{"version":3,"file":"azure_documentdb.cjs","names":["VectorStore","MongoClient","ObjectId","Document"],"sources":["../src/azure_documentdb.ts"],"sourcesContent":["import {\n ObjectId,\n Collection,\n Document as MongoDBDocument,\n MongoClient,\n Db,\n Filter,\n} from \"mongodb\";\nimport type { EmbeddingsInterface } from \"@langchain/core/embeddings\";\nimport {\n MaxMarginalRelevanceSearchOptions,\n VectorStore,\n} from \"@langchain/core/vectorstores\";\nimport { Document, DocumentInterface } from \"@langchain/core/documents\";\nimport { maximalMarginalRelevance } from \"@langchain/core/utils/math\";\nimport { getEnvironmentVariable } from \"@langchain/core/utils/env\";\n\n/** Azure DocumentDB Similarity type. */\nexport const AzureDocumentDBSimilarityType = {\n /** Cosine similarity */\n COS: \"COS\",\n /** Inner - product */\n IP: \"IP\",\n /** Euclidian distance */\n L2: \"L2\",\n} as const;\n\n/** Azure DocumentDB Similarity type. */\nexport type AzureDocumentDBSimilarityType =\n (typeof AzureDocumentDBSimilarityType)[keyof typeof AzureDocumentDBSimilarityType];\n\n/** Azure DocumentDB Index Options. */\nexport type AzureDocumentDBIndexOptions = {\n /** Skips automatic index creation. */\n readonly skipCreate?: boolean;\n\n readonly indexType?: \"ivf\" | \"hnsw\" | \"diskann\";\n /** Number of clusters that the inverted file (IVF) index uses to group the vector data. */\n readonly numLists?: number;\n /** Number of dimensions for vector similarity. */\n readonly dimensions?: number;\n /** Similarity metric to use with the IVF index. */\n readonly similarity?: AzureDocumentDBSimilarityType;\n /** The max number of connections per layer with the HNSW index. */\n readonly m?: number;\n /** The size of the dynamic candidate list for constructing the graph with the HNSW index. */\n readonly efConstruction?: number;\n /** Max number of neighbors with the DiskANN index */\n readonly maxDegree?: number;\n /** L value for index building with the DiskANN index */\n readonly lBuild?: number;\n /** L value for index searching with the DiskANN index */\n readonly lSearch?: number;\n};\n\n/** Azure DocumentDB delete Parameters. */\nexport type AzureDocumentDBDeleteParams = {\n /** List of IDs for the documents to be removed. */\n readonly ids?: string | string[];\n /** MongoDB filter object or list of IDs for the documents to be removed. */\n readonly filter?: Filter;\n};\n\n/** Configuration options for the `AzureDocumentDBVectorStore` constructor. */\nexport interface AzureDocumentDBConfig {\n readonly client?: MongoClient;\n readonly connectionString?: string;\n readonly databaseName?: string;\n readonly collectionName?: string;\n readonly indexName?: string;\n readonly textKey?: string;\n readonly embeddingKey?: string;\n readonly indexOptions?: AzureDocumentDBIndexOptions;\n}\n\n/**\n * Azure DocumentDB vector store.\n * To use this, you should have both:\n * - the `mongodb` NPM package installed\n * - a connection string associated with an Azure DocumentDB cluster\n *\n * You do not need to create a database or collection, it will be created\n * automatically.\n *\n * You also need an index on the collection, which is by default created\n * automatically using the `createIndex` method.