{ "fullyEncodeReservedExpansion": true, "baseUrl": "https://documentai.googleapis.com/", "parameters": { "prettyPrint": { "default": "true", "location": "query", "type": "boolean", "description": "Returns response with indentations and line breaks." }, "upload_protocol": { "description": "Upload protocol for media (e.g. \"raw\", \"multipart\").", "type": "string", "location": "query" }, "alt": { "type": "string", "enum": [ "json", "media", "proto" ], "enumDescriptions": [ "Responses with Content-Type of application/json", "Media download with context-dependent Content-Type", "Responses with Content-Type of application/x-protobuf" ], "location": "query", "description": "Data format for response.", "default": "json" }, "access_token": { "type": "string", "location": "query", "description": "OAuth access token." }, "uploadType": { "description": "Legacy upload protocol for media (e.g. \"media\", \"multipart\").", "location": "query", "type": "string" }, "key": { "location": "query", "type": "string", "description": "API key. Your API key identifies your project and provides you with API access, quota, and reports. Required unless you provide an OAuth 2.0 token." }, "oauth_token": { "type": "string", "description": "OAuth 2.0 token for the current user.", "location": "query" }, "quotaUser": { "description": "Available to use for quota purposes for server-side applications. Can be any arbitrary string assigned to a user, but should not exceed 40 characters.", "type": "string", "location": "query" }, "$.xgafv": { "enumDescriptions": [ "v1 error format", "v2 error format" ], "description": "V1 error format.", "type": "string", "enum": [ "1", "2" ], "location": "query" }, "fields": { "description": "Selector specifying which fields to include in a partial response.", "type": "string", "location": "query" }, "callback": { "location": "query", "description": "JSONP", "type": "string" } }, "auth": { "oauth2": { "scopes": { "https://www.googleapis.com/auth/cloud-platform": { "description": "See, edit, configure, and delete your Google Cloud data and see the email address for your Google Account." } } } }, "title": "Cloud Document AI API", "basePath": "", "protocol": "rest", "kind": "discovery#restDescription", "id": "documentai:v1beta3", "resources": { "projects": { "resources": { "locations": { "methods": { "list": { "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameterOrder": [ "name" ], "parameters": { "extraLocationTypes": { "repeated": true, "type": "string", "description": "Optional. Do not use this field unless explicitly documented otherwise. This is primarily for internal usage.", "location": "query" }, "filter": { "description": "A filter to narrow down results to a preferred subset. The filtering language accepts strings like `\"displayName=tokyo\"`, and is documented in more detail in [AIP-160](https://google.aip.dev/160).", "type": "string", "location": "query" }, "name": { "required": true, "location": "path", "description": "The resource that owns the locations collection, if applicable.", "pattern": "^projects/[^/]+$", "type": "string" }, "pageToken": { "location": "query", "type": "string", "description": "A page token received from the `next_page_token` field in the response. Send that page token to receive the subsequent page." }, "pageSize": { "format": "int32", "type": "integer", "location": "query", "description": "The maximum number of results to return. If not set, the service selects a default." } }, "description": "Lists information about the supported locations for this service. This method lists locations based on the resource scope provided in the ListLocationsRequest.name field: * **Global locations**: If `name` is empty, the method lists the public locations available to all projects. * **Project-specific locations**: If `name` follows the format `projects/{project}`, the method lists locations visible to that specific project. This includes public, private, or other project-specific locations enabled for the project. For gRPC and client library implementations, the resource name is passed as the `name` field. For direct service calls, the resource name is incorporated into the request path based on the specific service implementation and version.", "path": "v1beta3/{+name}/locations", "flatPath": "v1beta3/projects/{projectsId}/locations", "id": "documentai.projects.locations.list", "response": { "$ref": "GoogleCloudLocationListLocationsResponse" }, "httpMethod": "GET" }, "fetchProcessorTypes": { "parameters": { "parent": { "description": "Required. The location of processor types to list. Format: `projects/{project}/locations/{location}`.", "type": "string", "pattern": "^projects/[^/]+/locations/[^/]+$", "required": true, "location": "path" } }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "httpMethod": "GET", "id": "documentai.projects.locations.fetchProcessorTypes", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}:fetchProcessorTypes", "response": { "$ref": "GoogleCloudDocumentaiV1beta3FetchProcessorTypesResponse" }, "description": "Fetches processor types. Note that we don't use ListProcessorTypes here, because it isn't paginated.", "parameterOrder": [ "parent" ], "path": "v1beta3/{+parent}:fetchProcessorTypes" }, "get": { "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}", "httpMethod": "GET", "parameterOrder": [ "name" ], "response": { "$ref": "GoogleCloudLocationLocation" }, "description": "Gets information about a location.", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameters": { "name": { "description": "Resource name for the location.", "type": "string", "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+$", "required": true } }, "id": "documentai.projects.locations.get", "path": "v1beta3/{+name}" } }, "resources": { "operations": { "methods": { "get": { "response": { "$ref": "GoogleLongrunningOperation" }, "path": "v1beta3/{+name}", "parameters": { "name": { "description": "The name of the operation resource.", "pattern": "^projects/[^/]+/locations/[^/]+/operations/[^/]+$", "type": "string", "required": true, "location": "path" } }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/operations/{operationsId}", "id": "documentai.projects.locations.operations.get", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "description": "Gets the latest state of a long-running operation. Clients can use this method to poll the operation result at intervals as recommended by the API service.", "httpMethod": "GET", "parameterOrder": [ "name" ] }, "cancel": { "parameterOrder": [ "name" ], "httpMethod": "POST", "id": "documentai.projects.locations.operations.cancel", "response": { "$ref": "GoogleProtobufEmpty" }, "path": "v1beta3/{+name}:cancel", "description": "Starts asynchronous cancellation on a long-running operation. The server makes a best effort to cancel the operation, but success is not guaranteed. If the server doesn't support this method, it returns `google.rpc.Code.UNIMPLEMENTED`. Clients can use Operations.GetOperation or other methods to check whether the cancellation succeeded or whether the operation completed despite cancellation. On successful cancellation, the operation is not deleted; instead, it becomes an operation with an Operation.error value with a google.rpc.Status.code of `1`, corresponding to `Code.CANCELLED`.", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameters": { "name": { "type": "string", "description": "The name of the operation resource to be cancelled.", "location": "path", "required": true, "pattern": "^projects/[^/]+/locations/[^/]+/operations/[^/]+$" } }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/operations/{operationsId}:cancel" }, "list": { "response": { "$ref": "GoogleLongrunningListOperationsResponse" }, "parameters": { "returnPartialSuccess": { "location": "query", "description": "When set to `true`, operations that are reachable are returned as normal, and those that are unreachable are returned in the ListOperationsResponse.unreachable field. This can only be `true` when reading across collections. For example, when `parent` is set to `\"projects/example/locations/-\"`. This field is not supported by default and will result in an `UNIMPLEMENTED` error if set unless explicitly documented otherwise in service or product specific documentation.", "type": "boolean" }, "filter": { "location": "query", "type": "string", "description": "The standard list filter." }, "name": { "pattern": "^projects/[^/]+/locations/[^/]+/operations$", "type": "string", "required": true, "location": "path", "description": "The name of the operation's parent resource." }, "pageSize": { "format": "int32", "location": "query", "type": "integer", "description": "The standard list page size." }, "pageToken": { "location": "query", "type": "string", "description": "The standard list page token." } }, "path": "v1beta3/{+name}", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/operations", "description": "Lists operations that match the specified filter in the request. If the server doesn't support this method, it returns `UNIMPLEMENTED`.", "httpMethod": "GET", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameterOrder": [ "name" ], "id": "documentai.projects.locations.operations.list" } } }, "processors": { "methods": { "create": { "description": "Creates a processor from the ProcessorType provided. The processor will be at `ENABLED` state by default after its creation. Note that this method requires the `documentai.processors.create` permission on the project, which is highly privileged. A user or service account with this permission can create new processors that can interact with any gcs bucket in your project.", "id": "documentai.projects.locations.processors.create", "request": { "$ref": "GoogleCloudDocumentaiV1beta3Processor" }, "path": "v1beta3/{+parent}/processors", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameters": { "parent": { "location": "path", "type": "string", "required": true, "pattern": "^projects/[^/]+/locations/[^/]+$", "description": "Required. The parent (project and location) under which to create the processor. Format: `projects/{project}/locations/{location}`" } }, "response": { "$ref": "GoogleCloudDocumentaiV1beta3Processor" }, "parameterOrder": [ "parent" ], "httpMethod": "POST" }, "process": { "parameters": { "name": { "required": true, "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+$", "location": "path", "description": "Required. The resource name of the Processor or ProcessorVersion to use for processing. If a Processor is specified, the server will use its default version. Format: `projects/{project}/locations/{location}/processors/{processor}`, or `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`", "type": "string" } }, "parameterOrder": [ "name" ], "request": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessRequest" }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}:process", "httpMethod": "POST", "id": "documentai.projects.locations.processors.process", "description": "Processes a single document.", "response": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessResponse" }, "path": "v1beta3/{+name}:process" }, "batchProcess": { "path": "v1beta3/{+name}:batchProcess", "parameterOrder": [ "name" ], "id": "documentai.projects.locations.processors.batchProcess", "request": { "$ref": "GoogleCloudDocumentaiV1beta3BatchProcessRequest" }, "httpMethod": "POST", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}:batchProcess", "parameters": { "name": { "required": true, "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+$", "description": "Required. The resource name of Processor or ProcessorVersion. Format: `projects/{project}/locations/{location}/processors/{processor}`, or `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`", "type": "string" } }, "response": { "$ref": "GoogleLongrunningOperation" }, "description": "LRO endpoint to batch process many documents. The output is written to Cloud Storage as JSON in the [Document] format." }, "disable": { "parameters": { "name": { "type": "string", "required": true, "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+$", "description": "Required. The processor resource name to be disabled." } }, "request": { "$ref": "GoogleCloudDocumentaiV1beta3DisableProcessorRequest" }, "parameterOrder": [ "name" ], "id": "documentai.projects.locations.processors.disable", "path": "v1beta3/{+name}:disable", "httpMethod": "POST", "description": "Disables a processor", "response": { "$ref": "GoogleLongrunningOperation" }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}:disable", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ] }, "updateDataset": { "description": "Updates metadata associated with a dataset. Note that this method requires the `documentai.googleapis.com/datasets.update` permission on the project, which is highly privileged. A user or service account with this permission can create new processors that can interact with any gcs bucket in your project.", "httpMethod": "PATCH", "id": "documentai.projects.locations.processors.updateDataset", "path": "v1beta3/{+name}", "parameters": { "name": { "location": "path", "required": true, "description": "Dataset resource name. Format: `projects/{project}/locations/{location}/processors/{processor}/dataset`", "type": "string", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/dataset$" }, "updateMask": { "description": "The update mask applies to the resource.", "format": "google-fieldmask", "type": "string", "location": "query" } }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/dataset", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "response": { "$ref": "GoogleLongrunningOperation" }, "parameterOrder": [ "name" ], "request": { "$ref": "GoogleCloudDocumentaiV1beta3Dataset" } }, "enable": { "id": "documentai.projects.locations.processors.enable", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}:enable", "parameterOrder": [ "name" ], "httpMethod": "POST", "response": { "$ref": "GoogleLongrunningOperation" }, "path": "v1beta3/{+name}:enable", "description": "Enables a processor", "parameters": { "name": { "type": "string", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+$", "required": true, "description": "Required. The processor resource name to be enabled.", "location": "path" } }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "request": { "$ref": "GoogleCloudDocumentaiV1beta3EnableProcessorRequest" } }, "list": { "httpMethod": "GET", "path": "v1beta3/{+parent}/processors", "id": "documentai.projects.locations.processors.list", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors", "parameters": { "pageToken": { "description": "We will return the processors sorted by creation time. The page token will point to the next processor.", "location": "query", "type": "string" }, "parent": { "description": "Required. The parent (project and location) which owns this collection of Processors. Format: `projects/{project}/locations/{location}`", "pattern": "^projects/[^/]+/locations/[^/]+$", "location": "path", "required": true, "type": "string" }, "pageSize": { "type": "integer", "location": "query", "description": "The maximum number of processors to return. If unspecified, at most `50` processors will be returned. The maximum value is `100`. Values above `100` will be coerced to `100`.", "format": "int32" } }, "description": "Lists all processors which belong to this project.", "parameterOrder": [ "parent" ], "response": { "$ref": "GoogleCloudDocumentaiV1beta3ListProcessorsResponse" } }, "setDefaultProcessorVersion": { "parameters": { "processor": { "required": true, "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+$", "description": "Required. The resource name of the Processor to change default version.", "type": "string" } }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "path": "v1beta3/{+processor}:setDefaultProcessorVersion", "description": "Set the default (active) version of a Processor that will be used in ProcessDocument and BatchProcessDocuments.", "id": "documentai.projects.locations.processors.setDefaultProcessorVersion", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}:setDefaultProcessorVersion", "request": { "$ref": "GoogleCloudDocumentaiV1beta3SetDefaultProcessorVersionRequest" }, "httpMethod": "POST", "parameterOrder": [ "processor" ], "response": { "$ref": "GoogleLongrunningOperation" } }, "get": { "path": "v1beta3/{+name}", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameters": { "name": { "description": "Required. The processor resource name.", "required": true, "type": "string", "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+$" } }, "parameterOrder": [ "name" ], "description": "Gets a processor detail.", "id": "documentai.projects.locations.processors.get", "httpMethod": "GET", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}", "response": { "$ref": "GoogleCloudDocumentaiV1beta3Processor" } }, "delete": { "description": "Deletes the processor, unloads all deployed model artifacts if it was enabled and then deletes all artifacts associated with this processor.", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameterOrder": [ "name" ], "httpMethod": "DELETE", "path": "v1beta3/{+name}", "response": { "$ref": "GoogleLongrunningOperation" }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}", "parameters": { "name": { "description": "Required. The processor resource name to be deleted.", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+$", "location": "path", "required": true, "type": "string" } }, "id": "documentai.projects.locations.processors.delete" } }, "resources": { "dataset": { "methods": { "updateDatasetSchema": { "path": "v1beta3/{+name}", "id": "documentai.projects.locations.processors.dataset.updateDatasetSchema", "parameters": { "name": { "description": "Dataset schema resource name. Format: `projects/{project}/locations/{location}/processors/{processor}/dataset/datasetSchema`", "type": "string", "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/dataset/datasetSchema$", "required": true }, "updateMask": { "type": "string", "description": "The update mask applies to the resource.", "format": "google-fieldmask", "location": "query" } }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameterOrder": [ "name" ], "description": "Updates a `DatasetSchema`.", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/dataset/datasetSchema", "httpMethod": "PATCH", "response": { "$ref": "GoogleCloudDocumentaiV1beta3DatasetSchema" }, "request": { "$ref": "GoogleCloudDocumentaiV1beta3DatasetSchema" } }, "batchDeleteDocuments": { "id": "documentai.projects.locations.processors.dataset.batchDeleteDocuments", "request": { "$ref": "GoogleCloudDocumentaiV1beta3BatchDeleteDocumentsRequest" }, "httpMethod": "POST", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "response": { "$ref": "GoogleLongrunningOperation" }, "path": "v1beta3/{+dataset}:batchDeleteDocuments", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/dataset:batchDeleteDocuments", "parameterOrder": [ "dataset" ], "parameters": { "dataset": { "type": "string", "location": "path", "required": true, "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/dataset$", "description": "Required. The dataset resource name. Format: projects/{project}/locations/{location}/processors/{processor}/dataset" } }, "description": "Deletes a set of documents." }, "importDocuments": { "id": "documentai.projects.locations.processors.dataset.importDocuments", "parameterOrder": [ "dataset" ], "response": { "$ref": "GoogleLongrunningOperation" }, "description": "Import documents into a dataset.", "parameters": { "dataset": { "location": "path", "required": true, "type": "string", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/dataset$", "description": "Required. The dataset resource name. Format: projects/{project}/locations/{location}/processors/{processor}/dataset" } }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/dataset:importDocuments", "request": { "$ref": "GoogleCloudDocumentaiV1beta3ImportDocumentsRequest" }, "httpMethod": "POST", "path": "v1beta3/{+dataset}:importDocuments" }, "getDatasetSchema": { "parameterOrder": [ "name" ], "httpMethod": "GET", "description": "Gets the `DatasetSchema` of a `Dataset`.", "path": "v1beta3/{+name}", "id": "documentai.projects.locations.processors.dataset.getDatasetSchema", "response": { "$ref": "GoogleCloudDocumentaiV1beta3DatasetSchema" }, "parameters": { "visibleFieldsOnly": { "description": "If set, only returns the visible fields of the schema.", "type": "boolean", "location": "query" }, "name": { "description": "Required. The dataset schema resource name. Format: projects/{project}/locations/{location}/processors/{processor}/dataset/datasetSchema", "required": true, "type": "string", "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/dataset/datasetSchema$" } }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/dataset/datasetSchema", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ] }, "getDocument": { "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/dataset:getDocument", "httpMethod": "GET", "path": "v1beta3/{+dataset}:getDocument", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "response": { "$ref": "GoogleCloudDocumentaiV1beta3GetDocumentResponse" }, "id": "documentai.projects.locations.processors.dataset.getDocument", "description": "Returns relevant fields present in the requested document.", "parameters": { "readMask": { "format": "google-fieldmask", "description": "If set, only fields listed here will be returned. Otherwise, all fields will be returned by default.", "type": "string", "location": "query" }, "documentId.revisionRef.latestProcessorVersion": { "type": "string", "description": "Reads the revision generated by the processor version. The format takes the full resource name of processor version. `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`", "location": "query" }, "documentId.revisionRef.revisionCase": { "description": "Reads the revision by the predefined case.", "location": "query", "enumDescriptions": [ "Unspecified case, fall back to read the `LATEST_HUMAN_REVIEW`.", "The latest revision made by a human.", "The latest revision based on timestamp.", "The first (OCR) revision." ], "enum": [ "REVISION_CASE_UNSPECIFIED", "LATEST_HUMAN_REVIEW", "LATEST_TIMESTAMP", "BASE_OCR_REVISION" ], "type": "string" }, "pageRange.end": { "format": "int32", "location": "query", "type": "integer", "description": "Last page number (one-based index) to be returned." }, "documentId.unmanagedDocId.docId": { "location": "query", "description": "Required. The ID of the document.", "type": "string" }, "pageRange.start": { "description": "First page number (one-based index) to be returned.", "location": "query", "format": "int32", "type": "integer" }, "documentId.gcsManagedDocId.gcsUri": { "location": "query", "description": "Required. The Cloud Storage URI where the actual document is stored.", "type": "string" }, "documentId.revisionRef.revisionId": { "description": "Reads the revision given by the id.", "location": "query", "type": "string" }, "documentId.gcsManagedDocId.cwDocId": { "type": "string", "description": "ID of the document (indexed) managed by Content Warehouse.", "deprecated": true, "location": "query" }, "dataset": { "description": "Required. The resource name of the dataset that the document belongs to . Format: projects/{project}/locations/{location}/processors/{processor}/dataset", "location": "path", "type": "string", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/dataset$", "required": true } }, "parameterOrder": [ "dataset" ] }, "listDocuments": { "parameterOrder": [ "dataset" ], "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "httpMethod": "POST", "request": { "$ref": "GoogleCloudDocumentaiV1beta3ListDocumentsRequest" }, "parameters": { "dataset": { "description": "Required. The resource name of the dataset to be listed. Format: projects/{project}/locations/{location}/processors/{processor}/dataset", "required": true, "location": "path", "type": "string", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/dataset$" } }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/dataset:listDocuments", "description": "Returns a list of documents present in the dataset.", "response": { "$ref": "GoogleCloudDocumentaiV1beta3ListDocumentsResponse" }, "path": "v1beta3/{+dataset}:listDocuments", "id": "documentai.projects.locations.processors.dataset.listDocuments" } } }, "processorVersions": { "resources": { "evaluations": { "methods": { "list": { "description": "Retrieves a set of evaluations for a given processor version.", "parameterOrder": [ "parent" ], "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "httpMethod": "GET", "id": "documentai.projects.locations.processors.processorVersions.evaluations.list", "response": { "$ref": "GoogleCloudDocumentaiV1beta3ListEvaluationsResponse" }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions/{processorVersionsId}/evaluations", "parameters": { "pageToken": { "location": "query", "description": "A page token, received from a previous `ListEvaluations` call. Provide this to retrieve the subsequent page.", "type": "string" }, "pageSize": { "type": "integer", "format": "int32", "location": "query", "description": "The standard list page size. If unspecified, at most `5` evaluations are returned. The maximum value is `100`. Values above `100` are coerced to `100`." }, "parent": { "location": "path", "type": "string", "required": true, "description": "Required. The resource name of the ProcessorVersion to list evaluations for. `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/processorVersions/[^/]+$" } }, "path": "v1beta3/{+parent}/evaluations" }, "get": { "parameters": { "name": { "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/processorVersions/[^/]+/evaluations/[^/]+$", "description": "Required. The resource name of the Evaluation to get. `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}/evaluations/{evaluation}`", "required": true, "location": "path", "type": "string" } }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "httpMethod": "GET", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions/{processorVersionsId}/evaluations/{evaluationsId}", "id": "documentai.projects.locations.processors.processorVersions.evaluations.get", "path": "v1beta3/{+name}", "response": { "$ref": "GoogleCloudDocumentaiV1beta3Evaluation" }, "description": "Retrieves a specific evaluation.", "parameterOrder": [ "name" ] } } } }, "methods": { "delete": { "parameterOrder": [ "name" ], "path": "v1beta3/{+name}", "id": "documentai.projects.locations.processors.processorVersions.delete", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameters": { "name": { "type": "string", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/processorVersions/[^/]+$", "description": "Required. The processor version resource name to be deleted.", "required": true, "location": "path" } }, "description": "Deletes the processor version, all artifacts under the processor version will be deleted.", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions/{processorVersionsId}", "response": { "$ref": "GoogleLongrunningOperation" }, "httpMethod": "DELETE" }, "get": { "httpMethod": "GET", "response": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessorVersion" }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "description": "Gets a processor version detail.", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions/{processorVersionsId}", "parameterOrder": [ "name" ], "parameters": { "name": { "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/processorVersions/[^/]+$", "required": true, "location": "path", "description": "Required. The processor resource name.", "type": "string" } }, "id": "documentai.projects.locations.processors.processorVersions.get", "path": "v1beta3/{+name}" }, "process": { "parameterOrder": [ "name" ], "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions/{processorVersionsId}:process", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameters": { "name": { "type": "string", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/processorVersions/[^/]+$", "description": "Required. The resource name of the Processor or ProcessorVersion to use for processing. If a Processor is specified, the server will use its default version. Format: `projects/{project}/locations/{location}/processors/{processor}`, or `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`", "required": true, "location": "path" } }, "response": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessResponse" }, "request": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessRequest" }, "description": "Processes a single document.", "httpMethod": "POST", "path": "v1beta3/{+name}:process", "id": "documentai.projects.locations.processors.processorVersions.process" }, "undeploy": { "parameterOrder": [ "name" ], "response": { "$ref": "GoogleLongrunningOperation" }, "id": "documentai.projects.locations.processors.processorVersions.undeploy", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions/{processorVersionsId}:undeploy", "httpMethod": "POST", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "request": { "$ref": "GoogleCloudDocumentaiV1beta3UndeployProcessorVersionRequest" }, "path": "v1beta3/{+name}:undeploy", "description": "Undeploys the processor version.", "parameters": { "name": { "description": "Required. The processor version resource name to be undeployed.", "type": "string", "required": true, "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/processorVersions/[^/]+$" } } }, "train": { "parameters": { "parent": { "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+$", "required": true, "location": "path", "type": "string", "description": "Required. The parent (project, location and processor) to create the new version for. Format: `projects/{project}/locations/{location}/processors/{processor}`." } }, "description": "Trains a new processor version. Operation metadata is returned as TrainProcessorVersionMetadata.", "path": "v1beta3/{+parent}/processorVersions:train", "httpMethod": "POST", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions:train", "id": "documentai.projects.locations.processors.processorVersions.train", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "response": { "$ref": "GoogleLongrunningOperation" }, "parameterOrder": [ "parent" ], "request": { "$ref": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequest" } }, "importProcessorVersion": { "request": { "$ref": "GoogleCloudDocumentaiV1beta3ImportProcessorVersionRequest" }, "response": { "$ref": "GoogleLongrunningOperation" }, "parameters": { "parent": { "description": "Required. The destination processor name to create the processor version in. Format: `projects/{project}/locations/{location}/processors/{processor}`", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+$", "location": "path", "required": true, "type": "string" } }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions:importProcessorVersion", "parameterOrder": [ "parent" ], "description": "Imports a processor version from source processor version.", "path": "v1beta3/{+parent}/processorVersions:importProcessorVersion", "id": "documentai.projects.locations.processors.processorVersions.importProcessorVersion", "httpMethod": "POST", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ] }, "deploy": { "id": "documentai.projects.locations.processors.processorVersions.deploy", "description": "Deploys the processor version.", "response": { "$ref": "GoogleLongrunningOperation" }, "httpMethod": "POST", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions/{processorVersionsId}:deploy", "request": { "$ref": "GoogleCloudDocumentaiV1beta3DeployProcessorVersionRequest" }, "parameters": { "name": { "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/processorVersions/[^/]+$", "location": "path", "description": "Required. The processor version resource name to be deployed.", "required": true, "type": "string" } }, "path": "v1beta3/{+name}:deploy", "parameterOrder": [ "name" ], "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ] }, "list": { "path": "v1beta3/{+parent}/processorVersions", "parameters": { "pageSize": { "type": "integer", "location": "query", "format": "int32", "description": "The maximum number of processor versions to return. If unspecified, at most `10` processor versions will be returned. The maximum value is `20`. Values above `20` will be coerced to `20`." }, "pageToken": { "description": "We will return the processor versions sorted by creation time. The page token will point to the next processor version.", "type": "string", "location": "query" }, "parent": { "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+$", "description": "Required. The parent (project, location and processor) to list all versions. Format: `projects/{project}/locations/{location}/processors/{processor}`", "type": "string", "required": true, "location": "path" } }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "id": "documentai.projects.locations.processors.processorVersions.list", "response": { "$ref": "GoogleCloudDocumentaiV1beta3ListProcessorVersionsResponse" }, "description": "Lists all versions of a processor.", "httpMethod": "GET", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions", "parameterOrder": [ "parent" ] }, "evaluateProcessorVersion": { "id": "documentai.projects.locations.processors.processorVersions.evaluateProcessorVersion", "response": { "$ref": "GoogleLongrunningOperation" }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions/{processorVersionsId}:evaluateProcessorVersion", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameters": { "processorVersion": { "required": true, "type": "string", "description": "Required. The resource name of the ProcessorVersion to evaluate. `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`", "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/processorVersions/[^/]+$" } }, "parameterOrder": [ "processorVersion" ], "description": "Evaluates a ProcessorVersion against annotated documents, producing an Evaluation.", "request": { "$ref": "GoogleCloudDocumentaiV1beta3EvaluateProcessorVersionRequest" }, "path": "v1beta3/{+processorVersion}:evaluateProcessorVersion", "httpMethod": "POST" }, "batchProcess": { "httpMethod": "POST", "id": "documentai.projects.locations.processors.processorVersions.batchProcess", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "request": { "$ref": "GoogleCloudDocumentaiV1beta3BatchProcessRequest" }, "parameters": { "name": { "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/processorVersions/[^/]+$", "description": "Required. The resource name of Processor or ProcessorVersion. Format: `projects/{project}/locations/{location}/processors/{processor}`, or `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`", "location": "path", "type": "string", "required": true } }, "response": { "$ref": "GoogleLongrunningOperation" }, "path": "v1beta3/{+name}:batchProcess", "parameterOrder": [ "name" ], "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/processorVersions/{processorVersionsId}:batchProcess", "description": "LRO endpoint to batch process many documents. The output is written to Cloud Storage as JSON in the [Document] format." } } }, "humanReviewConfig": { "methods": { "reviewDocument": { "description": "Send a document for Human Review. The input document should be processed by the specified processor.", "id": "documentai.projects.locations.processors.humanReviewConfig.reviewDocument", "path": "v1beta3/{+humanReviewConfig}:reviewDocument", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "httpMethod": "POST", "deprecated": true, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processors/{processorsId}/humanReviewConfig:reviewDocument", "parameters": { "humanReviewConfig": { "type": "string", "description": "Required. The resource name of the HumanReviewConfig that the document will be reviewed with.", "required": true, "pattern": "^projects/[^/]+/locations/[^/]+/processors/[^/]+/humanReviewConfig$", "location": "path" } }, "response": { "$ref": "GoogleLongrunningOperation" }, "request": { "$ref": "GoogleCloudDocumentaiV1beta3ReviewDocumentRequest" }, "parameterOrder": [ "humanReviewConfig" ] } }, "deprecated": true } } }, "processorTypes": { "methods": { "get": { "parameters": { "name": { "type": "string", "description": "Required. The processor type resource name.", "pattern": "^projects/[^/]+/locations/[^/]+/processorTypes/[^/]+$", "required": true, "location": "path" } }, "description": "Gets a processor type detail.", "path": "v1beta3/{+name}", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processorTypes/{processorTypesId}", "parameterOrder": [ "name" ], "response": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessorType" }, "httpMethod": "GET", "id": "documentai.projects.locations.processorTypes.get" }, "list": { "response": { "$ref": "GoogleCloudDocumentaiV1beta3ListProcessorTypesResponse" }, "id": "documentai.projects.locations.processorTypes.list", "parameters": { "parent": { "type": "string", "pattern": "^projects/[^/]+/locations/[^/]+$", "location": "path", "required": true, "description": "Required. The location of processor types to list. Format: `projects/{project}/locations/{location}`." }, "pageToken": { "type": "string", "location": "query", "description": "Used to retrieve the next page of results, empty if at the end of the list." }, "pageSize": { "location": "query", "description": "The maximum number of processor types to return. If unspecified, at most `100` processor types will be returned. The maximum value is `500`. Values above `500` will be coerced to `500`.", "type": "integer", "format": "int32" } }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/processorTypes", "description": "Lists the processor types that exist.", "path": "v1beta3/{+parent}/processorTypes", "parameterOrder": [ "parent" ], "httpMethod": "GET", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ] } } }, "schemas": { "resources": { "schemaVersions": { "methods": { "delete": { "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas/{schemasId}/schemaVersions/{schemaVersionsId}", "parameters": { "name": { "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+/schemas/[^/]+/schemaVersions/[^/]+$", "required": true, "description": "Required. The name of the SchemaVersion to delete. Format: `projects/{project}/locations/{location}/schemas/{schema}/schemaVersions/{schema_version}`", "type": "string" } }, "response": { "$ref": "GoogleLongrunningOperation" }, "parameterOrder": [ "name" ], "id": "documentai.projects.locations.schemas.schemaVersions.delete", "httpMethod": "DELETE", "description": "Deletes a schema version.", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "path": "v1beta3/{+name}" }, "get": { "parameterOrder": [ "name" ], "id": "documentai.projects.locations.schemas.schemaVersions.get", "response": { "$ref": "GoogleCloudDocumentaiV1beta3SchemaVersion" }, "parameters": { "name": { "pattern": "^projects/[^/]+/locations/[^/]+/schemas/[^/]+/schemaVersions/[^/]+$", "description": "Required. The name of the SchemaVersion to get. Format: `projects/{project}/locations/{location}/schemas/{schema}/schemaVersions/{schema_version}`", "required": true, "location": "path", "type": "string" } }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas/{schemasId}/schemaVersions/{schemaVersionsId}", "path": "v1beta3/{+name}", "httpMethod": "GET", "description": "Gets a schema version." }, "patch": { "response": { "$ref": "GoogleCloudDocumentaiV1beta3SchemaVersion" }, "request": { "$ref": "GoogleCloudDocumentaiV1beta3SchemaVersion" }, "description": "Updates a schema version. Editable fields are: - `display_name` - `labels`", "parameters": { "updateMask": { "type": "string", "format": "google-fieldmask", "location": "query", "description": "Optional. The update mask to apply to the resource. **Note:** Only the following fields can be updated: - `display_name` - `labels`" }, "name": { "type": "string", "pattern": "^projects/[^/]+/locations/[^/]+/schemas/[^/]+/schemaVersions/[^/]+$", "required": true, "location": "path", "description": "Identifier. The resource name of the SchemaVersion. Format: `projects/{project}/locations/{location}/schemas/{schema}/schemaVersions/{schema_version}`" } }, "parameterOrder": [ "name" ], "id": "documentai.projects.locations.schemas.schemaVersions.patch", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas/{schemasId}/schemaVersions/{schemaVersionsId}", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "path": "v1beta3/{+name}", "httpMethod": "PATCH" }, "generate": { "parameterOrder": [ "parent" ], "httpMethod": "POST", "description": "Generates a schema version.", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "path": "v1beta3/{+parent}/schemaVersions:generate", "id": "documentai.projects.locations.schemas.schemaVersions.generate", "parameters": { "parent": { "pattern": "^projects/[^/]+/locations/[^/]+/schemas/[^/]+$", "type": "string", "description": "Required. The parent (project, location and schema) under which to generate the SchemaVersion. Format: `projects/{project}/locations/{location}/schemas/{schema}`", "required": true, "location": "path" } }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas/{schemasId}/schemaVersions:generate", "response": { "$ref": "GoogleCloudDocumentaiV1beta3GenerateSchemaVersionResponse" }, "request": { "$ref": "GoogleCloudDocumentaiV1beta3GenerateSchemaVersionRequest" } }, "list": { "path": "v1beta3/{+parent}/schemaVersions", "id": "documentai.projects.locations.schemas.schemaVersions.list", "parameters": { "parent": { "type": "string", "location": "path", "required": true, "pattern": "^projects/[^/]+/locations/[^/]+/schemas/[^/]+$", "description": "Required. Format: `projects/{project}/locations/{location}/schemas/{schema}`" }, "pageSize": { "format": "int32", "type": "integer", "location": "query", "description": "Optional. The maximum number of SchemaVersion to return. If unspecified, at most `10` SchemaVersion will be returned. The maximum value is `20`. Values above `20` will be coerced to `20`." }, "pageToken": { "location": "query", "description": "Optional. Returns the SchemaVersion sorted by creation time. The page token will point to the next SchemaVersion.", "type": "string" } }, "description": "Lists SchemaVersions.", "response": { "$ref": "GoogleCloudDocumentaiV1beta3ListSchemaVersionsResponse" }, "parameterOrder": [ "parent" ], "httpMethod": "GET", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas/{schemasId}/schemaVersions", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ] }, "create": { "parameterOrder": [ "parent" ], "response": { "$ref": "GoogleCloudDocumentaiV1beta3SchemaVersion" }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas/{schemasId}/schemaVersions", "path": "v1beta3/{+parent}/schemaVersions", "parameters": { "parent": { "description": "Required. The parent (project and location) under which to create the SchemaVersion. Format: `projects/{project}/locations/{location}/schemas/{schema}`", "pattern": "^projects/[^/]+/locations/[^/]+/schemas/[^/]+$", "required": true, "type": "string", "location": "path" } }, "request": { "$ref": "GoogleCloudDocumentaiV1beta3SchemaVersion" }, "description": "Creates a schema version.", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "id": "documentai.projects.locations.schemas.schemaVersions.create", "httpMethod": "POST" } } } }, "methods": { "patch": { "parameterOrder": [ "name" ], "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameters": { "updateMask": { "description": "Optional. The update mask to apply to the resource. **Note:** Only the following fields can be updated: - `display_name` - `labels`", "format": "google-fieldmask", "location": "query", "type": "string" }, "name": { "description": "Identifier. The resource name of the Schema. Format: `projects/{project}/locations/{location}/schemas/{schema}`", "pattern": "^projects/[^/]+/locations/[^/]+/schemas/[^/]+$", "required": true, "type": "string", "location": "path" } }, "id": "documentai.projects.locations.schemas.patch", "description": "Updates a schema. Editable fields are: - `display_name` - `labels`", "response": { "$ref": "GoogleCloudDocumentaiV1beta3NextSchema" }, "httpMethod": "PATCH", "request": { "$ref": "GoogleCloudDocumentaiV1beta3NextSchema" }, "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas/{schemasId}", "path": "v1beta3/{+name}" }, "list": { "parameters": { "pageToken": { "location": "query", "type": "string", "description": "Optional. Returns the schema groups sorted by creation time. The page token will point to the next Schema." }, "pageSize": { "type": "integer", "description": "Optional. The maximum number of schema groups to return. If unspecified, at most `10` Schema will be returned. The maximum value is `20`. Values above `20` will be coerced to `20`.", "format": "int32", "location": "query" }, "parent": { "required": true, "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+$", "description": "Required. Format: `projects/{project}/locations/{location}`", "type": "string" } }, "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "path": "v1beta3/{+parent}/schemas", "description": "Lists Schemas.", "id": "documentai.projects.locations.schemas.list", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas", "httpMethod": "GET", "parameterOrder": [ "parent" ], "response": { "$ref": "GoogleCloudDocumentaiV1beta3ListSchemasResponse" } }, "delete": { "id": "documentai.projects.locations.schemas.delete", "response": { "$ref": "GoogleLongrunningOperation" }, "parameters": { "name": { "description": "Required. The name of the Schema to be deleted. Format: `projects/{project}/locations/{location}/schemas/{schema}`", "pattern": "^projects/[^/]+/locations/[^/]+/schemas/[^/]+$", "type": "string", "required": true, "location": "path" }, "force": { "description": "Optional. If set to true, any child resources of this Schema will also be deleted. (Otherwise, the request will only work if the Schema has no child resources.)", "type": "boolean", "location": "query" } }, "description": "Deletes a schema.", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas/{schemasId}", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "parameterOrder": [ "name" ], "httpMethod": "DELETE", "path": "v1beta3/{+name}" }, "create": { "path": "v1beta3/{+parent}/schemas", "id": "documentai.projects.locations.schemas.create", "httpMethod": "POST", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "description": "Creates a schema.", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas", "parameters": { "parent": { "pattern": "^projects/[^/]+/locations/[^/]+$", "required": true, "location": "path", "description": "Required. The parent (project and location) under which to create the Schema. Format: `projects/{project}/locations/{location}`", "type": "string" } }, "request": { "$ref": "GoogleCloudDocumentaiV1beta3NextSchema" }, "response": { "$ref": "GoogleCloudDocumentaiV1beta3NextSchema" }, "parameterOrder": [ "parent" ] }, "get": { "response": { "$ref": "GoogleCloudDocumentaiV1beta3NextSchema" }, "httpMethod": "GET", "flatPath": "v1beta3/projects/{projectsId}/locations/{locationsId}/schemas/{schemasId}", "id": "documentai.projects.locations.schemas.get", "parameters": { "name": { "location": "path", "pattern": "^projects/[^/]+/locations/[^/]+/schemas/[^/]+$", "required": true, "type": "string", "description": "Required. The name of the Schema to get. Format: `projects/{project}/locations/{location}/schemas/{schema}`" } }, "path": "v1beta3/{+name}", "scopes": [ "https://www.googleapis.com/auth/cloud-platform" ], "description": "Gets a schema.", "parameterOrder": [ "name" ] } } } } } } } }, "description": "Service to parse structured information from unstructured or semi-structured documents using state-of-the-art Google AI such as natural language, computer vision, translation, and AutoML.", "name": "documentai", "mtlsRootUrl": "https://documentai.mtls.googleapis.com/", "schemas": { "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkImageChunkField": { "type": "object", "properties": { "annotations": { "description": "Annotation of the image chunk field.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentAnnotations" }, "dataUri": { "description": "Optional. Data URI of the image. It is composed of four parts: a prefix (data:), a MIME type indicating the type of data, an optional base64 token if non-textual, and the data itself: data:,", "type": "string" }, "blobAssetId": { "type": "string", "description": "Optional. Asset ID of the inline image. If set, find the image content in the blob_assets field." }, "gcsUri": { "description": "Optional. Google Cloud Storage URI of the image.", "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkImageChunkField", "description": "The image chunk field in the chunk." }, "GoogleCloudDocumentaiUiv1beta3TrainProcessorVersionMetadataDatasetValidation": { "description": "The dataset validation information. This includes any and all errors with documents and the dataset.", "id": "GoogleCloudDocumentaiUiv1beta3TrainProcessorVersionMetadataDatasetValidation", "properties": { "datasetErrorCount": { "format": "int32", "description": "The total number of dataset errors.", "type": "integer" }, "datasetErrors": { "description": "Error information for the dataset as a whole. A maximum of 10 dataset errors will be returned. A single dataset error is terminal for training.", "items": { "$ref": "GoogleRpcStatus" }, "type": "array" }, "documentErrorCount": { "type": "integer", "description": "The total number of document errors.", "format": "int32" }, "documentErrors": { "items": { "$ref": "GoogleRpcStatus" }, "description": "Error information pertaining to specific documents. A maximum of 10 document errors will be returned. Any document with errors will not be used throughout training.", "type": "array" } }, "type": "object" }, "GoogleCloudDocumentaiV1DeleteProcessorMetadata": { "type": "object", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "id": "GoogleCloudDocumentaiV1DeleteProcessorMetadata", "description": "The long-running operation metadata for the DeleteProcessor method." }, "GoogleCloudDocumentaiV1beta3ProcessOptionsIndividualPageSelector": { "properties": { "pages": { "type": "array", "description": "Optional. Indices of the pages (starting from 1).", "items": { "type": "integer", "format": "int32" } } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3ProcessOptionsIndividualPageSelector", "description": "A list of individual page numbers." }, "GoogleCloudDocumentaiV1beta3EvaluateProcessorVersionResponse": { "properties": { "evaluation": { "type": "string", "description": "The resource name of the created evaluation." } }, "id": "GoogleCloudDocumentaiV1beta3EvaluateProcessorVersionResponse", "type": "object", "description": "Response of the EvaluateProcessorVersion method." }, "GoogleCloudDocumentaiUiv1beta3DocumentId": { "description": "Document Identifier.", "properties": { "gcsManagedDocId": { "description": "A document ID within user-managed Cloud Storage.", "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentIdGCSManagedDocumentId" }, "unmanagedDocId": { "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentIdUnmanagedDocumentId", "description": "A document ID within unmanaged dataset." }, "revisionRef": { "$ref": "GoogleCloudDocumentaiUiv1beta3RevisionRef", "description": "Points to a specific revision of the document if set." } }, "id": "GoogleCloudDocumentaiUiv1beta3DocumentId", "type": "object" }, "GoogleCloudLocationLocation": { "description": "A resource that represents a Google Cloud location.", "properties": { "name": { "description": "Resource name for the location, which may vary between implementations. For example: `\"projects/example-project/locations/us-east1\"`", "type": "string" }, "metadata": { "additionalProperties": { "type": "any", "description": "Properties of the object. Contains field @type with type URL." }, "description": "Service-specific metadata. For example the available capacity at the given location.", "type": "object" }, "labels": { "type": "object", "additionalProperties": { "type": "string" }, "description": "Cross-service attributes for the location. For example {\"cloud.googleapis.com/region\": \"us-east1\"}" }, "locationId": { "type": "string", "description": "The canonical id for this location. For example: `\"us-east1\"`." }, "displayName": { "type": "string", "description": "The friendly name for this location, typically a nearby city name. For example, \"Tokyo\"." } }, "id": "GoogleCloudLocationLocation", "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentPageTableTableCell": { "description": "A cell representation inside the table.", "properties": { "detectedLanguages": { "type": "array", "description": "A list of detected languages together with confidence.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" } }, "rowSpan": { "description": "How many rows this cell spans.", "type": "integer", "format": "int32" }, "colSpan": { "type": "integer", "description": "How many columns this cell spans.", "format": "int32" }, "layout": { "description": "Layout for TableCell.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout" } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentPageTableTableCell" }, "GoogleCloudDocumentaiUiv1beta3DeleteProcessorVersionMetadata": { "id": "GoogleCloudDocumentaiUiv1beta3DeleteProcessorVersionMetadata", "description": "The long-running operation metadata for the DeleteProcessorVersion method.", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3EnableProcessorResponse": { "description": "Response message for the EnableProcessor method. Intentionally empty proto for adding fields in future.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3EnableProcessorResponse", "properties": {} }, "GoogleCloudDocumentaiV1beta3PropertyMetadata": { "properties": { "inactive": { "description": "Whether the property should be considered as \"inactive\".", "type": "boolean" }, "fieldExtractionMetadata": { "description": "Field extraction metadata on the property.", "$ref": "GoogleCloudDocumentaiV1beta3FieldExtractionMetadata" } }, "type": "object", "description": "Metadata about a property.", "id": "GoogleCloudDocumentaiV1beta3PropertyMetadata" }, "GoogleCloudDocumentaiV1beta3ProcessResponse": { "properties": { "humanReviewStatus": { "description": "The status of human review on the processed document.", "$ref": "GoogleCloudDocumentaiV1beta3HumanReviewStatus" }, "humanReviewOperation": { "deprecated": true, "description": "The name of the operation triggered by the processed document. If the human review process isn't triggered, this field is empty. It has the same response type and metadata as the long-running operation returned by ReviewDocument.", "type": "string" }, "document": { "description": "The document payload, will populate fields based on the processor's behavior.", "$ref": "GoogleCloudDocumentaiV1beta3Document" } }, "description": "Response message for the ProcessDocument method.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3ProcessResponse" }, "GoogleCloudDocumentaiV1beta3UndeployProcessorVersionRequest": { "description": "Request message for the UndeployProcessorVersion method.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3UndeployProcessorVersionRequest", "properties": {} }, "GoogleCloudDocumentaiV1beta3ImportDocumentsMetadata": { "type": "object", "properties": { "importConfigValidationResults": { "description": "Validation statuses of the batch documents import config.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3ImportDocumentsMetadataImportConfigValidationResult" } }, "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata" }, "individualImportStatuses": { "description": "The list of response details of each document.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3ImportDocumentsMetadataIndividualImportStatus" } }, "totalDocumentCount": { "format": "int32", "description": "Total number of the documents that are qualified for importing.", "type": "integer" } }, "description": "Metadata of the import document operation.", "id": "GoogleCloudDocumentaiV1beta3ImportDocumentsMetadata" }, "GoogleCloudDocumentaiV1beta3DocumentPageAnchor": { "properties": { "pageRefs": { "description": "One or more references to visual page elements", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageAnchorPageRef" }, "type": "array" } }, "type": "object", "description": "Referencing the visual context of the entity in the Document.pages. Page anchors can be cross-page, consist of multiple bounding polygons and optionally reference specific layout element types.", "id": "GoogleCloudDocumentaiV1beta3DocumentPageAnchor" }, "GoogleCloudDocumentaiUiv1beta3EvaluateProcessorVersionResponse": { "properties": { "evaluation": { "description": "The resource name of the created evaluation.", "type": "string" } }, "description": "Response of the EvaluateProcessorVersion method.", "id": "GoogleCloudDocumentaiUiv1beta3EvaluateProcessorVersionResponse", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3DocumentSchemaEntityType": { "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3DocumentSchemaEntityType", "description": "EntityType is the wrapper of a label of the corresponding model with detailed attributes and limitations for entity-based processors. Multiple types can also compose a dependency tree to represent nested types.", "properties": { "properties": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentSchemaEntityTypeProperty" }, "description": "Description the nested structure, or composition of an entity." }, "enumValues": { "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentSchemaEntityTypeEnumValues", "description": "If specified, lists all the possible values for this entity. This should not be more than a handful of values. If the number of values is \u003e10 or could change frequently, use the `EntityType.value_ontology` field and specify a list of all possible values in a value ontology file." }, "name": { "type": "string", "description": "Name of the type. It must be unique within the schema file and cannot be a \"Common Type\". The following naming conventions are used: - Use `snake_casing`. - Name matching is case-sensitive. - Maximum 64 characters. - Must start with a letter. - Allowed characters: ASCII letters `[a-z0-9_-]`. (For backward compatibility, internal infrastructure and tooling can handle any ASCII character.) - The `/` is sometimes used to denote a property of a type. For example, `line_item/amount`. This convention is deprecated, but will still be honored for backward compatibility." }, "baseTypes": { "type": "array", "description": "The entity type that this type is derived from. For now, one and only one should be set.", "items": { "type": "string" } }, "displayName": { "description": "User defined name for the type.", "type": "string" }, "entityTypeMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3EntityTypeMetadata", "description": "Metadata for the entity type." }, "description": { "type": "string", "description": "The description of the entity type. Could be used to provide more information about the entity type for model calls." } } }, "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageFooter": { "properties": { "text": { "type": "string", "description": "Footer in text format." }, "pageSpan": { "description": "Page span of the footer.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageSpan" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageFooter", "type": "object", "description": "Represents the page footer associated with the chunk." }, "GoogleCloudDocumentaiV1beta3ProcessorVersionDeprecationInfo": { "description": "Information about the upcoming deprecation of this processor version.", "id": "GoogleCloudDocumentaiV1beta3ProcessorVersionDeprecationInfo", "properties": { "replacementProcessorVersion": { "type": "string", "description": "If set, the processor version that will be used as a replacement." }, "deprecationTime": { "format": "google-datetime", "description": "The time at which this processor version will be deprecated.", "type": "string" } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentPageTable": { "type": "object", "description": "A table representation similar to HTML table structure.", "properties": { "bodyRows": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageTableTableRow" }, "description": "Body rows of the table." }, "headerRows": { "description": "Header rows of the table.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageTableTableRow" } }, "detectedLanguages": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" }, "description": "A list of detected languages together with confidence." }, "provenance": { "deprecated": true, "description": "The history of this table.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenance" }, "layout": { "description": "Layout for Table.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentPageTable" }, "GoogleCloudDocumentaiV1beta3SchemaGenerationIteration": { "type": "object", "properties": { "generatedSchema": { "description": "Required. The schema version generated by the model.", "$ref": "GoogleCloudDocumentaiV1beta3SchemaVersion" }, "adjustedSchema": { "$ref": "GoogleCloudDocumentaiV1beta3SchemaVersion", "description": "Optional. The previous schema version adjusted by the model." }, "prompt": { "description": "Optional. The prompt used for the iteration.", "type": "string" } }, "description": "A single iteration of the schema generation.", "id": "GoogleCloudDocumentaiV1beta3SchemaGenerationIteration" }, "GoogleCloudDocumentaiUiv1beta3SchemaEditabilityMetadata": { "properties": { "editable": { "type": "boolean", "description": "Explicit flag that controls whether the label is editable." }, "processorVersions": { "items": { "type": "string" }, "description": "Full resource name of processor versions that contain this label. e.g. `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`", "type": "array" } }, "id": "GoogleCloudDocumentaiUiv1beta3SchemaEditabilityMetadata", "type": "object", "description": "Metadata that specifies whether a label is editable and reasons why. These fields are read-only. Changing these fields has no impact on the backend." }, "GoogleCloudDocumentaiUiv1beta3DocumentSchemaEntityTypeEnumValues": { "id": "GoogleCloudDocumentaiUiv1beta3DocumentSchemaEntityTypeEnumValues", "type": "object", "description": "Defines the a list of enum values.", "properties": { "values": { "description": "The individual values that this enum values type can include.", "items": { "type": "string" }, "type": "array" } } }, "GoogleCloudDocumentaiV1beta3BatchDeleteDocumentsResponse": { "description": "Response of the delete documents operation.", "id": "GoogleCloudDocumentaiV1beta3BatchDeleteDocumentsResponse", "properties": {}, "type": "object" }, "GoogleCloudDocumentaiV1beta3ProcessOptionsLayoutConfigChunkingConfig": { "properties": { "includeAncestorHeadings": { "type": "boolean", "description": "Optional. Whether or not to include ancestor headings when splitting." }, "semanticChunkingGroupSize": { "description": "Optional. The number of tokens to group together when evaluating semantic similarity. **Note:** This field is not yet used.", "type": "boolean" }, "breakpointPercentileThreshold": { "format": "int32", "type": "integer", "description": "Optional. The percentile of cosine dissimilarity that must be exceeded between a group of tokens and the next. The smaller this number is, the more chunks will be generated. **Note:** This field is not yet used." }, "chunkSize": { "type": "integer", "format": "int32", "description": "Optional. The chunk sizes to use when splitting documents, in order of level." } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3ProcessOptionsLayoutConfigChunkingConfig", "description": "Serving config for chunking." }, "GoogleCloudDocumentaiV1DeployProcessorVersionMetadata": { "description": "The long-running operation metadata for the DeployProcessorVersion method.", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "id": "GoogleCloudDocumentaiV1DeployProcessorVersionMetadata", "type": "object" }, "GoogleCloudDocumentaiV1beta3ListDocumentsResponse": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3ListDocumentsResponse", "properties": { "nextPageToken": { "description": "A token, which can be sent as ListDocumentsRequest.page_token to retrieve the next page. If this field is omitted, there are no subsequent pages.", "type": "string" }, "documentMetadata": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentMetadata" }, "description": "Document metadata corresponding to the listed documents." }, "totalSize": { "type": "integer", "description": "Total count of documents queried.", "format": "int32" } } }, "GoogleCloudDocumentaiV1beta3BatchProcessRequestBatchOutputConfig": { "properties": { "gcsDestination": { "description": "The output Cloud Storage directory to put the processed documents.", "type": "string" } }, "description": "The output configuration in the BatchProcessDocuments method.", "id": "GoogleCloudDocumentaiV1beta3BatchProcessRequestBatchOutputConfig", "type": "object", "deprecated": true }, "GoogleCloudDocumentaiV1beta3BatchDeleteDocumentsMetadata": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3BatchDeleteDocumentsMetadata", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata" }, "individualBatchDeleteStatuses": { "type": "array", "description": "The list of response details of each document.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3BatchDeleteDocumentsMetadataIndividualBatchDeleteStatus" } }, "totalDocumentCount": { "description": "Total number of documents deleting from dataset.", "type": "integer", "format": "int32" }, "errorDocumentCount": { "type": "integer", "format": "int32", "description": "Total number of documents that failed to be deleted in storage." } } }, "GoogleCloudDocumentaiV1beta3BatchDeleteDocumentsRequest": { "id": "GoogleCloudDocumentaiV1beta3BatchDeleteDocumentsRequest", "properties": { "datasetDocuments": { "description": "Required. Dataset documents input. If given `filter`, all documents satisfying the filter will be deleted. If given documentIds, a maximum of 50 documents can be deleted in a batch. The request will be rejected if more than 50 document_ids are provided.", "$ref": "GoogleCloudDocumentaiV1beta3BatchDatasetDocuments" } }, "type": "object" }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInput": { "type": "object", "description": "Definition of the validation rules. Those are the input to the validator logic and they are used to validate a document. Next ID: 3", "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInput", "properties": { "validationRules": { "type": "array", "items": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRule" } } } }, "GoogleCloudDocumentaiV1beta3DocumentEntity": { "id": "GoogleCloudDocumentaiV1beta3DocumentEntity", "properties": { "confidence": { "format": "float", "description": "Optional. Confidence of detected Schema entity. Range `[0, 1]`.", "type": "number" }, "properties": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentEntity" }, "type": "array", "description": "Optional. Entities can be nested to form a hierarchical data structure representing the content in the document." }, "pageAnchor": { "description": "Optional. Represents the provenance of this entity wrt. the location on the page where it was found.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageAnchor" }, "type": { "description": "Required. Entity type from a schema, for example, `Address`.", "type": "string" }, "id": { "type": "string", "description": "Optional. Canonical id. This will be a unique value in the entity list for this document." }, "textAnchor": { "description": "Optional. Provenance of the entity. Text anchor indexing into the Document.text.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentTextAnchor" }, "mentionText": { "description": "Optional. Text value of the entity, for example, `1600 Amphitheatre Pkwy`.", "type": "string" }, "normalizedValue": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentEntityNormalizedValue", "description": "Optional. Normalized entity value. Absent if the extracted value could not be converted or the type (for example, address) is not supported for certain parsers. This field is also only populated for certain supported document types." }, "method": { "description": "Optional. Specifies how the entity's value is obtained.", "type": "string", "enumDescriptions": [ "When the method is not specified, it should be treated as `EXTRACT`.", "The entity's value is directly extracted as-is from the document text.", "The entity's value is derived through inference and is not necessarily an exact text extraction from the document." ], "enum": [ "METHOD_UNSPECIFIED", "EXTRACT", "DERIVE" ] }, "mentionId": { "type": "string", "description": "Optional. Deprecated. Use `id` field instead." }, "redacted": { "type": "boolean", "description": "Optional. Whether the entity will be redacted for de-identification purposes." }, "provenance": { "description": "Optional. The history of this annotation.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenance" } }, "type": "object", "description": "An entity that could be a phrase in the text or a property that belongs to the document. It is a known entity type, such as a person, an organization, or location." }, "GoogleCloudDocumentaiV1beta3BoundingPoly": { "description": "A bounding polygon for the detected image annotation.", "type": "object", "properties": { "normalizedVertices": { "type": "array", "description": "The bounding polygon normalized vertices.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3NormalizedVertex" } }, "vertices": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3Vertex" }, "type": "array", "description": "The bounding polygon vertices." } }, "id": "GoogleCloudDocumentaiV1beta3BoundingPoly" }, "GoogleCloudDocumentaiV1beta3DisableProcessorRequest": { "description": "Request message for the DisableProcessor method.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3DisableProcessorRequest", "properties": {} }, "GoogleCloudDocumentaiV1beta3Documents": { "description": "A set of inline documents.", "properties": { "documents": { "type": "array", "description": "The list of documents.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3Document" } } }, "id": "GoogleCloudDocumentaiV1beta3Documents", "type": "object" }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFieldOccurrences": { "properties": { "maxOccurrences": { "format": "uint32", "type": "integer" }, "minOccurrences": { "description": "Min and max occurrences of the field. If not set, there is limit set. The defined interval is a closed-closed interval, i.e. [min, max].", "format": "uint32", "type": "integer" }, "field": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleField" } }, "type": "object", "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFieldOccurrences" }, "GoogleCloudDocumentaiV1beta3DatasetSchema": { "properties": { "satisfiesPzi": { "readOnly": true, "type": "boolean", "description": "Output only. Reserved for future use." }, "satisfiesPzs": { "type": "boolean", "readOnly": true, "description": "Output only. Reserved for future use." }, "documentSchema": { "description": "Optional. Schema of the dataset.