folders)
         payload = self._client._logic._prepare_retrieve_chunks_request(
-            query, filters, k, min_score, use_colpali, effective_folder, None
+            query, filters, k, min_score, use_colpali, effective_folder, None, padding
         )
         response = await self._client._request("POST", "retrieve/chunks", data=payload)
         return self._client._logic._parse_chunk_result_list_response(response)
@@ -357,6 +359,7 @@ class AsyncFolder:
         schema: Optional[Union[Type[BaseModel], Dict[str, Any]]] = None,
         chat_id: Optional[str] = None,
         llm_config: Optional[Dict[str, Any]] = None,
+        padding: int = 0,
     ) -> CompletionResponse:
         """
         Generate completion using relevant chunks as context within this folder.
@@ -375,6 +378,7 @@ class AsyncFolder:
             prompt_overrides: Optional customizations for entity extraction, resolution, and query prompts
             schema: Optional schema for structured output
             additional_folders: Optional list of additional folder names to further scope operations
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
 
         Returns:
             CompletionResponse: Generated completion or structured output
@@ -397,6 +401,7 @@ class AsyncFolder:
             chat_id,
             schema,
             llm_config,
+            padding,
         )
 
         # Add schema to payload if provided
@@ -826,6 +831,7 @@ class AsyncUserScope:
         min_score: float = 0.0,
         use_colpali: bool = True,
         additional_folders: Optional[List[str]] = None,
+        padding: int = 0,
     ) -> List[FinalChunkResult]:
         """
         Retrieve relevant chunks as this end user.
@@ -837,13 +843,14 @@ class AsyncUserScope:
             min_score: Minimum similarity threshold (default: 0.0)
             use_colpali: Whether to use ColPali-style embedding model
             additional_folders: Optional list of additional folder names to further scope operations
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
 
         Returns:
             List[FinalChunkResult]: List of relevant chunks
         """
         effective_folder = self._merge_folders(additional_folders)
         payload = self._client._logic._prepare_retrieve_chunks_request(
-            query, filters, k, min_score, use_colpali, effective_folder, self._end_user_id
+            query, filters, k, min_score, use_colpali, effective_folder, self._end_user_id, padding
         )
         response = await self._client._request("POST", "retrieve/chunks", data=payload)
         return self._client._logic._parse_chunk_result_list_response(response)
@@ -895,6 +902,7 @@ class AsyncUserScope:
         schema: Optional[Union[Type[BaseModel], Dict[str, Any]]] = None,
         chat_id: Optional[str] = None,
         llm_config: Optional[Dict[str, Any]] = None,
+        padding: int = 0,
     ) -> CompletionResponse:
         """
         Generate completion using relevant chunks as context, scoped to the end user.
@@ -913,6 +921,7 @@ class AsyncUserScope:
             prompt_overrides: Optional customizations for entity extraction, resolution, and query prompts
             schema: Optional schema for structured output
             additional_folders: Optional list of additional folder names to further scope operations
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
 
         Returns:
             CompletionResponse: Generated completion or structured output
@@ -935,6 +944,7 @@ class AsyncUserScope:
             chat_id,
             schema,
             llm_config,
+            padding,
         )
 
         # Add schema to payload if provided
@@ -1478,6 +1488,7 @@ class AsyncMorphik:
         min_score: float = 0.0,
         use_colpali: bool = True,
         folder_name: Optional[Union[str, List[str]]] = None,
+        padding: int = 0,
     ) -> List[FinalChunkResult]:
         """
         Search for relevant chunks.
@@ -1489,6 +1500,7 @@ class AsyncMorphik:
             min_score: Minimum similarity threshold (default: 0.0)
             use_colpali: Whether to use ColPali-style embedding model to retrieve chunks
                 (only works for documents ingested with `use_colpali=True`)
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
         Returns:
             List[FinalChunkResult]
 
