langchain 1.0.0-alpha.4 → 1.0.0-alpha.6

package/dist/chains/router/multi_prompt.js.map

@@ -1 +1 @@
-{"version":3,"file":"multi_prompt.js","names":["llm: BaseLanguageModelInterface","promptNames: string[]","promptDescriptions: string[]","promptTemplates: string[] | PromptTemplate[]","defaultChain?: BaseChain","options?: Omit<MultiRouteChainInput, \"defaultChain\">","myPrompt: string | PromptTemplate"],"sources":["../../../src/chains/router/multi_prompt.ts"],"sourcesContent":["import type { BaseLanguageModelInterface } from \"@langchain/core/language_models/base\";\nimport { z } from \"zod\";\nimport { interpolateFString, PromptTemplate } from \"@langchain/core/prompts\";\nimport { MultiRouteChain, MultiRouteChainInput } from \"./multi_route.js\";\nimport { STRUCTURED_MULTI_PROMPT_ROUTER_TEMPLATE } from \"./multi_prompt_prompt.js\";\nimport { BaseChain } from \"../../chains/base.js\";\nimport { LLMChain, LLMChainInput } from \"../../chains/llm_chain.js\";\nimport { LLMRouterChain } from \"./llm_router.js\";\nimport { ConversationChain } from \"../../chains/conversation.js\";\nimport { zipEntries } from \"./utils.js\";\nimport { RouterOutputParser } from \"../../output_parsers/router.js\";\n\n/**\n * A class that represents a multi-prompt chain in the LangChain\n * framework. It extends the MultiRouteChain class and provides additional\n * functionality specific to multi-prompt chains.\n * @example\n * ```typescript\n * const multiPromptChain = MultiPromptChain.fromLLMAndPrompts(\n *   new ChatOpenAI({ model: \"gpt-4o-mini\" }),\n *   {\n *     promptNames: [\"physics\", \"math\", \"history\"],\n *     promptDescriptions: [\n *       \"Good for answering questions about physics\",\n *       \"Good for answering math questions\",\n *       \"Good for answering questions about history\",\n *     ],\n *     promptTemplates: [\n *       `You are a very smart physics professor. Here is a question:\\n{input}\\n`,\n *       `You are a very good mathematician. Here is a question:\\n{input}\\n`,\n *       `You are a very smart history professor. Here is a question:\\n{input}\\n`,\n *     ],\n *   }\n * );\n * const result = await multiPromptChain.call({\n *   input: \"What is the speed of light?\",\n * });\n * ```\n */\nexport class MultiPromptChain extends MultiRouteChain {\n  /**\n   * @deprecated Use `fromLLMAndPrompts` instead\n   */\n  static fromPrompts(\n    llm: BaseLanguageModelInterface,\n    promptNames: string[],\n    promptDescriptions: string[],\n    promptTemplates: string[] | PromptTemplate[],\n    defaultChain?: BaseChain,\n    options?: Omit<MultiRouteChainInput, \"defaultChain\">\n  ) {\n    return MultiPromptChain.fromLLMAndPrompts(llm, {\n      promptNames,\n      promptDescriptions,\n      promptTemplates,\n      defaultChain,\n      multiRouteChainOpts: options,\n    });\n  }\n\n  /**\n   * A static method that creates an instance of MultiPromptChain from a\n   * BaseLanguageModel and a set of prompts. It takes in optional parameters\n   * for the default chain and additional options.\n   * @param llm A BaseLanguageModel instance.\n   * @param promptNames An array of prompt names.\n   * @param promptDescriptions An array of prompt descriptions.\n   * @param promptTemplates An array of prompt templates.\n   * @param defaultChain An optional BaseChain instance to be used as the default chain.\n   * @param llmChainOpts Optional parameters for the LLMChainInput, excluding 'llm' and 'prompt'.\n   * @param conversationChainOpts Optional parameters for the LLMChainInput, excluding 'llm' and 'outputKey'.\n   * @param multiRouteChainOpts Optional parameters for the MultiRouteChainInput, excluding 'defaultChain'.\n   * @returns An instance of MultiPromptChain.\n   */\n  static fromLLMAndPrompts(\n    llm: BaseLanguageModelInterface,\n    {\n      promptNames,\n      promptDescriptions,\n      promptTemplates,\n      defaultChain,\n      llmChainOpts,\n      conversationChainOpts,\n      multiRouteChainOpts,\n    }: {\n      promptNames: string[];\n      promptDescriptions: string[];\n      promptTemplates: string[] | PromptTemplate[];\n      defaultChain?: BaseChain;\n      llmChainOpts?: Omit<LLMChainInput, \"llm\" | \"prompt\">;\n      conversationChainOpts?: Omit<LLMChainInput, \"llm\" | \"outputKey\">;\n      multiRouteChainOpts?: Omit<MultiRouteChainInput, \"defaultChain\">;\n    }\n  ): MultiPromptChain {\n    const destinations = zipEntries(promptNames, promptDescriptions).map(\n      ([name, desc]) => `${name}: ${desc}`\n    );\n\n    const structuredOutputParserSchema = z.object({\n      destination: z\n        .string()\n        .optional()\n        .describe('name of the question answering system to use or \"DEFAULT\"'),\n      next_inputs: z\n        .object({\n          input: z\n            .string()\n            .describe(\"a potentially modified version of the original input\"),\n        })\n        .describe(\"input to be fed to the next model\"),\n    });\n\n    const outputParser = new RouterOutputParser(structuredOutputParserSchema);\n\n    const destinationsStr = destinations.join(\"\\n\");\n    const routerTemplate = interpolateFString(\n      STRUCTURED_MULTI_PROMPT_ROUTER_TEMPLATE(\n        outputParser.getFormatInstructions({ interpolationDepth: 4 })\n      ),\n      {\n        destinations: destinationsStr,\n      }\n    );\n\n    const routerPrompt = new PromptTemplate({\n      template: routerTemplate,\n      inputVariables: [\"input\"],\n      outputParser,\n    });\n\n    const routerChain = LLMRouterChain.fromLLM(llm, routerPrompt);\n    const destinationChains = zipEntries<[string, string | PromptTemplate]>(\n      promptNames,\n      promptTemplates\n    ).reduce((acc, [name, template]) => {\n      let myPrompt: string | PromptTemplate;\n      if (typeof template === \"object\") {\n        myPrompt = template;\n      } else if (typeof template === \"string\") {\n        myPrompt = new PromptTemplate({\n          template: template as string,\n          inputVariables: [\"input\"],\n        });\n      } else {\n        throw new Error(\"Invalid prompt template\");\n      }\n      acc[name as string] = new LLMChain({\n        ...llmChainOpts,\n        llm,\n        prompt: myPrompt,\n      });\n      return acc;\n    }, {} as { [name: string]: LLMChain });\n\n    const convChain = new ConversationChain({\n      ...conversationChainOpts,\n      llm,\n      outputKey: \"text\",\n    });\n\n    return new MultiPromptChain({\n      ...multiRouteChainOpts,\n      routerChain,\n      destinationChains,\n      defaultChain: defaultChain ?? convChain,\n    });\n  }\n\n  _chainType(): string {\n    return \"multi_prompt_chain\";\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAuCA,IAAa,mBAAb,MAAa,yBAAyB,gBAAgB;;;;CAIpD,OAAO,YACLA,KACAC,aACAC,oBACAC,iBACAC,cACAC,SACA;AACA,SAAO,iBAAiB,kBAAkB,KAAK;GAC7C;GACA;GACA;GACA;GACA,qBAAqB;EACtB,EAAC;CACH;;;;;;;;;;;;;;;CAgBD,OAAO,kBACLL,KACA,EACE,aACA,oBACA,iBACA,cACA,cACA,uBACA,qBASD,EACiB;EAClB,MAAM,eAAe,WAAW,aAAa,mBAAmB,CAAC,IAC/D,CAAC,CAAC,MAAM,KAAK,KAAK,GAAG,KAAK,EAAE,EAAE,MAAM,CACrC;EAED,MAAM,+BAA+B,EAAE,OAAO;GAC5C,aAAa,EACV,QAAQ,CACR,UAAU,CACV,SAAS,8DAA4D;GACxE,aAAa,EACV,OAAO,EACN,OAAO,EACJ,QAAQ,CACR,SAAS,uDAAuD,CACpE,EAAC,CACD,SAAS,oCAAoC;EACjD,EAAC;EAEF,MAAM,eAAe,IAAI,mBAAmB;EAE5C,MAAM,kBAAkB,aAAa,KAAK,KAAK;EAC/C,MAAM,iBAAiB,mBACrB,wCACE,aAAa,sBAAsB,EAAE,oBAAoB,EAAG,EAAC,CAC9D,EACD,EACE,cAAc,gBACf,EACF;EAED,MAAM,eAAe,IAAI,eAAe;GACtC,UAAU;GACV,gBAAgB,CAAC,OAAQ;GACzB;EACD;EAED,MAAM,cAAc,eAAe,QAAQ,KAAK,aAAa;EAC7D,MAAM,oBAAoB,WACxB,aACA,gBACD,CAAC,OAAO,CAAC,KAAK,CAAC,MAAM,SAAS,KAAK;GAClC,IAAIM;AACJ,OAAI,OAAO,aAAa,UACtB,WAAW;YACF,OAAO,aAAa,UAC7B,WAAW,IAAI,eAAe;IAClB;IACV,gBAAgB,CAAC,OAAQ;GAC1B;OAED,OAAM,IAAI,MAAM;GAElB,IAAI,QAAkB,IAAI,SAAS;IACjC,GAAG;IACH;IACA,QAAQ;GACT;AACD,UAAO;EACR,GAAE,CAAE,EAAiC;EAEtC,MAAM,YAAY,IAAI,kBAAkB;GACtC,GAAG;GACH;GACA,WAAW;EACZ;AAED,SAAO,IAAI,iBAAiB;GAC1B,GAAG;GACH;GACA;GACA,cAAc,gBAAgB;EAC/B;CACF;CAED,aAAqB;AACnB,SAAO;CACR;AACF"}
+{"version":3,"file":"multi_prompt.js","names":["llm: BaseLanguageModelInterface","promptNames: string[]","promptDescriptions: string[]","promptTemplates: string[] | PromptTemplate[]","defaultChain?: BaseChain","options?: Omit<MultiRouteChainInput, \"defaultChain\">","myPrompt: string | PromptTemplate"],"sources":["../../../src/chains/router/multi_prompt.ts"],"sourcesContent":["import type { BaseLanguageModelInterface } from \"@langchain/core/language_models/base\";\nimport { z } from \"zod/v3\";\nimport { interpolateFString, PromptTemplate } from \"@langchain/core/prompts\";\nimport { MultiRouteChain, MultiRouteChainInput } from \"./multi_route.js\";\nimport { STRUCTURED_MULTI_PROMPT_ROUTER_TEMPLATE } from \"./multi_prompt_prompt.js\";\nimport { BaseChain } from \"../../chains/base.js\";\nimport { LLMChain, LLMChainInput } from \"../../chains/llm_chain.js\";\nimport { LLMRouterChain } from \"./llm_router.js\";\nimport { ConversationChain } from \"../../chains/conversation.js\";\nimport { zipEntries } from \"./utils.js\";\nimport { RouterOutputParser } from \"../../output_parsers/router.js\";\n\n/**\n * A class that represents a multi-prompt chain in the LangChain\n * framework. It extends the MultiRouteChain class and provides additional\n * functionality specific to multi-prompt chains.\n * @example\n * ```typescript\n * const multiPromptChain = MultiPromptChain.fromLLMAndPrompts(\n *   new ChatOpenAI({ model: \"gpt-4o-mini\" }),\n *   {\n *     promptNames: [\"physics\", \"math\", \"history\"],\n *     promptDescriptions: [\n *       \"Good for answering questions about physics\",\n *       \"Good for answering math questions\",\n *       \"Good for answering questions about history\",\n *     ],\n *     promptTemplates: [\n *       `You are a very smart physics professor. Here is a question:\\n{input}\\n`,\n *       `You are a very good mathematician. Here is a question:\\n{input}\\n`,\n *       `You are a very smart history professor. Here is a question:\\n{input}\\n`,\n *     ],\n *   }\n * );\n * const result = await multiPromptChain.call({\n *   input: \"What is the speed of light?\",\n * });\n * ```\n */\nexport class MultiPromptChain extends MultiRouteChain {\n  /**\n   * @deprecated Use `fromLLMAndPrompts` instead\n   */\n  static fromPrompts(\n    llm: BaseLanguageModelInterface,\n    promptNames: string[],\n    promptDescriptions: string[],\n    promptTemplates: string[] | PromptTemplate[],\n    defaultChain?: BaseChain,\n    options?: Omit<MultiRouteChainInput, \"defaultChain\">\n  ) {\n    return MultiPromptChain.fromLLMAndPrompts(llm, {\n      promptNames,\n      promptDescriptions,\n      promptTemplates,\n      defaultChain,\n      multiRouteChainOpts: options,\n    });\n  }\n\n  /**\n   * A static method that creates an instance of MultiPromptChain from a\n   * BaseLanguageModel and a set of prompts. It takes in optional parameters\n   * for the default chain and additional options.\n   * @param llm A BaseLanguageModel instance.\n   * @param promptNames An array of prompt names.\n   * @param promptDescriptions An array of prompt descriptions.\n   * @param promptTemplates An array of prompt templates.\n   * @param defaultChain An optional BaseChain instance to be used as the default chain.\n   * @param llmChainOpts Optional parameters for the LLMChainInput, excluding 'llm' and 'prompt'.\n   * @param conversationChainOpts Optional parameters for the LLMChainInput, excluding 'llm' and 'outputKey'.\n   * @param multiRouteChainOpts Optional parameters for the MultiRouteChainInput, excluding 'defaultChain'.\n   * @returns An instance of MultiPromptChain.\n   */\n  static fromLLMAndPrompts(\n    llm: BaseLanguageModelInterface,\n    {\n      promptNames,\n      promptDescriptions,\n      promptTemplates,\n      defaultChain,\n      llmChainOpts,\n      conversationChainOpts,\n      multiRouteChainOpts,\n    }: {\n      promptNames: string[];\n      promptDescriptions: string[];\n      promptTemplates: string[] | PromptTemplate[];\n      defaultChain?: BaseChain;\n      llmChainOpts?: Omit<LLMChainInput, \"llm\" | \"prompt\">;\n      conversationChainOpts?: Omit<LLMChainInput, \"llm\" | \"outputKey\">;\n      multiRouteChainOpts?: Omit<MultiRouteChainInput, \"defaultChain\">;\n    }\n  ): MultiPromptChain {\n    const destinations = zipEntries(promptNames, promptDescriptions).map(\n      ([name, desc]) => `${name}: ${desc}`\n    );\n\n    const structuredOutputParserSchema = z.object({\n      destination: z\n        .string()\n        .optional()\n        .describe('name of the question answering system to use or \"DEFAULT\"'),\n      next_inputs: z\n        .object({\n          input: z\n            .string()\n            .describe(\"a potentially modified version of the original input\"),\n        })\n        .describe(\"input to be fed to the next model\"),\n    });\n\n    const outputParser = new RouterOutputParser(structuredOutputParserSchema);\n\n    const destinationsStr = destinations.join(\"\\n\");\n    const routerTemplate = interpolateFString(\n      STRUCTURED_MULTI_PROMPT_ROUTER_TEMPLATE(\n        outputParser.getFormatInstructions({ interpolationDepth: 4 })\n      ),\n      {\n        destinations: destinationsStr,\n      }\n    );\n\n    const routerPrompt = new PromptTemplate({\n      template: routerTemplate,\n      inputVariables: [\"input\"],\n      outputParser,\n    });\n\n    const routerChain = LLMRouterChain.fromLLM(llm, routerPrompt);\n    const destinationChains = zipEntries<[string, string | PromptTemplate]>(\n      promptNames,\n      promptTemplates\n    ).reduce((acc, [name, template]) => {\n      let myPrompt: string | PromptTemplate;\n      if (typeof template === \"object\") {\n        myPrompt = template;\n      } else if (typeof template === \"string\") {\n        myPrompt = new PromptTemplate({\n          template: template as string,\n          inputVariables: [\"input\"],\n        });\n      } else {\n        throw new Error(\"Invalid prompt template\");\n      }\n      acc[name as string] = new LLMChain({\n        ...llmChainOpts,\n        llm,\n        prompt: myPrompt,\n      });\n      return acc;\n    }, {} as { [name: string]: LLMChain });\n\n    const convChain = new ConversationChain({\n      ...conversationChainOpts,\n      llm,\n      outputKey: \"text\",\n    });\n\n    return new MultiPromptChain({\n      ...multiRouteChainOpts,\n      routerChain,\n      destinationChains,\n      defaultChain: defaultChain ?? convChain,\n    });\n  }\n\n  _chainType(): string {\n    return \"multi_prompt_chain\";\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAuCA,IAAa,mBAAb,MAAa,yBAAyB,gBAAgB;;;;CAIpD,OAAO,YACLA,KACAC,aACAC,oBACAC,iBACAC,cACAC,SACA;AACA,SAAO,iBAAiB,kBAAkB,KAAK;GAC7C;GACA;GACA;GACA;GACA,qBAAqB;EACtB,EAAC;CACH;;;;;;;;;;;;;;;CAgBD,OAAO,kBACLL,KACA,EACE,aACA,oBACA,iBACA,cACA,cACA,uBACA,qBASD,EACiB;EAClB,MAAM,eAAe,WAAW,aAAa,mBAAmB,CAAC,IAC/D,CAAC,CAAC,MAAM,KAAK,KAAK,GAAG,KAAK,EAAE,EAAE,MAAM,CACrC;EAED,MAAM,+BAA+B,EAAE,OAAO;GAC5C,aAAa,EACV,QAAQ,CACR,UAAU,CACV,SAAS,8DAA4D;GACxE,aAAa,EACV,OAAO,EACN,OAAO,EACJ,QAAQ,CACR,SAAS,uDAAuD,CACpE,EAAC,CACD,SAAS,oCAAoC;EACjD,EAAC;EAEF,MAAM,eAAe,IAAI,mBAAmB;EAE5C,MAAM,kBAAkB,aAAa,KAAK,KAAK;EAC/C,MAAM,iBAAiB,mBACrB,wCACE,aAAa,sBAAsB,EAAE,oBAAoB,EAAG,EAAC,CAC9D,EACD,EACE,cAAc,gBACf,EACF;EAED,MAAM,eAAe,IAAI,eAAe;GACtC,UAAU;GACV,gBAAgB,CAAC,OAAQ;GACzB;EACD;EAED,MAAM,cAAc,eAAe,QAAQ,KAAK,aAAa;EAC7D,MAAM,oBAAoB,WACxB,aACA,gBACD,CAAC,OAAO,CAAC,KAAK,CAAC,MAAM,SAAS,KAAK;GAClC,IAAIM;AACJ,OAAI,OAAO,aAAa,UACtB,WAAW;YACF,OAAO,aAAa,UAC7B,WAAW,IAAI,eAAe;IAClB;IACV,gBAAgB,CAAC,OAAQ;GAC1B;OAED,OAAM,IAAI,MAAM;GAElB,IAAI,QAAkB,IAAI,SAAS;IACjC,GAAG;IACH;IACA,QAAQ;GACT;AACD,UAAO;EACR,GAAE,CAAE,EAAiC;EAEtC,MAAM,YAAY,IAAI,kBAAkB;GACtC,GAAG;GACH;GACA,WAAW;EACZ;AAED,SAAO,IAAI,iBAAiB;GAC1B,GAAG;GACH;GACA;GACA,cAAc,gBAAgB;EAC/B;CACF;CAED,aAAqB;AACnB,SAAO;CACR;AACF"}

