usecortex-ai 0.4.0__py3-none-any.whl → 0.5.1__py3-none-any.whl

usecortex_ai/search/client.py

@@ -38,31 +38,18 @@ class SearchClient:
         tenant_id: str,
         query: str,
         sub_tenant_id: typing.Optional[str] = OMIT,
-        max_chunks: typing.Optional[int] = OMIT,
+        max_results: typing.Optional[int] = OMIT,
         mode: typing.Optional[RetrieveMode] = OMIT,
         alpha: typing.Optional[Alpha] = OMIT,
         recency_bias: typing.Optional[float] = OMIT,
-        num_related_chunks: typing.Optional[int] = OMIT,
-        personalise_search: typing.Optional[bool] = OMIT,
+        personalise_recall: typing.Optional[bool] = OMIT,
         graph_context: typing.Optional[bool] = OMIT,
-        extra_context: typing.Optional[str] = OMIT,
+        additional_context: typing.Optional[str] = OMIT,
         search_mode: typing.Optional[SearchMode] = OMIT,
+        metadata_filters: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> RetrievalResult:
         """
-        Search for relevant content within your indexed sources or user memories.
-
-            Results are ranked by relevance and can be customized with parameters like
-            result limits, alpha weighting, and recency preferences.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories (uses inferred content)
-
-            Use `mode` to control retrieval quality:
-            - "fast" (default): Single query, faster response
-            - "accurate": Multi-query generation with reranking, higher quality
-
         Parameters
         ----------
         tenant_id : str
@@ -74,7 +61,7 @@ class SearchClient:
         sub_tenant_id : typing.Optional[str]
             Optional sub-tenant identifier used to organize data within a tenant. If omitted, the default sub-tenant created during tenant setup will be used.
 
-        max_chunks : typing.Optional[int]
+        max_results : typing.Optional[int]
             Maximum number of results to return
 
         mode : typing.Optional[RetrieveMode]
@@ -86,21 +73,21 @@ class SearchClient:
         recency_bias : typing.Optional[float]
             Preference for newer content (0.0 = no bias, 1.0 =             strong recency preference)
 
-        num_related_chunks : typing.Optional[int]
-            Number of related content chunks to include
-
-        personalise_search : typing.Optional[bool]
+        personalise_recall : typing.Optional[bool]
             Enable personalized search results based on user preferences
 
         graph_context : typing.Optional[bool]
             Enable graph context for search results
 
-        extra_context : typing.Optional[str]
+        additional_context : typing.Optional[str]
             Additional context provided by the user to guide retrieval
 
         search_mode : typing.Optional[SearchMode]
             What to search: 'sources' for documents or 'memories' for user memories
 
+        metadata_filters : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
+            Optional key-value pairs to filter search results by tenant metadata fields. Keys must match fields defined in tenant_metadata_schema during tenant creation. Supports exact match filtering on indexed metadata fields. Example: {'category': 'engineering', 'priority': 'high'}
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -114,21 +101,21 @@ class SearchClient:
         from usecortex-ai import CortexAI
 
         client = CortexAI(token="YOUR_TOKEN", )
-        client.search.retrieve(tenant_id='tenant_id', query='query', )
+        client.search.retrieve(tenant_id='tenant_1234', query='Which mode does user prefer', )
         """
         _response = self._raw_client.retrieve(
             tenant_id=tenant_id,
             query=query,
             sub_tenant_id=sub_tenant_id,
-            max_chunks=max_chunks,
+            max_results=max_results,
             mode=mode,
             alpha=alpha,
             recency_bias=recency_bias,
-            num_related_chunks=num_related_chunks,
-            personalise_search=personalise_search,
+            personalise_recall=personalise_recall,
             graph_context=graph_context,
-            extra_context=extra_context,
+            additional_context=additional_context,
             search_mode=search_mode,
+            metadata_filters=metadata_filters,
             request_options=request_options,
         )
         return _response.data
@@ -140,19 +127,11 @@ class SearchClient:
         query: str,
         sub_tenant_id: typing.Optional[str] = OMIT,
         operator: typing.Optional[Bm25OperatorType] = OMIT,
-        max_chunks: typing.Optional[int] = OMIT,
+        max_results: typing.Optional[int] = OMIT,
         search_mode: typing.Optional[SearchMode] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> RetrievalResult:
         """
-        Perform full text search for exact matches within your indexed sources or memories.
-            Choose between 'OR' and 'AND' operators to control how search terms are combined
-            for precise text matching.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories
-
         Parameters
         ----------
         tenant_id : str
@@ -167,7 +146,7 @@ class SearchClient:
         operator : typing.Optional[Bm25OperatorType]
             How to combine search terms (OR or AND)
 