@@ -1496,13 +1508,14 @@ class AsyncMorphik:
             ```python
             chunks = await db.retrieve_chunks(
                 "What are the key findings?",
-                filters={"department": "research"}
+                filters={"department": "research"},
+                padding=2  # Get 2 pages before and after each matched page
             )
             ```
         """
         effective_folder = folder_name if folder_name is not None else None
         payload = self._logic._prepare_retrieve_chunks_request(
-            query, filters, k, min_score, use_colpali, effective_folder, None
+            query, filters, k, min_score, use_colpali, effective_folder, None, padding
         )
         response = await self._request("POST", "retrieve/chunks", data=payload)
         return self._logic._parse_chunk_result_list_response(response)
@@ -1561,6 +1574,7 @@ class AsyncMorphik:
         chat_id: Optional[str] = None,
         schema: Optional[Union[Type[BaseModel], Dict[str, Any]]] = None,
         llm_config: Optional[Dict[str, Any]] = None,
+        padding: int = 0,
     ) -> CompletionResponse:
         """
         Generate completion using relevant chunks as context.
@@ -1581,6 +1595,7 @@ class AsyncMorphik:
                 Either a QueryPromptOverrides object or a dictionary with the same structure
             schema: Optional schema for structured output, can be a Pydantic model or a JSON schema dict
             llm_config: Optional LiteLLM-compatible model configuration (e.g., model name, API key, base URL)
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
         Returns:
             CompletionResponse
 
@@ -1669,6 +1684,7 @@ class AsyncMorphik:
             chat_id,
             schema,
             llm_config,
+            padding,
         )
 
         # Add schema to payload if provided
@@ -2575,7 +2591,7 @@ class AsyncMorphik:
         self,
         graph_name: str,
         timeout_seconds: int = 300,
-        check_interval_seconds: int = 5,
+        check_interval_seconds: int = 2,
     ) -> Graph:
         """Block until the specified graph finishes processing (async).
 
@@ -2662,6 +2678,30 @@ class AsyncMorphik:
         params = {"run_id": run_id} if run_id else None
         return await self._request("GET", f"graph/workflow/{workflow_id}/status", params=params)
 
+    async def get_graph_status(
+        self, graph_name: str, folder_name: Optional[str] = None, end_user_id: Optional[str] = None
+    ) -> Dict[str, Any]:
+        """Get the current status of a graph with pipeline stage information.
+
+        This is a lightweight endpoint that checks local database status and
+        optionally syncs with external workflow status if the graph is processing.
+
+        Args:
+            graph_name: Name of the graph to check
+            folder_name: Optional folder name for scoping
+            end_user_id: Optional end user ID for scoping
+
+        Returns:
+            Dict containing status, pipeline_stage (if processing), and other metadata
+        """
+        params = {}
+        if folder_name:
+            params["folder_name"] = folder_name
+        if end_user_id:
+            params["end_user_id"] = end_user_id
+
+        return await self._request("GET", f"graph/{graph_name}/status", params=params if params else None)
+
     # ------------------------------------------------------------------
     # Document download helpers ----------------------------------------
     # ------------------------------------------------------------------

{morphik-0.2.4 → morphik-0.2.6}/morphik/models.py

@@ -317,7 +317,7 @@ class Graph(BaseModel):
     def error(self) -> str | None:
         return self.system_metadata.get("error") if self.system_metadata else None
 
-    def wait_for_completion(self, timeout_seconds: int = 300, check_interval_seconds: int = 5) -> "Graph":
+    def wait_for_completion(self, timeout_seconds: int = 300, check_interval_seconds: int = 2) -> "Graph":
         """Poll the server until the graph processing is finished."""
         import time
 

{morphik-0.2.4 → morphik-0.2.6}/morphik/sync.py

@@ -289,6 +289,7 @@ class Folder:
         min_score: float = 0.0,
         use_colpali: bool = True,
         additional_folders: Optional[List[str]] = None,
+        padding: int = 0,
     ) -> List[FinalChunkResult]:
         """
         Retrieve relevant chunks within this folder.
@@ -300,21 +301,16 @@ class Folder:
             min_score: Minimum similarity threshold (default: 0.0)
             use_colpali: Whether to use ColPali-style embedding model
             additional_folders: Optional list of extra folders to include in the scope
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
 
