Regenerate SDK client core from Otari OpenAPI spec

sdk-endpoints.txt

@@ -119,6 +119,7 @@ GET /v1/usage/series                            # not yet wrapped
 DELETE /v1/usage                                # not yet wrapped
 POST /v1/usage/external-events                  # not yet wrapped
 POST /v1/usage/set-price                        # not yet wrapped
+GET /v1/usage/in-flight                         # dashboard-only, per-worker live view
 # Routing policies
 GET /v1/routing/policies                        # not yet wrapped
 POST /v1/routing/policies                       # not yet wrapped

src/otari/_client/__init__.py

@@ -148,6 +148,8 @@
     "IMGUsageOutputTokensDetails",
     "ImageGenerationRequest",
     "ImagesResponse",
+    "InFlightEntry",
+    "InFlightResponse",
     "Input",
     "Input1",
     "KeyInfo",
@@ -505,6 +507,8 @@
 from otari._client.models.img_usage_output_tokens_details import IMGUsageOutputTokensDetails as IMGUsageOutputTokensDetails
 from otari._client.models.image_generation_request import ImageGenerationRequest as ImageGenerationRequest
 from otari._client.models.images_response import ImagesResponse as ImagesResponse
+from otari._client.models.in_flight_entry import InFlightEntry as InFlightEntry
+from otari._client.models.in_flight_response import InFlightResponse as InFlightResponse
 from otari._client.models.input import Input as Input
 from otari._client.models.input1 import Input1 as Input1
 from otari._client.models.key_info import KeyInfo as KeyInfo

src/otari/_client/api/routing_api.py

@@ -870,7 +870,7 @@ def rank_candidates_v1_routing_preferences_rank_post(
     ) -> RankResponse:
         """Rank Candidates
 
-        Record scored examples: one routing-memory record each, plus an audit row.  The routing-memory record is written before its audit row for each example, because it is the load-bearing one (the router votes over it) and embedding it can fail; writing the audit row only afterwards means a failed embedding never leaves an orphan audit row.  A failed embedding is a 502 that names the model, not a 500. Every example in the batch is embedded, so this is the call an operator makes most often and the one most likely to meet a misconfigured ``router_embedding_model``.  Score keys are stored in the spelling the policy uses (see :func:`_validated_scores`), because the router matches them against its candidate selectors by exact string.
+        Record scored examples: one routing-memory record each, plus an audit row.  The routing-memory record is written before its audit row for each example, because it is the load-bearing one (the router votes over it) and embedding it can fail; writing the audit row only afterwards means a failed embedding never leaves an orphan audit row.  A failed embedding is a 502 that names the model, not a 500. Every example in the batch is embedded, so this is the call an operator makes most often and the one most likely to meet a misconfigured ``router_embedding_model``.  Score keys are stored canonically as ``instance:model`` (see :func:`_validated_scores`), which is the form the router canonicalizes its candidates to, so how a policy spells a candidate cannot decide whether it matches.
 
         :param rank_request: (required)
         :type rank_request: RankRequest
@@ -938,7 +938,7 @@ def rank_candidates_v1_routing_preferences_rank_post_with_http_info(
     ) -> ApiResponse[RankResponse]:
         """Rank Candidates
 
-        Record scored examples: one routing-memory record each, plus an audit row.  The routing-memory record is written before its audit row for each example, because it is the load-bearing one (the router votes over it) and embedding it can fail; writing the audit row only afterwards means a failed embedding never leaves an orphan audit row.  A failed embedding is a 502 that names the model, not a 500. Every example in the batch is embedded, so this is the call an operator makes most often and the one most likely to meet a misconfigured ``router_embedding_model``.  Score keys are stored in the spelling the policy uses (see :func:`_validated_scores`), because the router matches them against its candidate selectors by exact string.
+        Record scored examples: one routing-memory record each, plus an audit row.  The routing-memory record is written before its audit row for each example, because it is the load-bearing one (the router votes over it) and embedding it can fail; writing the audit row only afterwards means a failed embedding never leaves an orphan audit row.  A failed embedding is a 502 that names the model, not a 500. Every example in the batch is embedded, so this is the call an operator makes most often and the one most likely to meet a misconfigured ``router_embedding_model``.  Score keys are stored canonically as ``instance:model`` (see :func:`_validated_scores`), which is the form the router canonicalizes its candidates to, so how a policy spells a candidate cannot decide whether it matches.
 