package/dist/chains/router/multi_retrieval_qa.cjs

@@ -6,8 +6,8 @@ const require_llm_router = require('./llm_router.cjs');
 const require_utils = require('./utils.cjs');
 const require_router = require('../../output_parsers/router.cjs');
 const require_multi_retrieval_prompt = require('./multi_retrieval_prompt.cjs');
+const zod_v3 = require_rolldown_runtime.__toESM(require("zod/v3"));
 const __langchain_core_prompts = require_rolldown_runtime.__toESM(require("@langchain/core/prompts"));
-const zod = require_rolldown_runtime.__toESM(require("zod"));
 
 //#region src/chains/router/multi_retrieval_qa.ts
 /**
@@ -82,9 +82,9 @@ var MultiRetrievalQAChain = class MultiRetrievalQAChain extends require_multi_ro
 		const { defaultRetriever, defaultPrompt, defaultChain } = defaults ?? {};
 		if (defaultPrompt && !defaultRetriever) throw new Error("`default_retriever` must be specified if `default_prompt` is \nprovided. Received only `default_prompt`.");
 		const destinations = require_utils.zipEntries(retrieverNames, retrieverDescriptions).map(([name, desc]) => `${name}: ${desc}`);
-		const structuredOutputParserSchema = zod.z.object({
-			destination: zod.z.string().optional().describe("name of the question answering system to use or \"DEFAULT\""),
-			next_inputs: zod.z.object({ query: zod.z.string().describe("a potentially modified version of the original input") }).describe("input to be fed to the next model")
+		const structuredOutputParserSchema = zod_v3.z.object({
+			destination: zod_v3.z.string().optional().describe("name of the question answering system to use or \"DEFAULT\""),
+			next_inputs: zod_v3.z.object({ query: zod_v3.z.string().describe("a potentially modified version of the original input") }).describe("input to be fed to the next model")
 		});
 		const outputParser = new require_router.RouterOutputParser(structuredOutputParserSchema);
 		const destinationsStr = destinations.join("\n");