\n */\nexport class AzureDocumentDBVectorStore extends VectorStore {\n get lc_secrets(): { [key: string]: string } {\n return {\n connectionString: \"AZURE_DOCUMENTDB_CONNECTION_STRING\",\n };\n }\n\n private connectPromise: Promise;\n\n private initPromise?: Promise;\n\n private readonly client: MongoClient | undefined;\n\n private database: Db;\n\n private collection: Collection;\n\n readonly indexName: string;\n\n readonly textKey: string;\n\n readonly embeddingKey: string;\n\n private readonly indexOptions: AzureDocumentDBIndexOptions;\n\n /**\n * Initializes the AzureDocumentDBVectorStore.\n * Connect the client to the database and create the container, creating them if needed.\n * @returns A promise that resolves when the AzureDocumentDBVectorStore has been initialized.\n */\n initialize: () => Promise;\n\n _vectorstoreType(): string {\n return \"azure_documentdb\";\n }\n\n constructor(\n embeddings: EmbeddingsInterface,\n dbConfig: AzureDocumentDBConfig\n ) {\n super(embeddings, dbConfig);\n\n const connectionString =\n dbConfig.connectionString ??\n getEnvironmentVariable(\"AZURE_DOCUMENTDB_CONNECTION_STRING\") ??\n getEnvironmentVariable(\"AZURE_COSMOSDB_MONGODB_CONNECTION_STRING\");\n\n if (!dbConfig.client && !connectionString) {\n throw new Error(\n \"AzureDocumentDBVectorStore client or connection string must be set.\"\n );\n }\n\n if (!dbConfig.client) {\n this.client = new MongoClient(connectionString!, {\n appName: \"langchainjs\",\n });\n }\n\n const client = dbConfig.client || this.client!;\n const databaseName = dbConfig.databaseName ?? \"documentsDB\";\n const collectionName = dbConfig.collectionName ?? \"documents\";\n this.indexName = dbConfig.indexName ?? \"vectorSearchIndex\";\n this.textKey = dbConfig.textKey ?? \"textContent\";\n this.embeddingKey = dbConfig.embeddingKey ?? \"vectorContent\";\n this.indexOptions = dbConfig.indexOptions ?? {};\n\n // Deferring initialization to the first call to `initialize`\n this.initialize = () => {\n if (this.initPromise === undefined) {\n this.initPromise = this.init(\n client,\n databaseName,\n collectionName\n ).catch((error) => {\n console.error(\n \"Error during AzureDocumentDBVectorStore initialization:\",\n error\n );\n });\n }\n\n return this.initPromise;\n };\n }\n\n /**\n * Checks if the specified index name during instance construction exists\n * on the collection.\n * @returns A promise that resolves to a boolean indicating if the index exists.\n */\n async checkIndexExists(): Promise {\n await this.initialize();\n const indexes = await this.collection.listIndexes().toArray();\n return indexes.some((index) => index.name === this.indexName);\n }\n\n /**\n * Deletes the index specified during instance construction if it exists.\n * @returns A promise that resolves when the index has been deleted.\n */\n async deleteIndex(): Promise {\n await this.initialize();\n if (await this.checkIndexExists()) {\n await this.collection.dropIndex(this.indexName);\n }\n }\n\n /**\n * Creates an index on the collection with the specified index name during\n * instance construction.\n *\n * Setting the numLists parameter correctly is important for achieving good\n * accuracy and performance.\n * Since the vector store uses IVF as the indexing strategy, you should\n * create the index only after you have loaded a large enough sample\n * documents to ensure that the centroids for the respective buckets are\n * fairly distributed.\n *\n * We recommend that numLists is set to documentCount/1000 for up to\n * 1 million documents and to sqrt(documentCount) for more than 1 million\n * documents.