-        max_chunks : typing.Optional[int]
+        max_results : typing.Optional[int]
             Maximum number of results to return
 
         search_mode : typing.Optional[SearchMode]
@@ -186,14 +165,14 @@ class SearchClient:
         from usecortex-ai import CortexAI
 
         client = CortexAI(token="YOUR_TOKEN", )
-        client.search.full_text_search(tenant_id='tenant_id', query='query', )
+        client.search.full_text_search(tenant_id='tenant_1234', query='John Smith Jake', )
         """
         _response = self._raw_client.full_text_search(
             tenant_id=tenant_id,
             query=query,
             sub_tenant_id=sub_tenant_id,
             operator=operator,
-            max_chunks=max_chunks,
+            max_results=max_results,
             search_mode=search_mode,
             request_options=request_options,
         )
@@ -218,19 +197,6 @@ class SearchClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> QnASearchResponse:
         """
-        Ask a question and get an AI-generated answer based on your indexed sources or memories.
-
-            The response includes both the AI answer and the source chunks used to generate it,
-            enabling full transparency and citation capabilities.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories
-
-            Use `mode` to control retrieval quality:
-            - "fast" (default): Single query, faster response
-            - "accurate": Multi-query generation with reranking, higher quality
-
         Parameters
         ----------
         tenant_id : str
@@ -285,7 +251,7 @@ class SearchClient:
         from usecortex-ai import CortexAI
 
         client = CortexAI(token="YOUR_TOKEN", )
-        client.search.qna(tenant_id='tenant_id', question='question', )
+        client.search.qna(tenant_id='tenant_1234', question='What is Cortex AI', )
         """
         _response = self._raw_client.qna(
             tenant_id=tenant_id,
@@ -327,31 +293,18 @@ class AsyncSearchClient:
         tenant_id: str,
         query: str,
         sub_tenant_id: typing.Optional[str] = OMIT,
-        max_chunks: typing.Optional[int] = OMIT,
+        max_results: typing.Optional[int] = OMIT,
         mode: typing.Optional[RetrieveMode] = OMIT,
         alpha: typing.Optional[Alpha] = OMIT,
         recency_bias: typing.Optional[float] = OMIT,
-        num_related_chunks: typing.Optional[int] = OMIT,
-        personalise_search: typing.Optional[bool] = OMIT,
+        personalise_recall: typing.Optional[bool] = OMIT,
         graph_context: typing.Optional[bool] = OMIT,
-        extra_context: typing.Optional[str] = OMIT,
+        additional_context: typing.Optional[str] = OMIT,
         search_mode: typing.Optional[SearchMode] = OMIT,
+        metadata_filters: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> RetrievalResult:
         """
-        Search for relevant content within your indexed sources or user memories.
-
-            Results are ranked by relevance and can be customized with parameters like
-            result limits, alpha weighting, and recency preferences.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories (uses inferred content)
-
-            Use `mode` to control retrieval quality:
-            - "fast" (default): Single query, faster response
-            - "accurate": Multi-query generation with reranking, higher quality
-
         Parameters
         ----------
         tenant_id : str
@@ -363,7 +316,7 @@ class AsyncSearchClient:
         sub_tenant_id : typing.Optional[str]
             Optional sub-tenant identifier used to organize data within a tenant. If omitted, the default sub-tenant created during tenant setup will be used.
 