package/dist/chains/router/multi_retrieval_qa.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"multi_retrieval_qa.cjs","names":["MultiRouteChain","llm: BaseLanguageModelInterface","retrieverNames: string[]","retrieverDescriptions: string[]","retrievers: BaseRetrieverInterface[]","retrieverPrompts?: PromptTemplate[]","defaults?: MultiRetrievalDefaults","options?: Omit<MultiRouteChainInput, \"defaultChain\">","zipEntries","z","RouterOutputParser","STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE","PromptTemplate","LLMRouterChain","opt: Partial<RetrievalQAChainInput> & {\n          prompt?: PromptTemplate;\n        }","RetrievalQAChain","DEFAULT_TEMPLATE","ConversationChain"],"sources":["../../../src/chains/router/multi_retrieval_qa.ts"],"sourcesContent":["import { z } from \"zod\";\nimport type { BaseLanguageModelInterface } from \"@langchain/core/language_models/base\";\nimport type { BaseRetrieverInterface } from \"@langchain/core/retrievers\";\nimport { interpolateFString, PromptTemplate } from \"@langchain/core/prompts\";\nimport { MultiRouteChain, MultiRouteChainInput } from \"./multi_route.js\";\nimport { BaseChain } from \"../../chains/base.js\";\nimport { LLMRouterChain } from \"./llm_router.js\";\nimport {\n  ConversationChain,\n  DEFAULT_TEMPLATE,\n} from \"../../chains/conversation.js\";\nimport { STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE } from \"./multi_retrieval_prompt.js\";\nimport { zipEntries } from \"./utils.js\";\nimport {\n  RetrievalQAChain,\n  RetrievalQAChainInput,\n} from \"../../chains/retrieval_qa.js\";\nimport { RouterOutputParser } from \"../../output_parsers/router.js\";\n\n/**\n * A type that represents the default values for the MultiRetrievalQAChain\n * class. It includes optional properties for the default retriever,\n * default prompt, and default chain.\n */\nexport type MultiRetrievalDefaults = {\n  defaultRetriever?: BaseRetrieverInterface;\n  defaultPrompt?: PromptTemplate;\n  defaultChain?: BaseChain;\n};\n\n/**\n * A class that represents a multi-retrieval question answering chain in\n * the LangChain framework. It extends the MultiRouteChain class and\n * provides additional functionality specific to multi-retrieval QA\n * chains.\n * @example\n * ```typescript\n * const multiRetrievalQAChain = MultiRetrievalQAChain.fromLLMAndRetrievers(\n *   new ChatOpenAI({ model: \"gpt-4o-mini\" }),\n *   {\n *     retrieverNames: [\"aqua teen\", \"mst3k\", \"animaniacs\"],\n *     retrieverDescriptions: [\n *       \"Good for answering questions about Aqua Teen Hunger Force theme song\",\n *       \"Good for answering questions about Mystery Science Theater 3000 theme song\",\n *       \"Good for answering questions about Animaniacs theme song\",\n *     ],\n *     retrievers: [\n *       new MemoryVectorStore().asRetriever(3),\n *       new MemoryVectorStore().asRetriever(3),\n *       new MemoryVectorStore().asRetriever(3),\n *     ],\n *     retrievalQAChainOpts: {\n *       returnSourceDocuments: true,\n *     },\n *   },\n * );\n *\n * const result = await multiRetrievalQAChain.call({\n *   input:\n *     \"In the Aqua Teen Hunger Force theme song, who calls himself the mike rula?\",\n * });\n *\n * console.log(result.sourceDocuments, result.text);\n * ```\n */\nexport class MultiRetrievalQAChain extends MultiRouteChain {\n  get outputKeys(): string[] {\n    return [\"result\"];\n  }\n\n  /**\n   * @deprecated Use `fromRetrieversAndPrompts` instead\n   */\n  static fromRetrievers(\n    llm: BaseLanguageModelInterface,\n    retrieverNames: string[],\n    retrieverDescriptions: string[],\n    retrievers: BaseRetrieverInterface[],\n    retrieverPrompts?: PromptTemplate[],\n    defaults?: MultiRetrievalDefaults,\n    options?: Omit<MultiRouteChainInput, \"defaultChain\">\n  ) {\n    return MultiRetrievalQAChain.fromLLMAndRetrievers(llm, {\n      retrieverNames,\n      retrieverDescriptions,\n      retrievers,\n      retrieverPrompts,\n      defaults,\n      multiRetrievalChainOpts: options,\n    });\n  }\n\n  /**\n   * A static method that creates an instance of MultiRetrievalQAChain from\n   * a BaseLanguageModel and a set of retrievers. It takes in optional\n   * parameters for the retriever names, descriptions, prompts, defaults,\n   * and additional options. It is an alternative method to fromRetrievers\n   * and provides more flexibility in configuring the underlying chains.\n   * @param llm A BaseLanguageModel instance.\n   * @param retrieverNames An array of retriever names.\n   * @param retrieverDescriptions An array of retriever descriptions.\n   * @param retrievers An array of BaseRetrieverInterface instances.\n   * @param retrieverPrompts An optional array of PromptTemplate instances for the retrievers.\n   * @param defaults An optional MultiRetrievalDefaults instance.\n   * @param multiRetrievalChainOpts Additional optional parameters for the multi-retrieval chain.\n   * @param retrievalQAChainOpts Additional optional parameters for the retrieval QA chain.\n   * @returns A new instance of MultiRetrievalQAChain.\n   */\n  static fromLLMAndRetrievers(\n    llm: BaseLanguageModelInterface,\n    {\n      retrieverNames,\n      retrieverDescriptions,\n      retrievers,\n      retrieverPrompts,\n      defaults,\n      multiRetrievalChainOpts,\n      retrievalQAChainOpts,\n    }: {\n      retrieverNames: string[];\n      retrieverDescriptions: string[];\n      retrievers: BaseRetrieverInterface[];\n      retrieverPrompts?: PromptTemplate[];\n      defaults?: MultiRetrievalDefaults;\n      multiRetrievalChainOpts?: Omit<MultiRouteChainInput, \"defaultChain\">;\n      retrievalQAChainOpts?: Partial<\n        Omit<RetrievalQAChainInput, \"retriever\" | \"combineDocumentsChain\">\n      > & {\n        prompt?: PromptTemplate;\n      };\n    }\n  ): MultiRetrievalQAChain {\n    const { defaultRetriever, defaultPrompt, defaultChain } = defaults ?? {};\n    if (defaultPrompt && !defaultRetriever) {\n      throw new Error(\n        \"`default_retriever` must be specified if `default_prompt` is \\nprovided. Received only `default_prompt`.\"\n      );\n    }\n    const destinations = zipEntries<[string, string]>(\n      retrieverNames,\n      retrieverDescriptions\n    ).map(([name, desc]) => `${name}: ${desc}`);\n\n    const structuredOutputParserSchema = z.object({\n      destination: z\n        .string()\n        .optional()\n        .describe('name of the question answering system to use or \"DEFAULT\"'),\n      next_inputs: z\n        .object({\n          query: z\n            .string()\n            .describe(\"a potentially modified version of the original input\"),\n        })\n        .describe(\"input to be fed to the next model\"),\n    });\n\n    const outputParser = new RouterOutputParser<\n      typeof structuredOutputParserSchema\n    >(structuredOutputParserSchema);\n\n    const destinationsStr = destinations.join(\"\\n\");\n    const routerTemplate = interpolateFString(\n      STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE(\n        outputParser.getFormatInstructions({ interpolationDepth: 4 })\n      ),\n      {\n        destinations: destinationsStr,\n      }\n    );\n    const routerPrompt = new PromptTemplate({\n      template: routerTemplate,\n      inputVariables: [\"input\"],\n      outputParser,\n    });\n\n    const routerChain = LLMRouterChain.fromLLM(llm, routerPrompt);\n    const prompts = retrieverPrompts ?? retrievers.map(() => null);\n    const destinationChains = zipEntries<\n      [string, BaseRetrieverInterface, PromptTemplate | null]\n    >(retrieverNames, retrievers, prompts).reduce(\n      (acc, [name, retriever, prompt]) => {\n        const opt: Partial<RetrievalQAChainInput> & {\n          prompt?: PromptTemplate;\n        } = retrievalQAChainOpts ?? {};\n        if (prompt) {\n          opt.prompt = prompt;\n        }\n        acc[name] = RetrievalQAChain.fromLLM(llm, retriever, opt);\n        return acc;\n      },\n      {} as { [name: string]: RetrievalQAChain }\n    );\n\n    let _defaultChain;\n    if (defaultChain) {\n      _defaultChain = defaultChain;\n    } else if (defaultRetriever) {\n      _defaultChain = RetrievalQAChain.fromLLM(llm, defaultRetriever, {\n        ...retrievalQAChainOpts,\n        prompt: defaultPrompt,\n      });\n    } else {\n      const promptTemplate = DEFAULT_TEMPLATE.replace(\"input\", \"query\");\n      const prompt = new PromptTemplate({\n        template: promptTemplate,\n        inputVariables: [\"history\", \"query\"],\n      });\n      _defaultChain = new ConversationChain({\n        llm,\n        prompt,\n        outputKey: \"result\",\n      });\n    }\n\n    return new MultiRetrievalQAChain({\n      ...multiRetrievalChainOpts,\n      routerChain,\n      destinationChains,\n      defaultChain: _defaultChain,\n    });\n  }\n\n  _chainType(): string {\n    return \"multi_retrieval_qa_chain\";\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAiEA,IAAa,wBAAb,MAAa,8BAA8BA,oCAAgB;CACzD,IAAI,aAAuB;AACzB,SAAO,CAAC,QAAS;CAClB;;;;CAKD,OAAO,eACLC,KACAC,gBACAC,uBACAC,YACAC,kBACAC,UACAC,SACA;AACA,SAAO,sBAAsB,qBAAqB,KAAK;GACrD;GACA;GACA;GACA;GACA;GACA,yBAAyB;EAC1B,EAAC;CACH;;;;;;;;;;;;;;;;;CAkBD,OAAO,qBACLN,KACA,EACE,gBACA,uBACA,YACA,kBACA,UACA,yBACA,sBAaD,EACsB;EACvB,MAAM,EAAE,kBAAkB,eAAe,cAAc,GAAG,YAAY,CAAE;AACxE,MAAI,iBAAiB,CAAC,iBACpB,OAAM,IAAI,MACR;EAGJ,MAAM,eAAeO,yBACnB,gBACA,sBACD,CAAC,IAAI,CAAC,CAAC,MAAM,KAAK,KAAK,GAAG,KAAK,EAAE,EAAE,MAAM,CAAC;EAE3C,MAAM,+BAA+BC,MAAE,OAAO;GAC5C,aAAaA,MACV,QAAQ,CACR,UAAU,CACV,SAAS,8DAA4D;GACxE,aAAaA,MACV,OAAO,EACN,OAAOA,MACJ,QAAQ,CACR,SAAS,uDAAuD,CACpE,EAAC,CACD,SAAS,oCAAoC;EACjD,EAAC;EAEF,MAAM,eAAe,IAAIC,kCAEvB;EAEF,MAAM,kBAAkB,aAAa,KAAK,KAAK;EAC/C,MAAM,kEACJC,0EACE,aAAa,sBAAsB,EAAE,oBAAoB,EAAG,EAAC,CAC9D,EACD,EACE,cAAc,gBACf,EACF;EACD,MAAM,eAAe,IAAIC,wCAAe;GACtC,UAAU;GACV,gBAAgB,CAAC,OAAQ;GACzB;EACD;EAED,MAAM,cAAcC,kCAAe,QAAQ,KAAK,aAAa;EAC7D,MAAM,UAAU,oBAAoB,WAAW,IAAI,MAAM,KAAK;EAC9D,MAAM,oBAAoBL,yBAExB,gBAAgB,YAAY,QAAQ,CAAC,OACrC,CAAC,KAAK,CAAC,MAAM,WAAW,OAAO,KAAK;GAClC,MAAMM,MAEF,wBAAwB,CAAE;AAC9B,OAAI,QACF,IAAI,SAAS;GAEf,IAAI,QAAQC,sCAAiB,QAAQ,KAAK,WAAW,IAAI;AACzD,UAAO;EACR,GACD,CAAE,EACH;EAED,IAAI;AACJ,MAAI,cACF,gBAAgB;WACP,kBACT,gBAAgBA,sCAAiB,QAAQ,KAAK,kBAAkB;GAC9D,GAAG;GACH,QAAQ;EACT,EAAC;OACG;GACL,MAAM,iBAAiBC,sCAAiB,QAAQ,SAAS,QAAQ;GACjE,MAAM,SAAS,IAAIJ,wCAAe;IAChC,UAAU;IACV,gBAAgB,CAAC,WAAW,OAAQ;GACrC;GACD,gBAAgB,IAAIK,uCAAkB;IACpC;IACA;IACA,WAAW;GACZ;EACF;AAED,SAAO,IAAI,sBAAsB;GAC/B,GAAG;GACH;GACA;GACA,cAAc;EACf;CACF;CAED,aAAqB;AACnB,SAAO;CACR;AACF"}
+{"version":3,"file":"multi_retrieval_qa.cjs","names":["MultiRouteChain","llm: BaseLanguageModelInterface","retrieverNames: string[]","retrieverDescriptions: string[]","retrievers: BaseRetrieverInterface[]","retrieverPrompts?: PromptTemplate[]","defaults?: MultiRetrievalDefaults","options?: Omit<MultiRouteChainInput, \"defaultChain\">","zipEntries","z","RouterOutputParser","STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE","PromptTemplate","LLMRouterChain","opt: Partial<RetrievalQAChainInput> & {\n          prompt?: PromptTemplate;\n        }","RetrievalQAChain","DEFAULT_TEMPLATE","ConversationChain"],"sources":["../../../src/chains/router/multi_retrieval_qa.ts"],"sourcesContent":["import { z } from \"zod/v3\";\nimport type { BaseLanguageModelInterface } from \"@langchain/core/language_models/base\";\nimport type { BaseRetrieverInterface } from \"@langchain/core/retrievers\";\nimport { interpolateFString, PromptTemplate } from \"@langchain/core/prompts\";\nimport { MultiRouteChain, MultiRouteChainInput } from \"./multi_route.js\";\nimport { BaseChain } from \"../../chains/base.js\";\nimport { LLMRouterChain } from \"./llm_router.js\";\nimport {\n  ConversationChain,\n  DEFAULT_TEMPLATE,\n} from \"../../chains/conversation.js\";\nimport { STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE } from \"./multi_retrieval_prompt.js\";\nimport { zipEntries } from \"./utils.js\";\nimport {\n  RetrievalQAChain,\n  RetrievalQAChainInput,\n} from \"../../chains/retrieval_qa.js\";\nimport { RouterOutputParser } from \"../../output_parsers/router.js\";\n\n/**\n * A type that represents the default values for the MultiRetrievalQAChain\n * class. It includes optional properties for the default retriever,\n * default prompt, and default chain.\n */\nexport type MultiRetrievalDefaults = {\n  defaultRetriever?: BaseRetrieverInterface;\n  defaultPrompt?: PromptTemplate;\n  defaultChain?: BaseChain;\n};\n\n/**\n * A class that represents a multi-retrieval question answering chain in\n * the LangChain framework. It extends the MultiRouteChain class and\n * provides additional functionality specific to multi-retrieval QA\n * chains.\n * @example\n * ```typescript\n * const multiRetrievalQAChain = MultiRetrievalQAChain.fromLLMAndRetrievers(\n *   new ChatOpenAI({ model: \"gpt-4o-mini\" }),\n *   {\n *     retrieverNames: [\"aqua teen\", \"mst3k\", \"animaniacs\"],\n *     retrieverDescriptions: [\n *       \"Good for answering questions about Aqua Teen Hunger Force theme song\",\n *       \"Good for answering questions about Mystery Science Theater 3000 theme song\",\n *       \"Good for answering questions about Animaniacs theme song\",\n *     ],\n *     retrievers: [\n *       new MemoryVectorStore().asRetriever(3),\n *       new MemoryVectorStore().asRetriever(3),\n *       new MemoryVectorStore().asRetriever(3),\n *     ],\n *     retrievalQAChainOpts: {\n *       returnSourceDocuments: true,\n *     },\n *   },\n * );\n *\n * const result = await multiRetrievalQAChain.call({\n *   input:\n *     \"In the Aqua Teen Hunger Force theme song, who calls himself the mike rula?\",\n * });\n *\n * console.log(result.sourceDocuments, result.text);\n * ```\n */\nexport class MultiRetrievalQAChain extends MultiRouteChain {\n  get outputKeys(): string[] {\n    return [\"result\"];\n  }\n\n  /**\n   * @deprecated Use `fromRetrieversAndPrompts` instead\n   */\n  static fromRetrievers(\n    llm: BaseLanguageModelInterface,\n    retrieverNames: string[],\n    retrieverDescriptions: string[],\n    retrievers: BaseRetrieverInterface[],\n    retrieverPrompts?: PromptTemplate[],\n    defaults?: MultiRetrievalDefaults,\n    options?: Omit<MultiRouteChainInput, \"defaultChain\">\n  ) {\n    return MultiRetrievalQAChain.fromLLMAndRetrievers(llm, {\n      retrieverNames,\n      retrieverDescriptions,\n      retrievers,\n      retrieverPrompts,\n      defaults,\n      multiRetrievalChainOpts: options,\n    });\n  }\n\n  /**\n   * A static method that creates an instance of MultiRetrievalQAChain from\n   * a BaseLanguageModel and a set of retrievers. It takes in optional\n   * parameters for the retriever names, descriptions, prompts, defaults,\n   * and additional options. It is an alternative method to fromRetrievers\n   * and provides more flexibility in configuring the underlying chains.\n   * @param llm A BaseLanguageModel instance.\n   * @param retrieverNames An array of retriever names.\n   * @param retrieverDescriptions An array of retriever descriptions.\n   * @param retrievers An array of BaseRetrieverInterface instances.\n   * @param retrieverPrompts An optional array of PromptTemplate instances for the retrievers.\n   * @param defaults An optional MultiRetrievalDefaults instance.\n   * @param multiRetrievalChainOpts Additional optional parameters for the multi-retrieval chain.\n   * @param retrievalQAChainOpts Additional optional parameters for the retrieval QA chain.\n   * @returns A new instance of MultiRetrievalQAChain.\n   */\n  static fromLLMAndRetrievers(\n    llm: BaseLanguageModelInterface,\n    {\n      retrieverNames,\n      retrieverDescriptions,\n      retrievers,\n      retrieverPrompts,\n      defaults,\n      multiRetrievalChainOpts,\n      retrievalQAChainOpts,\n    }: {\n      retrieverNames: string[];\n      retrieverDescriptions: string[];\n      retrievers: BaseRetrieverInterface[];\n      retrieverPrompts?: PromptTemplate[];\n      defaults?: MultiRetrievalDefaults;\n      multiRetrievalChainOpts?: Omit<MultiRouteChainInput, \"defaultChain\">;\n      retrievalQAChainOpts?: Partial<\n        Omit<RetrievalQAChainInput, \"retriever\" | \"combineDocumentsChain\">\n      > & {\n        prompt?: PromptTemplate;\n      };\n    }\n  ): MultiRetrievalQAChain {\n    const { defaultRetriever, defaultPrompt, defaultChain } = defaults ?? {};\n    if (defaultPrompt && !defaultRetriever) {\n      throw new Error(\n        \"`default_retriever` must be specified if `default_prompt` is \\nprovided. Received only `default_prompt`.\"\n      );\n    }\n    const destinations = zipEntries<[string, string]>(\n      retrieverNames,\n      retrieverDescriptions\n    ).map(([name, desc]) => `${name}: ${desc}`);\n\n    const structuredOutputParserSchema = z.object({\n      destination: z\n        .string()\n        .optional()\n        .describe('name of the question answering system to use or \"DEFAULT\"'),\n      next_inputs: z\n        .object({\n          query: z\n            .string()\n            .describe(\"a potentially modified version of the original input\"),\n        })\n        .describe(\"input to be fed to the next model\"),\n    });\n\n    const outputParser = new RouterOutputParser<\n      typeof structuredOutputParserSchema\n    >(structuredOutputParserSchema);\n\n    const destinationsStr = destinations.join(\"\\n\");\n    const routerTemplate = interpolateFString(\n      STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE(\n        outputParser.getFormatInstructions({ interpolationDepth: 4 })\n      ),\n      {\n        destinations: destinationsStr,\n      }\n    );\n    const routerPrompt = new PromptTemplate({\n      template: routerTemplate,\n      inputVariables: [\"input\"],\n      outputParser,\n    });\n\n    const routerChain = LLMRouterChain.fromLLM(llm, routerPrompt);\n    const prompts = retrieverPrompts ?? retrievers.map(() => null);\n    const destinationChains = zipEntries<\n      [string, BaseRetrieverInterface, PromptTemplate | null]\n    >(retrieverNames, retrievers, prompts).reduce(\n      (acc, [name, retriever, prompt]) => {\n        const opt: Partial<RetrievalQAChainInput> & {\n          prompt?: PromptTemplate;\n        } = retrievalQAChainOpts ?? {};\n        if (prompt) {\n          opt.prompt = prompt;\n        }\n        acc[name] = RetrievalQAChain.fromLLM(llm, retriever, opt);\n        return acc;\n      },\n      {} as { [name: string]: RetrievalQAChain }\n    );\n\n    let _defaultChain;\n    if (defaultChain) {\n      _defaultChain = defaultChain;\n    } else if (defaultRetriever) {\n      _defaultChain = RetrievalQAChain.fromLLM(llm, defaultRetriever, {\n        ...retrievalQAChainOpts,\n        prompt: defaultPrompt,\n      });\n    } else {\n      const promptTemplate = DEFAULT_TEMPLATE.replace(\"input\", \"query\");\n      const prompt = new PromptTemplate({\n        template: promptTemplate,\n        inputVariables: [\"history\", \"query\"],\n      });\n      _defaultChain = new ConversationChain({\n        llm,\n        prompt,\n        outputKey: \"result\",\n      });\n    }\n\n    return new MultiRetrievalQAChain({\n      ...multiRetrievalChainOpts,\n      routerChain,\n      destinationChains,\n      defaultChain: _defaultChain,\n    });\n  }\n\n  _chainType(): string {\n    return \"multi_retrieval_qa_chain\";\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAiEA,IAAa,wBAAb,MAAa,8BAA8BA,oCAAgB;CACzD,IAAI,aAAuB;AACzB,SAAO,CAAC,QAAS;CAClB;;;;CAKD,OAAO,eACLC,KACAC,gBACAC,uBACAC,YACAC,kBACAC,UACAC,SACA;AACA,SAAO,sBAAsB,qBAAqB,KAAK;GACrD;GACA;GACA;GACA;GACA;GACA,yBAAyB;EAC1B,EAAC;CACH;;;;;;;;;;;;;;;;;CAkBD,OAAO,qBACLN,KACA,EACE,gBACA,uBACA,YACA,kBACA,UACA,yBACA,sBAaD,EACsB;EACvB,MAAM,EAAE,kBAAkB,eAAe,cAAc,GAAG,YAAY,CAAE;AACxE,MAAI,iBAAiB,CAAC,iBACpB,OAAM,IAAI,MACR;EAGJ,MAAM,eAAeO,yBACnB,gBACA,sBACD,CAAC,IAAI,CAAC,CAAC,MAAM,KAAK,KAAK,GAAG,KAAK,EAAE,EAAE,MAAM,CAAC;EAE3C,MAAM,+BAA+BC,SAAE,OAAO;GAC5C,aAAaA,SACV,QAAQ,CACR,UAAU,CACV,SAAS,8DAA4D;GACxE,aAAaA,SACV,OAAO,EACN,OAAOA,SACJ,QAAQ,CACR,SAAS,uDAAuD,CACpE,EAAC,CACD,SAAS,oCAAoC;EACjD,EAAC;EAEF,MAAM,eAAe,IAAIC,kCAEvB;EAEF,MAAM,kBAAkB,aAAa,KAAK,KAAK;EAC/C,MAAM,kEACJC,0EACE,aAAa,sBAAsB,EAAE,oBAAoB,EAAG,EAAC,CAC9D,EACD,EACE,cAAc,gBACf,EACF;EACD,MAAM,eAAe,IAAIC,wCAAe;GACtC,UAAU;GACV,gBAAgB,CAAC,OAAQ;GACzB;EACD;EAED,MAAM,cAAcC,kCAAe,QAAQ,KAAK,aAAa;EAC7D,MAAM,UAAU,oBAAoB,WAAW,IAAI,MAAM,KAAK;EAC9D,MAAM,oBAAoBL,yBAExB,gBAAgB,YAAY,QAAQ,CAAC,OACrC,CAAC,KAAK,CAAC,MAAM,WAAW,OAAO,KAAK;GAClC,MAAMM,MAEF,wBAAwB,CAAE;AAC9B,OAAI,QACF,IAAI,SAAS;GAEf,IAAI,QAAQC,sCAAiB,QAAQ,KAAK,WAAW,IAAI;AACzD,UAAO;EACR,GACD,CAAE,EACH;EAED,IAAI;AACJ,MAAI,cACF,gBAAgB;WACP,kBACT,gBAAgBA,sCAAiB,QAAQ,KAAK,kBAAkB;GAC9D,GAAG;GACH,QAAQ;EACT,EAAC;OACG;GACL,MAAM,iBAAiBC,sCAAiB,QAAQ,SAAS,QAAQ;GACjE,MAAM,SAAS,IAAIJ,wCAAe;IAChC,UAAU;IACV,gBAAgB,CAAC,WAAW,OAAQ;GACrC;GACD,gBAAgB,IAAIK,uCAAkB;IACpC;IACA;IACA,WAAW;GACZ;EACF;AAED,SAAO,IAAI,sBAAsB;GAC/B,GAAG;GACH;GACA;GACA,cAAc;EACf;CACF;CAED,aAAqB;AACnB,SAAO;CACR;AACF"}