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentSchema" }, "name": { "description": "Dataset schema resource name. Format: `projects/{project}/locations/{location}/processors/{processor}/dataset/datasetSchema`", "type": "string" } }, "description": "Dataset Schema.", "id": "GoogleCloudDocumentaiV1beta3DatasetSchema", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3ProcessorVersionGenAiModelInfoCustomGenAiModelInfo": { "description": "Information for a custom Generative AI model created by the user. These are created with `Create New Version` in either the `Call foundation model` or `Fine tuning` tabs.", "id": "GoogleCloudDocumentaiUiv1beta3ProcessorVersionGenAiModelInfoCustomGenAiModelInfo", "properties": { "customModelType": { "enum": [ "CUSTOM_MODEL_TYPE_UNSPECIFIED", "VERSIONED_FOUNDATION", "FINE_TUNED" ], "enumDescriptions": [ "The model type is unspecified.", "The model is a versioned foundation model.", "The model is a finetuned foundation model." ], "type": "string", "description": "The type of custom model created by the user." }, "baseProcessorVersionId": { "description": "The base processor version ID for the custom model.", "type": "string" } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentIdGCSManagedDocumentId": { "properties": { "cwDocId": { "deprecated": true, "type": "string", "description": "ID of the document (indexed) managed by Content Warehouse." }, "gcsUri": { "type": "string", "description": "Required. The Cloud Storage URI where the actual document is stored." } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentIdGCSManagedDocumentId", "description": "Identifies a document uniquely within the scope of a dataset in the user-managed Cloud Storage option." }, "GoogleCloudDocumentaiV1beta3ProcessorVersion": { "type": "object", "description": "A processor version is an implementation of a processor. Each processor can have multiple versions, pretrained by Google internally or uptrained by the customer. A processor can only have one default version at a time. Its document-processing behavior is defined by that version.", "properties": { "kmsKeyName": { "description": "Output only. The KMS key name used for encryption.", "readOnly": true, "type": "string" }, "state": { "readOnly": true, "enum": [ "STATE_UNSPECIFIED", "DEPLOYED", "DEPLOYING", "UNDEPLOYED", "UNDEPLOYING", "CREATING", "DELETING", "FAILED", "IMPORTING" ], "enumDescriptions": [ "The processor version is in an unspecified state.", "The processor version is deployed and can be used for processing.", "The processor version is being deployed.", "The processor version is not deployed and cannot be used for processing.", "The processor version is being undeployed.", "The processor version is being created.", "The processor version is being deleted.", "The processor version failed and is in an indeterminate state.", "The processor version is being imported." ], "type": "string", "description": "Output only. The state of the processor version." }, "googleManaged": { "readOnly": true, "type": "boolean", "description": "Output only. Denotes that this `ProcessorVersion` is managed by Google." }, "createTime": { "description": "Output only. The time the processor version was created.", "type": "string", "format": "google-datetime", "readOnly": true }, "name": { "description": "Identifier. The resource name of the processor version. Format: `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processor_version}`", "type": "string" }, "satisfiesPzs": { "type": "boolean", "readOnly": true, "description": "Output only. Reserved for future use." }, "deprecationInfo": { "readOnly": true, "description": "Output only. If set, information about the eventual deprecation of this version.", "$ref": "GoogleCloudDocumentaiV1beta3ProcessorVersionDeprecationInfo" }, "satisfiesPzi": { "description": "Output only. Reserved for future use.", "type": "boolean", "readOnly": true }, "documentSchema": { "description": "Output only. The schema of the processor version. Describes the output.", "readOnly": true, "$ref": "GoogleCloudDocumentaiV1beta3DocumentSchema" }, "modelType": { "enumDescriptions": [ "The processor version has unspecified model type.", "The processor version has generative model type.", "The processor version has custom model type." ], "enum": [ "MODEL_TYPE_UNSPECIFIED", "MODEL_TYPE_GENERATIVE", "MODEL_TYPE_CUSTOM" ], "readOnly": true, "type": "string", "description": "Output only. The model type of this processor version." }, "genAiModelInfo": { "description": "Output only. Information about Generative AI model-based processor versions.", "readOnly": true, "$ref": "GoogleCloudDocumentaiV1beta3ProcessorVersionGenAiModelInfo" }, "latestEvaluation": { "$ref": "GoogleCloudDocumentaiV1beta3EvaluationReference", "readOnly": true, "description": "Output only. The most recently invoked evaluation for the processor version." }, "groundingSettings": { "$ref": "GoogleCloudDocumentaiV1beta3GroundingSettings", "description": "Output only. The grounding settings of the processor version. This can only be set using TrainProcessorVersionRequest to override the default grounding settings.", "readOnly": true }, "displayName": { "description": "The display name of the processor version.", "type": "string" }, "kmsKeyVersionName": { "description": "Output only. The KMS key version with which data is encrypted.", "readOnly": true, "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3ProcessorVersion" }, "GoogleCloudDocumentaiV1beta3Vertex": { "description": "A vertex represents a 2D point in the image. NOTE: the vertex coordinates are in the same scale as the original image.", "properties": { "x": { "description": "X coordinate.", "type": "integer", "format": "int32" }, "y": { "format": "int32", "description": "Y coordinate (starts from the top of the image).", "type": "integer" } }, "id": "GoogleCloudDocumentaiV1beta3Vertex", "type": "object" }, "GoogleCloudDocumentaiV1beta3UpdateDatasetOperationMetadata": { "type": "object", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "id": "GoogleCloudDocumentaiV1beta3UpdateDatasetOperationMetadata" }, "GoogleCloudDocumentaiV1beta3ImportDocumentsResponse": { "type": "object", "description": "Response of the import document operation.", "id": "GoogleCloudDocumentaiV1beta3ImportDocumentsResponse", "properties": {} }, "GoogleCloudDocumentaiV1beta3DatasetGCSManagedConfig": { "properties": { "gcsPrefix": { "$ref": "GoogleCloudDocumentaiV1beta3GcsPrefix", "description": "Required. The Cloud Storage URI (a directory) where the documents belonging to the dataset must be stored." } }, "id": "GoogleCloudDocumentaiV1beta3DatasetGCSManagedConfig", "type": "object", "description": "Configuration specific to the Cloud Storage-based implementation." }, "GoogleCloudDocumentaiUiv1beta3ResyncDatasetMetadataDatasetResyncStatus": { "id": "GoogleCloudDocumentaiUiv1beta3ResyncDatasetMetadataDatasetResyncStatus", "properties": { "datasetInconsistencyType": { "type": "string", "description": "The type of the inconsistency of the dataset.", "enumDescriptions": [ "Default value.", "The marker file under the dataset folder is not found." ], "enum": [ "DATASET_INCONSISTENCY_TYPE_UNSPECIFIED", "DATASET_INCONSISTENCY_TYPE_NO_STORAGE_MARKER" ] }, "status": { "description": "The status of resyncing the dataset with regards to the detected inconsistency. Empty if ResyncDatasetRequest.validate_only is `true`.", "$ref": "GoogleRpcStatus" } }, "type": "object", "description": "Resync status against inconsistency types on the dataset level." }, "GoogleCloudDocumentaiUiv1beta3EvaluationMetrics": { "type": "object", "description": "Evaluation metrics, either in aggregate or about a specific entity.", "properties": { "recall": { "type": "number", "description": "The calculated recall.", "format": "float" }, "totalDocumentsCount": { "format": "int32", "type": "integer", "description": "The amount of documents that had an occurrence of this label." }, "f1Score": { "type": "number", "format": "float", "description": "The calculated F1 score." }, "groundTruthOccurrencesCount": { "type": "integer", "format": "int32", "description": "The amount of occurrences in ground truth documents." }, "predictedDocumentCount": { "description": "The amount of documents with a predicted occurrence.", "format": "int32", "type": "integer" }, "truePositivesCount": { "type": "integer", "format": "int32", "description": "The amount of true positives." }, "falsePositivesCount": { "format": "int32", "type": "integer", "description": "The amount of false positives." }, "groundTruthDocumentCount": { "description": "The amount of documents with a ground truth occurrence.", "type": "integer", "format": "int32" }, "predictedOccurrencesCount": { "type": "integer", "description": "The amount of occurrences in predicted documents.", "format": "int32" }, "precision": { "format": "float", "type": "number", "description": "The calculated precision." }, "falseNegativesCount": { "format": "int32", "type": "integer", "description": "The amount of false negatives." } }, "id": "GoogleCloudDocumentaiUiv1beta3EvaluationMetrics" }, "GoogleTypePostalAddress": { "type": "object", "properties": { "postalCode": { "type": "string", "description": "Optional. Postal code of the address. Not all countries use or require postal codes to be present, but where they are used, they may trigger additional validation with other parts of the address (for example, state or zip code validation in the United States)." }, "regionCode": { "type": "string", "description": "Required. CLDR region code of the country/region of the address. This is never inferred and it is up to the user to ensure the value is correct. See https://cldr.unicode.org/ and https://www.unicode.org/cldr/charts/30/supplemental/territory_information.html for details. Example: \"CH\" for Switzerland." }, "revision": { "description": "The schema revision of the `PostalAddress`. This must be set to 0, which is the latest revision. All new revisions **must** be backward compatible with old revisions.", "type": "integer", "format": "int32" }, "languageCode": { "description": "Optional. BCP-47 language code of the contents of this address (if known). This is often the UI language of the input form or is expected to match one of the languages used in the address' country/region, or their transliterated equivalents. This can affect formatting in certain countries, but is not critical to the correctness of the data and will never affect any validation or other non-formatting related operations. If this value is not known, it should be omitted (rather than specifying a possibly incorrect default). Examples: \"zh-Hant\", \"ja\", \"ja-Latn\", \"en\".", "type": "string" }, "recipients": { "description": "Optional. The recipient at the address. This field may, under certain circumstances, contain multiline information. For example, it might contain \"care of\" information.", "type": "array", "items": { "type": "string" } }, "sortingCode": { "description": "Optional. Additional, country-specific, sorting code. This is not used in most regions. Where it is used, the value is either a string like \"CEDEX\", optionally followed by a number (for example, \"CEDEX 7\"), or just a number alone, representing the \"sector code\" (Jamaica), \"delivery area indicator\" (Malawi) or \"post office indicator\" (Côte d'Ivoire).", "type": "string" }, "sublocality": { "type": "string", "description": "Optional. Sublocality of the address. For example, this can be a neighborhood, borough, or district." }, "locality": { "description": "Optional. Generally refers to the city or town portion of the address. Examples: US city, IT comune, UK post town. In regions of the world where localities are not well defined or do not fit into this structure well, leave `locality` empty and use `address_lines`.", "type": "string" }, "addressLines": { "items": { "type": "string" }, "description": "Unstructured address lines describing the lower levels of an address. Because values in `address_lines` do not have type information and may sometimes contain multiple values in a single field (for example, \"Austin, TX\"), it is important that the line order is clear. The order of address lines should be \"envelope order\" for the country or region of the address. In places where this can vary (for example, Japan), `address_language` is used to make it explicit (for example, \"ja\" for large-to-small ordering and \"ja-Latn\" or \"en\" for small-to-large). In this way, the most specific line of an address can be selected based on the language. The minimum permitted structural representation of an address consists of a `region_code` with all remaining information placed in the `address_lines`. It would be possible to format such an address very approximately without geocoding, but no semantic reasoning could be made about any of the address components until it was at least partially resolved. Creating an address only containing a `region_code` and `address_lines` and then geocoding is the recommended way to handle completely unstructured addresses (as opposed to guessing which parts of the address should be localities or administrative areas).", "type": "array" }, "organization": { "description": "Optional. The name of the organization at the address.", "type": "string" }, "administrativeArea": { "type": "string", "description": "Optional. Highest administrative subdivision which is used for postal addresses of a country or region. For example, this can be a state, a province, an oblast, or a prefecture. For Spain, this is the province and not the autonomous community (for example, \"Barcelona\" and not \"Catalonia\"). Many countries don't use an administrative area in postal addresses. For example, in Switzerland, this should be left unpopulated." } }, "id": "GoogleTypePostalAddress", "description": "Represents a postal address, such as for postal delivery or payments addresses. With a postal address, a postal service can deliver items to a premise, P.O. box, or similar. A postal address is not intended to model geographical locations like roads, towns, or mountains. In typical usage, an address would be created by user input or from importing existing data, depending on the type of process. Advice on address input or editing: - Use an internationalization-ready address widget such as https://github.com/google/libaddressinput. - Users should not be presented with UI elements for input or editing of fields outside countries where that field is used. For more guidance on how to use this schema, see: https://support.google.com/business/answer/6397478." }, "GoogleCloudDocumentaiV1beta3DocumentOutputConfigGcsOutputConfig": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentOutputConfigGcsOutputConfig", "description": "The configuration used when outputting documents.", "properties": { "shardingConfig": { "description": "Specifies the sharding config for the output document.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentOutputConfigGcsOutputConfigShardingConfig" }, "fieldMask": { "description": "Specifies which fields to include in the output documents. Only supports top level document and pages field so it must be in the form of `{document_field_name}` or `pages.{page_field_name}`.", "type": "string", "format": "google-fieldmask" }, "gcsUri": { "description": "The Cloud Storage uri (a directory) of the output.", "type": "string" } } }, "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTableRow": { "id": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTableRow", "type": "object", "description": "Represents a row in a table.", "properties": { "cells": { "description": "A table row is a list of table cells.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTableCell" } } } }, "GoogleCloudDocumentaiV1beta3DocumentEntityRelation": { "properties": { "objectId": { "type": "string", "description": "Object entity id." }, "relation": { "type": "string", "description": "Relationship description." }, "subjectId": { "description": "Subject entity id.", "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentEntityRelation", "type": "object", "description": "Relationship between Entities." }, "GoogleCloudDocumentaiV1EnableProcessorResponse": { "description": "Response message for the EnableProcessor method. Intentionally empty proto for adding fields in future.", "properties": {}, "id": "GoogleCloudDocumentaiV1EnableProcessorResponse", "type": "object" }, "GoogleCloudDocumentaiV1beta3Barcode": { "properties": { "format": { "description": "Format of a barcode. The supported formats are: - `CODE_128`: Code 128 type. - `CODE_39`: Code 39 type. - `CODE_93`: Code 93 type. - `CODABAR`: Codabar type. - `DATA_MATRIX`: 2D Data Matrix type. - `ITF`: ITF type. - `EAN_13`: EAN-13 type. - `EAN_8`: EAN-8 type. - `QR_CODE`: 2D QR code type. - `UPC_A`: UPC-A type. - `UPC_E`: UPC-E type. - `PDF417`: PDF417 type. - `AZTEC`: 2D Aztec code type. - `DATABAR`: GS1 DataBar code type.", "type": "string" }, "rawValue": { "description": "Raw value encoded in the barcode. For example: `'MEBKM:TITLE:Google;URL:https://www.google.com;;'`.", "type": "string" }, "valueFormat": { "description": "Value format describes the format of the value that a barcode encodes. The supported formats are: - `CONTACT_INFO`: Contact information. - `EMAIL`: Email address. - `ISBN`: ISBN identifier. - `PHONE`: Phone number. - `PRODUCT`: Product. - `SMS`: SMS message. - `TEXT`: Text string. - `URL`: URL address. - `WIFI`: Wifi information. - `GEO`: Geo-localization. - `CALENDAR_EVENT`: Calendar event. - `DRIVER_LICENSE`: Driver's license.", "type": "string" } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3Barcode", "description": "Encodes the detailed information of a barcode." }, "GoogleCloudDocumentaiV1beta3SummaryOptions": { "properties": { "format": { "type": "string", "enum": [ "FORMAT_UNSPECIFIED", "PARAGRAPH", "BULLETS" ], "enumDescriptions": [ "Default.", "Format the output in paragraphs.", "Format the output in bullets." ], "description": "The format the summary should be in." }, "length": { "enum": [ "LENGTH_UNSPECIFIED", "BRIEF", "MODERATE", "COMPREHENSIVE" ], "description": "How long the summary should be.", "type": "string", "enumDescriptions": [ "Default.", "A brief summary of one or two sentences.", "A paragraph-length summary.", "The longest option available." ] } }, "description": "Metadata for document summarization.", "id": "GoogleCloudDocumentaiV1beta3SummaryOptions", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3Processor": { "description": "The first-class citizen for Document AI. Each processor defines how to extract structural information from a document.", "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3Processor", "properties": { "activeSchemaVersion": { "type": "string", "description": "Optional. SchemaVersion used by the Processor. It is the same as Processor's DatasetSchema.schema_version Format is `projects/{project}/locations/{location}/schemas/{schema}/schemaVersions/{schema_version}." }, "processEndpoint": { "description": "Output only. Immutable. The http endpoint that can be called to invoke processing.", "readOnly": true, "type": "string" }, "displayName": { "description": "The display name of the processor.", "type": "string" }, "name": { "type": "string", "readOnly": true, "description": "Output only. Immutable. The resource name of the processor. Format: `projects/{project}/locations/{location}/processors/{processor}`" }, "createTime": { "description": "Output only. The time the processor was created.", "type": "string", "readOnly": true, "format": "google-datetime" }, "satisfiesPzi": { "description": "Output only. Reserved for future use.", "readOnly": true, "type": "boolean" }, "defaultProcessorVersion": { "description": "The default processor version.", "type": "string" }, "state": { "type": "string", "enum": [ "STATE_UNSPECIFIED", "ENABLED", "DISABLED", "ENABLING", "DISABLING", "CREATING", "FAILED", "DELETING" ], "enumDescriptions": [ "The processor is in an unspecified state.", "The processor is enabled, i.e., has an enabled version which can currently serve processing requests and all the feature dependencies have been successfully initialized.", "The processor is disabled.", "The processor is being enabled, will become `ENABLED` if successful.", "The processor is being disabled, will become `DISABLED` if successful.", "The processor is being created, will become either `ENABLED` (for successful creation) or `FAILED` (for failed ones). Once a processor is in this state, it can then be used for document processing, but the feature dependencies of the processor might not be fully created yet.", "The processor failed during creation or initialization of feature dependencies. The user should delete the processor and recreate one as all the functionalities of the processor are disabled.", "The processor is being deleted, will be removed if successful." ], "readOnly": true, "description": "Output only. The state of the processor." }, "type": { "description": "The processor type, such as: `OCR_PROCESSOR`, `INVOICE_PROCESSOR`. To get a list of processor types, see FetchProcessorTypes.", "type": "string" }, "satisfiesPzs": { "type": "boolean", "readOnly": true, "description": "Output only. Reserved for future use." }, "processorVersionAliases": { "type": "array", "readOnly": true, "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3ProcessorVersionAlias" }, "description": "Output only. The processor version aliases." }, "kmsKeyName": { "type": "string", "description": "The [KMS key](https://cloud.google.com/security-key-management) used for encryption and decryption in CMEK scenarios." } } }, "GoogleCloudDocumentaiV1beta3DocumentEntitiesRevision": { "id": "GoogleCloudDocumentaiV1beta3DocumentEntitiesRevision", "properties": { "provenance": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenance", "description": "Optional. The history of this revision." }, "revisionId": { "description": "The revision id.", "type": "string" }, "entityValidationOutput": { "description": "The entity validation output for this revision.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentEntityValidationOutput" }, "entities": { "description": "The entities in this revision.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentEntity" }, "type": "array" } }, "type": "object", "description": "Entity revision." }, "GoogleCloudDocumentaiUiv1beta3ImportDocumentsResponse": { "type": "object", "properties": {}, "description": "Response of the import document operation.", "id": "GoogleCloudDocumentaiUiv1beta3ImportDocumentsResponse" }, "GoogleCloudDocumentaiV1beta3DocumentChunkedDocument": { "description": "Represents the chunks that the document is divided into.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocument", "properties": { "chunks": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunk" }, "description": "List of chunks." } } }, "GoogleCloudDocumentaiV1beta3DocumentPageBlock": { "properties": { "provenance": { "description": "The history of this annotation.", "deprecated": true, "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenance" }, "detectedLanguages": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" }, "description": "A list of detected languages together with confidence." }, "layout": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout", "description": "Layout for Block." } }, "description": "A block has a set of lines (collected into paragraphs) that have a common line-spacing and orientation.", "id": "GoogleCloudDocumentaiV1beta3DocumentPageBlock", "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentEntityNormalizedValue": { "type": "object", "description": "Parsed and normalized entity value.", "id": "GoogleCloudDocumentaiV1beta3DocumentEntityNormalizedValue", "properties": { "datetimeValue": { "description": "DateTime value. Includes date, time, and timezone. See also: https://github.com/googleapis/googleapis/blob/master/google/type/datetime.proto", "$ref": "GoogleTypeDateTime" }, "floatValue": { "format": "float", "description": "Float value.", "type": "number" }, "moneyValue": { "$ref": "GoogleTypeMoney", "description": "Money value. See also: https://github.com/googleapis/googleapis/blob/master/google/type/money.proto" }, "addressValue": { "$ref": "GoogleTypePostalAddress", "description": "Postal address. See also: https://github.com/googleapis/googleapis/blob/master/google/type/postal_address.proto" }, "dateValue": { "description": "Date value. Includes year, month, day. See also: https://github.com/googleapis/googleapis/blob/master/google/type/date.proto", "$ref": "GoogleTypeDate" }, "booleanValue": { "description": "Boolean value. Can be used for entities with binary values, or for checkboxes.", "type": "boolean" }, "text": { "description": "Optional. An optional field to store a normalized string. For some entity types, one of respective `structured_value` fields may also be populated. Also not all the types of `structured_value` will be normalized. For example, some processors may not generate `float` or `integer` normalized text by default. Below are sample formats mapped to structured values. - Money/Currency type (`money_value`) is in the ISO 4217 text format. - Date type (`date_value`) is in the ISO 8601 text format. - Datetime type (`datetime_value`) is in the ISO 8601 text format.", "type": "string" }, "integerValue": { "description": "Integer value.", "format": "int32", "type": "integer" }, "signatureValue": { "description": "A signature, which is a graphical representation of a person's name, often used to sign a document.", "type": "boolean" } } }, "GoogleCloudDocumentaiV1beta3TrainProcessorVersionMetadataDatasetValidation": { "properties": { "datasetErrors": { "type": "array", "description": "Error information for the dataset as a whole. A maximum of 10 dataset errors will be returned. A single dataset error is terminal for training.", "items": { "$ref": "GoogleRpcStatus" } }, "documentErrorCount": { "description": "The total number of document errors.", "format": "int32", "type": "integer" }, "documentErrors": { "type": "array", "description": "Error information pertaining to specific documents. A maximum of 10 document errors will be returned. Any document with errors will not be used throughout training.", "items": { "$ref": "GoogleRpcStatus" } }, "datasetErrorCount": { "type": "integer", "description": "The total number of dataset errors.", "format": "int32" } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionMetadataDatasetValidation", "description": "The dataset validation information. This includes any and all errors with documents and the dataset." }, "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutListEntry": { "properties": { "blocks": { "description": "A list entry is a list of blocks. Repeated blocks support further hierarchies and nested blocks.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlock" } } }, "description": "Represents an entry in the list.", "id": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutListEntry", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3HumanReviewValidationMetadata": { "properties": { "confidenceThreshold": { "type": "number", "description": "The confidence threshold if human review validation is enabled.", "format": "float" }, "enableValidation": { "type": "boolean", "description": "Whether to enable human review validation." } }, "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3HumanReviewValidationMetadata", "description": "Metadata for Human Review config." }, "GoogleCloudDocumentaiV1beta3SetDefaultProcessorVersionRequest": { "properties": { "defaultProcessorVersion": { "type": "string", "description": "Required. The resource name of child ProcessorVersion to use as default. Format: `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{version}`" } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3SetDefaultProcessorVersionRequest", "description": "Request message for the SetDefaultProcessorVersion method." }, "GoogleCloudDocumentaiV1beta3ReviewDocumentOperationMetadata": { "description": "The long-running operation metadata for the ReviewDocument method.", "properties": { "questionId": { "description": "The Crowd Compute question ID.", "type": "string" }, "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata" }, "state": { "type": "string", "enumDescriptions": [ "Unspecified state.", "Operation is still running.", "Operation is being cancelled.", "Operation succeeded.", "Operation failed.", "Operation is cancelled." ], "enum": [ "STATE_UNSPECIFIED", "RUNNING", "CANCELLING", "SUCCEEDED", "FAILED", "CANCELLED" ], "description": "Used only when Operation.done is false." }, "createTime": { "format": "google-datetime", "type": "string", "description": "The creation time of the operation." }, "updateTime": { "format": "google-datetime", "type": "string", "description": "The last update time of the operation." }, "stateMessage": { "description": "A message providing more details about the current state of processing. For example, the error message if the operation is failed.", "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3ReviewDocumentOperationMetadata", "type": "object" }, "GoogleCloudDocumentaiV1beta3BatchDatasetDocumentsIndividualDocumentIds": { "id": "GoogleCloudDocumentaiV1beta3BatchDatasetDocumentsIndividualDocumentIds", "description": "List of individual DocumentIds.", "properties": { "documentIds": { "type": "array", "description": "Required. List of Document IDs indicating where the actual documents are stored.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentId" } } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3FieldExtractionMetadata": { "properties": { "summaryOptions": { "$ref": "GoogleCloudDocumentaiV1beta3SummaryOptions", "description": "Summary options config." } }, "description": "Metadata for how this field value is extracted.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3FieldExtractionMetadata" }, "GoogleCloudDocumentaiUiv1beta3ExportDocumentsMetadata": { "description": "Metadata of the batch export documents operation.", "id": "GoogleCloudDocumentaiUiv1beta3ExportDocumentsMetadata", "type": "object", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" }, "individualExportStatuses": { "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3ExportDocumentsMetadataIndividualExportStatus" }, "description": "The list of response details of each document.", "type": "array" }, "splitExportStats": { "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3ExportDocumentsMetadataSplitExportStat" }, "description": "The list of statistics for each dataset split type.", "type": "array" } } }, "GoogleCloudDocumentaiV1beta3EntityTypeMetadata": { "type": "object", "description": "Metadata about an entity type.", "id": "GoogleCloudDocumentaiV1beta3EntityTypeMetadata", "properties": { "inactive": { "type": "boolean", "description": "Whether the entity type should be considered inactive." } } }, "GoogleCloudDocumentaiV1beta3DatasetUnmanagedDatasetConfig": { "description": "Configuration specific to an unmanaged dataset.", "type": "object", "properties": {}, "id": "GoogleCloudDocumentaiV1beta3DatasetUnmanagedDatasetConfig" }, "GoogleCloudDocumentaiV1TrainProcessorVersionMetadataDatasetValidation": { "id": "GoogleCloudDocumentaiV1TrainProcessorVersionMetadataDatasetValidation", "type": "object", "properties": { "datasetErrors": { "type": "array", "items": { "$ref": "GoogleRpcStatus" }, "description": "Error information for the dataset as a whole. A maximum of 10 dataset errors will be returned. A single dataset error is terminal for training." }, "documentErrorCount": { "format": "int32", "type": "integer", "description": "The total number of document errors." }, "datasetErrorCount": { "description": "The total number of dataset errors.", "type": "integer", "format": "int32" }, "documentErrors": { "type": "array", "description": "Error information pertaining to specific documents. A maximum of 10 document errors will be returned. Any document with errors will not be used throughout training.", "items": { "$ref": "GoogleRpcStatus" } } }, "description": "The dataset validation information. This includes any and all errors with documents and the dataset." }, "GoogleCloudDocumentaiUiv1beta3BatchDeleteDocumentsResponse": { "description": "Response of the delete documents operation.", "properties": {}, "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3BatchDeleteDocumentsResponse" }, "GoogleCloudDocumentaiV1TrainProcessorVersionResponse": { "properties": { "processorVersion": { "description": "The resource name of the processor version produced by training.", "type": "string" } }, "type": "object", "description": "The response for TrainProcessorVersion.", "id": "GoogleCloudDocumentaiV1TrainProcessorVersionResponse" }, "GoogleCloudDocumentaiUiv1beta3TrainProcessorVersionResponse": { "type": "object", "description": "The response for TrainProcessorVersion.", "id": "GoogleCloudDocumentaiUiv1beta3TrainProcessorVersionResponse", "properties": { "processorVersion": { "description": "The resource name of the processor version produced by training.", "type": "string" } } }, "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTableCell": { "id": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTableCell", "description": "Represents a cell in a table row.", "properties": { "colSpan": { "format": "int32", "type": "integer", "description": "How many columns this cell spans." }, "rowSpan": { "type": "integer", "description": "How many rows this cell spans.", "format": "int32" }, "blocks": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlock" }, "description": "A table cell is a list of blocks. Repeated blocks support further hierarchies and nested blocks.", "type": "array" } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3ProcessorVersionOverrides": { "id": "GoogleCloudDocumentaiV1beta3ProcessorVersionOverrides", "description": "Overrides to apply when creating a new processor version.", "type": "object", "properties": { "groundingSettings": { "description": "Optional. Grounding settings to override the default values.", "$ref": "GoogleCloudDocumentaiV1beta3GroundingSettings" } } }, "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutImageBlock": { "description": "Represents an image type block.", "properties": { "annotations": { "description": "Annotation of the image block.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentAnnotations" }, "imageText": { "type": "string", "description": "Text extracted from the image using OCR or alt text describing the image." }, "dataUri": { "description": "Optional. Data URI of the image. It is composed of four parts: a prefix (data:), a MIME type indicating the type of data, an optional base64 token if non-textual, and the data itself: data:,", "type": "string" }, "mimeType": { "description": "Mime type of the image. An IANA published [media type (MIME type)] (https://www.iana.org/assignments/media-types/media-types.xhtml).", "type": "string" }, "blobAssetId": { "description": "Optional. Asset ID of the inline image. If set, find the image content in the blob_assets field.", "type": "string" }, "gcsUri": { "description": "Optional. Google Cloud Storage URI of the image.", "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutImageBlock", "type": "object" }, "GoogleCloudDocumentaiV1beta3UndeployProcessorVersionResponse": { "description": "Response message for the UndeployProcessorVersion method.", "id": "GoogleCloudDocumentaiV1beta3UndeployProcessorVersionResponse", "type": "object", "properties": {} }, "GoogleCloudDocumentaiUiv1beta3ExportProcessorVersionMetadata": { "id": "GoogleCloudDocumentaiUiv1beta3ExportProcessorVersionMetadata", "description": "Metadata message associated with the ExportProcessorVersion operation.", "type": "object", "properties": { "commonMetadata": { "description": "The common metadata about the operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" } } }, "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunk": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunk", "properties": { "sourceBlockIds": { "description": "Unused.", "type": "array", "items": { "type": "string" } }, "pageSpan": { "description": "Page span of the chunk.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageSpan" }, "pageFooters": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageFooter" }, "type": "array", "description": "Page footers associated with the chunk." }, "chunkFields": { "description": "Chunk fields inside this chunk.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkField" }, "type": "array" }, "content": { "type": "string", "description": "Text content of the chunk." }, "chunkId": { "description": "ID of the chunk.", "type": "string" }, "pageHeaders": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageHeader" }, "type": "array", "description": "Page headers associated with the chunk." } }, "description": "Represents a chunk." }, "GoogleCloudDocumentaiUiv1beta3SchemaEntityType": { "id": "GoogleCloudDocumentaiUiv1beta3SchemaEntityType", "type": "object", "description": "EntityType is the wrapper of a label of the corresponding model with detailed attributes and limitations for entity-based processors. Multiple types can also compose a dependency tree to represent nested types.", "properties": { "hide": { "type": "boolean", "description": "If the entity type is hidden in the schema. This provides the functionality to temporally \"disable\" an entity without deleting it." }, "enumValues": { "description": "If specified, lists all the possible values for this entity.", "type": "array", "items": { "type": "string" } }, "description": { "description": "Description of the entity type.", "type": "string" }, "baseType": { "type": "string" }, "occurrenceType": { "enumDescriptions": [ "Unspecified occurrence type.", "The entity type will appear zero times or once.", "The entity type will appear zero or multiple times.", "The entity type will only appear exactly once.", "The entity type will appear once or more times." ], "type": "string", "description": "Occurrence type limits the number of times an entity type appears in the document.", "enum": [ "OCCURRENCE_TYPE_UNSPECIFIED", "OPTIONAL_ONCE", "OPTIONAL_MULTIPLE", "REQUIRED_ONCE", "REQUIRED_MULTIPLE" ] }, "source": { "enum": [ "SOURCE_UNSPECIFIED", "PREDEFINED", "USER_INPUT" ], "type": "string", "enumDescriptions": [ "Unspecified source.", "The entity type is in the predefined schema of a pretrained version of a processor.", "The entity type is added by the users either: - during an uptraining of an existing processor, or - during the process of creating a customized processor." ], "description": "Source of this entity type." }, "method": { "enumDescriptions": [ "Unspecified method. It defaults to `EXTRACT`.", "The entity's value is directly extracted as-is from the document text.", "The entity's value is derived through inference and is not necessarily an exact text extraction from the document.", "Attempts to extract the position and value as-is from the document text. If the value cannot be grounded in the text, it falls back to derivation through inference, for both position and value. Note: The inference fallback is currently only supported for boolean entities." ], "enum": [ "METHOD_UNSPECIFIED", "EXTRACT", "DERIVE", "RELAXED_EXTRACT" ], "type": "string", "description": "Specifies how the entity's value is obtained." }, "properties": { "description": "Describing the nested structure of an entity. An EntityType may consist of several other EntityTypes. For example, in a document there can be an EntityType `ID`, which consists of EntityType `name` and `address`, with corresponding attributes, such as TEXT for both types and ONCE for occurrence types.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3SchemaEntityType" } }, "type": { "type": "string", "description": "Name of the type. It must satisfy the following constraints: 1. Must be unique within the set of same level types (with case-insensitive match). 2. Maximum 64 characters. 3. Must start with a letter. 4. Allowed characters: ASCII letters [a-zA-Z], ASCII digits [0-9], or one of the following punctuation characters: * underscore '_' (recommended) * hyphen '-' (allowed, not recommended) * colon ':' (allowed, not recommended) NOTE: Whitespace characters are not allowed. 5. Cannot end with a punctuation character. 6. Cannot contain the following restricted strings: \"google\", \"DocumentAI\" (case-insensitive match). 7. A slash character '/' is reserved as a separator in flattened representations of nested entity types (e.g., \"line_item/amount\") in which case each part (e.g., \"line_item\", \"amount\") must comply with the rules defined above. We recommend using the snake case (\"snake_case\") in entity type names." } } }, "GoogleCloudDocumentaiV1beta3DocumentPageSymbol": { "type": "object", "properties": { "layout": { "description": "Layout for Symbol.