-        max_chunks : typing.Optional[int]
+        max_results : typing.Optional[int]
             Maximum number of results to return
 
         mode : typing.Optional[RetrieveMode]
@@ -375,21 +328,21 @@ class AsyncSearchClient:
         recency_bias : typing.Optional[float]
             Preference for newer content (0.0 = no bias, 1.0 =             strong recency preference)
 
-        num_related_chunks : typing.Optional[int]
-            Number of related content chunks to include
-
-        personalise_search : typing.Optional[bool]
+        personalise_recall : typing.Optional[bool]
             Enable personalized search results based on user preferences
 
         graph_context : typing.Optional[bool]
             Enable graph context for search results
 
-        extra_context : typing.Optional[str]
+        additional_context : typing.Optional[str]
             Additional context provided by the user to guide retrieval
 
         search_mode : typing.Optional[SearchMode]
             What to search: 'sources' for documents or 'memories' for user memories
 
+        metadata_filters : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
+            Optional key-value pairs to filter search results by tenant metadata fields. Keys must match fields defined in tenant_metadata_schema during tenant creation. Supports exact match filtering on indexed metadata fields. Example: {'category': 'engineering', 'priority': 'high'}
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -406,22 +359,22 @@ class AsyncSearchClient:
 
         client = AsyncCortexAI(token="YOUR_TOKEN", )
         async def main() -> None:
-            await client.search.retrieve(tenant_id='tenant_id', query='query', )
+            await client.search.retrieve(tenant_id='tenant_1234', query='Which mode does user prefer', )
         asyncio.run(main())
         """
         _response = await self._raw_client.retrieve(
             tenant_id=tenant_id,
             query=query,
             sub_tenant_id=sub_tenant_id,
-            max_chunks=max_chunks,
+            max_results=max_results,
             mode=mode,
             alpha=alpha,
             recency_bias=recency_bias,
-            num_related_chunks=num_related_chunks,
-            personalise_search=personalise_search,
+            personalise_recall=personalise_recall,
             graph_context=graph_context,
-            extra_context=extra_context,
+            additional_context=additional_context,
             search_mode=search_mode,
+            metadata_filters=metadata_filters,
             request_options=request_options,
         )
         return _response.data
@@ -433,19 +386,11 @@ class AsyncSearchClient:
         query: str,
         sub_tenant_id: typing.Optional[str] = OMIT,
         operator: typing.Optional[Bm25OperatorType] = OMIT,
-        max_chunks: typing.Optional[int] = OMIT,
+        max_results: typing.Optional[int] = OMIT,
         search_mode: typing.Optional[SearchMode] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> RetrievalResult:
         """
-        Perform full text search for exact matches within your indexed sources or memories.
-            Choose between 'OR' and 'AND' operators to control how search terms are combined
-            for precise text matching.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories
-
         Parameters
         ----------
         tenant_id : str
@@ -460,7 +405,7 @@ class AsyncSearchClient:
         operator : typing.Optional[Bm25OperatorType]
             How to combine search terms (OR or AND)
 
-        max_chunks : typing.Optional[int]
+        max_results : typing.Optional[int]
             Maximum number of results to return
 
         search_mode : typing.Optional[SearchMode]
@@ -482,7 +427,7 @@ class AsyncSearchClient:
 
         client = AsyncCortexAI(token="YOUR_TOKEN", )
         async def main() -> None:
-            await client.search.full_text_search(tenant_id='tenant_id', query='query', )
+            await client.search.full_text_search(tenant_id='tenant_1234', query='John Smith Jake', )
         asyncio.run(main())
         """
         _response = await self._raw_client.full_text_search(
@@ -490,7 +435,7 @@ class AsyncSearchClient:
             query=query,
             sub_tenant_id=sub_tenant_id,
             operator=operator,
-            max_chunks=max_chunks,
+            max_results=max_results,
             search_mode=search_mode,
             request_options=request_options,
         )
@@ -515,19 +460,6 @@ class AsyncSearchClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> QnASearchResponse:
         """
-        Ask a question and get an AI-generated answer based on your indexed sources or memories.
-
-            The response includes both the AI answer and the source chunks used to generate it,
-            enabling full transparency and citation capabilities.
-
-            Use `search_mode` to specify what to search:
-            - "sources" (default): Search over indexed documents
-            - "memories": Search over user memories
-
-            Use `mode` to control retrieval quality:
-            - "fast" (default): Single query, faster response
-            - "accurate": Multi-query generation with reranking, higher quality
-
         Parameters
         ----------
         tenant_id : str
@@ -585,7 +517,7 @@ class AsyncSearchClient:
 
         client = AsyncCortexAI(token="YOUR_TOKEN", )
         async def main() -> None:
-            await client.search.qna(tenant_id='tenant_id', question='question', )
+            await client.search.qna(tenant_id='tenant_1234', question='What is Cortex AI', )
         asyncio.run(main())
         """
         _response = await self._raw_client.qna(