package/dist/chains/router/multi_retrieval_qa.js

@@ -5,8 +5,8 @@ import { LLMRouterChain } from "./llm_router.js";
 import { zipEntries } from "./utils.js";
 import { RouterOutputParser } from "../../output_parsers/router.js";
 import { STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE } from "./multi_retrieval_prompt.js";
+import { z } from "zod/v3";
 import { PromptTemplate, interpolateFString } from "@langchain/core/prompts";
-import { z } from "zod";
 
 //#region src/chains/router/multi_retrieval_qa.ts
 /**

package/dist/chains/router/multi_retrieval_qa.js.map

@@ -1 +1 @@
-{"version":3,"file":"multi_retrieval_qa.js","names":["llm: BaseLanguageModelInterface","retrieverNames: string[]","retrieverDescriptions: string[]","retrievers: BaseRetrieverInterface[]","retrieverPrompts?: PromptTemplate[]","defaults?: MultiRetrievalDefaults","options?: Omit<MultiRouteChainInput, \"defaultChain\">","opt: Partial<RetrievalQAChainInput> & {\n          prompt?: PromptTemplate;\n        }"],"sources":["../../../src/chains/router/multi_retrieval_qa.ts"],"sourcesContent":["import { z } from \"zod\";\nimport type { BaseLanguageModelInterface } from \"@langchain/core/language_models/base\";\nimport type { BaseRetrieverInterface } from \"@langchain/core/retrievers\";\nimport { interpolateFString, PromptTemplate } from \"@langchain/core/prompts\";\nimport { MultiRouteChain, MultiRouteChainInput } from \"./multi_route.js\";\nimport { BaseChain } from \"../../chains/base.js\";\nimport { LLMRouterChain } from \"./llm_router.js\";\nimport {\n  ConversationChain,\n  DEFAULT_TEMPLATE,\n} from \"../../chains/conversation.js\";\nimport { STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE } from \"./multi_retrieval_prompt.js\";\nimport { zipEntries } from \"./utils.js\";\nimport {\n  RetrievalQAChain,\n  RetrievalQAChainInput,\n} from \"../../chains/retrieval_qa.js\";\nimport { RouterOutputParser } from \"../../output_parsers/router.js\";\n\n/**\n * A type that represents the default values for the MultiRetrievalQAChain\n * class. It includes optional properties for the default retriever,\n * default prompt, and default chain.\n */\nexport type MultiRetrievalDefaults = {\n  defaultRetriever?: BaseRetrieverInterface;\n  defaultPrompt?: PromptTemplate;\n  defaultChain?: BaseChain;\n};\n\n/**\n * A class that represents a multi-retrieval question answering chain in\n * the LangChain framework. It extends the MultiRouteChain class and\n * provides additional functionality specific to multi-retrieval QA\n * chains.\n * @example\n * ```typescript\n * const multiRetrievalQAChain = MultiRetrievalQAChain.fromLLMAndRetrievers(\n *   new ChatOpenAI({ model: \"gpt-4o-mini\" }),\n *   {\n *     retrieverNames: [\"aqua teen\", \"mst3k\", \"animaniacs\"],\n *     retrieverDescriptions: [\n *       \"Good for answering questions about Aqua Teen Hunger Force theme song\",\n *       \"Good for answering questions about Mystery Science Theater 3000 theme song\",\n *       \"Good for answering questions about Animaniacs theme song\",\n *     ],\n *     retrievers: [\n *       new MemoryVectorStore().asRetriever(3),\n *       new MemoryVectorStore().asRetriever(3),\n *       new MemoryVectorStore().asRetriever(3),\n *     ],\n *     retrievalQAChainOpts: {\n *       returnSourceDocuments: true,\n *     },\n *   },\n * );\n *\n * const result = await multiRetrievalQAChain.call({\n *   input:\n *     \"In the Aqua Teen Hunger Force theme song, who calls himself the mike rula?\",\n * });\n *\n * console.log(result.sourceDocuments, result.text);\n * ```\n */\nexport class MultiRetrievalQAChain extends MultiRouteChain {\n  get outputKeys(): string[] {\n    return [\"result\"];\n  }\n\n  /**\n   * @deprecated Use `fromRetrieversAndPrompts` instead\n   */\n  static fromRetrievers(\n    llm: BaseLanguageModelInterface,\n    retrieverNames: string[],\n    retrieverDescriptions: string[],\n    retrievers: BaseRetrieverInterface[],\n    retrieverPrompts?: PromptTemplate[],\n    defaults?: MultiRetrievalDefaults,\n    options?: Omit<MultiRouteChainInput, \"defaultChain\">\n  ) {\n    return MultiRetrievalQAChain.fromLLMAndRetrievers(llm, {\n      retrieverNames,\n      retrieverDescriptions,\n      retrievers,\n      retrieverPrompts,\n      defaults,\n      multiRetrievalChainOpts: options,\n    });\n  }\n\n  /**\n   * A static method that creates an instance of MultiRetrievalQAChain from\n   * a BaseLanguageModel and a set of retrievers. It takes in optional\n   * parameters for the retriever names, descriptions, prompts, defaults,\n   * and additional options. It is an alternative method to fromRetrievers\n   * and provides more flexibility in configuring the underlying chains.\n   * @param llm A BaseLanguageModel instance.\n   * @param retrieverNames An array of retriever names.\n   * @param retrieverDescriptions An array of retriever descriptions.\n   * @param retrievers An array of BaseRetrieverInterface instances.\n   * @param retrieverPrompts An optional array of PromptTemplate instances for the retrievers.\n   * @param defaults An optional MultiRetrievalDefaults instance.\n   * @param multiRetrievalChainOpts Additional optional parameters for the multi-retrieval chain.\n   * @param retrievalQAChainOpts Additional optional parameters for the retrieval QA chain.\n   * @returns A new instance of MultiRetrievalQAChain.\n   */\n  static fromLLMAndRetrievers(\n    llm: BaseLanguageModelInterface,\n    {\n      retrieverNames,\n      retrieverDescriptions,\n      retrievers,\n      retrieverPrompts,\n      defaults,\n      multiRetrievalChainOpts,\n      retrievalQAChainOpts,\n    }: {\n      retrieverNames: string[];\n      retrieverDescriptions: string[];\n      retrievers: BaseRetrieverInterface[];\n      retrieverPrompts?: PromptTemplate[];\n      defaults?: MultiRetrievalDefaults;\n      multiRetrievalChainOpts?: Omit<MultiRouteChainInput, \"defaultChain\">;\n      retrievalQAChainOpts?: Partial<\n        Omit<RetrievalQAChainInput, \"retriever\" | \"combineDocumentsChain\">\n      > & {\n        prompt?: PromptTemplate;\n      };\n    }\n  ): MultiRetrievalQAChain {\n    const { defaultRetriever, defaultPrompt, defaultChain } = defaults ?? {};\n    if (defaultPrompt && !defaultRetriever) {\n      throw new Error(\n        \"`default_retriever` must be specified if `default_prompt` is \\nprovided. Received only `default_prompt`.\"\n      );\n    }\n    const destinations = zipEntries<[string, string]>(\n      retrieverNames,\n      retrieverDescriptions\n    ).map(([name, desc]) => `${name}: ${desc}`);\n\n    const structuredOutputParserSchema = z.object({\n      destination: z\n        .string()\n        .optional()\n        .describe('name of the question answering system to use or \"DEFAULT\"'),\n      next_inputs: z\n        .object({\n          query: z\n            .string()\n            .describe(\"a potentially modified version of the original input\"),\n        })\n        .describe(\"input to be fed to the next model\"),\n    });\n\n    const outputParser = new RouterOutputParser<\n      typeof structuredOutputParserSchema\n    >(structuredOutputParserSchema);\n\n    const destinationsStr = destinations.join(\"\\n\");\n    const routerTemplate = interpolateFString(\n      STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE(\n        outputParser.getFormatInstructions({ interpolationDepth: 4 })\n      ),\n      {\n        destinations: destinationsStr,\n      }\n    );\n    const routerPrompt = new PromptTemplate({\n      template: routerTemplate,\n      inputVariables: [\"input\"],\n      outputParser,\n    });\n\n    const routerChain = LLMRouterChain.fromLLM(llm, routerPrompt);\n    const prompts = retrieverPrompts ?? retrievers.map(() => null);\n    const destinationChains = zipEntries<\n      [string, BaseRetrieverInterface, PromptTemplate | null]\n    >(retrieverNames, retrievers, prompts).reduce(\n      (acc, [name, retriever, prompt]) => {\n        const opt: Partial<RetrievalQAChainInput> & {\n          prompt?: PromptTemplate;\n        } = retrievalQAChainOpts ?? {};\n        if (prompt) {\n          opt.prompt = prompt;\n        }\n        acc[name] = RetrievalQAChain.fromLLM(llm, retriever, opt);\n        return acc;\n      },\n      {} as { [name: string]: RetrievalQAChain }\n    );\n\n    let _defaultChain;\n    if (defaultChain) {\n      _defaultChain = defaultChain;\n    } else if (defaultRetriever) {\n      _defaultChain = RetrievalQAChain.fromLLM(llm, defaultRetriever, {\n        ...retrievalQAChainOpts,\n        prompt: defaultPrompt,\n      });\n    } else {\n      const promptTemplate = DEFAULT_TEMPLATE.replace(\"input\", \"query\");\n      const prompt = new PromptTemplate({\n        template: promptTemplate,\n        inputVariables: [\"history\", \"query\"],\n      });\n      _defaultChain = new ConversationChain({\n        llm,\n        prompt,\n        outputKey: \"result\",\n      });\n    }\n\n    return new MultiRetrievalQAChain({\n      ...multiRetrievalChainOpts,\n      routerChain,\n      destinationChains,\n      defaultChain: _defaultChain,\n    });\n  }\n\n  _chainType(): string {\n    return \"multi_retrieval_qa_chain\";\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAiEA,IAAa,wBAAb,MAAa,8BAA8B,gBAAgB;CACzD,IAAI,aAAuB;AACzB,SAAO,CAAC,QAAS;CAClB;;;;CAKD,OAAO,eACLA,KACAC,gBACAC,uBACAC,YACAC,kBACAC,UACAC,SACA;AACA,SAAO,sBAAsB,qBAAqB,KAAK;GACrD;GACA;GACA;GACA;GACA;GACA,yBAAyB;EAC1B,EAAC;CACH;;;;;;;;;;;;;;;;;CAkBD,OAAO,qBACLN,KACA,EACE,gBACA,uBACA,YACA,kBACA,UACA,yBACA,sBAaD,EACsB;EACvB,MAAM,EAAE,kBAAkB,eAAe,cAAc,GAAG,YAAY,CAAE;AACxE,MAAI,iBAAiB,CAAC,iBACpB,OAAM,IAAI,MACR;EAGJ,MAAM,eAAe,WACnB,gBACA,sBACD,CAAC,IAAI,CAAC,CAAC,MAAM,KAAK,KAAK,GAAG,KAAK,EAAE,EAAE,MAAM,CAAC;EAE3C,MAAM,+BAA+B,EAAE,OAAO;GAC5C,aAAa,EACV,QAAQ,CACR,UAAU,CACV,SAAS,8DAA4D;GACxE,aAAa,EACV,OAAO,EACN,OAAO,EACJ,QAAQ,CACR,SAAS,uDAAuD,CACpE,EAAC,CACD,SAAS,oCAAoC;EACjD,EAAC;EAEF,MAAM,eAAe,IAAI,mBAEvB;EAEF,MAAM,kBAAkB,aAAa,KAAK,KAAK;EAC/C,MAAM,iBAAiB,mBACrB,2CACE,aAAa,sBAAsB,EAAE,oBAAoB,EAAG,EAAC,CAC9D,EACD,EACE,cAAc,gBACf,EACF;EACD,MAAM,eAAe,IAAI,eAAe;GACtC,UAAU;GACV,gBAAgB,CAAC,OAAQ;GACzB;EACD;EAED,MAAM,cAAc,eAAe,QAAQ,KAAK,aAAa;EAC7D,MAAM,UAAU,oBAAoB,WAAW,IAAI,MAAM,KAAK;EAC9D,MAAM,oBAAoB,WAExB,gBAAgB,YAAY,QAAQ,CAAC,OACrC,CAAC,KAAK,CAAC,MAAM,WAAW,OAAO,KAAK;GAClC,MAAMO,MAEF,wBAAwB,CAAE;AAC9B,OAAI,QACF,IAAI,SAAS;GAEf,IAAI,QAAQ,iBAAiB,QAAQ,KAAK,WAAW,IAAI;AACzD,UAAO;EACR,GACD,CAAE,EACH;EAED,IAAI;AACJ,MAAI,cACF,gBAAgB;WACP,kBACT,gBAAgB,iBAAiB,QAAQ,KAAK,kBAAkB;GAC9D,GAAG;GACH,QAAQ;EACT,EAAC;OACG;GACL,MAAM,iBAAiB,iBAAiB,QAAQ,SAAS,QAAQ;GACjE,MAAM,SAAS,IAAI,eAAe;IAChC,UAAU;IACV,gBAAgB,CAAC,WAAW,OAAQ;GACrC;GACD,gBAAgB,IAAI,kBAAkB;IACpC;IACA;IACA,WAAW;GACZ;EACF;AAED,SAAO,IAAI,sBAAsB;GAC/B,GAAG;GACH;GACA;GACA,cAAc;EACf;CACF;CAED,aAAqB;AACnB,SAAO;CACR;AACF"}