         :param rank_request: (required)
         :type rank_request: RankRequest
@@ -1006,7 +1006,7 @@ def rank_candidates_v1_routing_preferences_rank_post_without_preload_content(
     ) -> RESTResponseType:
         """Rank Candidates
 
-        Record scored examples: one routing-memory record each, plus an audit row.  The routing-memory record is written before its audit row for each example, because it is the load-bearing one (the router votes over it) and embedding it can fail; writing the audit row only afterwards means a failed embedding never leaves an orphan audit row.  A failed embedding is a 502 that names the model, not a 500. Every example in the batch is embedded, so this is the call an operator makes most often and the one most likely to meet a misconfigured ``router_embedding_model``.  Score keys are stored in the spelling the policy uses (see :func:`_validated_scores`), because the router matches them against its candidate selectors by exact string.
+        Record scored examples: one routing-memory record each, plus an audit row.  The routing-memory record is written before its audit row for each example, because it is the load-bearing one (the router votes over it) and embedding it can fail; writing the audit row only afterwards means a failed embedding never leaves an orphan audit row.  A failed embedding is a 502 that names the model, not a 500. Every example in the batch is embedded, so this is the call an operator makes most often and the one most likely to meet a misconfigured ``router_embedding_model``.  Score keys are stored canonically as ``instance:model`` (see :func:`_validated_scores`), which is the form the router canonicalizes its candidates to, so how a policy spells a candidate cannot decide whether it matches.
 
         :param rank_request: (required)
         :type rank_request: RankRequest

src/otari/_client/api/usage_api.py

@@ -21,6 +21,7 @@
 from typing_extensions import Annotated
 from otari._client.models.external_events_request import ExternalEventsRequest
 from otari._client.models.external_ingest_result import ExternalIngestResult
+from otari._client.models.in_flight_response import InFlightResponse
 from otari._client.models.usage_count import UsageCount
 from otari._client.models.usage_delete_request import UsageDeleteRequest
 from otari._client.models.usage_delete_result import UsageDeleteResult
@@ -1131,6 +1132,253 @@ def _ingest_external_usage_v1_usage_external_events_post_serialize(
 
 
 