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout" }, "detectedLanguages": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" }, "description": "A list of detected languages together with confidence." } }, "id": "GoogleCloudDocumentaiV1beta3DocumentPageSymbol", "description": "A detected symbol." }, "GoogleCloudDocumentaiV1beta3ImportProcessorVersionMetadata": { "id": "GoogleCloudDocumentaiV1beta3ImportProcessorVersionMetadata", "description": "The long-running operation metadata for the ImportProcessorVersion method.", "type": "object", "properties": { "commonMetadata": { "description": "The basic metadata for the long-running operation.", "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata" } } }, "GoogleCloudDocumentaiUiv1beta3SetDefaultProcessorVersionMetadata": { "type": "object", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "description": "The long-running operation metadata for the SetDefaultProcessorVersion method.", "id": "GoogleCloudDocumentaiUiv1beta3SetDefaultProcessorVersionMetadata" }, "GoogleCloudDocumentaiV1beta3DocumentSchemaEntityTypeProperty": { "properties": { "valueType": { "description": "A reference to the value type of the property. This type is subject to the same conventions as the `Entity.base_types` field.", "type": "string" }, "displayName": { "description": "User defined name for the property.", "type": "string" }, "description": { "description": "The description of the property. Could be used to provide more information about the property for model calls.", "type": "string" }, "name": { "type": "string", "description": "The name of the property. Follows the same guidelines as the EntityType name." }, "propertyMetadata": { "$ref": "GoogleCloudDocumentaiV1beta3PropertyMetadata", "description": "Any additional metadata about the property can be added here." }, "method": { "description": "Specifies how the entity's value is obtained.", "type": "string", "enum": [ "METHOD_UNSPECIFIED", "EXTRACT", "DERIVE", "RELAXED_EXTRACT" ], "enumDescriptions": [ "Unspecified method. It defaults to `EXTRACT`.", "The entity's value is directly extracted as-is from the document text.", "The entity's value is derived through inference and is not necessarily an exact text extraction from the document.", "Attempts to extract the position and value as-is from the document text. If the value cannot be grounded in the text, it falls back to derivation through inference, for both position and value. Note: The inference fallback is currently only supported for boolean entities." ] }, "occurrenceType": { "description": "Occurrence type limits the number of instances an entity type appears in the document.", "enum": [ "OCCURRENCE_TYPE_UNSPECIFIED", "OPTIONAL_ONCE", "OPTIONAL_MULTIPLE", "REQUIRED_ONCE", "REQUIRED_MULTIPLE" ], "type": "string", "enumDescriptions": [ "Unspecified occurrence type.", "There will be zero or one instance of this entity type. The same entity instance may be mentioned multiple times.", "The entity type will appear zero or multiple times.", "The entity type will only appear exactly once. The same entity instance may be mentioned multiple times.", "The entity type will appear once or more times." ] } }, "description": "Defines properties that can be part of the entity type.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentSchemaEntityTypeProperty" }, "GoogleCloudDocumentaiV1beta3EvaluationMetrics": { "description": "Evaluation metrics, either in aggregate or about a specific entity.", "properties": { "predictedDocumentCount": { "description": "The amount of documents with a predicted occurrence.", "format": "int32", "type": "integer" }, "groundTruthDocumentCount": { "description": "The amount of documents with a ground truth occurrence.", "format": "int32", "type": "integer" }, "totalDocumentsCount": { "description": "The amount of documents that had an occurrence of this label.", "type": "integer", "format": "int32" }, "predictedOccurrencesCount": { "description": "The amount of occurrences in predicted documents.", "format": "int32", "type": "integer" }, "recall": { "type": "number", "format": "float", "description": "The calculated recall." }, "falseNegativesCount": { "type": "integer", "format": "int32", "description": "The amount of false negatives." }, "f1Score": { "description": "The calculated F1 score.", "format": "float", "type": "number" }, "precision": { "description": "The calculated precision.", "format": "float", "type": "number" }, "groundTruthOccurrencesCount": { "type": "integer", "format": "int32", "description": "The amount of occurrences in ground truth documents." }, "falsePositivesCount": { "format": "int32", "type": "integer", "description": "The amount of false positives." }, "truePositivesCount": { "type": "integer", "description": "The amount of true positives.", "format": "int32" } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3EvaluationMetrics" }, "GoogleCloudDocumentaiV1beta3DocumentProvenance": { "type": "object", "description": "Structure to identify provenance relationships between annotations in different revisions.", "properties": { "type": { "enum": [ "OPERATION_TYPE_UNSPECIFIED", "ADD", "REMOVE", "UPDATE", "REPLACE", "EVAL_REQUESTED", "EVAL_APPROVED", "EVAL_SKIPPED" ], "type": "string", "enumDeprecated": [ false, false, false, false, false, true, true, true ], "description": "The type of provenance operation.", "enumDescriptions": [ "Operation type unspecified. If no operation is specified a provenance entry is simply used to match against a `parent`.", "Add an element.", "Remove an element identified by `parent`.", "Updates any fields within the given provenance scope of the message. It overwrites the fields rather than replacing them. Use this when you want to update a field value of an entity without also updating all the child properties.", "Currently unused. Replace an element identified by `parent`.", "Deprecated. Request human review for the element identified by `parent`.", "Deprecated. Element is reviewed and approved at human review, confidence will be set to 1.0.", "Deprecated. Element is skipped in the validation process." ] }, "revision": { "type": "integer", "deprecated": true, "description": "The index of the revision that produced this element.", "format": "int32" }, "id": { "type": "integer", "description": "The ID of this operation. Needs to be unique within the scope of the revision.", "deprecated": true, "format": "int32" }, "parents": { "type": "array", "description": "References to the original elements that are replaced.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenanceParent" } } }, "id": "GoogleCloudDocumentaiV1beta3DocumentProvenance" }, "GoogleCloudDocumentaiV1beta3DocumentSchema": { "properties": { "documentPrompt": { "description": "Optional. Document-level prompt provided by the user. This custom text is injected into the AI model's prompt to provide extra, document-wide guidance for processing.", "type": "string" }, "entityTypes": { "type": "array", "description": "Entity types of the schema.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentSchemaEntityType" } }, "metadata": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentSchemaMetadata", "description": "Metadata of the schema." }, "description": { "description": "Description of the schema.", "type": "string" }, "displayName": { "type": "string", "description": "Display name to show users." } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentSchema", "description": "The schema defines the output of the processed document by a processor." }, "GoogleCloudDocumentaiV1beta3EvaluateProcessorVersionRequest": { "id": "GoogleCloudDocumentaiV1beta3EvaluateProcessorVersionRequest", "description": "Evaluates the given ProcessorVersion against the supplied documents.", "properties": { "evaluationDocuments": { "description": "Optional. The documents used in the evaluation. If unspecified, use the processor's dataset as evaluation input.", "$ref": "GoogleCloudDocumentaiV1beta3BatchDocumentsInputConfig" } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3ProcessorVersionGenAiModelInfo": { "properties": { "foundationGenAiModelInfo": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessorVersionGenAiModelInfoFoundationGenAiModelInfo", "description": "Information for a pretrained Google-managed foundation model." }, "customGenAiModelInfo": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessorVersionGenAiModelInfoCustomGenAiModelInfo", "description": "Information for a custom Generative AI model created by the user." } }, "type": "object", "description": "Information about Generative AI model-based processor versions.", "id": "GoogleCloudDocumentaiV1beta3ProcessorVersionGenAiModelInfo" }, "GoogleCloudDocumentaiUiv1beta3EvaluateProcessorVersionMetadata": { "id": "GoogleCloudDocumentaiUiv1beta3EvaluateProcessorVersionMetadata", "type": "object", "description": "Metadata of the EvaluateProcessorVersion method.", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" } } }, "GoogleCloudDocumentaiUiv1beta3TrainProcessorVersionMetadata": { "id": "GoogleCloudDocumentaiUiv1beta3TrainProcessorVersionMetadata", "properties": { "testDatasetValidation": { "description": "The test dataset validation information.", "$ref": "GoogleCloudDocumentaiUiv1beta3TrainProcessorVersionMetadataDatasetValidation" }, "trainingDatasetValidation": { "description": "The training dataset validation information.", "$ref": "GoogleCloudDocumentaiUiv1beta3TrainProcessorVersionMetadataDatasetValidation" }, "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "description": "The metadata that represents a processor version being created.", "type": "object" }, "GoogleTypeDateTime": { "description": "Represents civil time (or occasionally physical time). This type can represent a civil time in one of a few possible ways: * When utc_offset is set and time_zone is unset: a civil time on a calendar day with a particular offset from UTC. * When time_zone is set and utc_offset is unset: a civil time on a calendar day in a particular time zone. * When neither time_zone nor utc_offset is set: a civil time on a calendar day in local time. The date is relative to the Proleptic Gregorian Calendar. If year, month, or day are 0, the DateTime is considered not to have a specific year, month, or day respectively. This type may also be used to represent a physical time if all the date and time fields are set and either case of the `time_offset` oneof is set. Consider using `Timestamp` message for physical time instead. If your use case also would like to store the user's timezone, that can be done in another field. This type is more flexible than some applications may want. Make sure to document and validate your application's limitations.", "type": "object", "id": "GoogleTypeDateTime", "properties": { "utcOffset": { "type": "string", "description": "UTC offset. Must be whole seconds, between -18 hours and +18 hours. For example, a UTC offset of -4:00 would be represented as { seconds: -14400 }.", "format": "google-duration" }, "month": { "type": "integer", "format": "int32", "description": "Optional. Month of year. Must be from 1 to 12, or 0 if specifying a datetime without a month." }, "timeZone": { "$ref": "GoogleTypeTimeZone", "description": "Time zone." }, "nanos": { "format": "int32", "description": "Optional. Fractions of seconds in nanoseconds. Must be from 0 to 999,999,999, defaults to 0.", "type": "integer" }, "year": { "description": "Optional. Year of date. Must be from 1 to 9999, or 0 if specifying a datetime without a year.", "type": "integer", "format": "int32" }, "seconds": { "format": "int32", "description": "Optional. Seconds of minutes of the time. Must normally be from 0 to 59, defaults to 0. An API may allow the value 60 if it allows leap-seconds.", "type": "integer" }, "day": { "format": "int32", "type": "integer", "description": "Optional. Day of month. Must be from 1 to 31 and valid for the year and month, or 0 if specifying a datetime without a day." }, "hours": { "description": "Optional. Hours of day in 24 hour format. Should be from 0 to 23, defaults to 0 (midnight). An API may choose to allow the value \"24:00:00\" for scenarios like business closing time.", "format": "int32", "type": "integer" }, "minutes": { "type": "integer", "format": "int32", "description": "Optional. Minutes of hour of day. Must be from 0 to 59, defaults to 0." } } }, "GoogleCloudDocumentaiV1SetDefaultProcessorVersionResponse": { "properties": {}, "description": "Response message for the SetDefaultProcessorVersion method.", "type": "object", "id": "GoogleCloudDocumentaiV1SetDefaultProcessorVersionResponse" }, "GoogleCloudDocumentaiV1beta3ListDocumentsRequest": { "type": "object", "properties": { "pageToken": { "description": "A page token, received from a previous `ListDocuments` call. Provide this to retrieve the subsequent page. When paginating, all other parameters provided to `ListDocuments` must match the call that provided the page token.", "type": "string" }, "pageSize": { "description": "The maximum number of documents to return. The service may return fewer than this value. If unspecified, at most 20 documents will be returned. The maximum value is 100; values above 100 will be coerced to 100.", "format": "int32", "type": "integer" }, "returnTotalSize": { "description": "Optional. Controls if the request requires a total size of matched documents. See ListDocumentsResponse.total_size. Enabling this flag may adversely impact performance. Defaults to false.", "type": "boolean" }, "filter": { "description": "Optional. Query to filter the documents based on https://google.aip.dev/160. ## Currently support query strings are: `SplitType=DATASET_SPLIT_TEST|DATASET_SPLIT_TRAIN|DATASET_SPLIT_UNASSIGNED` - `LabelingState=DOCUMENT_LABELED|DOCUMENT_UNLABELED|DOCUMENT_AUTO_LABELED` - `DisplayName=\\\"file_name.pdf\\\"` - `EntityType=abc/def` - `TagName=\\\"auto-labeling-running\\\"|\\\"sampled\\\"` Note: - Only `AND`, `=` and `!=` are supported. e.g. `DisplayName=file_name AND EntityType!=abc` IS supported. - Wildcard `*` is supported only in `DisplayName` filter - No duplicate filter keys are allowed, e.g. `EntityType=a AND EntityType=b` is NOT supported. - String match is case sensitive (for filter `DisplayName` & `EntityType`).", "type": "string" }, "skip": { "format": "int32", "type": "integer", "description": "Optional. Number of results to skip beginning from the `page_token` if provided. https://google.aip.dev/158#skipping-results. It must be a non-negative integer. Negative values will be rejected. Note that this is not the number of pages to skip. If this value causes the cursor to move past the end of results, ListDocumentsResponse.document_metadata and ListDocumentsResponse.next_page_token will be empty." } }, "id": "GoogleCloudDocumentaiV1beta3ListDocumentsRequest" }, "GoogleCloudDocumentaiV1beta3EvaluationEvaluationRevision": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3EvaluationEvaluationRevision", "description": "A revision of the evaluation.", "properties": { "allEntitiesMetrics": { "readOnly": true, "description": "Output only. Metrics for all the entities in aggregate.", "$ref": "GoogleCloudDocumentaiV1beta3EvaluationMultiConfidenceMetrics" }, "entityMetrics": { "type": "object", "readOnly": true, "description": "Output only. Metrics across confidence levels, for different entities.", "additionalProperties": { "$ref": "GoogleCloudDocumentaiV1beta3EvaluationMultiConfidenceMetrics" } }, "documentCounters": { "readOnly": true, "description": "Output only. Counters for the documents used in the evaluation.", "$ref": "GoogleCloudDocumentaiV1beta3EvaluationCounters" }, "revisionId": { "readOnly": true, "type": "string", "description": "Output only. The revision ID of the evaluation." } } }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleChildAlignmentRule": { "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleChildAlignmentRule", "description": "A rule that aligns specified child fields with a parent field.", "type": "object", "properties": { "parentField": { "description": "The full path of the parent field.", "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleField" }, "childFields": { "items": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleField" }, "description": "The child fields to be aligned within the parent field.", "type": "array" }, "alignmentRule": { "description": "The alignment rule to apply to the child fields.", "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleAlignmentRule" } } }, "GoogleCloudDocumentaiUiv1beta3HumanReviewLabelingMetadata": { "description": "Metadata for human review labeling config.", "id": "GoogleCloudDocumentaiUiv1beta3HumanReviewLabelingMetadata", "type": "object", "properties": { "enableNormalizationEditing": { "description": "Whether to enable normalization editing.", "type": "boolean" } } }, "GoogleCloudDocumentaiV1beta3GcsDocuments": { "properties": { "documents": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3GcsDocument" }, "description": "The list of documents.", "type": "array" } }, "type": "object", "description": "Specifies a set of documents on Cloud Storage.", "id": "GoogleCloudDocumentaiV1beta3GcsDocuments" }, "GoogleCloudDocumentaiV1beta3ImportProcessorVersionResponse": { "description": "The response message for the ImportProcessorVersion method.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3ImportProcessorVersionResponse", "properties": { "processorVersion": { "description": "The destination processor version name.", "type": "string" } } }, "GoogleCloudDocumentaiUiv1beta3ProcessorVersionDeprecationInfo": { "id": "GoogleCloudDocumentaiUiv1beta3ProcessorVersionDeprecationInfo", "description": "Information about the upcoming deprecation of this processor version.", "properties": { "deprecationTime": { "type": "string", "format": "google-datetime", "description": "The time at which this processor version will be deprecated." }, "replacementProcessorVersion": { "type": "string", "description": "If set, the processor version that will be used as a replacement." } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3ListSchemasResponse": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3ListSchemasResponse", "description": "Response message for ListSchemas.", "properties": { "schemas": { "description": "The list of Schemas.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3NextSchema" } }, "nextPageToken": { "type": "string", "description": "Points to the next Schema, otherwise empty." } } }, "GoogleCloudDocumentaiUiv1beta3UndeployProcessorVersionMetadata": { "description": "The long-running operation metadata for the UndeployProcessorVersion method.", "id": "GoogleCloudDocumentaiUiv1beta3UndeployProcessorVersionMetadata", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentEntityValidationOutput": { "type": "object", "description": "The output of the validation given the document and the validation rules.", "properties": { "passAllRules": { "description": "The overall result of the validation, true if all applicable rules are valid.", "type": "boolean" }, "validationResults": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentEntityValidationOutputValidationResult" }, "description": "The result of each validation rule.", "type": "array" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentEntityValidationOutput" }, "GoogleCloudDocumentaiV1beta3DocumentStyleFontSize": { "id": "GoogleCloudDocumentaiV1beta3DocumentStyleFontSize", "type": "object", "properties": { "size": { "type": "number", "format": "float", "description": "Font size for the text." }, "unit": { "type": "string", "description": "Unit for the font size. Follows CSS naming (such as `in`, `px`, and `pt`)." } }, "description": "Font size with unit." }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleField": { "type": "object", "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleField", "properties": { "fieldName": { "type": "string", "description": "The field name to validate. This can be a simple field name or a nested field one using the ':' (meant as an aggregator) or '*' (meant as foreach) operators." }, "defaultValue": { "description": "Default value to use if the field is not present. If the field is missing and the default value is not set, the validation run as if the field is not present in the validation logic.", "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleConstant" } } }, "GoogleLongrunningOperation": { "description": "This resource represents a long-running operation that is the result of a network API call.", "type": "object", "properties": { "metadata": { "description": "Service-specific metadata associated with the operation. It typically contains progress information and common metadata such as create time. Some services might not provide such metadata. Any method that returns a long-running operation should document the metadata type, if any.", "type": "object", "additionalProperties": { "description": "Properties of the object. Contains field @type with type URL.", "type": "any" } }, "response": { "description": "The normal, successful response of the operation. If the original method returns no data on success, such as `Delete`, the response is `google.protobuf.Empty`. If the original method is standard `Get`/`Create`/`Update`, the response should be the resource. For other methods, the response should have the type `XxxResponse`, where `Xxx` is the original method name. For example, if the original method name is `TakeSnapshot()`, the inferred response type is `TakeSnapshotResponse`.", "additionalProperties": { "type": "any", "description": "Properties of the object. Contains field @type with type URL." }, "type": "object" }, "error": { "description": "The error result of the operation in case of failure or cancellation.", "$ref": "GoogleRpcStatus" }, "name": { "description": "The server-assigned name, which is only unique within the same service that originally returns it. If you use the default HTTP mapping, the `name` should be a resource name ending with `operations/{unique_id}`.", "type": "string" }, "done": { "type": "boolean", "description": "If the value is `false`, it means the operation is still in progress. If `true`, the operation is completed, and either `error` or `response` is available." } }, "id": "GoogleLongrunningOperation" }, "GoogleCloudDocumentaiV1BatchProcessMetadataIndividualProcessStatus": { "description": "The status of a each individual document in the batch process.", "id": "GoogleCloudDocumentaiV1BatchProcessMetadataIndividualProcessStatus", "type": "object", "properties": { "status": { "description": "The status processing the document.", "$ref": "GoogleRpcStatus" }, "outputGcsDestination": { "type": "string", "description": "The Cloud Storage output destination (in the request as DocumentOutputConfig.GcsOutputConfig.gcs_uri) of the processed document if it was successful, otherwise empty." }, "humanReviewStatus": { "$ref": "GoogleCloudDocumentaiV1HumanReviewStatus", "description": "The status of human review on the processed document." }, "inputGcsSource": { "description": "The source of the document, same as the input_gcs_source field in the request when the batch process started.", "type": "string" } } }, "GoogleCloudDocumentaiUiv1beta3ResyncDatasetMetadataIndividualDocumentResyncStatus": { "description": "Resync status for each document per inconsistency type.", "type": "object", "properties": { "documentInconsistencyType": { "type": "string", "enum": [ "DOCUMENT_INCONSISTENCY_TYPE_UNSPECIFIED", "DOCUMENT_INCONSISTENCY_TYPE_INVALID_DOCPROTO", "DOCUMENT_INCONSISTENCY_TYPE_MISMATCHED_METADATA", "DOCUMENT_INCONSISTENCY_TYPE_NO_PAGE_IMAGE" ], "description": "The type of document inconsistency.", "enumDescriptions": [ "Default value.", "The document proto is invalid.", "Indexed docproto metadata is mismatched.", "The page image or thumbnails are missing." ] }, "status": { "$ref": "GoogleRpcStatus", "description": "The status of resyncing the document with regards to the detected inconsistency. Empty if ResyncDatasetRequest.validate_only is `true`." }, "documentId": { "description": "The document identifier.", "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentId" } }, "id": "GoogleCloudDocumentaiUiv1beta3ResyncDatasetMetadataIndividualDocumentResyncStatus" }, "GoogleCloudDocumentaiV1beta3DeleteProcessorMetadata": { "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3DeleteProcessorMetadata", "description": "The long-running operation metadata for the DeleteProcessor method." }, "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlock": { "id": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlock", "properties": { "imageBlock": { "description": "Block consisting of image content.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutImageBlock" }, "tableBlock": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTableBlock", "description": "Block consisting of table content/structure." }, "textBlock": { "description": "Block consisting of text content.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTextBlock" }, "listBlock": { "description": "Block consisting of list content/structure.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutListBlock" }, "pageSpan": { "description": "Page span of the block.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutPageSpan" }, "blockId": { "type": "string", "description": "ID of the block." }, "boundingBox": { "description": "Identifies the bounding box for the block.", "$ref": "GoogleCloudDocumentaiV1beta3BoundingPoly" } }, "type": "object", "description": "Represents a block. A block could be one of the various types (text, table, list) supported." }, "GoogleCloudDocumentaiV1beta3EvaluationMultiConfidenceMetrics": { "type": "object", "properties": { "auprc": { "type": "number", "description": "The calculated area under the precision recall curve (AUPRC), computed by integrating over all confidence thresholds.", "format": "float" }, "estimatedCalibrationError": { "description": "The Estimated Calibration Error (ECE) of the confidence of the predicted entities.", "format": "float", "type": "number" }, "metricsType": { "type": "string", "description": "The metrics type for the label.", "enumDescriptions": [ "The metrics type is unspecified. By default, metrics without a particular specification are for leaf entity types (i.e., top-level entity types without child types, or child types which are not parent types themselves).", "Indicates whether metrics for this particular label type represent an aggregate of metrics for other types instead of being based on actual TP/FP/FN values for the label type. Metrics for parent (i.e., non-leaf) entity types are an aggregate of metrics for their children." ], "enum": [ "METRICS_TYPE_UNSPECIFIED", "AGGREGATE" ] }, "confidenceLevelMetrics": { "type": "array", "description": "Metrics across confidence levels with fuzzy matching enabled.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3EvaluationConfidenceLevelMetrics" } }, "estimatedCalibrationErrorExact": { "type": "number", "format": "float", "description": "The ECE for the predicted entities with fuzzy matching disabled, i.e., exact matching only." }, "confidenceLevelMetricsExact": { "type": "array", "description": "Metrics across confidence levels with only exact matching.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3EvaluationConfidenceLevelMetrics" } }, "auprcExact": { "type": "number", "description": "The AUPRC for metrics with fuzzy matching disabled, i.e., exact matching only.", "format": "float" } }, "description": "Metrics across multiple confidence levels.", "id": "GoogleCloudDocumentaiV1beta3EvaluationMultiConfidenceMetrics" }, "GoogleCloudDocumentaiV1CommonOperationMetadata": { "description": "The common metadata for long running operations.", "type": "object", "properties": { "resource": { "description": "A related resource to this operation.", "type": "string" }, "state": { "enumDescriptions": [ "Unspecified state.", "Operation is still running.", "Operation is being cancelled.", "Operation succeeded.", "Operation failed.", "Operation is cancelled." ], "type": "string", "description": "The state of the operation.", "enum": [ "STATE_UNSPECIFIED", "RUNNING", "CANCELLING", "SUCCEEDED", "FAILED", "CANCELLED" ] }, "stateMessage": { "type": "string", "description": "A message providing more details about the current state of processing." }, "createTime": { "format": "google-datetime", "type": "string", "description": "The creation time of the operation." }, "updateTime": { "format": "google-datetime", "type": "string", "description": "The last update time of the operation." } }, "id": "GoogleCloudDocumentaiV1CommonOperationMetadata" }, "GoogleCloudDocumentaiUiv1beta3SchemaInferenceMetadata": { "properties": { "inferred": { "description": "True if is inferred by schema inference.", "type": "boolean" } }, "id": "GoogleCloudDocumentaiUiv1beta3SchemaInferenceMetadata", "type": "object", "description": "Metadata for schema inference. Only used on dataset.schema for schema inference, can be safely ignored elsewhere." }, "GoogleCloudDocumentaiV1beta3DocumentPage": { "type": "object", "properties": { "dimension": { "description": "Physical dimension of the page.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDimension" }, "imageQualityScores": { "description": "Image quality scores.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageImageQualityScores" }, "detectedBarcodes": { "description": "A list of detected barcodes.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedBarcode" }, "type": "array" }, "pageNumber": { "type": "integer", "description": "1-based index for current Page in a parent Document. Useful when a page is taken out of a Document for individual processing.", "format": "int32" }, "transforms": { "description": "Transformation matrices that were applied to the original document image to produce Page.image.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageMatrix" } }, "symbols": { "type": "array", "description": "A list of visually detected symbols on the page.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageSymbol" } }, "blocks": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageBlock" }, "description": "A list of visually detected text blocks on the page. A block has a set of lines (collected into paragraphs) that have a common line-spacing and orientation.", "type": "array" }, "visualElements": { "description": "A list of detected non-text visual elements, for example, checkbox, signature etc. on the page.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageVisualElement" } }, "formFields": { "type": "array", "description": "A list of visually detected form fields on the page.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageFormField" } }, "paragraphs": { "description": "A list of visually detected text paragraphs on the page. A collection of lines that a human would perceive as a paragraph.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageParagraph" } }, "tables": { "description": "A list of visually detected tables on the page.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageTable" } }, "image": { "description": "Rendered image for this page. This image is preprocessed to remove any skew, rotation, and distortions such that the annotation bounding boxes can be upright and axis-aligned.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageImage" }, "tokens": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageToken" }, "description": "A list of visually detected tokens on the page." }, "lines": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLine" }, "type": "array", "description": "A list of visually detected text lines on the page. A collection of tokens that a human would perceive as a line." }, "provenance": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenance", "deprecated": true, "description": "The history of this page." }, "detectedLanguages": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" }, "description": "A list of detected languages together with confidence.", "type": "array" }, "layout": { "description": "Layout for the page.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentPage", "description": "A page in a Document." }, "GoogleCloudDocumentaiV1beta3SchemaGenerationHistory": { "type": "object", "properties": { "iterations": { "description": "Required. Previous prompt-answers in a chronological order.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3SchemaGenerationIteration" }, "type": "array" } }, "id": "GoogleCloudDocumentaiV1beta3SchemaGenerationHistory", "description": "The history of schema generation iterations." }, "GoogleCloudDocumentaiUiv1beta3UndeployProcessorVersionResponse": { "description": "Response message for the UndeployProcessorVersion method.", "type": "object", "properties": {}, "id": "GoogleCloudDocumentaiUiv1beta3UndeployProcessorVersionResponse" }, "GoogleCloudDocumentaiV1beta3DocumentTextChange": { "description": "This message is used for text changes aka. OCR corrections.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentTextChange", "properties": { "changedText": { "type": "string", "description": "The text that replaces the text identified in the `text_anchor`." }, "textAnchor": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentTextAnchor", "description": "Provenance of the correction. Text anchor indexing into the Document.text. There can only be a single `TextAnchor.text_segments` element. If the start and end index of the text segment are the same, the text change is inserted before that index." }, "provenance": { "deprecated": true, "description": "The history of this annotation.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenance" } } } }, "GoogleCloudDocumentaiV1beta3NormalizedVertex": { "type": "object", "properties": { "x": { "format": "float", "description": "X coordinate.", "type": "number" }, "y": { "type": "number", "description": "Y coordinate (starts from the top of the image).", "format": "float" } }, "id": "GoogleCloudDocumentaiV1beta3NormalizedVertex", "description": "A vertex represents a 2D point in the image. NOTE: the normalized vertex coordinates are relative to the original image and range from 0 to 1." }, "GoogleCloudDocumentaiUiv1beta3AutoLabelDocumentsMetadata": { "properties": { "totalDocumentCount": { "format": "int32", "description": "Total number of the auto-labeling documents.", "type": "integer" }, "individualAutoLabelStatuses": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3AutoLabelDocumentsMetadataIndividualAutoLabelStatus" }, "description": "The list of individual auto-labeling statuses of the dataset documents." }, "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3AutoLabelDocumentsMetadata", "description": "Metadata of the auto-labeling documents operation." }, "GoogleCloudDocumentaiV1beta3DocumentPageTableTableRow": { "id": "GoogleCloudDocumentaiV1beta3DocumentPageTableTableRow", "type": "object", "properties": { "cells": { "type": "array", "description": "Cells that make up this row.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageTableTableCell" } } }, "description": "A row of table cells." }, "GoogleCloudDocumentaiV1beta3ListProcessorsResponse": { "description": "Response message for the ListProcessors method.", "id": "GoogleCloudDocumentaiV1beta3ListProcessorsResponse", "type": "object", "properties": { "nextPageToken": { "description": "Points to the next processor, otherwise empty.", "type": "string" }, "processors": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3Processor" }, "description": "The list of processors.", "type": "array" } } }, "GoogleCloudDocumentaiV1beta3RawDocuments": { "properties": { "documents": { "type": "array", "description": "Specifies raw document content and mime type.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3RawDocument" } } }, "type": "object", "description": "Specifies a set of raw documents.", "id": "GoogleCloudDocumentaiV1beta3RawDocuments" }, "GoogleCloudDocumentaiV1beta3DocumentPageImageQualityScores": { "description": "Image quality scores for the page image.", "id": "GoogleCloudDocumentaiV1beta3DocumentPageImageQualityScores", "type": "object", "properties": { "detectedDefects": { "description": "A list of detected defects.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageImageQualityScoresDetectedDefect" } }, "qualityScore": { "description": "The overall quality score. Range `[0, 1]` where `1` is perfect quality.", "type": "number", "format": "float" } } }, "GoogleCloudDocumentaiV1beta3DocumentRevision": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentRevision", "description": "Contains past or forward revisions of this document.", "properties": { "id": { "type": "string", "description": "ID of the revision, internally generated by doc proto storage. Unique within the context of the document." }, "humanReview": { "description": "Human Review information of this revision.