+{"version":3,"file":"multi_retrieval_qa.js","names":["llm: BaseLanguageModelInterface","retrieverNames: string[]","retrieverDescriptions: string[]","retrievers: BaseRetrieverInterface[]","retrieverPrompts?: PromptTemplate[]","defaults?: MultiRetrievalDefaults","options?: Omit<MultiRouteChainInput, \"defaultChain\">","opt: Partial<RetrievalQAChainInput> & {\n          prompt?: PromptTemplate;\n        }"],"sources":["../../../src/chains/router/multi_retrieval_qa.ts"],"sourcesContent":["import { z } from \"zod/v3\";\nimport type { BaseLanguageModelInterface } from \"@langchain/core/language_models/base\";\nimport type { BaseRetrieverInterface } from \"@langchain/core/retrievers\";\nimport { interpolateFString, PromptTemplate } from \"@langchain/core/prompts\";\nimport { MultiRouteChain, MultiRouteChainInput } from \"./multi_route.js\";\nimport { BaseChain } from \"../../chains/base.js\";\nimport { LLMRouterChain } from \"./llm_router.js\";\nimport {\n  ConversationChain,\n  DEFAULT_TEMPLATE,\n} from \"../../chains/conversation.js\";\nimport { STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE } from \"./multi_retrieval_prompt.js\";\nimport { zipEntries } from \"./utils.js\";\nimport {\n  RetrievalQAChain,\n  RetrievalQAChainInput,\n} from \"../../chains/retrieval_qa.js\";\nimport { RouterOutputParser } from \"../../output_parsers/router.js\";\n\n/**\n * A type that represents the default values for the MultiRetrievalQAChain\n * class. It includes optional properties for the default retriever,\n * default prompt, and default chain.\n */\nexport type MultiRetrievalDefaults = {\n  defaultRetriever?: BaseRetrieverInterface;\n  defaultPrompt?: PromptTemplate;\n  defaultChain?: BaseChain;\n};\n\n/**\n * A class that represents a multi-retrieval question answering chain in\n * the LangChain framework. It extends the MultiRouteChain class and\n * provides additional functionality specific to multi-retrieval QA\n * chains.\n * @example\n * ```typescript\n * const multiRetrievalQAChain = MultiRetrievalQAChain.fromLLMAndRetrievers(\n *   new ChatOpenAI({ model: \"gpt-4o-mini\" }),\n *   {\n *     retrieverNames: [\"aqua teen\", \"mst3k\", \"animaniacs\"],\n *     retrieverDescriptions: [\n *       \"Good for answering questions about Aqua Teen Hunger Force theme song\",\n *       \"Good for answering questions about Mystery Science Theater 3000 theme song\",\n *       \"Good for answering questions about Animaniacs theme song\",\n *     ],\n *     retrievers: [\n *       new MemoryVectorStore().asRetriever(3),\n *       new MemoryVectorStore().asRetriever(3),\n *       new MemoryVectorStore().asRetriever(3),\n *     ],\n *     retrievalQAChainOpts: {\n *       returnSourceDocuments: true,\n *     },\n *   },\n * );\n *\n * const result = await multiRetrievalQAChain.call({\n *   input:\n *     \"In the Aqua Teen Hunger Force theme song, who calls himself the mike rula?\",\n * });\n *\n * console.log(result.sourceDocuments, result.text);\n * ```\n */\nexport class MultiRetrievalQAChain extends MultiRouteChain {\n  get outputKeys(): string[] {\n    return [\"result\"];\n  }\n\n  /**\n   * @deprecated Use `fromRetrieversAndPrompts` instead\n   */\n  static fromRetrievers(\n    llm: BaseLanguageModelInterface,\n    retrieverNames: string[],\n    retrieverDescriptions: string[],\n    retrievers: BaseRetrieverInterface[],\n    retrieverPrompts?: PromptTemplate[],\n    defaults?: MultiRetrievalDefaults,\n    options?: Omit<MultiRouteChainInput, \"defaultChain\">\n  ) {\n    return MultiRetrievalQAChain.fromLLMAndRetrievers(llm, {\n      retrieverNames,\n      retrieverDescriptions,\n      retrievers,\n      retrieverPrompts,\n      defaults,\n      multiRetrievalChainOpts: options,\n    });\n  }\n\n  /**\n   * A static method that creates an instance of MultiRetrievalQAChain from\n   * a BaseLanguageModel and a set of retrievers. It takes in optional\n   * parameters for the retriever names, descriptions, prompts, defaults,\n   * and additional options. It is an alternative method to fromRetrievers\n   * and provides more flexibility in configuring the underlying chains.\n   * @param llm A BaseLanguageModel instance.\n   * @param retrieverNames An array of retriever names.\n   * @param retrieverDescriptions An array of retriever descriptions.\n   * @param retrievers An array of BaseRetrieverInterface instances.\n   * @param retrieverPrompts An optional array of PromptTemplate instances for the retrievers.\n   * @param defaults An optional MultiRetrievalDefaults instance.\n   * @param multiRetrievalChainOpts Additional optional parameters for the multi-retrieval chain.\n   * @param retrievalQAChainOpts Additional optional parameters for the retrieval QA chain.\n   * @returns A new instance of MultiRetrievalQAChain.\n   */\n  static fromLLMAndRetrievers(\n    llm: BaseLanguageModelInterface,\n    {\n      retrieverNames,\n      retrieverDescriptions,\n      retrievers,\n      retrieverPrompts,\n      defaults,\n      multiRetrievalChainOpts,\n      retrievalQAChainOpts,\n    }: {\n      retrieverNames: string[];\n      retrieverDescriptions: string[];\n      retrievers: BaseRetrieverInterface[];\n      retrieverPrompts?: PromptTemplate[];\n      defaults?: MultiRetrievalDefaults;\n      multiRetrievalChainOpts?: Omit<MultiRouteChainInput, \"defaultChain\">;\n      retrievalQAChainOpts?: Partial<\n        Omit<RetrievalQAChainInput, \"retriever\" | \"combineDocumentsChain\">\n      > & {\n        prompt?: PromptTemplate;\n      };\n    }\n  ): MultiRetrievalQAChain {\n    const { defaultRetriever, defaultPrompt, defaultChain } = defaults ?? {};\n    if (defaultPrompt && !defaultRetriever) {\n      throw new Error(\n        \"`default_retriever` must be specified if `default_prompt` is \\nprovided. Received only `default_prompt`.\"\n      );\n    }\n    const destinations = zipEntries<[string, string]>(\n      retrieverNames,\n      retrieverDescriptions\n    ).map(([name, desc]) => `${name}: ${desc}`);\n\n    const structuredOutputParserSchema = z.object({\n      destination: z\n        .string()\n        .optional()\n        .describe('name of the question answering system to use or \"DEFAULT\"'),\n      next_inputs: z\n        .object({\n          query: z\n            .string()\n            .describe(\"a potentially modified version of the original input\"),\n        })\n        .describe(\"input to be fed to the next model\"),\n    });\n\n    const outputParser = new RouterOutputParser<\n      typeof structuredOutputParserSchema\n    >(structuredOutputParserSchema);\n\n    const destinationsStr = destinations.join(\"\\n\");\n    const routerTemplate = interpolateFString(\n      STRUCTURED_MULTI_RETRIEVAL_ROUTER_TEMPLATE(\n        outputParser.getFormatInstructions({ interpolationDepth: 4 })\n      ),\n      {\n        destinations: destinationsStr,\n      }\n    );\n    const routerPrompt = new PromptTemplate({\n      template: routerTemplate,\n      inputVariables: [\"input\"],\n      outputParser,\n    });\n\n    const routerChain = LLMRouterChain.fromLLM(llm, routerPrompt);\n    const prompts = retrieverPrompts ?? retrievers.map(() => null);\n    const destinationChains = zipEntries<\n      [string, BaseRetrieverInterface, PromptTemplate | null]\n    >(retrieverNames, retrievers, prompts).reduce(\n      (acc, [name, retriever, prompt]) => {\n        const opt: Partial<RetrievalQAChainInput> & {\n          prompt?: PromptTemplate;\n        } = retrievalQAChainOpts ?? {};\n        if (prompt) {\n          opt.prompt = prompt;\n        }\n        acc[name] = RetrievalQAChain.fromLLM(llm, retriever, opt);\n        return acc;\n      },\n      {} as { [name: string]: RetrievalQAChain }\n    );\n\n    let _defaultChain;\n    if (defaultChain) {\n      _defaultChain = defaultChain;\n    } else if (defaultRetriever) {\n      _defaultChain = RetrievalQAChain.fromLLM(llm, defaultRetriever, {\n        ...retrievalQAChainOpts,\n        prompt: defaultPrompt,\n      });\n    } else {\n      const promptTemplate = DEFAULT_TEMPLATE.replace(\"input\", \"query\");\n      const prompt = new PromptTemplate({\n        template: promptTemplate,\n        inputVariables: [\"history\", \"query\"],\n      });\n      _defaultChain = new ConversationChain({\n        llm,\n        prompt,\n        outputKey: \"result\",\n      });\n    }\n\n    return new MultiRetrievalQAChain({\n      ...multiRetrievalChainOpts,\n      routerChain,\n      destinationChains,\n      defaultChain: _defaultChain,\n    });\n  }\n\n  _chainType(): string {\n    return \"multi_retrieval_qa_chain\";\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAiEA,IAAa,wBAAb,MAAa,8BAA8B,gBAAgB;CACzD,IAAI,aAAuB;AACzB,SAAO,CAAC,QAAS;CAClB;;;;CAKD,OAAO,eACLA,KACAC,gBACAC,uBACAC,YACAC,kBACAC,UACAC,SACA;AACA,SAAO,sBAAsB,qBAAqB,KAAK;GACrD;GACA;GACA;GACA;GACA;GACA,yBAAyB;EAC1B,EAAC;CACH;;;;;;;;;;;;;;;;;CAkBD,OAAO,qBACLN,KACA,EACE,gBACA,uBACA,YACA,kBACA,UACA,yBACA,sBAaD,EACsB;EACvB,MAAM,EAAE,kBAAkB,eAAe,cAAc,GAAG,YAAY,CAAE;AACxE,MAAI,iBAAiB,CAAC,iBACpB,OAAM,IAAI,MACR;EAGJ,MAAM,eAAe,WACnB,gBACA,sBACD,CAAC,IAAI,CAAC,CAAC,MAAM,KAAK,KAAK,GAAG,KAAK,EAAE,EAAE,MAAM,CAAC;EAE3C,MAAM,+BAA+B,EAAE,OAAO;GAC5C,aAAa,EACV,QAAQ,CACR,UAAU,CACV,SAAS,8DAA4D;GACxE,aAAa,EACV,OAAO,EACN,OAAO,EACJ,QAAQ,CACR,SAAS,uDAAuD,CACpE,EAAC,CACD,SAAS,oCAAoC;EACjD,EAAC;EAEF,MAAM,eAAe,IAAI,mBAEvB;EAEF,MAAM,kBAAkB,aAAa,KAAK,KAAK;EAC/C,MAAM,iBAAiB,mBACrB,2CACE,aAAa,sBAAsB,EAAE,oBAAoB,EAAG,EAAC,CAC9D,EACD,EACE,cAAc,gBACf,EACF;EACD,MAAM,eAAe,IAAI,eAAe;GACtC,UAAU;GACV,gBAAgB,CAAC,OAAQ;GACzB;EACD;EAED,MAAM,cAAc,eAAe,QAAQ,KAAK,aAAa;EAC7D,MAAM,UAAU,oBAAoB,WAAW,IAAI,MAAM,KAAK;EAC9D,MAAM,oBAAoB,WAExB,gBAAgB,YAAY,QAAQ,CAAC,OACrC,CAAC,KAAK,CAAC,MAAM,WAAW,OAAO,KAAK;GAClC,MAAMO,MAEF,wBAAwB,CAAE;AAC9B,OAAI,QACF,IAAI,SAAS;GAEf,IAAI,QAAQ,iBAAiB,QAAQ,KAAK,WAAW,IAAI;AACzD,UAAO;EACR,GACD,CAAE,EACH;EAED,IAAI;AACJ,MAAI,cACF,gBAAgB;WACP,kBACT,gBAAgB,iBAAiB,QAAQ,KAAK,kBAAkB;GAC9D,GAAG;GACH,QAAQ;EACT,EAAC;OACG;GACL,MAAM,iBAAiB,iBAAiB,QAAQ,SAAS,QAAQ;GACjE,MAAM,SAAS,IAAI,eAAe;IAChC,UAAU;IACV,gBAAgB,CAAC,WAAW,OAAQ;GACrC;GACD,gBAAgB,IAAI,kBAAkB;IACpC;IACA;IACA,WAAW;GACZ;EACF;AAED,SAAO,IAAI,sBAAsB;GAC/B,GAAG;GACH;GACA;GACA,cAAc;EACf;CACF;CAED,aAAqB;AACnB,SAAO;CACR;AACF"}