         Returns:
             List[FinalChunkResult]: List of relevant chunks
         """
         effective_folder = self._merge_folders(additional_folders)
-        request = {
-            "query": query,
-            "filters": filters,
-            "k": k,
-            "min_score": min_score,
-            "use_colpali": use_colpali,
-            "folder_name": effective_folder,
-        }
-
-        response = self._client._request("POST", "retrieve/chunks", request)
+        payload = self._client._logic._prepare_retrieve_chunks_request(
+            query, filters, k, min_score, use_colpali, effective_folder, None, padding
+        )
+        response = self._client._request("POST", "retrieve/chunks", payload)
         return self._client._logic._parse_chunk_result_list_response(response)
 
     def retrieve_docs(
@@ -370,6 +366,7 @@ class Folder:
         schema: Optional[Union[Type[BaseModel], Dict[str, Any]]] = None,
         chat_id: Optional[str] = None,
         llm_config: Optional[Dict[str, Any]] = None,
+        padding: int = 0,
     ) -> CompletionResponse:
         """
         Generate completion using relevant chunks as context within this folder.
@@ -388,6 +385,7 @@ class Folder:
             prompt_overrides: Optional customizations for entity extraction, resolution, and query prompts
             additional_folders: Optional list of extra folders to include in the scope
             schema: Optional schema for structured output
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
 
         Returns:
             CompletionResponse: Generated completion
@@ -410,6 +408,7 @@ class Folder:
             chat_id,
             schema,
             llm_config,
+            padding,
         )
 
         # Add schema to payload if provided
@@ -864,6 +863,7 @@ class UserScope:
         min_score: float = 0.0,
         use_colpali: bool = True,
         additional_folders: Optional[List[str]] = None,
+        padding: int = 0,
     ) -> List[FinalChunkResult]:
         """
         Retrieve relevant chunks as this end user.
@@ -875,26 +875,16 @@ class UserScope:
             min_score: Minimum similarity threshold (default: 0.0)
             use_colpali: Whether to use ColPali-style embedding model
             additional_folders: Optional list of extra folders to include in the scope
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
 
         Returns:
             List[FinalChunkResult]: List of relevant chunks
         """
         effective_folder = self._merge_folders(additional_folders)
-        request = {
-            "query": query,
-            "filters": filters,
-            "k": k,
-            "min_score": min_score,
-            "use_colpali": use_colpali,
-            "end_user_id": self._end_user_id,  # Add end user ID here
-            "folder_name": effective_folder,  # Add folder name if provided
-        }
-
-        # Add folder name if scoped to a folder
-        if self._folder_name:
-            request["folder_name"] = self._folder_name
-
-        response = self._client._request("POST", "retrieve/chunks", request)
+        payload = self._client._logic._prepare_retrieve_chunks_request(
+            query, filters, k, min_score, use_colpali, effective_folder, self._end_user_id, padding
+        )
+        response = self._client._request("POST", "retrieve/chunks", payload)
         return self._client._logic._parse_chunk_result_list_response(response)
 
     def retrieve_docs(
@@ -955,6 +945,7 @@ class UserScope:
         schema: Optional[Union[Type[BaseModel], Dict[str, Any]]] = None,
         chat_id: Optional[str] = None,
         llm_config: Optional[Dict[str, Any]] = None,
+        padding: int = 0,
     ) -> CompletionResponse:
         """
         Generate completion using relevant chunks as context as this end user.
@@ -973,6 +964,7 @@ class UserScope:
             prompt_overrides: Optional customizations for entity extraction, resolution, and query prompts
             additional_folders: Optional list of extra folders to include in the scope
             schema: Optional schema for structured output
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
 
         Returns:
             CompletionResponse: Generated completion
@@ -995,6 +987,7 @@ class UserScope:
             chat_id,
             schema,
             llm_config,
+            padding,
         )
 