\n * As the number of items in your database grows, you should tune numLists\n * to be larger in order to achieve good latency performance for vector\n * search.\n *\n * If you're experimenting with a new scenario or creating a small demo,\n * you can start with numLists set to 1 to perform a brute-force search\n * across all vectors.\n * This should provide you with the most accurate results from the vector\n * search, however be aware that the search speed and latency will be slow.\n * After your initial setup, you should go ahead and tune the numLists\n * parameter using the above guidance.\n * @param numLists This integer is the number of clusters that the inverted\n * file (IVF) index uses to group the vector data.\n * We recommend that numLists is set to documentCount/1000 for up to\n * 1 million documents and to sqrt(documentCount) for more than 1 million\n * documents.\n * Using a numLists value of 1 is akin to performing brute-force search,\n * which has limited performance\n * @param indexType Index Type for Azure DocumentDB index.\n * @param dimensions Number of dimensions for vector similarity.\n * The maximum number of supported dimensions is 2000.\n * If no number is provided, it will be determined automatically by\n * embedding a short text.\n * @param similarity Similarity metric to use with the IVF index.\n * Possible options are:\n * - AzureDocumentDBSimilarityType.COS (cosine distance)\n * - AzureDocumentDBSimilarityType.L2 (Euclidean distance)\n * - AzureDocumentDBSimilarityType.IP (inner product)\n * @returns A promise that resolves when the index has been created.\n */\n async createIndex(\n dimensions: number | undefined = undefined,\n indexType: \"ivf\" | \"hnsw\" | \"diskann\" = \"ivf\",\n similarity: AzureDocumentDBSimilarityType = AzureDocumentDBSimilarityType.COS\n ): Promise {\n await this.connectPromise;\n\n let vectorLength = dimensions;\n\n if (vectorLength === undefined) {\n const queryEmbedding = await this.embeddings.embedQuery(\"test\");\n vectorLength = queryEmbedding.length;\n }\n\n // eslint-disable-next-line @typescript-eslint/no-explicit-any\n const cosmosSearchOptions: any = {\n kind: \"\",\n similarity,\n dimensions: vectorLength,\n };\n\n if (indexType === \"hnsw\") {\n cosmosSearchOptions.kind = \"vector-hnsw\";\n cosmosSearchOptions.m = this.indexOptions.m ?? 16;\n cosmosSearchOptions.efConstruction =\n this.indexOptions.efConstruction ?? 200;\n } else if (indexType === \"diskann\") {\n cosmosSearchOptions.kind = \"vector-diskann\";\n cosmosSearchOptions.maxDegree = this.indexOptions.maxDegree ?? 40;\n cosmosSearchOptions.lBuild = this.indexOptions.lBuild ?? 50;\n cosmosSearchOptions.lSearch = this.indexOptions.lSearch ?? 40;\n /** Default to IVF index */\n } else {\n cosmosSearchOptions.kind = \"vector-ivf\";\n cosmosSearchOptions.numLists = this.indexOptions.numLists ?? 100;\n }\n\n const createIndexCommands = {\n createIndexes: this.collection.collectionName,\n indexes: [\n {\n name: this.indexName,\n key: { [this.embeddingKey]: \"cosmosSearch\" },\n cosmosSearchOptions,\n },\n ],\n };\n\n await this.database.command(createIndexCommands);\n }\n\n /**\n * Removes specified documents from the AzureDocumentDBVectorStore.\n * If no IDs or filter are specified, all documents will be removed.\n * @param params Parameters for the delete operation.