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentRevisionHumanReview" }, "createTime": { "format": "google-datetime", "type": "string", "description": "The time that the revision was created, internally generated by doc proto storage at the time of create." }, "processor": { "description": "If the annotation was made by processor identify the processor by its resource name.", "type": "string" }, "parentIds": { "type": "array", "description": "The revisions that this revision is based on. Must include all the ids that have anything to do with this revision - eg. there are `provenance.parent.revision` fields that index into this field.", "items": { "type": "string" } }, "parent": { "items": { "format": "int32", "type": "integer" }, "description": "The revisions that this revision is based on. This can include one or more parent (when documents are merged.) This field represents the index into the `revisions` field.", "type": "array", "deprecated": true }, "agent": { "type": "string", "description": "If the change was made by a person specify the name or ID of that person." } } }, "GoogleCloudDocumentaiV1beta3ListProcessorTypesResponse": { "description": "Response message for the ListProcessorTypes method.", "id": "GoogleCloudDocumentaiV1beta3ListProcessorTypesResponse", "properties": { "processorTypes": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessorType" }, "description": "The processor types." }, "nextPageToken": { "type": "string", "description": "Points to the next page, otherwise empty." } }, "type": "object" }, "GoogleTypeMoney": { "type": "object", "id": "GoogleTypeMoney", "properties": { "currencyCode": { "description": "The three-letter currency code defined in ISO 4217.", "type": "string" }, "units": { "type": "string", "format": "int64", "description": "The whole units of the amount. For example if `currencyCode` is `\"USD\"`, then 1 unit is one US dollar." }, "nanos": { "description": "Number of nano (10^-9) units of the amount. The value must be between -999,999,999 and +999,999,999 inclusive. If `units` is positive, `nanos` must be positive or zero. If `units` is zero, `nanos` can be positive, zero, or negative. If `units` is negative, `nanos` must be negative or zero. For example $-1.75 is represented as `units`=-1 and `nanos`=-750,000,000.", "format": "int32", "type": "integer" } }, "description": "Represents an amount of money with its currency type." }, "GoogleCloudDocumentaiV1beta3NextSchema": { "properties": { "labels": { "type": "object", "description": "Optional. The {{gcp_name_short}} labels for the Schema.", "additionalProperties": { "type": "string" } }, "displayName": { "type": "string", "description": "Required. The user-defined name of the Schema." }, "updateTime": { "readOnly": true, "format": "google-datetime", "type": "string", "description": "Output only. The time when the Schema was last updated." }, "name": { "type": "string", "description": "Identifier. The resource name of the Schema. Format: `projects/{project}/locations/{location}/schemas/{schema}`" }, "createTime": { "description": "Output only. The time when the Schema was created.", "readOnly": true, "format": "google-datetime", "type": "string" } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3NextSchema", "description": "NextSchema is a collection of SchemaVersions." }, "GoogleCloudDocumentaiV1beta3DocumentTextAnchor": { "description": "Text reference indexing into the Document.text.", "type": "object", "properties": { "textSegments": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentTextAnchorTextSegment" }, "description": "The text segments from the Document.text." }, "content": { "type": "string", "description": "Contains the content of the text span so that users do not have to look it up in the text_segments. It is always populated for formFields." } }, "id": "GoogleCloudDocumentaiV1beta3DocumentTextAnchor" }, "GoogleCloudDocumentaiV1beta3EvaluationCounters": { "properties": { "failedDocumentsCount": { "type": "integer", "description": "How many documents were not included in the evaluation as Document AI failed to process them.", "format": "int32" }, "inputDocumentsCount": { "description": "How many documents were sent for evaluation.", "type": "integer", "format": "int32" }, "invalidDocumentsCount": { "type": "integer", "format": "int32", "description": "How many documents were not included in the evaluation as they didn't pass validation." }, "evaluatedDocumentsCount": { "format": "int32", "description": "How many documents were used in the evaluation.", "type": "integer" } }, "id": "GoogleCloudDocumentaiV1beta3EvaluationCounters", "type": "object", "description": "Evaluation counters for the documents that were used." }, "GoogleLongrunningListOperationsResponse": { "description": "The response message for Operations.ListOperations.", "id": "GoogleLongrunningListOperationsResponse", "type": "object", "properties": { "unreachable": { "items": { "type": "string" }, "description": "Unordered list. Unreachable resources. Populated when the request sets `ListOperationsRequest.return_partial_success` and reads across collections. For example, when attempting to list all resources across all supported locations.", "type": "array" }, "nextPageToken": { "type": "string", "description": "The standard List next-page token." }, "operations": { "items": { "$ref": "GoogleLongrunningOperation" }, "type": "array", "description": "A list of operations that matches the specified filter in the request." } } }, "GoogleCloudDocumentaiUiv1beta3DeployProcessorVersionMetadata": { "type": "object", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" } }, "id": "GoogleCloudDocumentaiUiv1beta3DeployProcessorVersionMetadata", "description": "The long-running operation metadata for the DeployProcessorVersion method." }, "GoogleCloudDocumentaiV1beta3UndeployProcessorVersionMetadata": { "description": "The long-running operation metadata for the UndeployProcessorVersion method.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3UndeployProcessorVersionMetadata", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } } }, "GoogleCloudDocumentaiUiv1beta3AutoLabelDocumentsResponse": { "description": "The response proto of AutoLabelDocuments method.", "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3AutoLabelDocumentsResponse", "properties": {} }, "GoogleCloudDocumentaiV1beta3DocumentPageLayout": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentPageLayout", "description": "Visual element describing a layout unit on a page.", "properties": { "textAnchor": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentTextAnchor", "description": "Text anchor indexing into the Document.text." }, "confidence": { "format": "float", "description": "Confidence of the current Layout within context of the object this layout is for. For example, confidence can be for a single token, a table, a visual element, etc. depending on context. Range `[0, 1]`.", "type": "number" }, "orientation": { "enumDescriptions": [ "Unspecified orientation.", "Orientation is aligned with page up.", "Orientation is aligned with page right. Turn the head 90 degrees clockwise from upright to read.", "Orientation is aligned with page down. Turn the head 180 degrees from upright to read.", "Orientation is aligned with page left. Turn the head 90 degrees counterclockwise from upright to read." ], "enum": [ "ORIENTATION_UNSPECIFIED", "PAGE_UP", "PAGE_RIGHT", "PAGE_DOWN", "PAGE_LEFT" ], "description": "Detected orientation for the Layout.", "type": "string" }, "boundingPoly": { "description": "The bounding polygon for the Layout.", "$ref": "GoogleCloudDocumentaiV1beta3BoundingPoly" } } }, "GoogleCloudDocumentaiV1beta3DocumentPageParagraph": { "properties": { "detectedLanguages": { "type": "array", "description": "A list of detected languages together with confidence.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" } }, "layout": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout", "description": "Layout for Paragraph." }, "provenance": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenance", "description": "The history of this annotation.", "deprecated": true } }, "id": "GoogleCloudDocumentaiV1beta3DocumentPageParagraph", "type": "object", "description": "A collection of lines that a human would perceive as a paragraph." }, "GoogleCloudDocumentaiV1beta3DocumentPageTokenStyleInfo": { "type": "object", "description": "Font and other text style attributes.", "id": "GoogleCloudDocumentaiV1beta3DocumentPageTokenStyleInfo", "properties": { "superscript": { "type": "boolean", "description": "Whether the text is a superscript. This feature is not supported yet." }, "handwritten": { "description": "Whether the text is handwritten.", "type": "boolean" }, "fontType": { "type": "string", "description": "Name or style of the font." }, "subscript": { "description": "Whether the text is a subscript. This feature is not supported yet.", "type": "boolean" }, "underlined": { "description": "Whether the text is underlined.", "type": "boolean" }, "fontSize": { "description": "Font size in points (`1` point is `¹⁄₇₂` inches).", "format": "int32", "type": "integer" }, "fontWeight": { "format": "int32", "description": "TrueType weight on a scale `100` (thin) to `1000` (ultra-heavy). Normal is `400`, bold is `700`.", "type": "integer" }, "bold": { "type": "boolean", "description": "Whether the text is bold (equivalent to font_weight is at least `700`)." }, "textColor": { "$ref": "GoogleTypeColor", "description": "Color of the text." }, "letterSpacing": { "type": "number", "description": "Letter spacing in points.", "format": "double" }, "italic": { "description": "Whether the text is italic.", "type": "boolean" }, "backgroundColor": { "description": "Color of the background.", "$ref": "GoogleTypeColor" }, "strikeout": { "description": "Whether the text is strikethrough. This feature is not supported yet.", "type": "boolean" }, "smallcaps": { "type": "boolean", "description": "Whether the text is in small caps. This feature is not supported yet." }, "pixelFontSize": { "description": "Font size in pixels, equal to _unrounded font_size_ * _resolution_ ÷ `72.0`.", "format": "double", "type": "number" } } }, "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTextBlock": { "id": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTextBlock", "properties": { "text": { "type": "string", "description": "Text content stored in the block." }, "type": { "description": "Type of the text in the block. Available options are: `paragraph`, `subtitle`, `heading-1`, `heading-2`, `heading-3`, `heading-4`, `heading-5`, `header`, `footer`.", "type": "string" }, "annotations": { "description": "Annotation of the text block.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentAnnotations" }, "blocks": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlock" }, "description": "A text block could further have child blocks. Repeated blocks support further hierarchies and nested blocks." } }, "type": "object", "description": "Represents a text type block." }, "GoogleCloudDocumentaiV1beta3GenerateSchemaVersionRequest": { "id": "GoogleCloudDocumentaiV1beta3GenerateSchemaVersionRequest", "type": "object", "description": "Request message for GenerateSchemaVersion.", "properties": { "gcsDocuments": { "description": "The set of documents placed on Cloud Storage.", "$ref": "GoogleCloudDocumentaiV1beta3GcsDocuments" }, "inlineDocuments": { "description": "The set of documents specified inline. For each document, its `uri` or `content` field must be set.", "$ref": "GoogleCloudDocumentaiV1beta3Documents" }, "rawDocuments": { "description": "The set of raw documents.", "$ref": "GoogleCloudDocumentaiV1beta3RawDocuments" }, "baseSchemaVersion": { "description": "The base schema version name to use for the schema generation. Format: `projects/{project}/locations/{location}/schemas/{schema}/schemaVersions/{schema_version}`", "type": "string" }, "generateSchemaVersionParams": { "description": "Optional. User specified parameters for the schema generation.", "$ref": "GoogleCloudDocumentaiV1beta3GenerateSchemaVersionRequestGenerateSchemaVersionParams" }, "gcsPrefix": { "description": "The common prefix of documents placed on Cloud Storage.", "$ref": "GoogleCloudDocumentaiV1beta3GcsPrefix" } } }, "GoogleCloudDocumentaiV1UndeployProcessorVersionResponse": { "type": "object", "description": "Response message for the UndeployProcessorVersion method.", "id": "GoogleCloudDocumentaiV1UndeployProcessorVersionResponse", "properties": {} }, "GoogleCloudDocumentaiV1beta3UpdateProcessorVersionMetadata": { "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata", "description": "The basic metadata for the long-running operation." } }, "description": "The long-running operation metadata for the UpdateProcessorVersion method.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3UpdateProcessorVersionMetadata" }, "GoogleCloudDocumentaiV1beta3Dataset": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3Dataset", "description": "A singleton resource under a Processor which configures a collection of documents.", "properties": { "state": { "type": "string", "description": "Required. State of the dataset. Ignored when updating dataset.", "enum": [ "STATE_UNSPECIFIED", "UNINITIALIZED", "INITIALIZING", "INITIALIZED" ], "enumDescriptions": [ "Default unspecified enum, should not be used.", "Dataset has not been initialized.", "Dataset is being initialized.", "Dataset has been initialized." ] }, "name": { "description": "Dataset resource name. Format: `projects/{project}/locations/{location}/processors/{processor}/dataset`", "type": "string" }, "unmanagedDatasetConfig": { "$ref": "GoogleCloudDocumentaiV1beta3DatasetUnmanagedDatasetConfig", "description": "Optional. Unmanaged dataset configuration. Use this configuration if the dataset documents are managed by the document service internally (not user-managed)." }, "satisfiesPzi": { "readOnly": true, "type": "boolean", "description": "Output only. Reserved for future use." }, "gcsManagedConfig": { "description": "Optional. User-managed Cloud Storage dataset configuration. Use this configuration if the dataset documents are stored under a user-managed Cloud Storage location.", "$ref": "GoogleCloudDocumentaiV1beta3DatasetGCSManagedConfig" }, "satisfiesPzs": { "type": "boolean", "description": "Output only. Reserved for future use.", "readOnly": true }, "documentWarehouseConfig": { "$ref": "GoogleCloudDocumentaiV1beta3DatasetDocumentWarehouseConfig", "deprecated": true, "description": "Optional. Deprecated. Warehouse-based dataset configuration is not supported." } } }, "GoogleCloudDocumentaiV1UndeployProcessorVersionMetadata": { "description": "The long-running operation metadata for the UndeployProcessorVersion method.", "type": "object", "id": "GoogleCloudDocumentaiV1UndeployProcessorVersionMetadata", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiV1CommonOperationMetadata" } } }, "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutListBlock": { "id": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutListBlock", "description": "Represents a list type block.", "properties": { "listEntries": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutListEntry" }, "description": "List entries that constitute a list block.", "type": "array" }, "type": { "description": "Type of the list_entries (if exist). Available options are `ordered` and `unordered`.", "type": "string" } }, "type": "object" }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleConstant": { "description": "The constant value used in the validation rules.", "type": "object", "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleConstant", "properties": { "floatValue": { "type": "number", "format": "float" } } }, "GoogleCloudDocumentaiV1DisableProcessorResponse": { "type": "object", "properties": {}, "id": "GoogleCloudDocumentaiV1DisableProcessorResponse", "description": "Response message for the DisableProcessor method. Intentionally empty proto for adding fields in future." }, "GoogleCloudDocumentaiV1beta3ListEvaluationsResponse": { "description": "The response from `ListEvaluations`.", "id": "GoogleCloudDocumentaiV1beta3ListEvaluationsResponse", "properties": { "evaluations": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3Evaluation" }, "description": "The evaluations requested." }, "nextPageToken": { "type": "string", "description": "A token, which can be sent as `page_token` to retrieve the next page. If this field is omitted, there are no subsequent pages." } }, "type": "object" }, "GoogleCloudDocumentaiV1HumanReviewStatus": { "properties": { "humanReviewOperation": { "type": "string", "description": "The name of the operation triggered by the processed document. This field is populated only when the state is `HUMAN_REVIEW_IN_PROGRESS`. It has the same response type and metadata as the long-running operation returned by ReviewDocument." }, "stateMessage": { "type": "string", "description": "A message providing more details about the human review state." }, "state": { "enumDescriptions": [ "Human review state is unspecified. Most likely due to an internal error.", "Human review is skipped for the document. This can happen because human review isn't enabled on the processor or the processing request has been set to skip this document.", "Human review validation is triggered and passed, so no review is needed.", "Human review validation is triggered and the document is under review.", "Some error happened during triggering human review, see the state_message for details." ], "type": "string", "description": "The state of human review on the processing request.", "enum": [ "STATE_UNSPECIFIED", "SKIPPED", "VALIDATION_PASSED", "IN_PROGRESS", "ERROR" ] } }, "description": "The status of human review on a processed document.", "type": "object", "id": "GoogleCloudDocumentaiV1HumanReviewStatus" }, "GoogleCloudDocumentaiV1beta3GcsPrefix": { "id": "GoogleCloudDocumentaiV1beta3GcsPrefix", "properties": { "gcsUriPrefix": { "type": "string", "description": "The URI prefix." } }, "type": "object", "description": "Specifies all documents on Cloud Storage with a common prefix." }, "GoogleCloudDocumentaiV1beta3DocumentMetadata": { "properties": { "datasetType": { "description": "Type of the dataset split to which the document belongs.", "enumDescriptions": [ "Default value if the enum is not set.", "Identifies the train documents.", "Identifies the test documents.", "Identifies the unassigned documents." ], "enum": [ "DATASET_SPLIT_TYPE_UNSPECIFIED", "DATASET_SPLIT_TRAIN", "DATASET_SPLIT_TEST", "DATASET_SPLIT_UNASSIGNED" ], "type": "string" }, "labelingState": { "description": "Labeling state of the document.", "enum": [ "DOCUMENT_LABELING_STATE_UNSPECIFIED", "DOCUMENT_LABELED", "DOCUMENT_UNLABELED", "DOCUMENT_AUTO_LABELED" ], "enumDescriptions": [ "Default value if the enum is not set.", "Document has been labeled.", "Document has not been labeled.", "Document has been auto-labeled." ], "type": "string" }, "displayName": { "description": "The display name of the document.", "type": "string" }, "documentId": { "description": "Document identifier.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentId" }, "pageCount": { "description": "Number of pages in the document.", "format": "int32", "type": "integer" } }, "description": "Metadata about a document.", "id": "GoogleCloudDocumentaiV1beta3DocumentMetadata", "type": "object" }, "GoogleCloudDocumentaiV1beta3GcsDocument": { "type": "object", "properties": { "gcsUri": { "description": "The Cloud Storage object uri.", "type": "string" }, "mimeType": { "type": "string", "description": "An IANA MIME type (RFC6838) of the content." } }, "id": "GoogleCloudDocumentaiV1beta3GcsDocument", "description": "Specifies a document stored on Cloud Storage." }, "GoogleCloudDocumentaiV1beta3DocumentPageImage": { "id": "GoogleCloudDocumentaiV1beta3DocumentPageImage", "type": "object", "description": "Rendered image contents for this page.", "properties": { "height": { "type": "integer", "description": "Height of the image in pixels.", "format": "int32" }, "content": { "type": "string", "format": "byte", "description": "Raw byte content of the image." }, "mimeType": { "description": "Encoding [media type (MIME type)](https://www.iana.org/assignments/media-types/media-types.xhtml) for the image.", "type": "string" }, "width": { "description": "Width of the image in pixels.", "type": "integer", "format": "int32" } } }, "GoogleCloudDocumentaiV1beta3ProcessorVersionGenAiModelInfoFoundationGenAiModelInfo": { "properties": { "minTrainLabeledDocuments": { "format": "int32", "type": "integer", "description": "The minimum number of labeled documents in the training dataset required for fine tuning." }, "finetuningAllowed": { "type": "boolean", "description": "Whether fine tuning is allowed for this base processor version." } }, "type": "object", "description": "Information for a pretrained Google-managed foundation model.", "id": "GoogleCloudDocumentaiV1beta3ProcessorVersionGenAiModelInfoFoundationGenAiModelInfo" }, "GoogleCloudDocumentaiV1beta3BatchDatasetDocuments": { "type": "object", "description": "Dataset documents that the batch operation will be applied to.", "id": "GoogleCloudDocumentaiV1beta3BatchDatasetDocuments", "properties": { "filter": { "description": "A filter matching the documents. Follows the same format and restriction as [google.cloud.documentai.master.ListDocumentsRequest.filter].", "type": "string" }, "individualDocumentIds": { "description": "Document identifiers.", "$ref": "GoogleCloudDocumentaiV1beta3BatchDatasetDocumentsIndividualDocumentIds" } } }, "GoogleCloudDocumentaiUiv1beta3FieldExtractionMetadata": { "type": "object", "description": "Metadata for how this field value is extracted.", "properties": { "entityQuery": { "description": "Entity query config.", "$ref": "GoogleCloudDocumentaiUiv1beta3FieldExtractionMetadataEntityQuery" }, "summaryOptions": { "description": "Summary options config.", "$ref": "GoogleCloudDocumentaiUiv1beta3SummaryOptions" } }, "id": "GoogleCloudDocumentaiUiv1beta3FieldExtractionMetadata" }, "GoogleRpcStatus": { "description": "The `Status` type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by [gRPC](https://github.com/grpc). Each `Status` message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the [API Design Guide](https://cloud.google.com/apis/design/errors).", "type": "object", "properties": { "message": { "description": "A developer-facing error message, which should be in English. Any user-facing error message should be localized and sent in the google.rpc.Status.details field, or localized by the client.", "type": "string" }, "code": { "description": "The status code, which should be an enum value of google.rpc.Code.", "format": "int32", "type": "integer" }, "details": { "type": "array", "items": { "type": "object", "additionalProperties": { "type": "any", "description": "Properties of the object. Contains field @type with type URL." } }, "description": "A list of messages that carry the error details. There is a common set of message types for APIs to use." } }, "id": "GoogleRpcStatus" }, "GoogleCloudDocumentaiV1beta3DeployProcessorVersionRequest": { "description": "Request message for the DeployProcessorVersion method.", "type": "object", "properties": {}, "id": "GoogleCloudDocumentaiV1beta3DeployProcessorVersionRequest" }, "GoogleCloudDocumentaiV1beta3ProcessorVersionAlias": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3ProcessorVersionAlias", "description": "Contains the alias and the aliased resource name of processor version.", "properties": { "alias": { "description": "The alias in the form of `processor_version` resource name.", "type": "string" }, "processorVersion": { "type": "string", "description": "The resource name of aliased processor version." } } }, "GoogleCloudDocumentaiV1beta3GenerateSchemaVersionRequestGenerateSchemaVersionParams": { "description": "The parameters for the schema generation.", "id": "GoogleCloudDocumentaiV1beta3GenerateSchemaVersionRequestGenerateSchemaVersionParams", "properties": { "prompt": { "type": "string", "description": "Optional. The prompt used for the schema generation." }, "history": { "$ref": "GoogleCloudDocumentaiV1beta3SchemaGenerationHistory", "description": "Optional. Previous prompt-answers in a chronological order." } }, "type": "object" }, "GoogleCloudDocumentaiUiv1beta3ImportDocumentsMetadataImportConfigValidationResult": { "description": "The validation status of each import config. Status is set to an error if there are no documents to import in the `import_config`, or `OK` if the operation will try to proceed with at least one document.", "properties": { "inputGcsSource": { "description": "The source Cloud Storage URI specified in the import config.", "type": "string" }, "status": { "$ref": "GoogleRpcStatus", "description": "The validation status of import config." } }, "id": "GoogleCloudDocumentaiUiv1beta3ImportDocumentsMetadataImportConfigValidationResult", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3DisableProcessorMetadata": { "description": "The long-running operation metadata for the DisableProcessor method.", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" } }, "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3DisableProcessorMetadata" }, "GoogleCloudDocumentaiV1beta3DocumentPageFormField": { "description": "A form field detected on the page.", "type": "object", "properties": { "fieldValue": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout", "description": "Layout for the FormField value." }, "provenance": { "description": "The history of this annotation.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenance" }, "nameDetectedLanguages": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" }, "type": "array", "description": "A list of detected languages for name together with confidence." }, "correctedKeyText": { "description": "Created for Labeling UI to export key text. If corrections were made to the text identified by the `field_name.text_anchor`, this field will contain the correction.", "type": "string" }, "fieldName": { "description": "Layout for the FormField name. For example, `Address`, `Email`, `Grand total`, `Phone number`, etc.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout" }, "valueType": { "type": "string", "description": "If the value is non-textual, this field represents the type. Current valid values are: - blank (this indicates the `field_value` is normal text) - `unfilled_checkbox` - `filled_checkbox`" }, "correctedValueText": { "description": "Created for Labeling UI to export value text. If corrections were made to the text identified by the `field_value.text_anchor`, this field will contain the correction.", "type": "string" }, "valueDetectedLanguages": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" }, "description": "A list of detected languages for value together with confidence.", "type": "array" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentPageFormField" }, "GoogleTypeDate": { "description": "Represents a whole or partial calendar date, such as a birthday. The time of day and time zone are either specified elsewhere or are insignificant. The date is relative to the Gregorian Calendar. This can represent one of the following: * A full date, with non-zero year, month, and day values. * A month and day, with a zero year (for example, an anniversary). * A year on its own, with a zero month and a zero day. * A year and month, with a zero day (for example, a credit card expiration date). Related types: * google.type.TimeOfDay * google.type.DateTime * google.protobuf.Timestamp", "id": "GoogleTypeDate", "properties": { "month": { "description": "Month of a year. Must be from 1 to 12, or 0 to specify a year without a month and day.", "type": "integer", "format": "int32" }, "day": { "type": "integer", "format": "int32", "description": "Day of a month. Must be from 1 to 31 and valid for the year and month, or 0 to specify a year by itself or a year and month where the day isn't significant." }, "year": { "description": "Year of the date. Must be from 1 to 9999, or 0 to specify a date without a year.", "type": "integer", "format": "int32" } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3ProcessRequest": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3ProcessRequest", "properties": { "skipHumanReview": { "type": "boolean", "description": "Whether human review should be skipped for this request. Default to `false`.", "deprecated": true }, "processOptions": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessOptions", "description": "Inference-time options for the process API" }, "imagelessMode": { "description": "Optional. Option to remove images from the document.", "type": "boolean" }, "inlineDocument": { "description": "An inline document proto.", "$ref": "GoogleCloudDocumentaiV1beta3Document" }, "labels": { "type": "object", "additionalProperties": { "type": "string" }, "description": "Optional. The labels with user-defined metadata for the request. Label keys and values can be no longer than 63 characters (Unicode codepoints) and can only contain lowercase letters, numeric characters, underscores, and dashes. International characters are allowed. Label values are optional. Label keys must start with a letter." }, "fieldMask": { "type": "string", "format": "google-fieldmask", "description": "Specifies which fields to include in the ProcessResponse.document output. Only supports top-level document and pages field, so it must be in the form of `{document_field_name}` or `pages.{page_field_name}`." }, "gcsDocument": { "$ref": "GoogleCloudDocumentaiV1beta3GcsDocument", "description": "A raw document on Google Cloud Storage." }, "document": { "$ref": "GoogleCloudDocumentaiV1beta3Document", "deprecated": true, "description": "The document payload, the content and mime_type fields must be set." }, "rawDocument": { "$ref": "GoogleCloudDocumentaiV1beta3RawDocument", "description": "A raw document content (bytes)." } }, "description": "Request message for the ProcessDocument method." }, "GoogleCloudDocumentaiV1beta3ListProcessorVersionsResponse": { "id": "GoogleCloudDocumentaiV1beta3ListProcessorVersionsResponse", "type": "object", "properties": { "processorVersions": { "type": "array", "description": "The list of processors.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessorVersion" } }, "nextPageToken": { "description": "Points to the next processor, otherwise empty.", "type": "string" } }, "description": "Response message for the ListProcessorVersions method." }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFieldRegex": { "properties": { "field": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleField" }, "pattern": { "type": "string", "description": "Python regex to validate the field values." } }, "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFieldRegex", "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage": { "id": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage", "type": "object", "description": "Detected language for a structural component.", "properties": { "confidence": { "format": "float", "description": "Confidence of detected language. Range `[0, 1]`.", "type": "number" }, "languageCode": { "description": "The [BCP-47 language code](https://www.unicode.org/reports/tr35/#Unicode_locale_identifier), such as `en-US` or `sr-Latn`.", "type": "string" } } }, "GoogleCloudDocumentaiV1beta3SetDefaultProcessorVersionResponse": { "type": "object", "description": "Response message for the SetDefaultProcessorVersion method.", "id": "GoogleCloudDocumentaiV1beta3SetDefaultProcessorVersionResponse", "properties": {} }, "GoogleCloudDocumentaiV1beta3ImportDocumentsRequestBatchDocumentsImportConfig": { "properties": { "datasetSplit": { "type": "string", "description": "Target dataset split where the documents must be stored.", "enumDescriptions": [ "Default value if the enum is not set.", "Identifies the train documents.", "Identifies the test documents.", "Identifies the unassigned documents." ], "enum": [ "DATASET_SPLIT_TYPE_UNSPECIFIED", "DATASET_SPLIT_TRAIN", "DATASET_SPLIT_TEST", "DATASET_SPLIT_UNASSIGNED" ] }, "batchInputConfig": { "description": "The common config to specify a set of documents used as input.", "$ref": "GoogleCloudDocumentaiV1beta3BatchDocumentsInputConfig" }, "autoSplitConfig": { "$ref": "GoogleCloudDocumentaiV1beta3ImportDocumentsRequestBatchDocumentsImportConfigAutoSplitConfig", "description": "If set, documents will be automatically split into training and test split category with the specified ratio." }, "documentType": { "type": "string", "description": "Optional. If set, determines the type of the documents to be imported in this batch. It can be used to auto-label the documents with a single entity of the provided type. This field can only be used with a classifier or splitter processor. Providing this field is mutually exclusive with `entities` and `auto_labeling_config`." } }, "description": "Config for importing documents. Each batch can have its own dataset split type.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3ImportDocumentsRequestBatchDocumentsImportConfig" }, "GoogleCloudDocumentaiV1beta3RevisionRef": { "properties": { "latestProcessorVersion": { "description": "Reads the revision generated by the processor version. The format takes the full resource name of processor version. `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`", "type": "string" }, "revisionId": { "type": "string", "description": "Reads the revision given by the id." }, "revisionCase": { "enumDescriptions": [ "Unspecified case, fall back to read the `LATEST_HUMAN_REVIEW`.", "The latest revision made by a human.", "The latest revision based on timestamp.", "The first (OCR) revision." ], "description": "Reads the revision by the predefined case.", "type": "string", "enum": [ "REVISION_CASE_UNSPECIFIED", "LATEST_HUMAN_REVIEW", "LATEST_TIMESTAMP", "BASE_OCR_REVISION" ] } }, "description": "The revision reference specifies which revision on the document to read.", "id": "GoogleCloudDocumentaiV1beta3RevisionRef", "type": "object" }, "GoogleCloudDocumentaiV1ReviewDocumentResponse": { "type": "object", "properties": { "state": { "enumDescriptions": [ "The default value. This value is used if the state is omitted.", "The review operation is rejected by the reviewer.", "The review operation is succeeded." ], "enum": [ "STATE_UNSPECIFIED", "REJECTED", "SUCCEEDED" ], "description": "The state of the review operation.", "type": "string" }, "rejectionReason": { "description": "The reason why the review is rejected by reviewer.", "type": "string" }, "gcsDestination": { "description": "The Cloud Storage uri for the human reviewed document if the review is succeeded.", "type": "string" } }, "description": "Response message for the ReviewDocument method.", "id": "GoogleCloudDocumentaiV1ReviewDocumentResponse" }, "GoogleCloudDocumentaiV1beta3SchemaVersion": { "properties": { "labels": { "additionalProperties": { "type": "string" }, "type": "object", "description": "Optional. The {{gcp_name_short}} labels for the SchemaVersion." }, "schema": { "description": "Required. The schema of the SchemaVersion.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentSchema" }, "name": { "type": "string", "description": "Identifier. The resource name of the SchemaVersion. Format: `projects/{project}/locations/{location}/schemas/{schema}/schemaVersions/{schema_version}`" }, "createTime": { "description": "Output only. The time when the SchemaVersion was created.", "type": "string", "readOnly": true, "format": "google-datetime" }, "displayName": { "description": "Required. The user-defined name of the SchemaVersion.", "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3SchemaVersion", "type": "object", "description": "SchemaVersion is a version of the Schema which is created in SchemaGroup." }, "GoogleCloudDocumentaiV1beta3BatchProcessResponse": { "description": "Response message for BatchProcessDocuments.", "id": "GoogleCloudDocumentaiV1beta3BatchProcessResponse", "type": "object", "properties": {} }, "GoogleCloudDocumentaiV1beta3ProcessorType": { "properties": { "sampleDocumentUris": { "items": { "type": "string" }, "type": "array", "description": "A set of Cloud Storage URIs of sample documents for this processor." }, "name": { "description": "The resource name of the processor type. Format: `projects/{project}/processorTypes/{processor_type}`", "type": "string" }, "allowCreation": { "type": "boolean", "description": "Whether the processor type allows creation. If true, users can create a processor of this processor type. Otherwise, users need to request access." }, "category": { "type": "string", "description": "The processor category, used by UI to group processor types." }, "type": { "description": "The processor type, such as: `OCR_PROCESSOR`, `INVOICE_PROCESSOR`.", "type": "string" }, "availableLocations": { "type": "array", "description": "The locations in which this processor is available.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessorTypeLocationInfo" } }, "launchStage": { "enumDescriptions": [ "Do not use this default value.", "The feature is not yet implemented. Users can not use it.", "Prelaunch features are hidden from users and are only visible internally.", "Early Access features are limited to a closed group of testers. To use these features, you must sign up in advance and sign a Trusted Tester agreement (which includes confidentiality provisions). These features may be unstable, changed in backward-incompatible ways, and are not guaranteed to be released.", "Alpha is a limited availability test for releases before they are cleared for widespread use. By Alpha, all significant design issues are resolved and we are in the process of verifying functionality. Alpha customers need to apply for access, agree to applicable terms, and have their projects allowlisted. Alpha releases don't have to be feature complete, no SLAs are provided, and there are no technical support obligations, but they will be far enough along that customers can actually use them in test environments or for limited-use tests -- just like they would in normal production cases.", "Beta is the point at which we are ready to open a release for any customer to use. There are no SLA or technical support obligations in a Beta release. Products will be complete from a feature perspective, but may have some open outstanding issues. Beta releases are suitable for limited production use cases.", "GA features are open to all developers and are considered stable and fully qualified for production use.", "Deprecated features are scheduled to be shut down and removed. For more information, see the \"Deprecation Policy\" section of our [Terms of Service](https://cloud.google.com/terms/) and the [Google Cloud Platform Subject to the Deprecation Policy](https://cloud.google.com/terms/deprecation) documentation." ], "description": "Launch stage of the processor type", "enum": [ "LAUNCH_STAGE_UNSPECIFIED", "UNIMPLEMENTED", "PRELAUNCH", "EARLY_ACCESS", "ALPHA", "BETA", "GA", "DEPRECATED" ], "type": "string" } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3ProcessorType", "description": "A processor type is responsible for performing a certain document understanding task on a certain type of document." }, "GoogleCloudDocumentaiV1beta3ImportDocumentsMetadataIndividualImportStatus": { "description": "The status of each individual document in the import process.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3ImportDocumentsMetadataIndividualImportStatus", "properties": { "status": { "$ref": "GoogleRpcStatus", "description": "The status of the importing of the document." }, "inputGcsSource": { "description": "The source Cloud Storage URI of the document.", "type": "string" }, "outputDocumentId": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentId", "description": "The document ID of imported document if it was successful, otherwise empty." } } }, "GoogleCloudDocumentaiUiv1beta3CreateLabelerPoolOperationMetadata": { "description": "The long-running operation metadata for the CreateLabelerPool method.", "id": "GoogleCloudDocumentaiUiv1beta3CreateLabelerPoolOperationMetadata", "type": "object", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" } } }, "GoogleCloudDocumentaiV1beta3DocumentSchemaEntityType": { "description": "EntityType is the wrapper of a label of the corresponding model with detailed attributes and limitations for entity-based processors. Multiple types can also compose a dependency tree to represent nested types.