package/dist/chains/sql_db/sql_db_chain.d.cts

@@ -2,9 +2,9 @@ import { BaseChain, ChainInputs } from "../base.cjs";
 import { SqlDialect } from "./sql_db_prompt.cjs";
 import { SqlDatabase } from "../../sql_db.cjs";
 import { BaseLanguageModel, BaseLanguageModelInterface } from "@langchain/core/language_models/base";
-import { BasePromptTemplate, PromptTemplate } from "@langchain/core/prompts";
-import { RunnableSequence } from "@langchain/core/runnables";
 import { ChainValues } from "@langchain/core/utils/types";
+import { RunnableSequence } from "@langchain/core/runnables";
+import { BasePromptTemplate, PromptTemplate } from "@langchain/core/prompts";
 import { CallbackManagerForChainRun } from "@langchain/core/callbacks/manager";
 
 //#region src/chains/sql_db/sql_db_chain.d.ts

package/dist/chains/sql_db/sql_db_prompt.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"sql_db_prompt.cjs","names":["PromptTemplate","SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate>"],"sources":["../../../src/chains/sql_db/sql_db_prompt.ts"],"sourcesContent":["/* eslint-disable spaced-comment */\nimport { BasePromptTemplate, PromptTemplate } from \"@langchain/core/prompts\";\n\nexport const DEFAULT_SQL_DATABASE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `Given an input question, first create a syntactically correct {dialect} query to run, then look at the results of the query and return the answer. Unless the user specifies in his question a specific number of examples he wishes to obtain, always limit your query to at most {top_k} results. You can order the results by a relevant column to return the most interesting examples in the database.\n\nNever query for all the columns from a specific table, only ask for a the few relevant columns given the question.\n\nPay attention to use only the column names that you can see in the schema description. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the tables listed below.\n\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_POSTGRES_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a PostgreSQL expert. Given an input question, first create a syntactically correct PostgreSQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per PostgreSQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_SQLITE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a SQLite expert. Given an input question, first create a syntactically correct SQLite query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per SQLite. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_MYSQL_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a MySQL expert. Given an input question, first create a syntactically correct MySQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per MySQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in backticks (\\`) to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_MSSQL_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are an MS SQL expert. Given an input question, first create a syntactically correct MS SQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the TOP clause as per MS SQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in square brackets ([]) to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_SAP_HANA_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a SAP HANA expert. Given an input question, first create a syntactically correct SAP HANA query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per SAP HANA. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\nAlways use a schema name when executing a query.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_ORACLE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a ORACLE expert. Given an input question, first create a syntactically correct ORACLE query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the ROWNUM clause as per ORACLE. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport type SqlDialect =\n  | \"oracle\"\n  | \"postgres\"\n  | \"sqlite\"\n  | \"mysql\"\n  | \"mssql\"\n  | \"sap hana\";\n\nexport const SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate> = {\n  oracle: SQL_ORACLE_PROMPT,\n  postgres: SQL_POSTGRES_PROMPT,\n  sqlite: SQL_SQLITE_PROMPT,\n  mysql: SQL_MYSQL_PROMPT,\n  mssql: SQL_MSSQL_PROMPT,\n  \"sap hana\": SQL_SAP_HANA_PROMPT,\n};\n"],"mappings":";;;;AAGA,MAAa,8CAA4C,IAAIA,wCAAe;CAC1E,UAAU,CAAC;;;;;;;;;;;;;;;;;iBAiBI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,sCAAoC,IAAIA,wCAAe;CAClE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,oCAAkC,IAAIA,wCAAe;CAChE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,mCAAiC,IAAIA,wCAAe;CAC/D,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,mCAAiC,IAAIA,wCAAe;CAC/D,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,sCAAoC,IAAIA,wCAAe;CAClE,UAAU,CAAC;;;;;;;;;;;;;;;;iBAgBI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,oCAAkC,IAAIA,wCAAe;CAChE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAUD,MAAaC,kBAA0D;CACrE,QAAQ;CACR,UAAU;CACV,QAAQ;CACR,OAAO;CACP,OAAO;CACP,YAAY;AACb"}
+{"version":3,"file":"sql_db_prompt.cjs","names":["PromptTemplate","SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate>"],"sources":["../../../src/chains/sql_db/sql_db_prompt.ts"],"sourcesContent":["import { BasePromptTemplate, PromptTemplate } from \"@langchain/core/prompts\";\n\nexport const DEFAULT_SQL_DATABASE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `Given an input question, first create a syntactically correct {dialect} query to run, then look at the results of the query and return the answer. Unless the user specifies in his question a specific number of examples he wishes to obtain, always limit your query to at most {top_k} results. You can order the results by a relevant column to return the most interesting examples in the database.\n\nNever query for all the columns from a specific table, only ask for a the few relevant columns given the question.\n\nPay attention to use only the column names that you can see in the schema description. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the tables listed below.\n\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_POSTGRES_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a PostgreSQL expert. Given an input question, first create a syntactically correct PostgreSQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per PostgreSQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_SQLITE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a SQLite expert. Given an input question, first create a syntactically correct SQLite query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per SQLite. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_MYSQL_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a MySQL expert. Given an input question, first create a syntactically correct MySQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per MySQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in backticks (\\`) to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_MSSQL_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are an MS SQL expert. Given an input question, first create a syntactically correct MS SQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the TOP clause as per MS SQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in square brackets ([]) to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_SAP_HANA_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a SAP HANA expert. Given an input question, first create a syntactically correct SAP HANA query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per SAP HANA. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\nAlways use a schema name when executing a query.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_ORACLE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a ORACLE expert. Given an input question, first create a syntactically correct ORACLE query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the ROWNUM clause as per ORACLE. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport type SqlDialect =\n  | \"oracle\"\n  | \"postgres\"\n  | \"sqlite\"\n  | \"mysql\"\n  | \"mssql\"\n  | \"sap hana\";\n\nexport const SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate> = {\n  oracle: SQL_ORACLE_PROMPT,\n  postgres: SQL_POSTGRES_PROMPT,\n  sqlite: SQL_SQLITE_PROMPT,\n  mysql: SQL_MYSQL_PROMPT,\n  mssql: SQL_MSSQL_PROMPT,\n  \"sap hana\": SQL_SAP_HANA_PROMPT,\n};\n"],"mappings":";;;;AAEA,MAAa,8CAA4C,IAAIA,wCAAe;CAC1E,UAAU,CAAC;;;;;;;;;;;;;;;;;iBAiBI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,sCAAoC,IAAIA,wCAAe;CAClE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,oCAAkC,IAAIA,wCAAe;CAChE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,mCAAiC,IAAIA,wCAAe;CAC/D,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,mCAAiC,IAAIA,wCAAe;CAC/D,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,sCAAoC,IAAIA,wCAAe;CAClE,UAAU,CAAC;;;;;;;;;;;;;;;;iBAgBI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,oCAAkC,IAAIA,wCAAe;CAChE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAUD,MAAaC,kBAA0D;CACrE,QAAQ;CACR,UAAU;CACV,QAAQ;CACR,OAAO;CACP,OAAO;CACP,YAAY;AACb"}