\n * @returns A promise that resolves when the documents have been removed.\n */\n async delete(\n params: AzureDocumentDBDeleteParams | string[] = {}\n ): Promise {\n await this.initialize();\n\n let ids: string | string[] | undefined;\n let filter: AzureDocumentDBDeleteParams[\"filter\"];\n if (Array.isArray(params)) {\n ids = params;\n } else {\n ids = params.ids;\n filter = params.filter;\n }\n const idsArray = Array.isArray(ids) ? ids : [ids];\n const deleteIds = ids && idsArray.length > 0 ? idsArray : undefined;\n let deleteFilter = filter ?? {};\n\n if (deleteIds) {\n const objectIds = deleteIds.map((id) => new ObjectId(id));\n deleteFilter = { _id: { $in: objectIds }, ...deleteFilter };\n }\n\n await this.collection.deleteMany(deleteFilter);\n }\n\n /**\n * Closes any newly instantiated Azure DocumentDB client.\n * If the client was passed in the constructor, it will not be closed.\n * @returns A promise that resolves when any newly instantiated Azure\n * DocumentDB client has been closed.\n */\n async close(): Promise {\n if (this.client) {\n await this.client.close();\n }\n }\n\n /**\n * Method for adding vectors to the AzureDocumentDBVectorStore.\n * @param vectors Vectors to be added.\n * @param documents Corresponding documents to be added.\n * @returns A promise that resolves to the added documents IDs.\n */\n async addVectors(\n vectors: number[][],\n documents: DocumentInterface[]\n ): Promise {\n const docs = vectors.map((embedding, idx) => ({\n [this.textKey]: documents[idx].pageContent,\n [this.embeddingKey]: embedding,\n ...documents[idx].metadata,\n }));\n await this.initialize();\n const result = await this.collection.insertMany(docs);\n return Object.values(result.insertedIds).map((id) => String(id));\n }\n\n /**\n * Method for adding documents to the AzureDocumentDBVectorStore. It first converts\n * the documents to texts and then adds them as vectors.\n * @param documents The documents to add.\n * @returns A promise that resolves to the added documents IDs.\n */\n async addDocuments(documents: DocumentInterface[]): Promise {\n const texts = documents.map(({ pageContent }) => pageContent);\n return this.addVectors(\n await this.embeddings.embedDocuments(texts),\n documents\n );\n }\n\n /**\n * Method that performs a similarity search on the vectors stored in the\n * collection. It returns a list of documents and their corresponding\n * similarity scores.\n * @param queryVector Query vector for the similarity search.\n * @param k=4 Number of nearest neighbors to return.\n * @returns Promise that resolves to a list of documents and their corresponding similarity scores.\n */\n async similaritySearchVectorWithScore(\n queryVector: number[],\n k: number,\n indexType?: \"ivf\" | \"hnsw\" | \"diskann\"\n ): Promise<[Document, number][]> {\n await this.initialize();\n\n const pipeline = [\n {\n $search: {\n cosmosSearch: {\n vector: queryVector,\n path: this.embeddingKey,\n k: k ?? 4,\n ...(indexType === \"diskann\"\n ? { lSearch: this.indexOptions.lSearch ?? 40 }\n : {}),\n },\n returnStoredSource: true,\n },\n },\n {\n $project: {\n similarityScore: { $meta: \"searchScore\" },\n document: \"$$ROOT\",\n },\n },\n ];\n const results = await this.collection\n .aggregate(pipeline)\n .map<[Document, number]>((result) => {\n const { similarityScore: score, document } = result;\n const text = document[this.textKey];\n return [new Document({ pageContent: text, metadata: document }), score];\n });\n\n return results.toArray();\n }\n\n /**\n * Return documents selected using the maximal marginal relevance.