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentSchemaEntityType", "properties": { "baseTypes": { "description": "The entity type that this type is derived from. For now, one and only one should be set.", "type": "array", "items": { "type": "string" } }, "properties": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentSchemaEntityTypeProperty" }, "description": "Description the nested structure, or composition of an entity." }, "entityTypeMetadata": { "description": "Metadata for the entity type.", "$ref": "GoogleCloudDocumentaiV1beta3EntityTypeMetadata" }, "enumValues": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentSchemaEntityTypeEnumValues", "description": "If specified, lists all the possible values for this entity. This should not be more than a handful of values. If the number of values is \u003e10 or could change frequently, use the `EntityType.value_ontology` field and specify a list of all possible values in a value ontology file." }, "displayName": { "description": "User defined name for the type.", "type": "string" }, "description": { "description": "The description of the entity type. Could be used to provide more information about the entity type for model calls.", "type": "string" }, "name": { "type": "string", "description": "Name of the type. It must be unique within the schema file and cannot be a \"Common Type\". The following naming conventions are used: - Use `snake_casing`. - Name matching is case-sensitive. - Maximum 64 characters. - Must start with a letter. - Allowed characters: ASCII letters `[a-z0-9_-]`. (For backward compatibility, internal infrastructure and tooling can handle any ASCII character.) - The `/` is sometimes used to denote a property of a type. For example, `line_item/amount`. This convention is deprecated, but will still be honored for backward compatibility." } } }, "GoogleCloudDocumentaiUiv1beta3EnableProcessorResponse": { "id": "GoogleCloudDocumentaiUiv1beta3EnableProcessorResponse", "description": "Response message for the EnableProcessor method. Intentionally empty proto for adding fields in future.", "properties": {}, "type": "object" }, "GoogleCloudDocumentaiUiv1beta3FieldExtractionMetadataEntityQuery": { "description": "Message for entity query.", "properties": { "userEntityQuery": { "description": "The original entity query inputed by the user.", "type": "string" } }, "id": "GoogleCloudDocumentaiUiv1beta3FieldExtractionMetadataEntityQuery", "type": "object" }, "GoogleCloudDocumentaiV1TrainProcessorVersionMetadata": { "properties": { "trainingDatasetValidation": { "description": "The training dataset validation information.", "$ref": "GoogleCloudDocumentaiV1TrainProcessorVersionMetadataDatasetValidation" }, "testDatasetValidation": { "description": "The test dataset validation information.", "$ref": "GoogleCloudDocumentaiV1TrainProcessorVersionMetadataDatasetValidation" }, "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiV1CommonOperationMetadata" } }, "id": "GoogleCloudDocumentaiV1TrainProcessorVersionMetadata", "description": "The metadata that represents a processor version being created.", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3DocumentSchemaMetadata": { "type": "object", "description": "Metadata for global schema behavior.", "properties": { "prefixedNamingOnProperties": { "type": "boolean", "description": "If set, all the nested entities must be prefixed with the parents." }, "documentSplitter": { "type": "boolean", "description": "If true, a `document` entity type can be applied to subdocument (splitting). Otherwise, it can only be applied to the entire document (classification)." }, "documentAllowMultipleLabels": { "description": "If true, on a given page, there can be multiple `document` annotations covering it.", "type": "boolean" }, "skipNamingValidation": { "description": "If set, this will skip the naming format validation in the schema. So the string values in `DocumentSchema.EntityType.name` and `DocumentSchema.EntityType.Property.name` will not be checked.", "type": "boolean" } }, "id": "GoogleCloudDocumentaiUiv1beta3DocumentSchemaMetadata" }, "GoogleCloudDocumentaiV1beta3Evaluation": { "properties": { "createTime": { "description": "The time that the evaluation was created.", "type": "string", "format": "google-datetime" }, "entityMetrics": { "description": "Metrics across confidence levels, for different entities.", "additionalProperties": { "$ref": "GoogleCloudDocumentaiV1beta3EvaluationMultiConfidenceMetrics" }, "type": "object" }, "kmsKeyVersionName": { "type": "string", "description": "The KMS key version with which data is encrypted." }, "kmsKeyName": { "description": "The KMS key name used for encryption.", "type": "string" }, "name": { "description": "The resource name of the evaluation. Format: `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processor_version}/evaluations/{evaluation}`", "type": "string" }, "documentCounters": { "description": "Counters for the documents used in the evaluation.", "$ref": "GoogleCloudDocumentaiV1beta3EvaluationCounters" }, "allEntitiesMetrics": { "$ref": "GoogleCloudDocumentaiV1beta3EvaluationMultiConfidenceMetrics", "description": "Metrics for all the entities in aggregate." }, "revisions": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3EvaluationEvaluationRevision" }, "description": "Contains all revisions of the evaluation, excluding the latest one." } }, "id": "GoogleCloudDocumentaiV1beta3Evaluation", "description": "An evaluation of a ProcessorVersion's performance.", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3ResyncDatasetMetadata": { "description": "The metadata proto of `ResyncDataset` method.", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" }, "datasetResyncStatuses": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3ResyncDatasetMetadataDatasetResyncStatus" }, "description": "The list of dataset resync statuses. Not checked when ResyncDatasetRequest.dataset_documents is specified." }, "individualDocumentResyncStatuses": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3ResyncDatasetMetadataIndividualDocumentResyncStatus" }, "description": "The list of document resync statuses. The same document could have multiple `individual_document_resync_statuses` if it has multiple inconsistencies." } }, "id": "GoogleCloudDocumentaiUiv1beta3ResyncDatasetMetadata", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3SummaryOptions": { "id": "GoogleCloudDocumentaiUiv1beta3SummaryOptions", "description": "Metadata for document summarization.", "type": "object", "properties": { "length": { "enum": [ "LENGTH_UNSPECIFIED", "BRIEF", "MODERATE", "COMPREHENSIVE" ], "type": "string", "enumDescriptions": [ "Default.", "A brief summary of one or two sentences.", "A paragraph-length summary.", "The longest option available." ], "description": "How long the summary should be." }, "format": { "enum": [ "FORMAT_UNSPECIFIED", "PARAGRAPH", "BULLETS" ], "type": "string", "enumDescriptions": [ "Default.", "Format the output in paragraphs.", "Format the output in bullets." ], "description": "The format the summary should be in." } } }, "GoogleCloudDocumentaiUiv1beta3ProcessorVersion": { "properties": { "state": { "description": "Output only. The state of the processor version.", "enum": [ "STATE_UNSPECIFIED", "DEPLOYED", "DEPLOYING", "UNDEPLOYED", "UNDEPLOYING", "CREATING", "DELETING", "FAILED", "IMPORTING" ], "type": "string", "enumDescriptions": [ "The processor version is in an unspecified state.", "The processor version is deployed and can be used for processing.", "The processor version is being deployed.", "The processor version is not deployed and cannot be used for processing.", "The processor version is being undeployed.", "The processor version is being created.", "The processor version is being deleted.", "The processor version failed and is in an indeterminate state.", "The processor version is being imported." ], "readOnly": true }, "satisfiesPzs": { "description": "Output only. Reserved for future use.", "readOnly": true, "type": "boolean" }, "deploymentAllowed": { "type": "boolean", "readOnly": true, "description": "Output only. Denotes that this `ProcessorVersion` can be deployed and undeployed." }, "createTime": { "readOnly": true, "description": "Output only. The time the processor version was created.", "format": "google-datetime", "type": "string" }, "schema": { "description": "The schema of the processor version. Describes the output.", "$ref": "GoogleCloudDocumentaiUiv1beta3Schema" }, "kmsKeyName": { "type": "string", "description": "Output only. The KMS key name used for encryption.", "readOnly": true }, "kmsKeyVersionName": { "type": "string", "readOnly": true, "description": "Output only. The KMS key version with which data is encrypted." }, "name": { "type": "string", "description": "Identifier. The resource name of the processor version. Format: `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processor_version}`" }, "displayName": { "type": "string", "description": "The display name of the processor version." }, "deprecationInfo": { "description": "Output only. If set, information about the eventual deprecation of this version.", "$ref": "GoogleCloudDocumentaiUiv1beta3ProcessorVersionDeprecationInfo", "readOnly": true }, "groundingSettings": { "$ref": "GoogleCloudDocumentaiUiv1beta3GroundingSettings", "description": "Output only. The grounding settings of the processor version. This can only be set using TrainProcessorVersionRequest to override the default grounding settings.", "readOnly": true }, "documentSchema": { "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentSchema", "readOnly": true, "description": "Output only. The schema of the processor version. Describes the output." }, "genAiModelInfo": { "description": "Output only. Information about Generative AI model-based processor versions.", "readOnly": true, "$ref": "GoogleCloudDocumentaiUiv1beta3ProcessorVersionGenAiModelInfo" }, "latestEvaluation": { "description": "Output only. The most recently invoked evaluation for the processor version.", "readOnly": true, "$ref": "GoogleCloudDocumentaiUiv1beta3EvaluationReference" }, "modelType": { "type": "string", "enum": [ "MODEL_TYPE_UNSPECIFIED", "MODEL_TYPE_GENERATIVE", "MODEL_TYPE_CUSTOM" ], "readOnly": true, "enumDescriptions": [ "The processor version has unspecified model type.", "The processor version has generative model type.", "The processor version has custom model type." ], "description": "Output only. The model type of this processor version." }, "satisfiesPzi": { "description": "Output only. Reserved for future use.", "readOnly": true, "type": "boolean" }, "googleManaged": { "readOnly": true, "type": "boolean", "description": "Output only. Denotes that this `ProcessorVersion` is managed by Google." } }, "description": "A processor version is an implementation of a processor. Each processor can have multiple versions, pretrained by Google internally or uptrained by the customer. A processor can only have one default version at a time. Its document-processing behavior is defined by that version.", "id": "GoogleCloudDocumentaiUiv1beta3ProcessorVersion", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3ResyncDatasetResponse": { "type": "object", "description": "The response proto of ResyncDataset method.", "id": "GoogleCloudDocumentaiUiv1beta3ResyncDatasetResponse", "properties": {} }, "GoogleTypeColor": { "description": "Represents a color in the RGBA color space. This representation is designed for simplicity of conversion to and from color representations in various languages over compactness. For example, the fields of this representation can be trivially provided to the constructor of `java.awt.Color` in Java; it can also be trivially provided to UIColor's `+colorWithRed:green:blue:alpha` method in iOS; and, with just a little work, it can be easily formatted into a CSS `rgba()` string in JavaScript. This reference page doesn't have information about the absolute color space that should be used to interpret the RGB value—for example, sRGB, Adobe RGB, DCI-P3, and BT.2020. By default, applications should assume the sRGB color space. When color equality needs to be decided, implementations, unless documented otherwise, treat two colors as equal if all their red, green, blue, and alpha values each differ by at most `1e-5`. Example (Java): import com.google.type.Color; // ... public static java.awt.Color fromProto(Color protocolor) { float alpha = protocolor.hasAlpha() ? protocolor.getAlpha().getValue() : 1.0; return new java.awt.Color( protocolor.getRed(), protocolor.getGreen(), protocolor.getBlue(), alpha); } public static Color toProto(java.awt.Color color) { float red = (float) color.getRed(); float green = (float) color.getGreen(); float blue = (float) color.getBlue(); float denominator = 255.0; Color.Builder resultBuilder = Color .newBuilder() .setRed(red / denominator) .setGreen(green / denominator) .setBlue(blue / denominator); int alpha = color.getAlpha(); if (alpha != 255) { result.setAlpha( FloatValue .newBuilder() .setValue(((float) alpha) / denominator) .build()); } return resultBuilder.build(); } // ... Example (iOS / Obj-C): // ... static UIColor* fromProto(Color* protocolor) { float red = [protocolor red]; float green = [protocolor green]; float blue = [protocolor blue]; FloatValue* alpha_wrapper = [protocolor alpha]; float alpha = 1.0; if (alpha_wrapper != nil) { alpha = [alpha_wrapper value]; } return [UIColor colorWithRed:red green:green blue:blue alpha:alpha]; } static Color* toProto(UIColor* color) { CGFloat red, green, blue, alpha; if (![color getRed:&red green:&green blue:&blue alpha:&alpha]) { return nil; } Color* result = [[Color alloc] init]; [result setRed:red]; [result setGreen:green]; [result setBlue:blue]; if (alpha \u003c= 0.9999) { [result setAlpha:floatWrapperWithValue(alpha)]; } [result autorelease]; return result; } // ... Example (JavaScript): // ... var protoToCssColor = function(rgb_color) { var redFrac = rgb_color.red || 0.0; var greenFrac = rgb_color.green || 0.0; var blueFrac = rgb_color.blue || 0.0; var red = Math.floor(redFrac * 255); var green = Math.floor(greenFrac * 255); var blue = Math.floor(blueFrac * 255); if (!('alpha' in rgb_color)) { return rgbToCssColor(red, green, blue); } var alphaFrac = rgb_color.alpha.value || 0.0; var rgbParams = [red, green, blue].join(','); return ['rgba(', rgbParams, ',', alphaFrac, ')'].join(''); }; var rgbToCssColor = function(red, green, blue) { var rgbNumber = new Number((red \u003c\u003c 16) | (green \u003c\u003c 8) | blue); var hexString = rgbNumber.toString(16); var missingZeros = 6 - hexString.length; var resultBuilder = ['#']; for (var i = 0; i \u003c missingZeros; i++) { resultBuilder.push('0'); } resultBuilder.push(hexString); return resultBuilder.join(''); }; // ...", "id": "GoogleTypeColor", "properties": { "red": { "description": "The amount of red in the color as a value in the interval [0, 1].", "format": "float", "type": "number" }, "alpha": { "type": "number", "format": "float", "description": "The fraction of this color that should be applied to the pixel. That is, the final pixel color is defined by the equation: `pixel color = alpha * (this color) + (1.0 - alpha) * (background color)` This means that a value of 1.0 corresponds to a solid color, whereas a value of 0.0 corresponds to a completely transparent color. This uses a wrapper message rather than a simple float scalar so that it is possible to distinguish between a default value and the value being unset. If omitted, this color object is rendered as a solid color (as if the alpha value had been explicitly given a value of 1.0)." }, "green": { "description": "The amount of green in the color as a value in the interval [0, 1].", "format": "float", "type": "number" }, "blue": { "description": "The amount of blue in the color as a value in the interval [0, 1].", "format": "float", "type": "number" } }, "type": "object" }, "GoogleCloudDocumentaiV1EvaluateProcessorVersionMetadata": { "type": "object", "description": "Metadata of the EvaluateProcessorVersion method.", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "id": "GoogleCloudDocumentaiV1EvaluateProcessorVersionMetadata" }, "GoogleCloudDocumentaiUiv1beta3DeleteLabelerPoolOperationMetadata": { "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3DeleteLabelerPoolOperationMetadata", "description": "The long-running operation metadata for DeleteLabelerPool." }, "GoogleCloudDocumentaiUiv1beta3ExportDocumentsResponse": { "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3ExportDocumentsResponse", "properties": {}, "description": "The response proto of ExportDocuments method." }, "GoogleCloudDocumentaiUiv1beta3ProcessorVersionAlias": { "id": "GoogleCloudDocumentaiUiv1beta3ProcessorVersionAlias", "type": "object", "properties": { "alias": { "description": "The alias in the form of `processor_version` resource name.", "type": "string" }, "processorVersion": { "description": "The resource name of aliased processor version.", "type": "string" } }, "description": "Contains the alias and the aliased resource name of processor version." }, "GoogleCloudDocumentaiUiv1beta3UpdateProcessorVersionMetadata": { "properties": { "commonMetadata": { "description": "The basic metadata for the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" } }, "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3UpdateProcessorVersionMetadata", "description": "The long-running operation metadata for the UpdateProcessorVersion method." }, "GoogleCloudDocumentaiV1beta3DocumentDocumentLayout": { "description": "Represents the parsed layout of a document as a collection of blocks that the document is divided into.", "id": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayout", "properties": { "blocks": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlock" }, "description": "List of blocks in the document." } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3OcrConfigPremiumFeatures": { "description": "Configurations for premium OCR features.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3OcrConfigPremiumFeatures", "properties": { "enableSelectionMarkDetection": { "type": "boolean", "description": "Turn on selection mark detector in OCR engine. Only available in OCR 2.0 (and later) processors." }, "enableMathOcr": { "description": "Turn on the model that can extract LaTeX math formulas.", "type": "boolean" }, "computeStyleInfo": { "description": "Turn on font identification model and return font style information.", "type": "boolean" } } }, "GoogleCloudDocumentaiV1beta3DocumentSchemaMetadata": { "properties": { "prefixedNamingOnProperties": { "type": "boolean", "description": "If set, all the nested entities must be prefixed with the parents." }, "skipNamingValidation": { "description": "If set, this will skip the naming format validation in the schema. So the string values in `DocumentSchema.EntityType.name` and `DocumentSchema.EntityType.Property.name` will not be checked.", "type": "boolean" }, "documentSplitter": { "description": "If true, a `document` entity type can be applied to subdocument (splitting). Otherwise, it can only be applied to the entire document (classification).", "type": "boolean" }, "documentAllowMultipleLabels": { "description": "If true, on a given page, there can be multiple `document` annotations covering it.", "type": "boolean" } }, "description": "Metadata for global schema behavior.", "id": "GoogleCloudDocumentaiV1beta3DocumentSchemaMetadata", "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentPageDimension": { "id": "GoogleCloudDocumentaiV1beta3DocumentPageDimension", "description": "Dimension for the page.", "properties": { "unit": { "type": "string", "description": "Dimension unit." }, "height": { "type": "number", "description": "Page height.", "format": "float" }, "width": { "format": "float", "type": "number", "description": "Page width." } }, "type": "object" }, "GoogleTypeTimeZone": { "properties": { "version": { "description": "Optional. IANA Time Zone Database version number. For example \"2019a\".", "type": "string" }, "id": { "type": "string", "description": "IANA Time Zone Database time zone. For example \"America/New_York\"." } }, "type": "object", "description": "Represents a time zone from the [IANA Time Zone Database](https://www.iana.org/time-zones).", "id": "GoogleTypeTimeZone" }, "GoogleCloudDocumentaiUiv1beta3DocumentIdGCSManagedDocumentId": { "properties": { "gcsUri": { "description": "Required. The Cloud Storage URI where the actual document is stored.", "type": "string" }, "cwDocId": { "type": "string", "deprecated": true, "description": "ID of the document (indexed) managed by Content Warehouse." } }, "id": "GoogleCloudDocumentaiUiv1beta3DocumentIdGCSManagedDocumentId", "description": "Identifies a document uniquely within the scope of a dataset in the user-managed Cloud Storage option.", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3DocumentSchema": { "properties": { "entityTypes": { "type": "array", "description": "Entity types of the schema.", "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentSchemaEntityType" } }, "metadata": { "description": "Metadata of the schema.", "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentSchemaMetadata" }, "documentPrompt": { "description": "Optional. Document-level prompt provided by the user. This custom text is injected into the AI model's prompt to provide extra, document-wide guidance for processing.", "type": "string" }, "description": { "description": "Description of the schema.", "type": "string" }, "displayName": { "description": "Display name to show users.", "type": "string" } }, "description": "The schema defines the output of the processed document by a processor.", "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3DocumentSchema" }, "GoogleCloudDocumentaiUiv1beta3BatchUpdateDocumentsMetadata": { "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" }, "individualBatchUpdateStatuses": { "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3BatchUpdateDocumentsMetadataIndividualBatchUpdateStatus" }, "description": "The list of response details of each document.", "type": "array" } }, "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3BatchUpdateDocumentsMetadata" }, "GoogleCloudDocumentaiUiv1beta3BatchDeleteDocumentsMetadata": { "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3BatchDeleteDocumentsMetadata", "properties": { "errorDocumentCount": { "type": "integer", "description": "Total number of documents that failed to be deleted in storage.", "format": "int32" }, "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" }, "individualBatchDeleteStatuses": { "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3BatchDeleteDocumentsMetadataIndividualBatchDeleteStatus" }, "description": "The list of response details of each document.", "type": "array" }, "totalDocumentCount": { "type": "integer", "description": "Total number of documents deleting from dataset.", "format": "int32" } } }, "GoogleCloudDocumentaiV1DisableProcessorMetadata": { "description": "The long-running operation metadata for the DisableProcessor method.", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiV1CommonOperationMetadata" } }, "type": "object", "id": "GoogleCloudDocumentaiV1DisableProcessorMetadata" }, "GoogleCloudDocumentaiV1beta3BatchDeleteDocumentsMetadataIndividualBatchDeleteStatus": { "id": "GoogleCloudDocumentaiV1beta3BatchDeleteDocumentsMetadataIndividualBatchDeleteStatus", "description": "The status of each individual document in the batch delete process.", "properties": { "documentId": { "description": "The document ID of the document.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentId" }, "status": { "description": "The status of deleting the document in storage.", "$ref": "GoogleRpcStatus" } }, "type": "object" }, "GoogleCloudDocumentaiUiv1beta3BatchMoveDocumentsResponse": { "id": "GoogleCloudDocumentaiUiv1beta3BatchMoveDocumentsResponse", "properties": {}, "type": "object", "description": "Response of the batch move documents operation." }, "GoogleCloudDocumentaiV1beta3ProcessOptionsLayoutConfig": { "type": "object", "properties": { "enableTableSplit": { "description": "Optional. Whether to split table.", "type": "boolean" }, "enableImageExtraction": { "description": "Optional. Whether to extract images in layout parser response.", "type": "boolean" }, "enableImageAnnotation": { "description": "Optional. Whether to include image annotations in layout parser response.", "type": "boolean" }, "returnImages": { "type": "boolean", "description": "Optional. Whether to include images in layout parser processor response." }, "enableLlmLayoutParsing": { "type": "boolean", "description": "Optional. Whether to refine PDF layout using LLM." }, "enableTableAnnotation": { "description": "Optional. Whether to include table annotations in layout parser response.", "type": "boolean" }, "returnBoundingBoxes": { "type": "boolean", "description": "Optional. Whether to include bounding boxes in layout parser processor response." }, "chunkingConfig": { "description": "Optional. Config for chunking in layout parser processor.", "$ref": "GoogleCloudDocumentaiV1beta3ProcessOptionsLayoutConfigChunkingConfig" } }, "id": "GoogleCloudDocumentaiV1beta3ProcessOptionsLayoutConfig", "description": "Serving config for layout parser processor." }, "GoogleCloudDocumentaiUiv1beta3UpdateDatasetOperationMetadata": { "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" } }, "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3UpdateDatasetOperationMetadata" }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRule": { "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRule", "type": "object", "properties": { "description": { "description": "Description of the validation rule. This has no use but for documentation.", "type": "string" }, "entityAlignmentRule": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleEntityAlignmentRule" }, "fieldOccurrences": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFieldOccurrences" }, "childAlignmentRule": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleChildAlignmentRule" }, "name": { "type": "string", "description": "Name of the validation rule." }, "ruleId": { "type": "string", "description": "Unique identifier of the rule. Optional." }, "formValidation": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFormValidation" }, "fieldRegex": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFieldRegex" } }, "description": "Next ID: 9" }, "GoogleCloudDocumentaiUiv1beta3SampleDocumentsResponseSelectedDocument": { "properties": { "documentId": { "type": "string", "description": "An internal identifier for document." } }, "id": "GoogleCloudDocumentaiUiv1beta3SampleDocumentsResponseSelectedDocument", "type": "object" }, "GoogleCloudDocumentaiV1beta3Document": { "properties": { "blobAssets": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentBlobAsset" }, "description": "Optional. The blob assets in this document. This is used to store the content of the inline blobs in this document, for example, image bytes, such that it can be referenced by other fields in the document via asset ID.", "type": "array" }, "textChanges": { "description": "Placeholder. A list of text corrections made to Document.text. This is usually used for annotating corrections to OCR mistakes. Text changes for a given revision may not overlap with each other.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentTextChange" } }, "text": { "type": "string", "description": "Optional. UTF-8 encoded text in reading order from the document." }, "docid": { "description": "Optional. An internal identifier for document. Should be loggable (no PII).", "type": "string" }, "pages": { "type": "array", "description": "Visual page layout for the Document.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPage" } }, "entitiesRevisions": { "description": "A list of entity revisions. The entity revisions are appended to the document in the processing order. This field can be used for comparing the entity extraction results at different stages of the processing.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentEntitiesRevision" } }, "entityRelations": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentEntityRelation" }, "description": "Placeholder. Relationship among Document.entities.", "type": "array" }, "entitiesRevisionId": { "type": "string", "description": "The entity revision ID that `document.entities` field is based on. If this field and `entities_revisions` are set, the entities in `document.entities` are the entities in the entity revision with this ID. The `document.entity_validation_output` field is the `entity_validation_output` field in this entity revision." }, "mimeType": { "description": "An IANA published [media type (MIME type)](https://www.iana.org/assignments/media-types/media-types.xhtml).", "type": "string" }, "chunkedDocument": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocument", "description": "Document chunked based on chunking config." }, "entities": { "type": "array", "description": "A list of entities detected on Document.text. For document shards, entities in this list may cross shard boundaries.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentEntity" } }, "textStyles": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentStyle" }, "type": "array", "deprecated": true, "description": "Styles for the Document.text." }, "documentLayout": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayout", "description": "Parsed layout of the document." }, "content": { "format": "byte", "description": "Optional. Inline document content, represented as a stream of bytes. Note: As with all `bytes` fields, protobuffers use a pure binary representation, whereas JSON representations use base64.", "type": "string" }, "error": { "$ref": "GoogleRpcStatus", "description": "Any error that occurred while processing this document." }, "entityValidationOutput": { "description": "The entity validation output for the document. This is the validation output for `document.entities` field.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentEntityValidationOutput" }, "revisions": { "description": "Placeholder. Revision history of this document.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentRevision" }, "type": "array" }, "shardInfo": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentShardInfo", "description": "Information about the sharding if this document is sharded part of a larger document. If the document is not sharded, this message is not specified." }, "uri": { "description": "Optional. Currently supports Google Cloud Storage URI of the form `gs://bucket_name/object_name`. Object versioning is not supported. For more information, refer to [Google Cloud Storage Request URIs](https://cloud.google.com/storage/docs/reference-uris).", "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3Document", "type": "object", "description": "Document represents the canonical document resource in Document AI. It is an interchange format that provides insights into documents and allows for collaboration between users and Document AI to iterate and optimize for quality." }, "GoogleCloudDocumentaiV1beta3DocumentPageAnchorPageRef": { "properties": { "layoutId": { "type": "string", "deprecated": true, "description": "Optional. Deprecated. Use PageRef.bounding_poly instead." }, "confidence": { "description": "Optional. Confidence of detected page element, if applicable. Range `[0, 1]`.", "format": "float", "type": "number" }, "layoutType": { "enum": [ "LAYOUT_TYPE_UNSPECIFIED", "BLOCK", "PARAGRAPH", "LINE", "TOKEN", "VISUAL_ELEMENT", "TABLE", "FORM_FIELD" ], "type": "string", "description": "Optional. The type of the layout element that is being referenced if any.", "enumDescriptions": [ "Layout Unspecified.", "References a Page.blocks element.", "References a Page.paragraphs element.", "References a Page.lines element.", "References a Page.tokens element.", "References a Page.visual_elements element.", "Refrrences a Page.tables element.", "References a Page.form_fields element." ] }, "boundingPoly": { "description": "Optional. Identifies the bounding polygon of a layout element on the page. If `layout_type` is set, the bounding polygon must be exactly the same to the layout element it's referring to.", "$ref": "GoogleCloudDocumentaiV1beta3BoundingPoly" }, "page": { "format": "int64", "description": "Required. Index into the Document.pages element, for example using `Document.pages` to locate the related page element. This field is skipped when its value is the default `0`. See https://developers.google.com/protocol-buffers/docs/proto3#json.", "type": "string" } }, "description": "Represents a weak reference to a page element within a document.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentPageAnchorPageRef" }, "GoogleCloudDocumentaiUiv1beta3SampleDocumentsMetadata": { "id": "GoogleCloudDocumentaiUiv1beta3SampleDocumentsMetadata", "type": "object", "description": "Metadata of the sample documents operation.", "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" } } }, "GoogleCloudDocumentaiV1beta3GenerateSchemaVersionResponse": { "description": "Response message for GenerateSchemaVersion.", "id": "GoogleCloudDocumentaiV1beta3GenerateSchemaVersionResponse", "type": "object", "properties": { "schemaVersion": { "$ref": "GoogleCloudDocumentaiV1beta3SchemaVersion", "description": "The schema version generated by the model." } } }, "GoogleCloudDocumentaiUiv1beta3UpdateLabelerPoolOperationMetadata": { "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata" } }, "type": "object", "description": "The long-running operation metadata for UpdateLabelerPool.", "id": "GoogleCloudDocumentaiUiv1beta3UpdateLabelerPoolOperationMetadata" }, "GoogleCloudDocumentaiV1beta3EnableProcessorMetadata": { "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata" } }, "description": "The long-running operation metadata for the EnableProcessor method.", "id": "GoogleCloudDocumentaiV1beta3EnableProcessorMetadata", "type": "object" }, "GoogleCloudDocumentaiV1beta3ListSchemaVersionsResponse": { "type": "object", "properties": { "schemaVersions": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3SchemaVersion" }, "description": "The list of SchemaVersions.", "type": "array" }, "nextPageToken": { "description": "Points to the next SchemaVersion, otherwise empty.", "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3ListSchemaVersionsResponse", "description": "Response message for ListSchemaVersions." }, "GoogleCloudDocumentaiUiv1beta3SetDefaultProcessorVersionResponse": { "type": "object", "description": "Response message for the SetDefaultProcessorVersion method.", "id": "GoogleCloudDocumentaiUiv1beta3SetDefaultProcessorVersionResponse", "properties": {} }, "GoogleCloudDocumentaiV1EvaluateProcessorVersionResponse": { "properties": { "evaluation": { "type": "string", "description": "The resource name of the created evaluation." } }, "type": "object", "description": "Response of the EvaluateProcessorVersion method.", "id": "GoogleCloudDocumentaiV1EvaluateProcessorVersionResponse" }, "GoogleCloudDocumentaiV1beta3BatchProcessRequestBatchInputConfig": { "deprecated": true, "id": "GoogleCloudDocumentaiV1beta3BatchProcessRequestBatchInputConfig", "type": "object", "properties": { "mimeType": { "type": "string", "description": "An IANA published [media type (MIME type)](https://www.iana.org/assignments/media-types/media-types.xhtml) of the input. If the input is a raw document, refer to [supported file types](https://cloud.google.com/document-ai/docs/file-types) for the list of media types. If the input is a Document, the type should be `application/json`." }, "gcsSource": { "type": "string", "description": "The Cloud Storage location as the source of the document." } }, "description": "The message for input config in batch process." }, "GoogleCloudDocumentaiV1beta3ReviewDocumentRequest": { "description": "Request message for the ReviewDocument method.", "properties": { "documentSchema": { "description": "The document schema of the human review task.