UNPKG

@pulumi/gcp

Version:

A Pulumi package for creating and managing Google Cloud Platform resources.

290 lines • 11.7 kB
"use strict"; // *** WARNING: this file was generated by pulumi-language-nodejs. *** // *** Do not edit by hand unless you're certain you know what you are doing! *** var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) { if (k2 === undefined) k2 = k; var desc = Object.getOwnPropertyDescriptor(m, k); if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) { desc = { enumerable: true, get: function() { return m[k]; } }; } Object.defineProperty(o, k2, desc); }) : (function(o, m, k, k2) { if (k2 === undefined) k2 = k; o[k2] = m[k]; })); var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) { Object.defineProperty(o, "default", { enumerable: true, value: v }); }) : function(o, v) { o["default"] = v; }); var __importStar = (this && this.__importStar) || function (mod) { if (mod && mod.__esModule) return mod; var result = {}; if (mod != null) for (var k in mod) if (k !== "default" && Object.prototype.hasOwnProperty.call(mod, k)) __createBinding(result, mod, k); __setModuleDefault(result, mod); return result; }; Object.defineProperty(exports, "__esModule", { value: true }); exports.DataObject = void 0; const pulumi = __importStar(require("@pulumi/pulumi")); const utilities = __importStar(require("../utilities")); /** * A DataObject is a single item of data (with optional vectors) stored in a * Vector Search Collection. Each DataObject conforms to the parent * Collection's `dataSchema` and `vectorSchema`. * * This resource always issues one `CreateDataObject` request per Terraform * resource block. It does NOT use the `batchCreate` REST endpoint -- * Terraform's resource lifecycle is inherently per-object, so batching * across resources is not modeled. When you use `forEach` or `count`, * Terraform will still issue individual requests, up to `-parallelism` * in parallel. * * For ingesting more than a few hundred items, prefer one of the * following out-of-band paths instead of Terraform: * * * `importDataObjects` (bulk ingest from Cloud Storage) -- highest * throughput, but only available *before* any Index is created on * the Collection. * * `batchCreate` (up to ~1000 items per call) -- available at any * time, but must be driven from your own client code, not Terraform. * * Once an Index exists on the Collection, `importDataObjects` is no * longer available and DataObjects must be created via `CreateDataObject` * (as this resource does) or via `batchCreate`. * * ## Example Usage * * ### Vectorsearch Data Object Basic * * ```typescript * import * as pulumi from "@pulumi/pulumi"; * import * as gcp from "@pulumi/gcp"; * * // NOTE: This resource issues one CreateDataObject request per block. * // It does NOT batch across resources. Terraform will parallelize a * // 'for_each' up to '-parallelism', but each item is still a separate * // HTTP call. * // * // For bulk ingestion of many items, prefer one of these out-of-band * // paths instead of Terraform: * // * 'importDataObjects' (from Cloud Storage) -- highest throughput, * // but only available *before* any Index is created on the Collection. * // * 'batchCreate' (up to ~1000 items per call) -- available at any * // time, but must be driven from client code, not Terraform. * const parent = new gcp.vectorsearch.Collection("parent", { * location: "us-central1", * collectionId: "example-collection", * displayName: "My Awesome Collection", * description: "This collection stores important data.", * dataSchema: `{ * \\"type\\": \\"object\\", * \\"properties\\": { * \\"title\\": { * \\"type\\": \\"string\\" * }, * \\"plot\\": { * \\"type\\": \\"string\\" * } * } * } * `, * vectorSchemas: [{ * fieldName: "text_embedding", * denseVector: { * dimensions: 768, * vertexEmbeddingConfig: { * modelId: "text-embedding-005", * taskType: "RETRIEVAL_DOCUMENT", * textTemplate: "Title: {title} ---- Plot: {plot}", * }, * }, * }], * }); * // Because the parent Collection's 'text_embedding' field is configured * // with a 'vertex_embedding_config', the server will populate the vector * // automatically from 'data.title' and 'data.plot' -- no explicit * // 'vectors' block is required. * const example_data_object = new gcp.vectorsearch.DataObject("example-data-object", { * location: "us-central1", * collectionId: parent.collectionId, * dataObjectId: "example-data-object", * data: JSON.stringify({ * title: "The Matrix", * plot: "A computer hacker learns about the true nature of reality.", * }), * }); * ``` * ### Vectorsearch Data Object With Vectors * * ```typescript * import * as pulumi from "@pulumi/pulumi"; * import * as gcp from "@pulumi/gcp"; * * // NOTE: This resource issues one CreateDataObject request per block. * // It does NOT batch across resources. Terraform will parallelize a * // 'for_each' up to '-parallelism', but each item is still a separate * // HTTP call. * // * // For bulk ingestion of many items, prefer one of these out-of-band * // paths