\n * Maximal marginal relevance optimizes for similarity to the query AND\n * diversity among selected documents.\n * @param query Text to look up documents similar to.\n * @param options.k Number of documents to return.\n * @param options.fetchK=20 Number of documents to fetch before passing to\n * the MMR algorithm.\n * @param options.lambda=0.5 Number between 0 and 1 that determines the\n * degree of diversity among the results, where 0 corresponds to maximum\n * diversity and 1 to minimum diversity.\n * @returns List of documents selected by maximal marginal relevance.\n */\n async maxMarginalRelevanceSearch(\n query: string,\n options: MaxMarginalRelevanceSearchOptions\n ): Promise;\n\n async maxMarginalRelevanceSearch(\n query: string,\n options: MaxMarginalRelevanceSearchOptions,\n indexType: \"ivf\" | \"hnsw\" | \"diskann\"\n ): Promise;\n\n async maxMarginalRelevanceSearch(\n query: string,\n options: MaxMarginalRelevanceSearchOptions,\n indexType?: \"ivf\" | \"hnsw\" | \"diskann\"\n ): Promise {\n const { k, fetchK = 20, lambda = 0.5 } = options;\n\n const queryEmbedding = await this.embeddings.embedQuery(query);\n const docs = await this.similaritySearchVectorWithScore(\n queryEmbedding,\n fetchK,\n indexType\n );\n const embeddingList = docs.map((doc) => doc[0].metadata[this.embeddingKey]);\n\n // Re-rank the results using MMR\n const mmrIndexes = maximalMarginalRelevance(\n queryEmbedding,\n embeddingList,\n lambda,\n k\n );\n\n const mmrDocs = mmrIndexes.map((index) => docs[index][0]);\n return mmrDocs;\n }\n\n /**\n * Initializes the AzureDocumentDBVectorStore by connecting to the database.\n * @param client The MongoClient to use for connecting to the database.\n * @param databaseName The name of the database to use.\n * @param collectionName The name of the collection to use.\n * @returns A promise that resolves when the AzureDocumentDBVectorStore has been initialized.\n */\n private async init(\n client: MongoClient,\n databaseName: string,\n collectionName: string\n ): Promise {\n this.connectPromise = (async () => {\n await client.connect();\n this.database = client.db(databaseName);\n this.collection = this.database.collection(collectionName);\n })();\n\n // Unless skipCreate is set, create the index\n // This operation is no-op if the index already exists\n if (!this.indexOptions.skipCreate) {\n const indexType = this.indexOptions.indexType || \"ivf\";\n await this.createIndex(\n this.indexOptions.dimensions,\n indexType,\n this.indexOptions.similarity\n );\n }\n }\n\n /**\n * Static method to create an instance of AzureDocumentDBVectorStore from a\n * list of texts. It first converts the texts to vectors and then adds\n * them to the collection.\n * @param texts List of texts to be converted to vectors.\n * @param metadatas Metadata for the texts.\n * @param embeddings Embeddings to be used for conversion.\n * @param dbConfig Database configuration for Azure DocumentDB.\n * @returns Promise that resolves to a new instance of AzureDocumentDBVectorStore.