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentSchema" }, "inlineDocument": { "description": "An inline document proto.", "$ref": "GoogleCloudDocumentaiV1beta3Document" }, "priority": { "type": "string", "enumDescriptions": [ "The default priority level.", "The urgent priority level. The labeling manager should allocate labeler resource to the urgent task queue to respect this priority level." ], "description": "The priority of the human review task.", "enum": [ "DEFAULT", "URGENT" ] }, "enableSchemaValidation": { "type": "boolean", "description": "Whether the validation should be performed on the ad-hoc review request." }, "document": { "description": "The document that needs human review.", "$ref": "GoogleCloudDocumentaiV1beta3Document", "deprecated": true } }, "id": "GoogleCloudDocumentaiV1beta3ReviewDocumentRequest", "type": "object" }, "GoogleCloudDocumentaiV1beta3Processor": { "properties": { "createTime": { "type": "string", "readOnly": true, "format": "google-datetime", "description": "Output only. The time the processor was created." }, "defaultProcessorVersion": { "type": "string", "description": "The default processor version." }, "activeSchemaVersion": { "description": "Optional. SchemaVersion used by the Processor. It is the same as Processor's DatasetSchema.schema_version Format is `projects/{project}/locations/{location}/schemas/{schema}/schemaVersions/{schema_version}.", "type": "string" }, "processorVersionAliases": { "readOnly": true, "items": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessorVersionAlias" }, "type": "array", "description": "Output only. The processor version aliases." }, "type": { "type": "string", "description": "The processor type, such as: `OCR_PROCESSOR`, `INVOICE_PROCESSOR`. To get a list of processor types, see FetchProcessorTypes." }, "satisfiesPzi": { "readOnly": true, "type": "boolean", "description": "Output only. Reserved for future use." }, "displayName": { "type": "string", "description": "The display name of the processor." }, "kmsKeyName": { "description": "The [KMS key](https://cloud.google.com/security-key-management) used for encryption and decryption in CMEK scenarios.", "type": "string" }, "state": { "readOnly": true, "type": "string", "enum": [ "STATE_UNSPECIFIED", "ENABLED", "DISABLED", "ENABLING", "DISABLING", "CREATING", "FAILED", "DELETING" ], "enumDescriptions": [ "The processor is in an unspecified state.", "The processor is enabled, i.e., has an enabled version which can currently serve processing requests and all the feature dependencies have been successfully initialized.", "The processor is disabled.", "The processor is being enabled, will become `ENABLED` if successful.", "The processor is being disabled, will become `DISABLED` if successful.", "The processor is being created, will become either `ENABLED` (for successful creation) or `FAILED` (for failed ones). Once a processor is in this state, it can then be used for document processing, but the feature dependencies of the processor might not be fully created yet.", "The processor failed during creation or initialization of feature dependencies. The user should delete the processor and recreate one as all the functionalities of the processor are disabled.", "The processor is being deleted, will be removed if successful." ], "description": "Output only. The state of the processor." }, "name": { "description": "Output only. Immutable. The resource name of the processor. Format: `projects/{project}/locations/{location}/processors/{processor}`", "readOnly": true, "type": "string" }, "satisfiesPzs": { "readOnly": true, "description": "Output only. Reserved for future use.", "type": "boolean" }, "processEndpoint": { "description": "Output only. Immutable. The http endpoint that can be called to invoke processing.", "type": "string", "readOnly": true } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3Processor", "description": "The first-class citizen for Document AI. Each processor defines how to extract structural information from a document." }, "GoogleCloudDocumentaiV1beta3DocumentId": { "type": "object", "description": "Document Identifier.", "properties": { "unmanagedDocId": { "description": "A document ID within unmanaged dataset.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentIdUnmanagedDocumentId" }, "revisionRef": { "description": "Points to a specific revision of the document if set.", "$ref": "GoogleCloudDocumentaiV1beta3RevisionRef" }, "gcsManagedDocId": { "description": "A document ID within user-managed Cloud Storage.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentIdGCSManagedDocumentId" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentId" }, "GoogleCloudDocumentaiUiv1beta3Schema": { "properties": { "displayName": { "type": "string", "description": "Display name to show users." }, "description": { "description": "Description of the schema.", "type": "string" }, "entityTypes": { "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3SchemaEntityType" }, "type": "array", "description": "Entity types of the schema." } }, "description": "The schema defines the output of the processed document by a processor.", "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3Schema" }, "GoogleCloudDocumentaiV1beta3RawDocument": { "description": "Payload message of raw document content (bytes).", "properties": { "content": { "format": "byte", "type": "string", "description": "Inline document content." }, "mimeType": { "type": "string", "description": "An IANA MIME type (RFC6838) indicating the nature and format of the content." }, "displayName": { "type": "string", "description": "The display name of the document, it supports all Unicode characters except the following: `*`, `?`, `[`, `]`, `%`, `{`, `}`,`'`, `\\\"`, `,` `~`, `=` and `:` are reserved. If not specified, a default ID is generated." } }, "id": "GoogleCloudDocumentaiV1beta3RawDocument", "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentBlobAsset": { "description": "Represents a blob asset. It's used to store the content of the inline blob in this document, for example, image bytes, such that it can be referenced by other fields in the document via asset ID.", "id": "GoogleCloudDocumentaiV1beta3DocumentBlobAsset", "type": "object", "properties": { "content": { "format": "byte", "type": "string", "description": "Optional. The content of the blob asset, for example, image bytes." }, "assetId": { "description": "Optional. The ID of the blob asset.", "type": "string" }, "mimeType": { "type": "string", "description": "The mime type of the blob asset. An IANA published [media type (MIME type)](https://www.iana.org/assignments/media-types/media-types.xhtml)." } } }, "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageSpan": { "properties": { "pageStart": { "type": "integer", "format": "int32", "description": "Page where chunk starts in the document." }, "pageEnd": { "description": "Page where chunk ends in the document.", "format": "int32", "type": "integer" } }, "description": "Represents where the chunk starts and ends in the document.", "id": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageSpan", "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentTextAnchorTextSegment": { "id": "GoogleCloudDocumentaiV1beta3DocumentTextAnchorTextSegment", "description": "A text segment in the Document.text. The indices may be out of bounds which indicate that the text extends into another document shard for large sharded documents. See ShardInfo.text_offset", "properties": { "startIndex": { "description": "TextSegment start UTF-8 char index in the Document.text.", "type": "string", "format": "int64" }, "endIndex": { "description": "TextSegment half open end UTF-8 char index in the Document.text.", "format": "int64", "type": "string" } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3BatchProcessMetadata": { "id": "GoogleCloudDocumentaiV1beta3BatchProcessMetadata", "properties": { "individualProcessStatuses": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3BatchProcessMetadataIndividualProcessStatus" }, "description": "The list of response details of each document.", "type": "array" }, "state": { "enumDescriptions": [ "The default value. This value is used if the state is omitted.", "Request operation is waiting for scheduling.", "Request is being processed.", "The batch processing completed successfully.", "The batch processing was being cancelled.", "The batch processing was cancelled.", "The batch processing has failed." ], "description": "The state of the current batch processing.", "type": "string", "enum": [ "STATE_UNSPECIFIED", "WAITING", "RUNNING", "SUCCEEDED", "CANCELLING", "CANCELLED", "FAILED" ] }, "createTime": { "type": "string", "description": "The creation time of the operation.", "format": "google-datetime" }, "stateMessage": { "type": "string", "description": "A message providing more details about the current state of processing. For example, the error message if the operation is failed." }, "updateTime": { "description": "The last update time of the operation.", "type": "string", "format": "google-datetime" } }, "type": "object", "description": "The long-running operation metadata for BatchProcessDocuments." }, "GoogleCloudDocumentaiUiv1beta3FieldTierMetadata": { "properties": { "tierLevel": { "type": "integer", "format": "int32", "description": "Integer that indicates the tier of a property. e.g. Invoice has entities that are classified as tier 1 which is the most important, while tier 2 and tier 3 less so. This attribute can be used to filter schema attributes before running eval. e.g. compute F1 score for only tier 1 entities. If not present this attribute should be inferred as 1." } }, "type": "object", "description": "Metadata for the field tier of a property.", "id": "GoogleCloudDocumentaiUiv1beta3FieldTierMetadata" }, "GoogleCloudDocumentaiUiv1beta3DeployProcessorVersionResponse": { "type": "object", "description": "Response message for the DeployProcessorVersion method.", "id": "GoogleCloudDocumentaiUiv1beta3DeployProcessorVersionResponse", "properties": {} }, "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata": { "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The common metadata for long running operations.", "properties": { "resource": { "description": "A related resource to this operation.", "type": "string" }, "stateMessage": { "type": "string", "description": "A message providing more details about the current state of processing." }, "state": { "enumDescriptions": [ "Unspecified state.", "Operation is still running.", "Operation is being cancelled.", "Operation succeeded.", "Operation failed.", "Operation is cancelled." ], "enum": [ "STATE_UNSPECIFIED", "RUNNING", "CANCELLING", "SUCCEEDED", "FAILED", "CANCELLED" ], "type": "string", "description": "The state of the operation." }, "createTime": { "format": "google-datetime", "description": "The creation time of the operation.", "type": "string" }, "updateTime": { "format": "google-datetime", "type": "string", "description": "The last update time of the operation." } } }, "GoogleProtobufEmpty": { "type": "object", "id": "GoogleProtobufEmpty", "description": "A generic empty message that you can re-use to avoid defining duplicated empty messages in your APIs. A typical example is to use it as the request or the response type of an API method. For instance: service Foo { rpc Bar(google.protobuf.Empty) returns (google.protobuf.Empty); }", "properties": {} }, "GoogleCloudDocumentaiUiv1beta3ProcessorVersionGenAiModelInfoFoundationGenAiModelInfo": { "description": "Information for a pretrained Google-managed foundation model.", "type": "object", "properties": { "finetuningAllowed": { "type": "boolean", "description": "Whether fine tuning is allowed for this base processor version." }, "minTrainLabeledDocuments": { "description": "The minimum number of labeled documents in the training dataset required for fine tuning.", "format": "int32", "type": "integer" } }, "id": "GoogleCloudDocumentaiUiv1beta3ProcessorVersionGenAiModelInfoFoundationGenAiModelInfo" }, "GoogleCloudDocumentaiUiv1beta3ImportProcessorVersionResponse": { "id": "GoogleCloudDocumentaiUiv1beta3ImportProcessorVersionResponse", "description": "The response message for the ImportProcessorVersion method.", "type": "object", "properties": { "processorVersion": { "description": "The destination processor version name.", "type": "string" } } }, "GoogleCloudDocumentaiV1BatchProcessResponse": { "type": "object", "properties": {}, "description": "Response message for BatchProcessDocuments.", "id": "GoogleCloudDocumentaiV1BatchProcessResponse" }, "GoogleCloudDocumentaiV1beta3ImportProcessorVersionRequest": { "properties": { "processorVersionSource": { "description": "The source processor version to import from. The source processor version and destination processor need to be in the same environment and region.", "type": "string" }, "externalProcessorVersionSource": { "$ref": "GoogleCloudDocumentaiV1beta3ImportProcessorVersionRequestExternalProcessorVersionSource", "description": "The source processor version to import from. It can be from a different environment and region than the destination processor." } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3ImportProcessorVersionRequest", "description": "The request message for the ImportProcessorVersion method. The Document AI [Service Agent](https://cloud.google.com/iam/docs/service-agents) of the destination project must have [Document AI Editor role](https://cloud.google.com/document-ai/docs/access-control/iam-roles) on the source project. The destination project is specified as part of the parent field. The source project is specified as part of the source or external_processor_version_source field." }, "GoogleCloudDocumentaiV1beta3DatasetDocumentWarehouseConfig": { "properties": { "collection": { "description": "Output only. The collection in Document AI Warehouse associated with the dataset.", "readOnly": true, "type": "string" }, "schema": { "type": "string", "description": "Output only. The schema in Document AI Warehouse associated with the dataset.", "readOnly": true } }, "id": "GoogleCloudDocumentaiV1beta3DatasetDocumentWarehouseConfig", "description": "Configuration specific to the Document AI Warehouse-based implementation.", "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkTableChunkField": { "id": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkTableChunkField", "description": "The table chunk field in the chunk.", "properties": { "annotations": { "description": "Annotation of the table chunk field.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentAnnotations" } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentPageMatrix": { "description": "Representation for transformation matrix, intended to be compatible and used with OpenCV format for image manipulation.", "properties": { "type": { "type": "integer", "format": "int32", "description": "This encodes information about what data type the matrix uses. For example, 0 (CV_8U) is an unsigned 8-bit image. For the full list of OpenCV primitive data types, please refer to https://docs.opencv.org/4.3.0/d1/d1b/group__core__hal__interface.html" }, "data": { "format": "byte", "type": "string", "description": "The matrix data." }, "rows": { "type": "integer", "description": "Number of rows in the matrix.", "format": "int32" }, "cols": { "type": "integer", "description": "Number of columns in the matrix.", "format": "int32" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentPageMatrix", "type": "object" }, "GoogleCloudDocumentaiV1beta3FetchProcessorTypesResponse": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3FetchProcessorTypesResponse", "description": "Response message for the FetchProcessorTypes method.", "properties": { "processorTypes": { "description": "The list of processor types.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessorType" }, "type": "array" } } }, "GoogleCloudDocumentaiUiv1beta3BatchMoveDocumentsMetadataIndividualBatchMoveStatus": { "properties": { "documentId": { "description": "The document ID of the document.", "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentId" }, "status": { "description": "The status of moving the document.", "$ref": "GoogleRpcStatus" } }, "type": "object", "description": "The status of each individual document in the batch move process.", "id": "GoogleCloudDocumentaiUiv1beta3BatchMoveDocumentsMetadataIndividualBatchMoveStatus" }, "GoogleCloudDocumentaiUiv1beta3ProcessorVersionGenAiModelInfo": { "properties": { "foundationGenAiModelInfo": { "$ref": "GoogleCloudDocumentaiUiv1beta3ProcessorVersionGenAiModelInfoFoundationGenAiModelInfo", "description": "Information for a pretrained Google-managed foundation model." }, "customGenAiModelInfo": { "description": "Information for a custom Generative AI model created by the user.", "$ref": "GoogleCloudDocumentaiUiv1beta3ProcessorVersionGenAiModelInfoCustomGenAiModelInfo" } }, "id": "GoogleCloudDocumentaiUiv1beta3ProcessorVersionGenAiModelInfo", "type": "object", "description": "Information about Generative AI model-based processor versions." }, "GoogleCloudDocumentaiV1beta3DeployProcessorVersionResponse": { "id": "GoogleCloudDocumentaiV1beta3DeployProcessorVersionResponse", "description": "Response message for the DeployProcessorVersion method.", "properties": {}, "type": "object" }, "GoogleCloudDocumentaiV1SetDefaultProcessorVersionMetadata": { "type": "object", "description": "The long-running operation metadata for the SetDefaultProcessorVersion method.", "id": "GoogleCloudDocumentaiV1SetDefaultProcessorVersionMetadata", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } } }, "GoogleCloudDocumentaiV1beta3CommonOperationMetadata": { "description": "The common metadata for long running operations.", "properties": { "stateMessage": { "description": "A message providing more details about the current state of processing.", "type": "string" }, "createTime": { "type": "string", "format": "google-datetime", "description": "The creation time of the operation." }, "updateTime": { "format": "google-datetime", "description": "The last update time of the operation.", "type": "string" }, "state": { "enumDescriptions": [ "Unspecified state.", "Operation is still running.", "Operation is being cancelled.", "Operation succeeded.", "Operation failed.", "Operation is cancelled." ], "description": "The state of the operation.", "type": "string", "enum": [ "STATE_UNSPECIFIED", "RUNNING", "CANCELLING", "SUCCEEDED", "FAILED", "CANCELLED" ] }, "resource": { "description": "A related resource to this operation.", "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata", "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentIdUnmanagedDocumentId": { "properties": { "docId": { "description": "Required. The ID of the document.", "type": "string" } }, "description": "Identifies a document uniquely within the scope of a dataset in unmanaged option.", "id": "GoogleCloudDocumentaiV1beta3DocumentIdUnmanagedDocumentId", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3BatchDeleteDocumentsMetadataIndividualBatchDeleteStatus": { "properties": { "documentId": { "description": "The document ID of the document.", "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentId" }, "status": { "$ref": "GoogleRpcStatus", "description": "The status of deleting the document in storage." } }, "description": "The status of each individual document in the batch delete process.", "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3BatchDeleteDocumentsMetadataIndividualBatchDeleteStatus" }, "GoogleCloudDocumentaiV1beta3DocumentAnnotations": { "id": "GoogleCloudDocumentaiV1beta3DocumentAnnotations", "description": "Represents the annotation of a block or a chunk.", "properties": { "description": { "description": "The description of the content with this annotation.", "type": "string" } }, "type": "object" }, "GoogleCloudDocumentaiV1BatchProcessMetadata": { "properties": { "state": { "description": "The state of the current batch processing.", "enumDescriptions": [ "The default value. This value is used if the state is omitted.", "Request operation is waiting for scheduling.", "Request is being processed.", "The batch processing completed successfully.", "The batch processing was being cancelled.", "The batch processing was cancelled.", "The batch processing has failed." ], "enum": [ "STATE_UNSPECIFIED", "WAITING", "RUNNING", "SUCCEEDED", "CANCELLING", "CANCELLED", "FAILED" ], "type": "string" }, "updateTime": { "format": "google-datetime", "type": "string", "description": "The last update time of the operation." }, "individualProcessStatuses": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1BatchProcessMetadataIndividualProcessStatus" }, "description": "The list of response details of each document." }, "stateMessage": { "description": "A message providing more details about the current state of processing. For example, the error message if the operation is failed.", "type": "string" }, "createTime": { "type": "string", "format": "google-datetime", "description": "The creation time of the operation." } }, "type": "object", "description": "The long-running operation metadata for BatchProcessDocuments.", "id": "GoogleCloudDocumentaiV1BatchProcessMetadata" }, "GoogleCloudDocumentaiV1beta3DocumentOutputConfigGcsOutputConfigShardingConfig": { "description": "The sharding config for the output document.", "type": "object", "properties": { "pagesOverlap": { "type": "integer", "description": "The number of overlapping pages between consecutive shards.", "format": "int32" }, "pagesPerShard": { "format": "int32", "type": "integer", "description": "The number of pages per shard." } }, "id": "GoogleCloudDocumentaiV1beta3DocumentOutputConfigGcsOutputConfigShardingConfig" }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFormValidationOperation": { "properties": { "fields": { "type": "array", "items": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleField" }, "description": "A list of fields to be used as operands." }, "operations": { "items": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFormValidationOperation" }, "type": "array", "description": "A list of recursive operations to be used as operands." }, "constants": { "type": "array", "description": "A list of constants to be used as operands.", "items": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleConstant" } }, "operationType": { "type": "string", "enumDescriptions": [ "", "", "", "", "", "", "", "", "", "" ], "enum": [ "OPERATION_TYPE_UNSPECIFIED", "OPERATION_TYPE_SUM", "OPERATION_TYPE_SUB", "OPERATION_TYPE_MUL", "OPERATION_TYPE_DIV", "OPERATION_TYPE_MAX", "OPERATION_TYPE_MIN", "OPERATION_TYPE_ABS", "OPERATION_TYPE_UNIQUE", "OPERATION_TYPE_COUNT" ], "description": "The operation type to be applied to all the operands." } }, "type": "object", "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFormValidationOperation" }, "GoogleCloudDocumentaiV1beta3TrainProcessorVersionMetadata": { "description": "The metadata that represents a processor version being created.", "id": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionMetadata", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." }, "trainingDatasetValidation": { "$ref": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionMetadataDatasetValidation", "description": "The training dataset validation information." }, "testDatasetValidation": { "description": "The test dataset validation information.", "$ref": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionMetadataDatasetValidation" } }, "type": "object" }, "GoogleCloudDocumentaiUiv1beta3DeleteProcessorMetadata": { "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "description": "The long-running operation metadata for the DeleteProcessor method.", "id": "GoogleCloudDocumentaiUiv1beta3DeleteProcessorMetadata", "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentPageDetectedBarcode": { "id": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedBarcode", "properties": { "layout": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout", "description": "Layout for DetectedBarcode." }, "barcode": { "description": "Detailed barcode information of the DetectedBarcode.", "$ref": "GoogleCloudDocumentaiV1beta3Barcode" } }, "type": "object", "description": "A detected barcode." }, "GoogleCloudDocumentaiV1beta3DocumentPageTokenDetectedBreak": { "properties": { "type": { "type": "string", "enumDescriptions": [ "Unspecified break type.", "A single whitespace.", "A wider whitespace.", "A hyphen that indicates that a token has been split across lines." ], "enum": [ "TYPE_UNSPECIFIED", "SPACE", "WIDE_SPACE", "HYPHEN" ], "description": "Detected break type." } }, "description": "Detected break at the end of a Token.", "id": "GoogleCloudDocumentaiV1beta3DocumentPageTokenDetectedBreak", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3ImportDocumentsMetadataIndividualImportStatus": { "description": "The status of each individual document in the import process.", "id": "GoogleCloudDocumentaiUiv1beta3ImportDocumentsMetadataIndividualImportStatus", "type": "object", "properties": { "status": { "description": "The status of the importing of the document.", "$ref": "GoogleRpcStatus" }, "outputDocumentId": { "description": "The document ID of imported document if it was successful, otherwise empty.", "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentId" }, "outputGcsDestination": { "type": "string", "description": "The output_gcs_destination of the processed document if it was successful, otherwise empty." }, "inputGcsSource": { "type": "string", "description": "The source Cloud Storage URI of the document." } } }, "GoogleCloudDocumentaiV1beta3SetDefaultProcessorVersionMetadata": { "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata" } }, "id": "GoogleCloudDocumentaiV1beta3SetDefaultProcessorVersionMetadata", "description": "The long-running operation metadata for the SetDefaultProcessorVersion method.", "type": "object" }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleEntityAlignmentRule": { "properties": { "fields": { "description": "The fields to be aligned.", "type": "array", "items": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleField" } }, "alignmentRule": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleAlignmentRule", "description": "The alignment rule to apply to the fields." } }, "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleEntityAlignmentRule", "type": "object", "description": "A rule that aligns specified fields with each other." }, "GoogleCloudDocumentaiV1beta3ImportDocumentsRequest": { "id": "GoogleCloudDocumentaiV1beta3ImportDocumentsRequest", "properties": { "batchDocumentsImportConfigs": { "type": "array", "description": "Required. The Cloud Storage uri containing raw documents that must be imported.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3ImportDocumentsRequestBatchDocumentsImportConfig" } } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentPageImageQualityScoresDetectedDefect": { "description": "Image Quality Defects", "id": "GoogleCloudDocumentaiV1beta3DocumentPageImageQualityScoresDetectedDefect", "properties": { "type": { "type": "string", "description": "Name of the defect type. Supported values are: - `quality/defect_blurry` - `quality/defect_noisy` - `quality/defect_dark` - `quality/defect_faint` - `quality/defect_text_too_small` - `quality/defect_document_cutoff` - `quality/defect_text_cutoff` - `quality/defect_glare`" }, "confidence": { "format": "float", "type": "number", "description": "Confidence of detected defect. Range `[0, 1]` where `1` indicates strong confidence that the defect exists." } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentStyle": { "id": "GoogleCloudDocumentaiV1beta3DocumentStyle", "type": "object", "description": "Annotation for common text style attributes. This adheres to CSS conventions as much as possible.", "properties": { "textStyle": { "description": "[Text style](https://www.w3schools.com/cssref/pr_font_font-style.asp). Possible values are `normal`, `italic`, and `oblique`.", "type": "string" }, "fontFamily": { "type": "string", "description": "Font family such as `Arial`, `Times New Roman`. https://www.w3schools.com/cssref/pr_font_font-family.asp" }, "fontSize": { "description": "Font size.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentStyleFontSize" }, "textAnchor": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentTextAnchor", "description": "Text anchor indexing into the Document.text." }, "color": { "description": "Text color.", "$ref": "GoogleTypeColor" }, "fontWeight": { "type": "string", "description": "[Font weight](https://www.w3schools.com/cssref/pr_font_weight.asp). Possible values are `normal`, `bold`, `bolder`, and `lighter`." }, "backgroundColor": { "description": "Text background color.", "$ref": "GoogleTypeColor" }, "textDecoration": { "description": "[Text decoration](https://www.w3schools.com/cssref/pr_text_text-decoration.asp). Follows CSS standard. ", "type": "string" } } }, "GoogleCloudDocumentaiV1beta3BatchProcessRequest": { "type": "object", "properties": { "inputDocuments": { "$ref": "GoogleCloudDocumentaiV1beta3BatchDocumentsInputConfig", "description": "The input documents for the BatchProcessDocuments method." }, "processOptions": { "description": "Inference-time options for the process API", "$ref": "GoogleCloudDocumentaiV1beta3ProcessOptions" }, "labels": { "description": "Optional. The labels with user-defined metadata for the request. Label keys and values can be no longer than 63 characters (Unicode codepoints) and can only contain lowercase letters, numeric characters, underscores, and dashes. International characters are allowed. Label values are optional. Label keys must start with a letter.", "type": "object", "additionalProperties": { "type": "string" } }, "documentOutputConfig": { "description": "The output configuration for the BatchProcessDocuments method.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentOutputConfig" }, "skipHumanReview": { "description": "Whether human review should be skipped for this request. Default to `false`.", "type": "boolean", "deprecated": true }, "inputConfigs": { "description": "The input config for each single document in the batch process.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3BatchProcessRequestBatchInputConfig" }, "deprecated": true, "type": "array" }, "outputConfig": { "$ref": "GoogleCloudDocumentaiV1beta3BatchProcessRequestBatchOutputConfig", "description": "The overall output config for batch process.", "deprecated": true } }, "id": "GoogleCloudDocumentaiV1beta3BatchProcessRequest", "description": "Request message for BatchProcessDocuments." }, "GoogleCloudDocumentaiUiv1beta3EntityTypeMetadata": { "type": "object", "properties": { "fieldTierMetadata": { "description": "Field tier metadata on the property", "$ref": "GoogleCloudDocumentaiUiv1beta3FieldTierMetadata" }, "inactive": { "type": "boolean", "description": "Whether the entity type should be considered inactive." }, "humanReviewLabelingMetadata": { "description": "Human review labeling config on the entity.", "$ref": "GoogleCloudDocumentaiUiv1beta3HumanReviewLabelingMetadata" }, "schemaEditabilityMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3SchemaEditabilityMetadata", "description": "Schema editability metadata on the entity." }, "humanReviewMetadata": { "description": "Human review config on the entity.", "$ref": "GoogleCloudDocumentaiUiv1beta3HumanReviewValidationMetadata" }, "schemaInferenceMetadata": { "description": "Schema inference metadata on the entity.", "$ref": "GoogleCloudDocumentaiUiv1beta3SchemaInferenceMetadata" } }, "description": "Metadata about an entity type.", "id": "GoogleCloudDocumentaiUiv1beta3EntityTypeMetadata" }, "GoogleCloudDocumentaiV1beta3ImportDocumentsMetadataImportConfigValidationResult": { "type": "object", "description": "The validation status of each import config. Status is set to an error if there are no documents to import in the `import_config`, or `OK` if the operation will try to proceed with at least one document.", "properties": { "inputGcsSource": { "description": "The source Cloud Storage URI specified in the import config.", "type": "string" }, "status": { "$ref": "GoogleRpcStatus", "description": "The validation status of import config." } }, "id": "GoogleCloudDocumentaiV1beta3ImportDocumentsMetadataImportConfigValidationResult" }, "GoogleCloudDocumentaiV1beta3DocumentRevisionHumanReview": { "type": "object", "properties": { "state": { "type": "string", "description": "Human review state. For example, `requested`, `succeeded`, `rejected`." }, "stateMessage": { "description": "A message providing more details about the current state of processing. For example, the rejection reason when the state is `rejected`.", "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentRevisionHumanReview", "description": "Human Review information of the document." }, "GoogleCloudDocumentaiV1beta3BatchDocumentsInputConfig": { "properties": { "gcsDocuments": { "$ref": "GoogleCloudDocumentaiV1beta3GcsDocuments", "description": "The set of documents individually specified on Cloud Storage." }, "gcsPrefix": { "$ref": "GoogleCloudDocumentaiV1beta3GcsPrefix", "description": "The set of documents that match the specified Cloud Storage `gcs_prefix`." } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3BatchDocumentsInputConfig", "description": "The common config to specify a set of documents used as input." }, "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequestInputData": { "id": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequestInputData", "properties": { "trainingDocuments": { "$ref": "GoogleCloudDocumentaiV1beta3BatchDocumentsInputConfig", "description": "The documents used for training the new version." }, "testDocuments": { "description": "The documents used for testing the trained version.", "$ref": "GoogleCloudDocumentaiV1beta3BatchDocumentsInputConfig" } }, "description": "The input data used to train a new ProcessorVersion.", "type": "object" }, "GoogleCloudDocumentaiV1beta3EvaluationReference": { "description": "Gives a short summary of an evaluation, and links to the evaluation itself.", "id": "GoogleCloudDocumentaiV1beta3EvaluationReference", "properties": { "aggregateMetricsExact": { "$ref": "GoogleCloudDocumentaiV1beta3EvaluationMetrics", "description": "An aggregate of the statistics for the evaluation with fuzzy matching off." }, "operation": { "description": "The resource name of the Long Running Operation for the evaluation.", "type": "string" }, "aggregateMetrics": { "description": "An aggregate of the statistics for the evaluation with fuzzy matching on.", "$ref": "GoogleCloudDocumentaiV1beta3EvaluationMetrics" }, "evaluation": { "type": "string", "description": "The resource name of the evaluation." } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3HumanReviewStatus": { "id": "GoogleCloudDocumentaiV1beta3HumanReviewStatus", "properties": { "stateMessage": { "description": "A message providing more details about the human review state.", "type": "string" }, "humanReviewOperation": { "description": "The name of the operation triggered by the processed document. This field is populated only when the state is `HUMAN_REVIEW_IN_PROGRESS`. It has the same response type and metadata as the long-running operation returned by ReviewDocument.", "type": "string" }, "state": { "type": "string", "enumDescriptions": [ "Human review state is unspecified. Most likely due to an internal error.", "Human review is skipped for the document. This can happen because human review isn't enabled on the processor or the processing request has been set to skip this document.", "Human review validation is triggered and passed, so no review is needed.", "Human review validation is triggered and the document is under review.", "Some error happened during triggering human review, see the state_message for details." ], "description": "The state of human review on the processing request.", "enum": [ "STATE_UNSPECIFIED", "SKIPPED", "VALIDATION_PASSED", "IN_PROGRESS", "ERROR" ] } }, "type": "object", "description": "The status of human review on a processed document." }, "GoogleCloudDocumentaiUiv1beta3DocumentIdUnmanagedDocumentId": { "id": "GoogleCloudDocumentaiUiv1beta3DocumentIdUnmanagedDocumentId", "type": "object", "description": "Identifies a document uniquely within the scope of a dataset in unmanaged option.", "properties": { "docId": { "description": "Required. The ID of the document.", "type": "string" } } }, "GoogleCloudDocumentaiV1beta3DocumentSchemaEntityTypeEnumValues": { "properties": { "values": { "items": { "type": "string" }, "description": "The individual values that this enum values type can include.", "type": "array" } }, "description": "Defines the a list of enum values.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentSchemaEntityTypeEnumValues" }, "GoogleCloudDocumentaiUiv1beta3RevisionRef": { "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3RevisionRef", "properties": { "latestProcessorVersion": { "type": "string", "description": "Reads the revision generated by the processor version. The format takes the full resource name of processor version. `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`" }, "revisionId": { "description": "Reads the revision given by the id.", "type": "string" }, "revisionCase": { "description": "Reads the revision by the predefined case.", "type": "string", "enumDescriptions": [ "Unspecified case, fall back to read the `LATEST_HUMAN_REVIEW`.", "The latest revision made by a human.", "The latest revision based on timestamp.", "The first (OCR) revision." ], "enum": [ "REVISION_CASE_UNSPECIFIED", "LATEST_HUMAN_REVIEW", "LATEST_TIMESTAMP", "BASE_OCR_REVISION" ] } }, "description": "The revision reference specifies which revision on the document to read." }, "GoogleCloudDocumentaiV1beta3OcrConfigHints": { "description": "Hints for OCR Engine", "type": "object", "id": "GoogleCloudDocumentaiV1beta3OcrConfigHints", "properties": { "languageHints": { "description": "List of BCP-47 language codes to use for OCR. In most cases, not specifying it yields the best results since it enables automatic language detection. For languages based on the Latin alphabet, setting hints is not needed. In rare cases, when the language of the text in the image is known, setting a hint will help get better results (although it will be a significant hindrance if the hint is wrong).", "items": { "type": "string" }, "type": "array" } } }, "GoogleCloudDocumentaiUiv1beta3DisableProcessorResponse": { "id": "GoogleCloudDocumentaiUiv1beta3DisableProcessorResponse", "type": "object", "description": "Response message for the DisableProcessor method. Intentionally empty proto for adding fields in future.", "properties": {} }, "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTableBlock": { "description": "Represents a table type block.", "id": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTableBlock", "properties": { "bodyRows": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTableRow" }, "description": "Body rows containing main table content." }, "headerRows": { "type": "array", "description": "Header rows at the top of the table.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutTableRow" } }, "caption": { "description": "Table caption/title.", "type": "string" }, "annotations": { "description": "Annotation of the table block.