package/dist/chains/sql_db/sql_db_prompt.d.cts.map

@@ -1 +1 @@
-{"version":3,"file":"sql_db_prompt.d.cts","names":["BasePromptTemplate","PromptTemplate","DEFAULT_SQL_DATABASE_PROMPT","SQL_POSTGRES_PROMPT","SQL_SQLITE_PROMPT","SQL_MYSQL_PROMPT","SQL_MSSQL_PROMPT","SQL_SAP_HANA_PROMPT","SQL_ORACLE_PROMPT","SqlDialect","SQL_PROMPTS_MAP","Record"],"sources":["../../../src/chains/sql_db/sql_db_prompt.d.ts"],"sourcesContent":["/* eslint-disable spaced-comment */\nimport { BasePromptTemplate, PromptTemplate } from \"@langchain/core/prompts\";\nexport declare const DEFAULT_SQL_DATABASE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_POSTGRES_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_SQLITE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_MYSQL_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_MSSQL_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_SAP_HANA_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_ORACLE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport type SqlDialect = \"oracle\" | \"postgres\" | \"sqlite\" | \"mysql\" | \"mssql\" | \"sap hana\";\nexport declare const SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate>;\n"],"mappings":";;;cAEqBE,6BAA6BD;EAA7BC,OAAAA,EAAAA,GAAAA;EAMAC,KAAAA,EAAAA,GAAAA;EAMAC,UAAAA,EAAAA,GAAAA;EAMAC,KAAAA,EAAAA,GAAAA;AAMrB,CAAA,EAAA,GAAqBC,CAAAA;AAMAC,cAxBAJ,mBAwBqBF,EAxBAA,cAwBc,CAAA;EAY5CQ,OAAAA,EAAAA,GAAAA;EACSC,KAAAA,EAAAA,GAAAA;EAAuD,UAAA,EAAA,GAAA;EAAA,KAA/BD,EAAAA,GAAAA;CAAU,EAAA,GAAA,CAAET;AAAnBW,cA/BjBP,iBA+BiBO,EA/BEV,cA+BFU,CAAAA;EAAM,OAAA,EAAA,GAAA;;;;;cAzBvBN,kBAAkBJ;;;;;;cAMlBK,kBAAkBL;;;;;;cAMlBM,qBAAqBN;;;;;;KAY9BQ,UAAAA;cACSC,iBAAiBC,OAAOF,YAAYT"}
+{"version":3,"file":"sql_db_prompt.d.cts","names":["BasePromptTemplate","PromptTemplate","DEFAULT_SQL_DATABASE_PROMPT","SQL_POSTGRES_PROMPT","SQL_SQLITE_PROMPT","SQL_MYSQL_PROMPT","SQL_MSSQL_PROMPT","SQL_SAP_HANA_PROMPT","SQL_ORACLE_PROMPT","SqlDialect","SQL_PROMPTS_MAP","Record"],"sources":["../../../src/chains/sql_db/sql_db_prompt.d.ts"],"sourcesContent":["import { BasePromptTemplate, PromptTemplate } from \"@langchain/core/prompts\";\nexport declare const DEFAULT_SQL_DATABASE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_POSTGRES_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_SQLITE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_MYSQL_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_MSSQL_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_SAP_HANA_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_ORACLE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport type SqlDialect = \"oracle\" | \"postgres\" | \"sqlite\" | \"mysql\" | \"mssql\" | \"sap hana\";\nexport declare const SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate>;\n"],"mappings":";;;cACqBE,6BAA6BD;;EAA7BC,KAAAA,EAAAA,GAAAA;EAMAC,UAAAA,EAAAA,GAAAA;EAMAC,KAAAA,EAAAA,GAAAA;AAMrB,CAAA,EAAA,GAAqBC,CAAAA;AAMAC,cAlBAH,mBAkBkBF,EAlBGA,cAkBW,CAAA;EAMhCM,OAAAA,EAAAA,GAAAA;EAYTE,KAAAA,EAAAA,GAAAA;EACSC,UAAAA,EAAAA,GAAAA;EAAuD,KAAA,EAAA,GAAA;CAAA,EAAA,GAAA,CAA/BD;AAAYT,cA/BpCI,iBA+BoCJ,EA/BjBC,cA+BiBD,CAAAA;EAAkB,OAArCW,EAAAA,GAAAA;EAAM,KAAA,EAAA,GAAA;;;;cAzBvBN,kBAAkBJ;;;;;;cAMlBK,kBAAkBL;;;;;;cAMlBM,qBAAqBN;;;;;;KAY9BQ,UAAAA;cACSC,iBAAiBC,OAAOF,YAAYT"}

package/dist/chains/sql_db/sql_db_prompt.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"sql_db_prompt.d.ts","names":["BasePromptTemplate","PromptTemplate","DEFAULT_SQL_DATABASE_PROMPT","SQL_POSTGRES_PROMPT","SQL_SQLITE_PROMPT","SQL_MYSQL_PROMPT","SQL_MSSQL_PROMPT","SQL_SAP_HANA_PROMPT","SQL_ORACLE_PROMPT","SqlDialect","SQL_PROMPTS_MAP","Record"],"sources":["../../../src/chains/sql_db/sql_db_prompt.d.ts"],"sourcesContent":["/* eslint-disable spaced-comment */\nimport { BasePromptTemplate, PromptTemplate } from \"@langchain/core/prompts\";\nexport declare const DEFAULT_SQL_DATABASE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_POSTGRES_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_SQLITE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_MYSQL_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_MSSQL_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_SAP_HANA_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_ORACLE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport type SqlDialect = \"oracle\" | \"postgres\" | \"sqlite\" | \"mysql\" | \"mssql\" | \"sap hana\";\nexport declare const SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate>;\n"],"mappings":";;;cAEqBE,6BAA6BD;EAA7BC,OAAAA,EAAAA,GAAAA;EAMAC,KAAAA,EAAAA,GAAAA;EAMAC,UAAAA,EAAAA,GAAAA;EAMAC,KAAAA,EAAAA,GAAAA;AAMrB,CAAA,EAAA,GAAqBC,CAAAA;AAMAC,cAxBAJ,mBAwBqBF,EAxBAA,cAwBc,CAAA;EAY5CQ,OAAAA,EAAAA,GAAAA;EACSC,KAAAA,EAAAA,GAAAA;EAAuD,UAAA,EAAA,GAAA;EAAA,KAA/BD,EAAAA,GAAAA;CAAU,EAAA,GAAA,CAAET;AAAnBW,cA/BjBP,iBA+BiBO,EA/BEV,cA+BFU,CAAAA;EAAM,OAAA,EAAA,GAAA;;;;;cAzBvBN,kBAAkBJ;;;;;;cAMlBK,kBAAkBL;;;;;;cAMlBM,qBAAqBN;;;;;;KAY9BQ,UAAAA;cACSC,iBAAiBC,OAAOF,YAAYT"}
+{"version":3,"file":"sql_db_prompt.d.ts","names":["BasePromptTemplate","PromptTemplate","DEFAULT_SQL_DATABASE_PROMPT","SQL_POSTGRES_PROMPT","SQL_SQLITE_PROMPT","SQL_MYSQL_PROMPT","SQL_MSSQL_PROMPT","SQL_SAP_HANA_PROMPT","SQL_ORACLE_PROMPT","SqlDialect","SQL_PROMPTS_MAP","Record"],"sources":["../../../src/chains/sql_db/sql_db_prompt.d.ts"],"sourcesContent":["import { BasePromptTemplate, PromptTemplate } from \"@langchain/core/prompts\";\nexport declare const DEFAULT_SQL_DATABASE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_POSTGRES_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_SQLITE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_MYSQL_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_MSSQL_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_SAP_HANA_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport declare const SQL_ORACLE_PROMPT: PromptTemplate<{\n    dialect: any;\n    input: any;\n    table_info: any;\n    top_k: any;\n}, any>;\nexport type SqlDialect = \"oracle\" | \"postgres\" | \"sqlite\" | \"mysql\" | \"mssql\" | \"sap hana\";\nexport declare const SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate>;\n"],"mappings":";;;cACqBE,6BAA6BD;;EAA7BC,KAAAA,EAAAA,GAAAA;EAMAC,UAAAA,EAAAA,GAAAA;EAMAC,KAAAA,EAAAA,GAAAA;AAMrB,CAAA,EAAA,GAAqBC,CAAAA;AAMAC,cAlBAH,mBAkBkBF,EAlBGA,cAkBW,CAAA;EAMhCM,OAAAA,EAAAA,GAAAA;EAYTE,KAAAA,EAAAA,GAAAA;EACSC,UAAAA,EAAAA,GAAAA;EAAuD,KAAA,EAAA,GAAA;CAAA,EAAA,GAAA,CAA/BD;AAAYT,cA/BpCI,iBA+BoCJ,EA/BjBC,cA+BiBD,CAAAA;EAAkB,OAArCW,EAAAA,GAAAA;EAAM,KAAA,EAAA,GAAA;;;;cAzBvBN,kBAAkBJ;;;;;;cAMlBK,kBAAkBL;;;;;;cAMlBM,qBAAqBN;;;;;;KAY9BQ,UAAAA;cACSC,iBAAiBC,OAAOF,YAAYT"}

package/dist/chains/sql_db/sql_db_prompt.js.map

@@ -1 +1 @@
-{"version":3,"file":"sql_db_prompt.js","names":["SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate>"],"sources":["../../../src/chains/sql_db/sql_db_prompt.ts"],"sourcesContent":["/* eslint-disable spaced-comment */\nimport { BasePromptTemplate, PromptTemplate } from \"@langchain/core/prompts\";\n\nexport const DEFAULT_SQL_DATABASE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `Given an input question, first create a syntactically correct {dialect} query to run, then look at the results of the query and return the answer. Unless the user specifies in his question a specific number of examples he wishes to obtain, always limit your query to at most {top_k} results. You can order the results by a relevant column to return the most interesting examples in the database.\n\nNever query for all the columns from a specific table, only ask for a the few relevant columns given the question.\n\nPay attention to use only the column names that you can see in the schema description. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the tables listed below.\n\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_POSTGRES_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a PostgreSQL expert. Given an input question, first create a syntactically correct PostgreSQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per PostgreSQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_SQLITE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a SQLite expert. Given an input question, first create a syntactically correct SQLite query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per SQLite. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_MYSQL_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a MySQL expert. Given an input question, first create a syntactically correct MySQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per MySQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in backticks (\\`) to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_MSSQL_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are an MS SQL expert. Given an input question, first create a syntactically correct MS SQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the TOP clause as per MS SQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in square brackets ([]) to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_SAP_HANA_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a SAP HANA expert. Given an input question, first create a syntactically correct SAP HANA query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per SAP HANA. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\nAlways use a schema name when executing a query.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_ORACLE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a ORACLE expert. Given an input question, first create a syntactically correct ORACLE query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the ROWNUM clause as per ORACLE. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport type SqlDialect =\n  | \"oracle\"\n  | \"postgres\"\n  | \"sqlite\"\n  | \"mysql\"\n  | \"mssql\"\n  | \"sap hana\";\n\nexport const SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate> = {\n  oracle: SQL_ORACLE_PROMPT,\n  postgres: SQL_POSTGRES_PROMPT,\n  sqlite: SQL_SQLITE_PROMPT,\n  mysql: SQL_MYSQL_PROMPT,\n  mssql: SQL_MSSQL_PROMPT,\n  \"sap hana\": SQL_SAP_HANA_PROMPT,\n};\n"],"mappings":";;;AAGA,MAAa,8CAA4C,IAAI,eAAe;CAC1E,UAAU,CAAC;;;;;;;;;;;;;;;;;iBAiBI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,sCAAoC,IAAI,eAAe;CAClE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,oCAAkC,IAAI,eAAe;CAChE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,mCAAiC,IAAI,eAAe;CAC/D,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,mCAAiC,IAAI,eAAe;CAC/D,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,sCAAoC,IAAI,eAAe;CAClE,UAAU,CAAC;;;;;;;;;;;;;;;;iBAgBI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,oCAAkC,IAAI,eAAe;CAChE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAUD,MAAaA,kBAA0D;CACrE,QAAQ;CACR,UAAU;CACV,QAAQ;CACR,OAAO;CACP,OAAO;CACP,YAAY;AACb"}
+{"version":3,"file":"sql_db_prompt.js","names":["SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate>"],"sources":["../../../src/chains/sql_db/sql_db_prompt.ts"],"sourcesContent":["import { BasePromptTemplate, PromptTemplate } from \"@langchain/core/prompts\";\n\nexport const DEFAULT_SQL_DATABASE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `Given an input question, first create a syntactically correct {dialect} query to run, then look at the results of the query and return the answer. Unless the user specifies in his question a specific number of examples he wishes to obtain, always limit your query to at most {top_k} results. You can order the results by a relevant column to return the most interesting examples in the database.\n\nNever query for all the columns from a specific table, only ask for a the few relevant columns given the question.\n\nPay attention to use only the column names that you can see in the schema description. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the tables listed below.\n\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_POSTGRES_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a PostgreSQL expert. Given an input question, first create a syntactically correct PostgreSQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per PostgreSQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_SQLITE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a SQLite expert. Given an input question, first create a syntactically correct SQLite query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per SQLite. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_MYSQL_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a MySQL expert. Given an input question, first create a syntactically correct MySQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per MySQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in backticks (\\`) to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_MSSQL_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are an MS SQL expert. Given an input question, first create a syntactically correct MS SQL query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the TOP clause as per MS SQL. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in square brackets ([]) to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_SAP_HANA_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a SAP HANA expert. Given an input question, first create a syntactically correct SAP HANA query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the LIMIT clause as per SAP HANA. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question. Wrap each column name in double quotes (\") to denote them as delimited identifiers.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\nAlways use a schema name when executing a query.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport const SQL_ORACLE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template: `You are a ORACLE expert. Given an input question, first create a syntactically correct ORACLE query to run, then look at the results of the query and return the answer to the input question.\nUnless the user specifies in the question a specific number of examples to obtain, query for at most {top_k} results using the ROWNUM clause as per ORACLE. You can order the results to return the most informative data in the database.\nNever query for all columns from a table. You must query only the columns that are needed to answer the question.\nPay attention to use only the column names you can see in the tables below. Be careful to not query for columns that do not exist. Also, pay attention to which column is in which table.\n\nUse the following format:\n\nQuestion: \"Question here\"\nSQLQuery: \"SQL Query to run\"\nSQLResult: \"Result of the SQLQuery\"\nAnswer: \"Final answer here\"\n\nOnly use the following tables:\n{table_info}\n\nQuestion: {input}`,\n  inputVariables: [\"dialect\", \"table_info\", \"input\", \"top_k\"],\n});\n\nexport type SqlDialect =\n  | \"oracle\"\n  | \"postgres\"\n  | \"sqlite\"\n  | \"mysql\"\n  | \"mssql\"\n  | \"sap hana\";\n\nexport const SQL_PROMPTS_MAP: Record<SqlDialect, BasePromptTemplate> = {\n  oracle: SQL_ORACLE_PROMPT,\n  postgres: SQL_POSTGRES_PROMPT,\n  sqlite: SQL_SQLITE_PROMPT,\n  mysql: SQL_MYSQL_PROMPT,\n  mssql: SQL_MSSQL_PROMPT,\n  \"sap hana\": SQL_SAP_HANA_PROMPT,\n};\n"],"mappings":";;;AAEA,MAAa,8CAA4C,IAAI,eAAe;CAC1E,UAAU,CAAC;;;;;;;;;;;;;;;;;iBAiBI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,sCAAoC,IAAI,eAAe;CAClE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,oCAAkC,IAAI,eAAe;CAChE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,mCAAiC,IAAI,eAAe;CAC/D,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,mCAAiC,IAAI,eAAe;CAC/D,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,sCAAoC,IAAI,eAAe;CAClE,UAAU,CAAC;;;;;;;;;;;;;;;;iBAgBI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAED,MAAa,oCAAkC,IAAI,eAAe;CAChE,UAAU,CAAC;;;;;;;;;;;;;;;iBAeI,CAAC;CAChB,gBAAgB;EAAC;EAAW;EAAc;EAAS;CAAQ;AAC5D;AAUD,MAAaA,kBAA0D;CACrE,QAAQ;CACR,UAAU;CACV,QAAQ;CACR,OAAO;CACP,OAAO;CACP,YAAY;AACb"}