\n */\n static async fromTexts(\n texts: string[],\n metadatas: object[] | object,\n embeddings: EmbeddingsInterface,\n dbConfig: AzureDocumentDBConfig\n ): Promise {\n const docs: Document[] = [];\n for (let i = 0; i < texts.length; i += 1) {\n const metadata = Array.isArray(metadatas) ? metadatas[i] : metadatas;\n const newDoc = new Document({\n pageContent: texts[i],\n metadata,\n });\n docs.push(newDoc);\n }\n return AzureDocumentDBVectorStore.fromDocuments(docs, embeddings, dbConfig);\n }\n\n /**\n * Static method to create an instance of AzureDocumentDBVectorStore from a\n * list of documents. It first converts the documents to vectors and then\n * adds them to the collection.\n * @param docs List of documents to be converted to vectors.\n * @param embeddings Embeddings to be used for conversion.\n * @param dbConfig Database configuration for Azure DocumentDB.\n * @returns Promise that resolves to a new instance of AzureDocumentDBVectorStore.\n */\n static async fromDocuments(\n docs: Document[],\n embeddings: EmbeddingsInterface,\n dbConfig: AzureDocumentDBConfig\n ): Promise {\n const instance = new this(embeddings, dbConfig);\n await instance.addDocuments(docs);\n return instance;\n }\n}\n\n// ============================================================================\n// Deprecated aliases for backwards compatibility\n// ============================================================================\n\n/**\n * @deprecated Use `AzureDocumentDBSimilarityType` instead. This alias will be removed in a future version.\n */\nexport const AzureCosmosDBMongoDBSimilarityType = AzureDocumentDBSimilarityType;\n\n/**\n * @deprecated Use `AzureDocumentDBSimilarityType` instead. This alias will be removed in a future version.\n */\nexport type AzureCosmosDBMongoDBSimilarityType = AzureDocumentDBSimilarityType;\n\n/**\n * @deprecated Use `AzureDocumentDBIndexOptions` instead. This alias will be removed in a future version.\n */\nexport type AzureCosmosDBMongoDBIndexOptions = AzureDocumentDBIndexOptions;\n\n/**\n * @deprecated Use `AzureDocumentDBDeleteParams` instead. This alias will be removed in a future version.\n */\nexport type AzureCosmosDBMongoDBDeleteParams = AzureDocumentDBDeleteParams;\n\n/**\n * @deprecated Use `AzureDocumentDBConfig` instead. This alias will be removed in a future version.\n */\nexport type AzureCosmosDBMongoDBConfig = AzureDocumentDBConfig;\n\n/**\n * @deprecated Use `AzureDocumentDBVectorStore` instead. This alias will be removed in a future version.\n */\nexport const AzureCosmosDBMongoDBVectorStore = AzureDocumentDBVectorStore;\n\n/**\n * @deprecated Use `AzureDocumentDBVectorStore` instead. This alias will be removed in a future version.\n */\nexport type AzureCosmosDBMongoDBVectorStore = AzureDocumentDBVectorStore;\n"],"mappings":";;;;;;;;AAkBA,MAAa,gCAAgC;CAE3C,KAAK;CAEL,IAAI;CAEJ,IAAI;CACL;;;;;;;;;;;;;AA8DD,IAAa,6BAAb,MAAa,mCAAmCA,yCAAY;CAC1D,IAAI,aAAwC;AAC1C,SAAO,EACL,kBAAkB,sCACnB;;CAGH,AAAQ;CAER,AAAQ;CAER,AAAiB;CAEjB,AAAQ;CAER,AAAQ;CAER,AAAS;CAET,AAAS;CAET,AAAS;CAET,AAAiB;;;;;;CAOjB;CAEA,mBAA2B;AACzB,SAAO;;CAGT,YACE,YACA,UACA;AACA,QAAM,YAAY,SAAS;EAE3B,MAAM,mBACJ,SAAS,0EACc,qCAAqC,0DACrC,2CAA2C;AAEpE,MAAI,CAAC,SAAS,UAAU,CAAC,iBACvB,OAAM,IAAI,MACR,sEACD;AAGH,MAAI,CAAC,SAAS,OACZ,MAAK,SAAS,IAAIC,oBAAY,kBAAmB,EAC/C,SAAS,eACV,CAAC;EAGJ,MAAM,SAAS,SAAS,UAAU,KAAK;EACvC,MAAM,eAAe,SAAS,gBAAgB;EAC9C,MAAM,iBAAiB,SAAS,kBAAkB;AAClD,OAAK,YAAY,SAAS,aAAa;AACvC,OAAK,UAAU,SAAS,WAAW;AACnC,OAAK,eAAe,SAAS,gBAAgB;AAC7C,OAAK,eAAe,SAAS,gBAAgB,EAAE;AAG/C,OAAK,mBAAmB;AACtB,OAAI,KAAK,gBAAgB,