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentAnnotations" } }, "type": "object" }, "GoogleCloudDocumentaiV1ReviewDocumentOperationMetadata": { "description": "The long-running operation metadata for the ReviewDocument method.", "properties": { "questionId": { "description": "The Crowd Compute question ID.", "type": "string" }, "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "id": "GoogleCloudDocumentaiV1ReviewDocumentOperationMetadata", "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentEntityValidationOutputValidationResult": { "description": "Validation result for a single validation rule.", "type": "object", "properties": { "rule": { "type": "string", "description": "Optional. The name of the rule resource that is used for validation. Format: `projects/{project}/locations/{location}/rules/{rule}`" }, "validationResultType": { "enum": [ "VALIDATION_RESULT_TYPE_UNSPECIFIED", "VALIDATION_RESULT_TYPE_VALID", "VALIDATION_RESULT_TYPE_INVALID", "VALIDATION_RESULT_TYPE_SKIPPED", "VALIDATION_RESULT_TYPE_NOT_APPLICABLE" ], "type": "string", "description": "The result of the validation rule.", "enumDescriptions": [ "The validation result type is unspecified.", "The validation is valid.", "The validation is invalid.", "The validation is skipped.", "The validation is not applicable." ] }, "ruleDescription": { "description": "The description of the validation rule.", "type": "string" }, "validationDetails": { "type": "string", "description": "The detailed information of the running the validation process using the entity from the document based on the validation rule." }, "ruleName": { "type": "string", "description": "The display name of the validation rule." } }, "id": "GoogleCloudDocumentaiV1beta3DocumentEntityValidationOutputValidationResult" }, "GoogleCloudDocumentaiV1beta3EvaluationConfidenceLevelMetrics": { "id": "GoogleCloudDocumentaiV1beta3EvaluationConfidenceLevelMetrics", "type": "object", "description": "Evaluations metrics, at a specific confidence level.", "properties": { "confidenceLevel": { "format": "float", "type": "number", "description": "The confidence level." }, "metrics": { "$ref": "GoogleCloudDocumentaiV1beta3EvaluationMetrics", "description": "The metrics at the specific confidence level." } } }, "GoogleCloudLocationListLocationsResponse": { "description": "The response message for Locations.ListLocations.", "properties": { "locations": { "items": { "$ref": "GoogleCloudLocationLocation" }, "type": "array", "description": "A list of locations that matches the specified filter in the request." }, "nextPageToken": { "description": "The standard List next-page token.", "type": "string" } }, "type": "object", "id": "GoogleCloudLocationListLocationsResponse" }, "GoogleCloudDocumentaiUiv1beta3ImportProcessorVersionMetadata": { "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3ImportProcessorVersionMetadata", "description": "The long-running operation metadata for the ImportProcessorVersion method.", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata for the long-running operation." } } }, "GoogleCloudDocumentaiV1beta3ProcessorVersionGenAiModelInfoCustomGenAiModelInfo": { "description": "Information for a custom Generative AI model created by the user. These are created with `Create New Version` in either the `Call foundation model` or `Fine tuning` tabs.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3ProcessorVersionGenAiModelInfoCustomGenAiModelInfo", "properties": { "customModelType": { "description": "The type of custom model created by the user.", "type": "string", "enum": [ "CUSTOM_MODEL_TYPE_UNSPECIFIED", "VERSIONED_FOUNDATION", "FINE_TUNED" ], "enumDescriptions": [ "The model type is unspecified.", "The model is a versioned foundation model.", "The model is a finetuned foundation model." ] }, "baseProcessorVersionId": { "description": "The base processor version ID for the custom model.", "type": "string" } } }, "GoogleCloudDocumentaiV1beta3DocumentPageToken": { "id": "GoogleCloudDocumentaiV1beta3DocumentPageToken", "description": "A detected token.", "properties": { "layout": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout", "description": "Layout for Token." }, "provenance": { "deprecated": true, "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenance", "description": "The history of this annotation." }, "styleInfo": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageTokenStyleInfo", "description": "Text style attributes." }, "detectedBreak": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageTokenDetectedBreak", "description": "Detected break at the end of a Token." }, "detectedLanguages": { "description": "A list of detected languages together with confidence.", "type": "array", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" } } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentShardInfo": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentShardInfo", "description": "For a large document, sharding may be performed to produce several document shards. Each document shard contains this field to detail which shard it is.", "properties": { "shardCount": { "type": "string", "format": "int64", "description": "Total number of shards." }, "shardIndex": { "description": "The 0-based index of this shard.", "type": "string", "format": "int64" }, "pageOffset": { "format": "int32", "type": "integer", "description": "The index of the first page in Document.pages in the overall document global pages. Available for document shards created by the document splitter." }, "textOffset": { "type": "string", "format": "int64", "description": "The index of the first character in Document.text in the overall document global text." } } }, "GoogleCloudDocumentaiV1beta3ProcessorTypeLocationInfo": { "description": "The location information about where the processor is available.", "properties": { "locationId": { "type": "string", "description": "The location ID. For supported locations, refer to [regional and multi-regional support](/document-ai/docs/regions)." } }, "id": "GoogleCloudDocumentaiV1beta3ProcessorTypeLocationInfo", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3ExportDocumentsMetadataIndividualExportStatus": { "id": "GoogleCloudDocumentaiUiv1beta3ExportDocumentsMetadataIndividualExportStatus", "description": "The status of each individual document in the export process.", "type": "object", "properties": { "status": { "description": "The status of the exporting of the document.", "$ref": "GoogleRpcStatus" }, "outputGcsDestination": { "description": "The output_gcs_destination of the exported document if it was successful, otherwise empty.", "type": "string" }, "documentId": { "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentId", "description": "The path to source docproto of the document." } } }, "GoogleCloudDocumentaiV1beta3DisableProcessorResponse": { "properties": {}, "description": "Response message for the DisableProcessor method. Intentionally empty proto for adding fields in future.", "type": "object", "id": "GoogleCloudDocumentaiV1beta3DisableProcessorResponse" }, "GoogleCloudDocumentaiV1DeployProcessorVersionResponse": { "properties": {}, "type": "object", "description": "Response message for the DeployProcessorVersion method.", "id": "GoogleCloudDocumentaiV1DeployProcessorVersionResponse" }, "GoogleCloudDocumentaiV1beta3ReviewDocumentResponse": { "description": "Response message for the ReviewDocument method.", "type": "object", "properties": { "rejectionReason": { "description": "The reason why the review is rejected by reviewer.", "type": "string" }, "gcsDestination": { "description": "The Cloud Storage uri for the human reviewed document if the review is succeeded.", "type": "string" }, "state": { "enumDescriptions": [ "The default value. This value is used if the state is omitted.", "The review operation is rejected by the reviewer.", "The review operation is succeeded." ], "type": "string", "description": "The state of the review operation.", "enum": [ "STATE_UNSPECIFIED", "REJECTED", "SUCCEEDED" ] } }, "id": "GoogleCloudDocumentaiV1beta3ReviewDocumentResponse" }, "GoogleCloudDocumentaiV1beta3EvaluateProcessorVersionMetadata": { "id": "GoogleCloudDocumentaiV1beta3EvaluateProcessorVersionMetadata", "description": "Metadata of the EvaluateProcessorVersion method.", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3DeleteProcessorVersionMetadata": { "id": "GoogleCloudDocumentaiV1beta3DeleteProcessorVersionMetadata", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "description": "The long-running operation metadata for the DeleteProcessorVersion method.", "type": "object" }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleAlignmentRule": { "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleAlignmentRule", "type": "object", "description": "A rule for checking field alignment. Horizontal alignment checks if fields are on the same row by comparing y-coordinates of bounding box centers, while vertical alignment checks if fields are on the same column by comparing x-coordinates of bounding box centers.", "properties": { "alignmentType": { "enum": [ "ALIGNMENT_TYPE_UNSPECIFIED", "ALIGNMENT_TYPE_HORIZONTAL", "ALIGNMENT_TYPE_VERTICAL" ], "type": "string", "enumDescriptions": [ "", "", "" ] }, "tolerance": { "type": "number", "format": "float", "description": "The tolerance to use when comparing coordinates." } } }, "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequestCustomDocumentExtractionOptions": { "id": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequestCustomDocumentExtractionOptions", "properties": { "trainingMethod": { "enumDescriptions": [ "", "", "" ], "type": "string", "enum": [ "TRAINING_METHOD_UNSPECIFIED", "MODEL_BASED", "TEMPLATE_BASED" ], "description": "Optional. Training method to use for CDE training." } }, "description": "Options to control the training of the Custom Document Extraction (CDE) Processor.", "type": "object" }, "GoogleCloudDocumentaiV1beta3ImportDocumentsRequestBatchDocumentsImportConfigAutoSplitConfig": { "description": "The config for auto-split.", "properties": { "trainingSplitRatio": { "type": "number", "description": "Ratio of training dataset split.", "format": "float" } }, "id": "GoogleCloudDocumentaiV1beta3ImportDocumentsRequestBatchDocumentsImportConfigAutoSplitConfig", "type": "object" }, "GoogleCloudDocumentaiV1beta3EnableProcessorRequest": { "properties": {}, "type": "object", "id": "GoogleCloudDocumentaiV1beta3EnableProcessorRequest", "description": "Request message for the EnableProcessor method." }, "GoogleCloudDocumentaiUiv1beta3UpdateHumanReviewConfigMetadata": { "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3UpdateHumanReviewConfigMetadata", "description": "The long-running operation metadata for updating the human review configuration.", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } } }, "GoogleCloudDocumentaiV1beta3DocumentOutputConfig": { "id": "GoogleCloudDocumentaiV1beta3DocumentOutputConfig", "properties": { "gcsOutputConfig": { "description": "Output config to write the results to Cloud Storage.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentOutputConfigGcsOutputConfig" } }, "description": "Config that controls the output of documents. All documents will be written as a JSON file.", "type": "object" }, "GoogleCloudDocumentaiV1DeleteProcessorVersionMetadata": { "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiV1CommonOperationMetadata" } }, "id": "GoogleCloudDocumentaiV1DeleteProcessorVersionMetadata", "type": "object", "description": "The long-running operation metadata for the DeleteProcessorVersion method." }, "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutPageSpan": { "properties": { "pageEnd": { "format": "int32", "description": "Page where block ends in the document.", "type": "integer" }, "pageStart": { "format": "int32", "type": "integer", "description": "Page where block starts in the document." } }, "id": "GoogleCloudDocumentaiV1beta3DocumentDocumentLayoutDocumentLayoutBlockLayoutPageSpan", "description": "Represents where the block starts and ends in the document.", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3BatchUpdateDocumentsMetadataIndividualBatchUpdateStatus": { "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3BatchUpdateDocumentsMetadataIndividualBatchUpdateStatus", "description": "The status of each individual document in the batch update process.", "properties": { "documentId": { "description": "The document ID of the document.", "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentId" }, "status": { "$ref": "GoogleRpcStatus", "description": "The status of updating the document in storage." } } }, "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFormValidation": { "properties": { "leftOperand": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFormValidationOperation" }, "rightOperand": { "$ref": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFormValidationOperation" }, "validationOperator": { "description": "The relational operator to be applied to the operands.", "enumDescriptions": [ "", "", "", "", "", "", "" ], "enum": [ "OPERATION_TYPE_UNSPECIFIED", "OPERATION_TYPE_EQ", "OPERATION_TYPE_NE", "OPERATION_TYPE_LT", "OPERATION_TYPE_LE", "OPERATION_TYPE_GT", "OPERATION_TYPE_GE" ], "type": "string" } }, "type": "object", "id": "CloudAiDocumentaiLabHifiaToolsValidationValidatorInputValidationRuleFormValidation" }, "GoogleCloudDocumentaiV1beta3ImportProcessorVersionRequestExternalProcessorVersionSource": { "type": "object", "description": "The external source processor version.", "properties": { "processorVersion": { "type": "string", "description": "Required. The processor version name. Format: `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`" }, "serviceEndpoint": { "description": "Optional. The Document AI service endpoint. For example, 'https://us-documentai.googleapis.com'", "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3ImportProcessorVersionRequestExternalProcessorVersionSource" }, "GoogleCloudDocumentaiV1beta3DocumentProvenanceParent": { "description": "The parent element the current element is based on. Used for referencing/aligning, removal and replacement operations.", "id": "GoogleCloudDocumentaiV1beta3DocumentProvenanceParent", "type": "object", "properties": { "index": { "format": "int32", "type": "integer", "description": "The index of the parent item in the corresponding item list (eg. list of entities, properties within entities, etc.) in the parent revision." }, "id": { "format": "int32", "description": "The ID of the parent provenance.", "deprecated": true, "type": "integer" }, "revision": { "type": "integer", "description": "The index of the index into current revision's parent_ids list.", "format": "int32" } } }, "GoogleCloudDocumentaiUiv1beta3EvaluationReference": { "id": "GoogleCloudDocumentaiUiv1beta3EvaluationReference", "properties": { "aggregateMetrics": { "description": "An aggregate of the statistics for the evaluation with fuzzy matching on.", "$ref": "GoogleCloudDocumentaiUiv1beta3EvaluationMetrics" }, "aggregateMetricsExact": { "description": "An aggregate of the statistics for the evaluation with fuzzy matching off.", "$ref": "GoogleCloudDocumentaiUiv1beta3EvaluationMetrics" }, "operation": { "description": "The resource name of the Long Running Operation for the evaluation.", "type": "string" }, "evaluation": { "type": "string", "description": "The resource name of the evaluation." } }, "type": "object", "description": "Gives a short summary of an evaluation, and links to the evaluation itself." }, "GoogleCloudDocumentaiUiv1beta3EnableProcessorMetadata": { "description": "The long-running operation metadata for the EnableProcessor method.", "id": "GoogleCloudDocumentaiUiv1beta3EnableProcessorMetadata", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3BatchProcessMetadataIndividualProcessStatus": { "description": "The status of a each individual document in the batch process.", "id": "GoogleCloudDocumentaiV1beta3BatchProcessMetadataIndividualProcessStatus", "properties": { "outputGcsDestination": { "description": "The Cloud Storage output destination (in the request as DocumentOutputConfig.GcsOutputConfig.gcs_uri) of the processed document if it was successful, otherwise empty.", "type": "string" }, "inputGcsSource": { "type": "string", "description": "The source of the document, same as the input_gcs_source field in the request when the batch process started." }, "humanReviewOperation": { "deprecated": true, "type": "string", "description": "The name of the operation triggered by the processed document. If the human review process isn't triggered, this field will be empty. It has the same response type and metadata as the long-running operation returned by the ReviewDocument method." }, "humanReviewStatus": { "description": "The status of human review on the processed document.", "$ref": "GoogleCloudDocumentaiV1beta3HumanReviewStatus" }, "status": { "$ref": "GoogleRpcStatus", "description": "The status processing the document." } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3GetDocumentResponse": { "type": "object", "properties": { "document": { "$ref": "GoogleCloudDocumentaiV1beta3Document" } }, "id": "GoogleCloudDocumentaiV1beta3GetDocumentResponse" }, "GoogleCloudDocumentaiUiv1beta3ExportProcessorVersionResponse": { "description": "Response message associated with the ExportProcessorVersion operation.", "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3ExportProcessorVersionResponse", "properties": { "gcsUri": { "description": "The Cloud Storage URI containing the output artifacts.", "type": "string" } } }, "GoogleCloudDocumentaiUiv1beta3DocumentSchemaEntityTypeProperty": { "description": "Defines properties that can be part of the entity type.", "id": "GoogleCloudDocumentaiUiv1beta3DocumentSchemaEntityTypeProperty", "type": "object", "properties": { "description": { "type": "string", "description": "The description of the property. Could be used to provide more information about the property for model calls." }, "occurrenceType": { "type": "string", "enumDescriptions": [ "Unspecified occurrence type.", "There will be zero or one instance of this entity type. The same entity instance may be mentioned multiple times.", "The entity type will appear zero or multiple times.", "The entity type will only appear exactly once. The same entity instance may be mentioned multiple times.", "The entity type will appear once or more times." ], "enum": [ "OCCURRENCE_TYPE_UNSPECIFIED", "OPTIONAL_ONCE", "OPTIONAL_MULTIPLE", "REQUIRED_ONCE", "REQUIRED_MULTIPLE" ], "description": "Occurrence type limits the number of instances an entity type appears in the document." }, "valueType": { "type": "string", "description": "A reference to the value type of the property. This type is subject to the same conventions as the `Entity.base_types` field." }, "displayName": { "description": "User defined name for the property.", "type": "string" }, "method": { "enum": [ "METHOD_UNSPECIFIED", "EXTRACT", "DERIVE", "RELAXED_EXTRACT" ], "type": "string", "description": "Specifies how the entity's value is obtained.", "enumDescriptions": [ "Unspecified method. It defaults to `EXTRACT`.", "The entity's value is directly extracted as-is from the document text.", "The entity's value is derived through inference and is not necessarily an exact text extraction from the document.", "Attempts to extract the position and value as-is from the document text. If the value cannot be grounded in the text, it falls back to derivation through inference, for both position and value. Note: The inference fallback is currently only supported for boolean entities." ] }, "name": { "description": "The name of the property. Follows the same guidelines as the EntityType name.", "type": "string" }, "propertyMetadata": { "description": "Any additional metadata about the property can be added here.", "$ref": "GoogleCloudDocumentaiUiv1beta3PropertyMetadata" } } }, "GoogleCloudDocumentaiUiv1beta3AutoLabelDocumentsMetadataIndividualAutoLabelStatus": { "id": "GoogleCloudDocumentaiUiv1beta3AutoLabelDocumentsMetadataIndividualAutoLabelStatus", "properties": { "status": { "description": "The status of the document auto-labeling.", "$ref": "GoogleRpcStatus" }, "documentId": { "$ref": "GoogleCloudDocumentaiUiv1beta3DocumentId", "description": "The document ID of the auto-labeled document. This will replace the gcs_uri." } }, "description": "The status of individual documents in the auto-labeling process.", "type": "object" }, "GoogleCloudDocumentaiV1EnableProcessorMetadata": { "id": "GoogleCloudDocumentaiV1EnableProcessorMetadata", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "type": "object", "description": "The long-running operation metadata for the EnableProcessor method." }, "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequestFoundationModelTuningOptions": { "id": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequestFoundationModelTuningOptions", "description": "Options to control foundation model tuning of the processor.", "properties": { "trainSteps": { "type": "integer", "format": "int32", "description": "Optional. The number of steps to run for model tuning. Valid values are between 1 and 400. If not provided, recommended steps will be used." }, "learningRateMultiplier": { "description": "Optional. The multiplier to apply to the recommended learning rate. Valid values are between 0.1 and 10. If not provided, recommended learning rate will be used.", "type": "number", "format": "float" }, "previousFineTunedProcessorVersionName": { "type": "string", "description": "Optional. Resource name of a previously fine tuned version ID to copy the overwritten configs from. The base_processor_version should be newer than the base processor version used to fine tune this provided processor version. Format: `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`." } }, "type": "object" }, "GoogleCloudDocumentaiUiv1beta3ImportDocumentsMetadata": { "id": "GoogleCloudDocumentaiUiv1beta3ImportDocumentsMetadata", "description": "Metadata of the import document operation.", "properties": { "totalDocumentCount": { "description": "Total number of the documents that are qualified for importing.", "format": "int32", "type": "integer" }, "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." }, "individualImportStatuses": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3ImportDocumentsMetadataIndividualImportStatus" }, "description": "The list of response details of each document." }, "importConfigValidationResults": { "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3ImportDocumentsMetadataImportConfigValidationResult" }, "type": "array", "description": "Validation statuses of the batch documents import config." } }, "type": "object" }, "GoogleCloudDocumentaiUiv1beta3BatchMoveDocumentsMetadata": { "type": "object", "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." }, "destSplitType": { "enum": [ "DATASET_SPLIT_TYPE_UNSPECIFIED", "DATASET_SPLIT_TRAIN", "DATASET_SPLIT_TEST", "DATASET_SPLIT_UNASSIGNED" ], "type": "string", "description": "The destination dataset split type.", "enumDescriptions": [ "Default value if the enum is not set.", "Identifies the train documents.", "Identifies the test documents.", "Identifies the unassigned documents." ] }, "destDatasetType": { "enumDescriptions": [ "Default value if the enum is not set.", "Identifies the train documents.", "Identifies the test documents.", "Identifies the unassigned documents." ], "deprecated": true, "type": "string", "description": "The destination dataset split type.", "enum": [ "DATASET_SPLIT_TYPE_UNSPECIFIED", "DATASET_SPLIT_TRAIN", "DATASET_SPLIT_TEST", "DATASET_SPLIT_UNASSIGNED" ] }, "individualBatchMoveStatuses": { "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3BatchMoveDocumentsMetadataIndividualBatchMoveStatus" }, "type": "array", "description": "The list of response details of each document." } }, "id": "GoogleCloudDocumentaiUiv1beta3BatchMoveDocumentsMetadata" }, "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkField": { "properties": { "tableChunkField": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkTableChunkField", "description": "The table chunk field in the chunk." }, "imageChunkField": { "description": "The image chunk field in the chunk.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkImageChunkField" } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkField", "description": "The chunk field in the chunk. A chunk field could be one of the various types (for example, image, table) supported." }, "GoogleCloudDocumentaiV1beta3DocumentPageLine": { "properties": { "layout": { "description": "Layout for Line.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout" }, "detectedLanguages": { "description": "A list of detected languages together with confidence.", "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" }, "type": "array" }, "provenance": { "deprecated": true, "description": "The history of this annotation.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentProvenance" } }, "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentPageLine", "description": "A collection of tokens that a human would perceive as a line. Does not cross column boundaries, can be horizontal, vertical, etc." }, "GoogleCloudDocumentaiV1beta3TrainProcessorVersionResponse": { "type": "object", "description": "The response for TrainProcessorVersion.", "properties": { "processorVersion": { "type": "string", "description": "The resource name of the processor version produced by training." } }, "id": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionResponse" }, "GoogleCloudDocumentaiUiv1beta3PropertyMetadata": { "description": "Metadata about a property.", "properties": { "humanReviewLabelingMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3HumanReviewLabelingMetadata", "description": "Human review labeling config on the property." }, "inactive": { "type": "boolean", "description": "Whether the property should be considered as \"inactive\"." }, "fieldTierMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3FieldTierMetadata", "description": "Field tier metadata on the property" }, "schemaInferenceMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3SchemaInferenceMetadata", "description": "Schema inference metadata on the property." }, "humanReviewMetadata": { "description": "Human review validation config on the property.", "$ref": "GoogleCloudDocumentaiUiv1beta3HumanReviewValidationMetadata" }, "schemaEditabilityMetadata": { "$ref": "GoogleCloudDocumentaiUiv1beta3SchemaEditabilityMetadata", "description": "Schema editability metadata on the property." }, "fieldExtractionMetadata": { "description": "Field extraction metadata on the property.", "$ref": "GoogleCloudDocumentaiUiv1beta3FieldExtractionMetadata" } }, "id": "GoogleCloudDocumentaiUiv1beta3PropertyMetadata", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3BatchUpdateDocumentsResponse": { "properties": {}, "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3BatchUpdateDocumentsResponse", "description": "Response of the batch update documents operation." }, "GoogleCloudDocumentaiV1beta3DeployProcessorVersionMetadata": { "properties": { "commonMetadata": { "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata", "description": "The basic metadata of the long-running operation." } }, "id": "GoogleCloudDocumentaiV1beta3DeployProcessorVersionMetadata", "description": "The long-running operation metadata for the DeployProcessorVersion method.", "type": "object" }, "GoogleCloudDocumentaiUiv1beta3GroundingSettings": { "description": "Settings for grounding extractions.", "properties": { "groundingType": { "description": "The type of grounding to apply.", "enum": [ "GROUNDING_TYPE_UNSPECIFIED", "HARD", "OCR_RELAXED", "NO_GROUNDING" ], "enumDescriptions": [ "The default value. Behaves like HARD.", "Requires exact match with OCR text for extractions.", "Allows for minor discrepancies between LLM output and OCR text.", "LLM extractions are used without strict OCR matching. Bounding boxes may be approximated or absent." ], "type": "string" } }, "type": "object", "id": "GoogleCloudDocumentaiUiv1beta3GroundingSettings" }, "GoogleCloudDocumentaiV1beta3ProcessOptions": { "id": "GoogleCloudDocumentaiV1beta3ProcessOptions", "type": "object", "properties": { "ocrConfig": { "$ref": "GoogleCloudDocumentaiV1beta3OcrConfig", "description": "Only applicable to `OCR_PROCESSOR` and `FORM_PARSER_PROCESSOR`. Returns error if set on other processor types." }, "fromEnd": { "type": "integer", "description": "Only process certain pages from the end, same as above.", "format": "int32" }, "schemaOverride": { "description": "Optional. Override the schema of the ProcessorVersion. Will return an Invalid Argument error if this field is set when the underlying ProcessorVersion doesn't support schema override.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentSchema" }, "individualPageSelector": { "description": "Which pages to process (1-indexed).", "$ref": "GoogleCloudDocumentaiV1beta3ProcessOptionsIndividualPageSelector" }, "layoutConfig": { "description": "Optional. Only applicable to `LAYOUT_PARSER_PROCESSOR`. Returns error if set on other processor types.", "$ref": "GoogleCloudDocumentaiV1beta3ProcessOptionsLayoutConfig" }, "fromStart": { "description": "Only process certain pages from the start. Process all if the document has fewer pages.", "format": "int32", "type": "integer" } }, "description": "Options for Process API" }, "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageHeader": { "properties": { "text": { "type": "string", "description": "Header in text format." }, "pageSpan": { "description": "Page span of the header.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageSpan" } }, "id": "GoogleCloudDocumentaiV1beta3DocumentChunkedDocumentChunkChunkPageHeader", "type": "object", "description": "Represents the page header associated with the chunk." }, "GoogleCloudDocumentaiV1beta3DisableProcessorMetadata": { "properties": { "commonMetadata": { "description": "The basic metadata of the long-running operation.", "$ref": "GoogleCloudDocumentaiV1beta3CommonOperationMetadata" } }, "id": "GoogleCloudDocumentaiV1beta3DisableProcessorMetadata", "type": "object", "description": "The long-running operation metadata for the DisableProcessor method." }, "GoogleCloudDocumentaiUiv1beta3SampleDocumentsResponse": { "id": "GoogleCloudDocumentaiUiv1beta3SampleDocumentsResponse", "description": "Response of the sample documents operation.", "type": "object", "properties": { "sampleTestStatus": { "description": "The status of sampling documents in test split.", "$ref": "GoogleRpcStatus" }, "sampleTrainingStatus": { "$ref": "GoogleRpcStatus", "description": "The status of sampling documents in training split." }, "selectedDocuments": { "type": "array", "items": { "$ref": "GoogleCloudDocumentaiUiv1beta3SampleDocumentsResponseSelectedDocument" }, "description": "The result of the sampling process." } } }, "GoogleCloudDocumentaiV1beta3GroundingSettings": { "description": "Settings for grounding extractions.", "type": "object", "properties": { "groundingType": { "description": "The type of grounding to apply.", "enumDescriptions": [ "The default value. Behaves like HARD.", "Requires exact match with OCR text for extractions.", "Allows for minor discrepancies between LLM output and OCR text.", "LLM extractions are used without strict OCR matching. Bounding boxes may be approximated or absent." ], "enum": [ "GROUNDING_TYPE_UNSPECIFIED", "HARD", "OCR_RELAXED", "NO_GROUNDING" ], "type": "string" } }, "id": "GoogleCloudDocumentaiV1beta3GroundingSettings" }, "GoogleCloudDocumentaiV1beta3OcrConfig": { "description": "Config for Document OCR.", "id": "GoogleCloudDocumentaiV1beta3OcrConfig", "properties": { "disableCharacterBoxesDetection": { "description": "Turn off character box detector in OCR engine. Character box detection is enabled by default in OCR 2.0 (and later) processors.", "type": "boolean" }, "hints": { "description": "Hints for the OCR model.", "$ref": "GoogleCloudDocumentaiV1beta3OcrConfigHints" }, "enableNativePdfParsing": { "description": "Enables special handling for PDFs with existing text information. Results in better text extraction quality in such PDF inputs.", "type": "boolean" }, "computeStyleInfo": { "deprecated": true, "description": "Turn on font identification model and return font style information. Deprecated, use PremiumFeatures.compute_style_info instead.", "type": "boolean" }, "enableImageQualityScores": { "type": "boolean", "description": "Enables intelligent document quality scores after OCR. Can help with diagnosing why OCR responses are of poor quality for a given input. Adds additional latency comparable to regular OCR to the process call." }, "advancedOcrOptions": { "type": "array", "items": { "type": "string" }, "description": "A list of advanced OCR options to further fine-tune OCR behavior. Current valid values are: - `legacy_layout`: a heuristics layout detection algorithm, which serves as an alternative to the current ML-based layout detection algorithm. Customers can choose the best suitable layout algorithm based on their situation." }, "enableSymbol": { "description": "Includes symbol level OCR information if set to true.", "type": "boolean" }, "premiumFeatures": { "$ref": "GoogleCloudDocumentaiV1beta3OcrConfigPremiumFeatures", "description": "Configurations for premium OCR features." } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequest": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequest", "properties": { "baseProcessorVersion": { "description": "Optional. The processor version to use as a base for training. This processor version must be a child of `parent`. Format: `projects/{project}/locations/{location}/processors/{processor}/processorVersions/{processorVersion}`.", "type": "string" }, "customDocumentExtractionOptions": { "description": "Options to control Custom Document Extraction (CDE) Processor.", "$ref": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequestCustomDocumentExtractionOptions" }, "inputData": { "description": "Optional. The input data used to train the ProcessorVersion.", "$ref": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequestInputData" }, "foundationModelTuningOptions": { "description": "Options to control foundation model tuning of a processor.", "$ref": "GoogleCloudDocumentaiV1beta3TrainProcessorVersionRequestFoundationModelTuningOptions" }, "processorVersionOverrides": { "$ref": "GoogleCloudDocumentaiV1beta3ProcessorVersionOverrides", "description": "Optional. Options to override structures in the base processor version." }, "processorVersion": { "description": "Required. The processor version to be created.", "$ref": "GoogleCloudDocumentaiV1beta3ProcessorVersion" }, "documentSchema": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentSchema", "description": "Optional. The schema the processor version will be trained with." } }, "description": "Request message for the TrainProcessorVersion method." }, "GoogleCloudDocumentaiUiv1beta3ExportDocumentsMetadataSplitExportStat": { "description": "The statistic representing a dataset split type for this export.", "id": "GoogleCloudDocumentaiUiv1beta3ExportDocumentsMetadataSplitExportStat", "properties": { "totalDocumentCount": { "type": "integer", "format": "int32", "description": "Total number of documents with the given dataset split type to be exported." }, "splitType": { "type": "string", "description": "The dataset split type.", "enum": [ "DATASET_SPLIT_TYPE_UNSPECIFIED", "DATASET_SPLIT_TRAIN", "DATASET_SPLIT_TEST", "DATASET_SPLIT_UNASSIGNED" ], "enumDescriptions": [ "Default value if the enum is not set.", "Identifies the train documents.", "Identifies the test documents.", "Identifies the unassigned documents." ] } }, "type": "object" }, "GoogleCloudDocumentaiV1beta3DocumentPageVisualElement": { "type": "object", "id": "GoogleCloudDocumentaiV1beta3DocumentPageVisualElement", "description": "Detected non-text visual elements, for example, checkbox, signature etc. on the page.", "properties": { "type": { "description": "Type of the VisualElement.", "type": "string" }, "detectedLanguages": { "items": { "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageDetectedLanguage" }, "type": "array", "description": "A list of detected languages together with confidence." }, "layout": { "description": "Layout for VisualElement.", "$ref": "GoogleCloudDocumentaiV1beta3DocumentPageLayout" } } } }, "batchPath": "batch", "version": "v1beta3", "rootUrl": "https://documentai.googleapis.com/", "revision": "20261026", "documentationLink": "https://cloud.google.com/document-ai/docs/", "servicePath": "", "canonicalName": "Document", "icons": { "x16": "http://www.google.com/images/icons/product/search-16.gif", "x32": "http://www.google.com/images/icons/product/search-32.gif" }, "endpoints": [ { "endpointUrl": "https://documentai.us.rep.googleapis.com/", "location": "us", "description": "Regional Endpoint" }, { "description": "Regional Endpoint", "location": "eu", "endpointUrl": "https://documentai.eu.rep.googleapis.com/" }, { "endpointUrl": "https://documentai.asia-south1.rep.googleapis.com/", "location": "asia-south1", "description": "Regional Endpoint" }, { "location": "asia-southeast1", "endpointUrl": "https://documentai.asia-southeast1.rep.googleapis.com/", "description": "Regional Endpoint" }, { "location": "northamerica-northeast1", "endpointUrl": "https://documentai.northamerica-northeast1.rep.googleapis.com/", "description": "Regional Endpoint" }, { "description": "Regional Endpoint", "location": "australia-southeast1", "endpointUrl": "https://documentai.australia-southeast1.rep.googleapis.com/" }, { "location": "europe-west2", "description": "Regional Endpoint", "endpointUrl": "https://documentai.europe-west2.rep.googleapis.com/" }, { "endpointUrl": "https://documentai.europe-west3.rep.googleapis.com/", "location": "europe-west3", "description": "Regional Endpoint" } ], "version_module": true, "discoveryVersion": "v1", "ownerName": "Google", "ownerDomain": "google.com" }