+    @validate_call
+    def list_in_flight_v1_usage_in_flight_get(
+        self,
+        _request_timeout: Union[
+            None,
+            Annotated[StrictFloat, Field(gt=0)],
+            Tuple[
+                Annotated[StrictFloat, Field(gt=0)],
+                Annotated[StrictFloat, Field(gt=0)]
+            ]
+        ] = None,
+        _request_auth: Optional[Dict[StrictStr, Any]] = None,
+        _content_type: Optional[StrictStr] = None,
+        _headers: Optional[Dict[StrictStr, Any]] = None,
+        _host_index: Annotated[StrictInt, Field(ge=0, le=0)] = 0,
+    ) -> InFlightResponse:
+        """List In Flight
+
+        Requests the gateway is currently serving, longest-running first.  A usage row is written when a request settles, so the log alone cannot answer \"is anything happening right now\": on a slow backend, a 30-second local model call is invisible until it finishes. This reports what is in progress.  Read from an in-memory registry, so it describes the process that answers this call and not the deployment: behind a load balancer, consecutive polls reach different otari processes, and there is no deployment-wide total to ask for. ``total`` is the true in-flight count for the answering process even when ``requests`` is capped.
+
+        :param _request_timeout: timeout setting for this request. If one
+                                 number provided, it will be total request
+                                 timeout. It can also be a pair (tuple) of
+                                 (connection, read) timeouts.
+        :type _request_timeout: int, tuple(int, int), optional
+        :param _request_auth: set to override the auth_settings for an a single
+                              request; this effectively ignores the
+                              authentication in the spec for a single request.
+        :type _request_auth: dict, optional
+        :param _content_type: force content-type for the request.
+        :type _content_type: str, Optional
+        :param _headers: set to override the headers for a single
+                         request; this effectively ignores the headers
+                         in the spec for a single request.
+        :type _headers: dict, optional
+        :param _host_index: set to override the host_index for a single
+                            request; this effectively ignores the host_index
+                            in the spec for a single request.
+        :type _host_index: int, optional
+        :return: Returns the result object.
+        """ # noqa: E501
+
+        _param = self._list_in_flight_v1_usage_in_flight_get_serialize(
+            _request_auth=_request_auth,
+            _content_type=_content_type,
+            _headers=_headers,
+            _host_index=_host_index
+        )
+
+        _response_types_map: Dict[str, Optional[str]] = {
+            '200': "InFlightResponse",
+        }
+        response_data = self.api_client.call_api(
+            *_param,
+            _request_timeout=_request_timeout
+        )
+        response_data.read()
+        return self.api_client.response_deserialize(
+            response_data=response_data,
+            response_types_map=_response_types_map,
+        ).data
+
+
+    @validate_call
+    def list_in_flight_v1_usage_in_flight_get_with_http_info(
+        self,
+        _request_timeout: Union[
+            None,
+            Annotated[StrictFloat, Field(gt=0)],
+            Tuple[
+                Annotated[StrictFloat, Field(gt=0)],
+                Annotated[StrictFloat, Field(gt=0)]
+            ]
+        ] = None,
+        _request_auth: Optional[Dict[StrictStr, Any]] = None,
+        _content_type: Optional[StrictStr] = None,
+        _headers: Optional[Dict[StrictStr, Any]] = None,
+        _host_index: Annotated[StrictInt, Field(ge=0, le=0)] = 0,
+    ) -> ApiResponse[InFlightResponse]:
+        """List In Flight
+
+        Requests the gateway is currently serving, longest-running first.  A usage row is written when a request settles, so the log alone cannot answer \"is anything happening right now\": on a slow backend, a 30-second local model call is invisible until it finishes. This reports what is in progress.  Read from an in-memory registry, so it describes the process that answers this call and not the deployment: behind a load balancer, consecutive polls reach different otari processes, and there is no deployment-wide total to ask for. ``total`` is the true in-flight count for the answering process even when ``requests`` is capped.
+
+        :param _request_timeout: timeout setting for this request. If one
+                                 number provided, it will be total request
+                                 timeout. It can also be a pair (tuple) of
+                                 (connection, read) timeouts.
+        :type _request_timeout: int, tuple(int, int), optional
+        :param _request_auth: set to override the auth_settings for an a single
+                              request; this effectively ignores the
+                              authentication in the spec for a single request.
+        :type _request_auth: dict, optional
+        :param _content_type: force content-type for the request.
+        :type _content_type: str, Optional
+        :param _headers: set to override the headers for a single
+                         request; this effectively ignores the headers
+                         in the spec for a single request.
+        :type _headers: dict, optional
+        :param _host_index: set to override the host_index for a single
+                            request; this effectively ignores the host_index
+                            in the spec for a single request.
+        :type _host_index: int, optional
+        :return: Returns the result object.
+        """ # noqa: E501
+
+        _param = self._list_in_flight_v1_usage_in_flight_get_serialize(
+            _request_auth=_request_auth,
+            _content_type=_content_type,
+            _headers=_headers,