OACvB,MAAK,cAAc,KAAK,KACtB,QACA,cACA,eACD,CAAC,OAAO,UAAU;AACjB,YAAQ,MACN,2DACA,MACD;KACD;AAGJ,UAAO,KAAK;;;;;;;;CAShB,MAAM,mBAAqC;AACzC,QAAM,KAAK,YAAY;AAEvB,UADgB,MAAM,KAAK,WAAW,aAAa,CAAC,SAAS,EAC9C,MAAM,UAAU,MAAM,SAAS,KAAK,UAAU;;;;;;CAO/D,MAAM,cAA6B;AACjC,QAAM,KAAK,YAAY;AACvB,MAAI,MAAM,KAAK,kBAAkB,CAC/B,OAAM,KAAK,WAAW,UAAU,KAAK,UAAU;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAgDnD,MAAM,YACJ,aAAiC,QACjC,YAAwC,OACxC,aAA4C,8BAA8B,KAC3D;AACf,QAAM,KAAK;EAEX,IAAI,eAAe;AAEnB,MAAI,iBAAiB,OAEnB,iBADuB,MAAM,KAAK,WAAW,WAAW,OAAO,EACjC;EAIhC,MAAM,sBAA2B;GAC/B,MAAM;GACN;GACA,YAAY;GACb;AAED,MAAI,cAAc,QAAQ;AACxB,uBAAoB,OAAO;AAC3B,uBAAoB,IAAI,KAAK,aAAa,KAAK;AAC/C,uBAAoB,iBAClB,KAAK,aAAa,kBAAkB;aAC7B,cAAc,WAAW;AAClC,uBAAoB,OAAO;AAC3B,uBAAoB,YAAY,KAAK,aAAa,aAAa;AAC/D,uBAAoB,SAAS,KAAK,aAAa,UAAU;AACzD,uBAAoB,UAAU,KAAK,aAAa,WAAW;SAEtD;AACL,uBAAoB,OAAO;AAC3B,uBAAoB,WAAW,KAAK,aAAa,YAAY;;EAG/D,MAAM,sBAAsB;GAC1B,eAAe,KAAK,WAAW;GAC/B,SAAS,CACP;IACE,MAAM,KAAK;IACX,KAAK,GAAG,KAAK,eAAe,gBAAgB;IAC5C;IACD,CACF;GACF;AAED,QAAM,KAAK,SAAS,QAAQ,oBAAoB;;;;;;;;CASlD,MAAM,OACJ,SAAiD,EAAE,EACpC;AACf,QAAM,KAAK,YAAY;EAEvB,IAAI;EACJ,IAAI;AACJ,MAAI,MAAM,QAAQ,OAAO,CACvB,OAAM;OACD;AACL,SAAM,OAAO;AACb,YAAS,OAAO;;EAElB,MAAM,WAAW,MAAM,QAAQ,IAAI,GAAG,MAAM,CAAC,IAAI;EACjD,MAAM,YAAY,OAAO,SAAS,SAAS,IAAI,WAAW;EAC1D,IAAI,eAAe,UAAU,EAAE;AAE/B,MAAI,UAEF,gBAAe;GAAE,KAAK,EAAE,KADN,UAAU,KAAK,OAAO,IAAIC,iBAAS,GAAG,CAAC,EACjB;GAAE,GAAG;GAAc;AAG7D,QAAM,KAAK,WAAW,WAAW,aAAa;;;;;;;;CAShD,MAAM,QAAuB;AAC3B,MAAI,KAAK,OACP,OAAM,KAAK,OAAO,OAAO;;;;;;;;CAU7B,MAAM,WACJ,SACA,WACmB;EACnB,MAAM,OAAO,QAAQ,KAAK,WAAW,SAAS;IAC3C,KAAK,UAAU,UAAU,KAAK;IAC9B,KAAK,eAAe;GACrB,GAAG,UAAU,KAAK;GACnB,EAAE;AACH,QAAM,KAAK,YAAY;EACvB,MAAM,SAAS,MAAM,KAAK,WAAW,WAAW,KAAK;AACrD,SAAO,OAAO,OAAO,OAAO,YAAY,CAAC,KAAK,OAAO,OAAO,GAAG,CAAC;;;;;;;;CASlE,MAAM,aAAa,WAAmD;EACpE,MAAM,QAAQ,UAAU,KAAK,EAAE,kBAAkB,YAAY;AAC7D,SAAO,KAAK,WACV,MAAM,KAAK,WAAW,eAAe,MAAM,EAC3C,UACD;;;;;;;;;;CAWH,MAAM,gCACJ,aACA,GACA,WAC+B;AAC/B,QAAM,KAAK,YAAY;EAEvB,MAAM,WAAW,CACf,EACE,SAAS;GACP,cAAc;IACZ,QAAQ;IACR,MAAM,KAAK;IACX,GAAG,KAAK;IACR,GAAI,cAAc,YACd,EAAE,SAAS,KAAK,aAAa,WAAW,IAAI,GAC5C,EAAE;IACP;GACD,oBAAoB;GACrB,EACF,EACD,EACE,UAAU;GACR,iBAAiB,EAAE,OAAO,eAAe;GACzC,UAAU;GACX,EACF,CACF;AASD,UARgB,MAAM,KAAK,WACxB,UAAU,SAAS,CACnB,KAAyB,WAAW;GACnC,MAAM,EAAE,iBAAiB,OAAO,aAAa;GAC7C,MAAM,OAAO,SAAS,KAAK;AAC3B,UAAO,CAAC,IAAIC,mCAAS;IAAE,aAAa;IAAM,UAAU;IAAU,CAAC,EAAE,MAAM;IACvE,EAEW,SAAS;;CA2B1B,MAAM,2BACJ,OACA,SACA,WACqB;EACrB,MAAM,EAAE,GAAG,SAAS,IAAI,SAAS,OAAQ;EAEzC,MAAM,iBAAiB,MAAM,KAAK,WAAW,WAAW,MAAM;EAC9D,MAAM,OAAO,MAAM,KAAK,gCACtB,gBACA,QACA,UACD;AAYD,kEAPE,gBAJoB,KAAK,KAAK,QAAQ,IAAI,GAAG,SAAS,KAAK,cAAc,EAMzE,QACA,EACD,CAE0B,KAAK,UAAU,KAAK,OAAO,GAAG;;;;;;;;;CAW3D,MAAc,KACZ,QACA,cACA,gBACe;AACf,OAAK,kBAAkB,YAAY;AACjC,SAAM,OAAO,SAAS;AACtB,QAAK,WAAW,OAAO,GAAG,aAAa;AACvC,QAAK,aAAa,KAAK,SAAS,WAAW,eAAe;MACxD;AAIJ,MAAI,CAAC,KAAK,aAAa,YAAY;GACjC,MAAM,YAAY,KAAK,aAAa,aAAa;AACjD,SAAM,KAAK,YACT,KAAK,aAAa,YAClB,WACA,KAAK,aAAa,WACnB;;;;;;;;;;;;;CAcL,aAAa,UACX,OACA,WACA,YACA,UACqC;EACrC,MAAM,OAAmB,EAAE;AAC3B,OAAK,IAAI,IAAI,GAAG,IAAI,MAAM,QAAQ,KAAK,GAAG;GACxC,MAAM,WAAW,MAAM,QAAQ,UAAU,GAAG,UAAU,KAAK;GAC3D,MAAM,SAAS,IAAIA,mCAAS;IAC1B,aAAa,MAAM;IACnB;IACD,CAAC;AACF,QAAK,KAAK,OAAO;;AAEnB,SAAO,2BAA2B,cAAc,MAAM,YAAY,SAAS;;;;;;;;;;;CAY7E,aAAa,cACX,MACA,YACA,UACqC;EACrC,MAAM,WAAW,IAAI,KAAK,YAAY,SAAS;AAC/C,QAAM,SAAS,aAAa,KAAK;AACjC,SAAO;;;;;;AAWX,MAAa,qCAAqC;;;;AAyBlD,MAAa,kCAAkC"}