instead of Terraform: * // * 'importDataObjects' (from Cloud Storage) -- highest throughput, * // but only available *before* any Index is created on the Collection. * // * 'batchCreate' (up to ~1000 items per call) -- available at any * // time, but must be driven from client code, not Terraform. * const parent = new gcp.vectorsearch.Collection("parent", { * location: "us-central1", * collectionId: "example-vectors-collection", * displayName: "My BYO-Embedding Collection", * description: "Collection whose vectors are supplied by the client.", * dataSchema: `{ * \\"type\\": \\"object\\", * \\"properties\\": { * \\"title\\": { * \\"type\\": \\"string\\" * }, * \\"category\\": { * \\"type\\": \\"string\\" * } * } * } * `, * vectorSchemas: [ * { * fieldName: "dense_embedding", * denseVector: { * dimensions: 4, * }, * }, * { * fieldName: "sparse_embedding", * sparseVector: {}, * }, * ], * }); * const example_vectors_data_object = new gcp.vectorsearch.DataObject("example-vectors-data-object", { * location: "us-central1", * collectionId: parent.collectionId, * dataObjectId: "example-vectors-data-object", * data: JSON.stringify({ * title: "The Matrix", * category: "movie", * }), * vectors: [ * { * fieldName: "dense_embedding", * dense: { * values: [ * 0.11, * 0.22, * 0.33, * 0.44, * ], * }, * }, * { * fieldName: "sparse_embedding", * sparse: { * values: [ * 0.9, * 0.5, * 0.1, * ], * indices: [ * 3, * 17, * 42, * ], * }, * }, * ], * }); * ``` * * ## Import * * DataObject can be imported using any of these accepted formats: * * * `projects/{{project}}/locations/{{location}}/collections/{{collection_id}}/dataObjects/{{data_object_id}}` * * `{{project}}/{{location}}/{{collection_id}}/{{data_object_id}}` * * `{{location}}/{{collection_id}}/{{data_object_id}}` * * When using the `pulumi import` command, DataObject can be imported using one of the formats above. For example: * * ```sh * $ pulumi import gcp:vectorsearch/dataObject:DataObject default projects/{{project}}/locations/{{location}}/collections/{{collection_id}}/dataObjects/{{data_object_id}} * $ pulumi import gcp:vectorsearch/dataObject:DataObject default {{project}}/{{location}}/{{collection_id}}/{{data_object_id}} * $ pulumi import gcp:vectorsearch/dataObject:DataObject default {{location}}/{{collection_id}}/{{data_object_id}} * ``` */ class DataObject extends pulumi.CustomResource { /** * Get an existing DataObject resource's state with the given name, ID, and optional extra * properties used to qualify the lookup. * * @param name The _unique_ name of the resulting resource. * @param id The _unique_ provider ID of the resource to lookup. * @param state Any extra arguments used during the lookup. * @param opts Optional settings to control the behavior of the CustomResource. */ static get(name, id, state, opts) { return new DataObject(name, state, { ...opts, id: id }); } /** @internal */ static __pulumiType = 'gcp:vectorsearch/dataObject:DataObject'; /** * Returns true if the given object is an instance of DataObject. This is designed to work even * when multiple copies of the Pulumi SDK have been loaded into the same process. */ static isInstance(obj) { if (obj === undefined || obj === null) { return false; } return obj['__pulumiType'] === DataObject.__pulumiType; } constructor(name, argsOrState, opts) { let resourceInputs = {}; opts = opts || {}; if (opts.id) { const state = argsOrState; resourceInputs["collectionId"] = state?.collectionId; resourceInputs["createTime"] = state?.createTime; resourceInputs["data"] = state?.data; resourceInputs["dataObjectId"] = state?.dataObjectId; resourceInputs["deletionPolicy"] = state?.deletionPolicy; resourceInputs["etag"] = state?.etag; resourceInputs["location"] = state?.location; resourceInputs["name"] = state?.name; resourceInputs["project"] = state?.project; resourceInputs["updateTime"] = state?.updateTime; resourceInputs["vectors"] = state?.vectors; } else { const args = argsOrState; if (args?.collectionId === undefined && !opts.urn) { throw new Error("Missing required property 'collectionId'"); } if (args?.dataObjectId === undefined && !opts.urn) { throw new Error("Missing required property 'dataObjectId'"); } if (args?.location === undefined && !opts.urn) { throw new Error("Missing required property 'location'"); } resourceInputs["collectionId"] = args?.collectionId; resourceInputs["data"] = args?.data; resourceInputs["dataObjectId"] = args?.dataObjectId; resourceInputs["deletionPolicy"] = args?.deletionPolicy; resourceInputs["etag"] = args?.etag; resourceInputs["location"] = args?.location; resourceInputs["project"] = args?.project; resourceInputs["vectors"] = args?.vectors; resourceInputs["createTime"] = undefined /*out*/; resourceInputs["name"] = undefined /*out*/; resourceInputs["updateTime"] = undefined /*out*/; } opts = pulumi.mergeOptions(utilities.resourceOptsDefaults(), opts); super(DataObject.__pulumiType, name, resourceInputs, opts); } } exports.DataObject = DataObject; //# sourceMappingURL=dataObject.js.map