         # Add schema to payload if provided
@@ -1623,6 +1616,7 @@ class Morphik:
         min_score: float = 0.0,
         use_colpali: bool = True,
         folder_name: Optional[Union[str, List[str]]] = None,
+        padding: int = 0,
     ) -> List[FinalChunkResult]:
         """
         Retrieve relevant chunks.
@@ -1634,6 +1628,7 @@ class Morphik:
             min_score: Minimum similarity threshold (default: 0.0)
             use_colpali: Whether to use ColPali-style embedding model to retrieve the chunks
                 (only works for documents ingested with `use_colpali=True`)
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
         Returns:
             List[ChunkResult]
 
@@ -1646,7 +1641,7 @@ class Morphik:
             ```
         """
         payload = self._logic._prepare_retrieve_chunks_request(
-            query, filters, k, min_score, use_colpali, folder_name, None
+            query, filters, k, min_score, use_colpali, folder_name, None, padding
         )
         response = self._request("POST", "retrieve/chunks", data=payload)
         return self._logic._parse_chunk_result_list_response(response)
@@ -1704,6 +1699,7 @@ class Morphik:
         chat_id: Optional[str] = None,
         schema: Optional[Union[Type[BaseModel], Dict[str, Any]]] = None,
         llm_config: Optional[Dict[str, Any]] = None,
+        padding: int = 0,
     ) -> CompletionResponse:
         """
         Generate completion using relevant chunks as context.
@@ -1725,6 +1721,7 @@ class Morphik:
             folder_name: Optional folder name to further scope operations
             schema: Optional schema for structured output, can be a Pydantic model or a JSON schema dict
             llm_config: Optional LiteLLM-compatible model configuration (e.g., model name, API key, base URL)
+            padding: Number of additional chunks/pages to retrieve before and after matched chunks (ColPali only, default: 0)
         Returns:
             CompletionResponse
 
@@ -1813,6 +1810,7 @@ class Morphik:
             chat_id,
             schema,
             llm_config,
+            padding,
         )
 
         # Add schema to payload if provided
@@ -2748,7 +2746,7 @@ class Morphik:
         self,
         graph_name: str,
         timeout_seconds: int = 300,
-        check_interval_seconds: int = 5,
+        check_interval_seconds: int = 2,
     ) -> Graph:
         """Block until the specified graph finishes processing.
 
@@ -2852,10 +2850,34 @@ class Morphik:
         return self._request("GET", f"graph/{name}/visualization", params=params)
 
     def check_workflow_status(self, workflow_id: str, run_id: Optional[str] = None) -> Dict[str, Any]:
-        """Poll the status of an asynchronous graph build/update workflow."""
+        """Poll the status of an async graph build/update workflow."""
         params = {"run_id": run_id} if run_id else None
         return self._request("GET", f"graph/workflow/{workflow_id}/status", params=params)
 
+    def get_graph_status(
+        self, graph_name: str, folder_name: Optional[str] = None, end_user_id: Optional[str] = None
+    ) -> Dict[str, Any]:
+        """Get the current status of a graph with pipeline stage information.
+
+        This is a lightweight endpoint that checks local database status and
+        optionally syncs with external workflow status if the graph is processing.
+
+        Args:
+            graph_name: Name of the graph to check
+            folder_name: Optional folder name for scoping
+            end_user_id: Optional end user ID for scoping
+
+        Returns:
+            Dict containing status, pipeline_stage (if processing), and other metadata
+        """
+        params = {}
+        if folder_name:
+            params["folder_name"] = folder_name
+        if end_user_id:
+            params["end_user_id"] = end_user_id
+
+        return self._request("GET", f"graph/{graph_name}/status", params=params if params else None)
+
     # ------------------------------------------------------------------
     # Document download helpers ----------------------------------------
     # ------------------------------------------------------------------

{morphik-0.2.4 → morphik-0.2.6}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "morphik"
-version = "0.2.4"
+version = "0.2.6"
 authors = [
     { name = "Morphik", email = "founders@morphik.ai" },
 ]