+            _host_index=_host_index
+        )
+
+        _response_types_map: Dict[str, Optional[str]] = {
+            '200': "InFlightResponse",
+        }
+        response_data = self.api_client.call_api(
+            *_param,
+            _request_timeout=_request_timeout
+        )
+        response_data.read()
+        return self.api_client.response_deserialize(
+            response_data=response_data,
+            response_types_map=_response_types_map,
+        )
+
+
+    @validate_call
+    def list_in_flight_v1_usage_in_flight_get_without_preload_content(
+        self,
+        _request_timeout: Union[
+            None,
+            Annotated[StrictFloat, Field(gt=0)],
+            Tuple[
+                Annotated[StrictFloat, Field(gt=0)],
+                Annotated[StrictFloat, Field(gt=0)]
+            ]
+        ] = None,
+        _request_auth: Optional[Dict[StrictStr, Any]] = None,
+        _content_type: Optional[StrictStr] = None,
+        _headers: Optional[Dict[StrictStr, Any]] = None,
+        _host_index: Annotated[StrictInt, Field(ge=0, le=0)] = 0,
+    ) -> RESTResponseType:
+        """List In Flight
+
+        Requests the gateway is currently serving, longest-running first.  A usage row is written when a request settles, so the log alone cannot answer \"is anything happening right now\": on a slow backend, a 30-second local model call is invisible until it finishes. This reports what is in progress.  Read from an in-memory registry, so it describes the process that answers this call and not the deployment: behind a load balancer, consecutive polls reach different otari processes, and there is no deployment-wide total to ask for. ``total`` is the true in-flight count for the answering process even when ``requests`` is capped.
+
+        :param _request_timeout: timeout setting for this request. If one
+                                 number provided, it will be total request
+                                 timeout. It can also be a pair (tuple) of
+                                 (connection, read) timeouts.
+        :type _request_timeout: int, tuple(int, int), optional
+        :param _request_auth: set to override the auth_settings for an a single
+                              request; this effectively ignores the
+                              authentication in the spec for a single request.
+        :type _request_auth: dict, optional
+        :param _content_type: force content-type for the request.
+        :type _content_type: str, Optional
+        :param _headers: set to override the headers for a single
+                         request; this effectively ignores the headers
+                         in the spec for a single request.
+        :type _headers: dict, optional
+        :param _host_index: set to override the host_index for a single
+                            request; this effectively ignores the host_index
+                            in the spec for a single request.
+        :type _host_index: int, optional
+        :return: Returns the result object.
+        """ # noqa: E501
+
+        _param = self._list_in_flight_v1_usage_in_flight_get_serialize(
+            _request_auth=_request_auth,
+            _content_type=_content_type,
+            _headers=_headers,
+            _host_index=_host_index
+        )
+
+        _response_types_map: Dict[str, Optional[str]] = {
+            '200': "InFlightResponse",
+        }
+        response_data = self.api_client.call_api(
+            *_param,
+            _request_timeout=_request_timeout
+        )
+        return response_data.response
+
+
+    def _list_in_flight_v1_usage_in_flight_get_serialize(
+        self,
+        _request_auth,
+        _content_type,
+        _headers,
+        _host_index,
+    ) -> RequestSerialized:
+
+        _host = None
+
+        _collection_formats: Dict[str, str] = {
+        }
+
+        _path_params: Dict[str, str] = {}
+        _query_params: List[Tuple[str, str]] = []
+        _header_params: Dict[str, Optional[str]] = _headers or {}
+        _form_params: List[Tuple[str, str]] = []
+        _files: Dict[
+            str, Union[str, bytes, List[str], List[bytes], List[Tuple[str, bytes]]]
+        ] = {}
+        _body_params: Optional[bytes] = None
+
+        # process the path parameters
+        # process the query parameters
+        # process the header parameters
+        # process the form parameters
+        # process the body parameter
+
+
+        # set the HTTP header `Accept`
+        if 'Accept' not in _header_params:
+            _header_params['Accept'] = self.api_client.select_header_accept(
+                [
+                    'application/json'
+                ]
+            )
+
+
+        # authentication setting
+        _auth_settings: List[str] = [
+            'XApiKeyAuth', 
+            'ApiKeyAuth'
+        ]
+
+        return self.api_client.param_serialize(
+            method='GET',
+            resource_path='/v1/usage/in-flight',
+            path_params=_path_params,
+            query_params=_query_params,
+            header_params=_header_params,
+            body=_body_params,
+            post_params=_form_params,
+            files=_files,
+            auth_settings=_auth_settings,
+            collection_formats=_collection_formats,
+            _host=_host,
+            _request_auth=_request_auth
+        )
+
+
+
+
     @validate_call
     def list_usage_v1_usage_get(
         self,

src/otari/_client/models/__init__.py

@@ -108,6 +108,8 @@
 from otari._client.models.img_usage_output_tokens_details import IMGUsageOutputTokensDetails
 from otari._client.models.image_generation_request import ImageGenerationRequest
 from otari._client.models.images_response import ImagesResponse
+from otari._client.models.in_flight_entry import InFlightEntry
+from otari._client.models.in_flight_response import InFlightResponse
 from otari._client.models.input import Input
 from otari._client.models.input1 import Input1
 from otari._client.models.key_info import KeyInfo