package/dist/chains/summarization/stuff_prompts.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"stuff_prompts.cjs","names":["PromptTemplate"],"sources":["../../../src/chains/summarization/stuff_prompts.ts"],"sourcesContent":["/* eslint-disable spaced-comment */\nimport { PromptTemplate } from \"@langchain/core/prompts\";\n\nconst template = `Write a concise summary of the following:\n\n\n\"{text}\"\n\n\nCONCISE SUMMARY:`;\n\nexport const DEFAULT_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template,\n  inputVariables: [\"text\"],\n});\n"],"mappings":";;;;AAGA,MAAM,WAAW,CAAC;;;;;;gBAMF,CAAC;AAEjB,MAAa,iCAA+B,IAAIA,wCAAe;CAC7D;CACA,gBAAgB,CAAC,MAAO;AACzB"}
+{"version":3,"file":"stuff_prompts.cjs","names":["PromptTemplate"],"sources":["../../../src/chains/summarization/stuff_prompts.ts"],"sourcesContent":["import { PromptTemplate } from \"@langchain/core/prompts\";\n\nconst template = `Write a concise summary of the following:\n\n\n\"{text}\"\n\n\nCONCISE SUMMARY:`;\n\nexport const DEFAULT_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template,\n  inputVariables: [\"text\"],\n});\n"],"mappings":";;;;AAEA,MAAM,WAAW,CAAC;;;;;;gBAMF,CAAC;AAEjB,MAAa,iCAA+B,IAAIA,wCAAe;CAC7D;CACA,gBAAgB,CAAC,MAAO;AACzB"}

package/dist/chains/summarization/stuff_prompts.js.map

@@ -1 +1 @@
-{"version":3,"file":"stuff_prompts.js","names":[],"sources":["../../../src/chains/summarization/stuff_prompts.ts"],"sourcesContent":["/* eslint-disable spaced-comment */\nimport { PromptTemplate } from \"@langchain/core/prompts\";\n\nconst template = `Write a concise summary of the following:\n\n\n\"{text}\"\n\n\nCONCISE SUMMARY:`;\n\nexport const DEFAULT_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template,\n  inputVariables: [\"text\"],\n});\n"],"mappings":";;;AAGA,MAAM,WAAW,CAAC;;;;;;gBAMF,CAAC;AAEjB,MAAa,iCAA+B,IAAI,eAAe;CAC7D;CACA,gBAAgB,CAAC,MAAO;AACzB"}
+{"version":3,"file":"stuff_prompts.js","names":[],"sources":["../../../src/chains/summarization/stuff_prompts.ts"],"sourcesContent":["import { PromptTemplate } from \"@langchain/core/prompts\";\n\nconst template = `Write a concise summary of the following:\n\n\n\"{text}\"\n\n\nCONCISE SUMMARY:`;\n\nexport const DEFAULT_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template,\n  inputVariables: [\"text\"],\n});\n"],"mappings":";;;AAEA,MAAM,WAAW,CAAC;;;;;;gBAMF,CAAC;AAEjB,MAAa,iCAA+B,IAAI,eAAe;CAC7D;CACA,gBAAgB,CAAC,MAAO;AACzB"}

package/dist/chat_models/universal.cjs

@@ -135,6 +135,7 @@ function _inferModelProvider(modelName) {
 	else if (modelName.startsWith("accounts/fireworks")) return "fireworks";
 	else if (modelName.startsWith("gemini")) return "google-vertexai";
 	else if (modelName.startsWith("amazon.")) return "bedrock";
+	else if (modelName.startsWith("mistral")) return "mistralai";
 	else return void 0;
 }
 /**
@@ -187,21 +188,23 @@ var ConfigurableModel = class ConfigurableModel extends __langchain_core_languag
 		return model._generate(messages, options ?? {}, runManager);
 	}
 	bindTools(tools, params) {
-		this._queuedMethodOperations.bindTools = [tools, params];
+		const newQueuedOperations = { ...this._queuedMethodOperations };
+		newQueuedOperations.bindTools = [tools, params];
 		return new ConfigurableModel({
 			defaultConfig: this._defaultConfig,
 			configurableFields: this._configurableFields,
 			configPrefix: this._configPrefix,
-			queuedMethodOperations: this._queuedMethodOperations
+			queuedMethodOperations: newQueuedOperations
 		});
 	}
 	withStructuredOutput = (schema, ...args) => {
-		this._queuedMethodOperations.withStructuredOutput = [schema, ...args];
+		const newQueuedOperations = { ...this._queuedMethodOperations };
+		newQueuedOperations.withStructuredOutput = [schema, ...args];
 		return new ConfigurableModel({
 			defaultConfig: this._defaultConfig,
 			configurableFields: this._configurableFields,
 			configPrefix: this._configPrefix,
-			queuedMethodOperations: this._queuedMethodOperations
+			queuedMethodOperations: newQueuedOperations
 		});
 	};
 	_modelParams(config) {
@@ -404,7 +407,7 @@ var ConfigurableModel = class ConfigurableModel extends __langchain_core_languag
 * @example Bind tools to a configurable model:
 * ```typescript
 * import { initChatModel } from "langchain/chat_models/universal";
-* import { z } from "zod";
+* import { z } from "zod/v3";
 * import { tool } from "@langchain/core/tools";
 *
 * const getWeatherTool = tool(

package/dist/chat_models/universal.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"universal.cjs","names":["className: string","config","module","e: unknown","model: string","modelProvider?: string","params: Record<string, any>","modelName: string","BaseChatModel","fields: ConfigurableModelFields","config?: RunnableConfig","messages: BaseMessage[]","options?: this[\"ParsedCallOptions\"]","runManager?: CallbackManagerForLLMRun","tools: BindToolsInput[]","params?: Record<string, any>","modelParams: Record<string, any>","str: string","prefix: string","mergedConfig: RunnableConfig","remainingConfig: RunnableConfig","RunnableBinding","input: RunInput","options?: CallOptions","AsyncGeneratorWithSetup","IterableReadableStream","inputs: RunInput[]","options?: Partial<CallOptions> | Partial<CallOptions>[]","batchOptions?: RunnableBatchOptions","generator: AsyncGenerator<RunInput>","options: CallOptions","options?: Partial<CallOptions>","streamOptions?: Omit<LogStreamCallbackHandlerInput, \"autoClose\">","options: Partial<CallOptions> & {\n      version: \"v1\" | \"v2\";\n      encoding?: \"text/event-stream\" | undefined;\n    }","streamOptions?: Omit<EventStreamCallbackHandlerInput, \"autoClose\">","model?: string","fields?: Partial<Record<string, any>> & {\n    modelProvider?: string;\n    configurableFields?: string[] | \"any\";\n    configPrefix?: string;\n  }","paramsCopy: Record<string, any>"],"sources":["../../src/chat_models/universal.ts"],"sourcesContent":["/* eslint-disable import/no-extraneous-dependencies */\nimport {\n  BaseLanguageModelInput,\n  ToolDefinition,\n} from \"@langchain/core/language_models/base\";\nimport {\n  BaseChatModel,\n  BaseChatModelParams,\n  BindToolsInput,\n  type BaseChatModelCallOptions,\n} from \"@langchain/core/language_models/chat_models\";\nimport { BaseMessage, type AIMessageChunk } from \"@langchain/core/messages\";\nimport {\n  type RunnableBatchOptions,\n  RunnableBinding,\n  type RunnableConfig,\n  type RunnableToolLike,\n  ensureConfig,\n} from \"@langchain/core/runnables\";\nimport {\n  AsyncGeneratorWithSetup,\n  IterableReadableStream,\n} from \"@langchain/core/utils/stream\";\nimport {\n  type LogStreamCallbackHandlerInput,\n  type RunLogPatch,\n  type StreamEvent,\n} from \"@langchain/core/tracers/log_stream\";\nimport { type StructuredToolInterface } from \"@langchain/core/tools\";\nimport { CallbackManagerForLLMRun } from \"@langchain/core/callbacks/manager\";\nimport { ChatResult } from \"@langchain/core/outputs\";\n\n// TODO: remove once `EventStreamCallbackHandlerInput` is exposed in core\ninterface EventStreamCallbackHandlerInput\n  extends Omit<LogStreamCallbackHandlerInput, \"_schemaFormat\"> {}\n\nexport interface ConfigurableChatModelCallOptions\n  extends BaseChatModelCallOptions {\n  tools?: (\n    | StructuredToolInterface\n    | Record<string, unknown>\n    | ToolDefinition\n    | RunnableToolLike\n  )[];\n}\n\n// Configuration map for model providers\nexport const MODEL_PROVIDER_CONFIG = {\n  openai: {\n    package: \"@langchain/openai\",\n    className: \"ChatOpenAI\",\n  },\n  anthropic: {\n    package: \"@langchain/anthropic\",\n    className: \"ChatAnthropic\",\n  },\n  azure_openai: {\n    package: \"@langchain/openai\",\n    className: \"AzureChatOpenAI\",\n  },\n  cohere: {\n    package: \"@langchain/cohere\",\n    className: \"ChatCohere\",\n  },\n  \"google-vertexai\": {\n    package: \"@langchain/google-vertexai\",\n    className: \"ChatVertexAI\",\n  },\n  \"google-vertexai-web\": {\n    package: \"@langchain/google-vertexai-web\",\n    className: \"ChatVertexAI\",\n  },\n  \"google-genai\": {\n    package: \"@langchain/google-genai\",\n    className: \"ChatGoogleGenerativeAI\",\n  },\n  ollama: {\n    package: \"@langchain/ollama\",\n    className: \"ChatOllama\",\n  },\n  mistralai: {\n    package: \"@langchain/mistralai\",\n    className: \"ChatMistralAI\",\n  },\n  groq: {\n    package: \"@langchain/groq\",\n    className: \"ChatGroq\",\n  },\n  cerebras: {\n    package: \"@langchain/cerebras\",\n    className: \"ChatCerebras\",\n  },\n  bedrock: {\n    package: \"@langchain/aws\",\n    className: \"ChatBedrockConverse\",\n  },\n  deepseek: {\n    package: \"@langchain/deepseek\",\n    className: \"ChatDeepSeek\",\n  },\n  xai: {\n    package: \"@langchain/xai\",\n    className: \"ChatXAI\",\n  },\n  fireworks: {\n    package: \"@langchain/community/chat_models/fireworks\",\n    className: \"ChatFireworks\",\n    hasCircularDependency: true,\n  },\n  together: {\n    package: \"@langchain/community/chat_models/togetherai\",\n    className: \"ChatTogetherAI\",\n    hasCircularDependency: true,\n  },\n} as const;\n\nconst SUPPORTED_PROVIDERS = Object.keys(\n  MODEL_PROVIDER_CONFIG\n) as (keyof typeof MODEL_PROVIDER_CONFIG)[];\nexport type ChatModelProvider = keyof typeof MODEL_PROVIDER_CONFIG;\ntype ModelProviderConfig = {\n  package: string;\n  className: string;\n  hasCircularDependency?: boolean;\n};\n\n/**\n * Helper function to get a chat model class by its class name\n * @param className The class name (e.g., \"ChatOpenAI\", \"ChatAnthropic\")\n * @returns The imported model class or undefined if not found\n */\nexport async function getChatModelByClassName(className: string) {\n  // Find the provider config that matches the class name\n  const providerEntry = Object.entries(MODEL_PROVIDER_CONFIG).find(\n    ([, config]) => config.className === className\n  );\n\n  if (!providerEntry) {\n    return undefined;\n  }\n\n  const [, config] = providerEntry;\n  try {\n    const module = await import(config.package);\n    return module[config.className];\n  } catch (e: unknown) {\n    const err = e as Error;\n    if (\n      \"code\" in err &&\n      err.code?.toString().includes(\"ERR_MODULE_NOT_FOUND\")\n    ) {\n      const attemptedPackage = err.message\n        .split(\"Error: Cannot find package '\")[1]\n        .split(\"'\")[0];\n      throw new Error(\n        `Unable to import ${attemptedPackage}. Please install with ` +\n          `\\`npm install ${attemptedPackage}\\` or \\`pnpm install ${attemptedPackage}\\``\n      );\n    }\n    throw e;\n  }\n}\n\nasync function _initChatModelHelper(\n  model: string,\n  modelProvider?: string,\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  params: Record<string, any> = {}\n): Promise<BaseChatModel> {\n  const modelProviderCopy = modelProvider || _inferModelProvider(model);\n  if (!modelProviderCopy) {\n    throw new Error(\n      `Unable to infer model provider for { model: ${model} }, please specify modelProvider directly.`\n    );\n  }\n\n  const config = MODEL_PROVIDER_CONFIG[\n    modelProviderCopy as keyof typeof MODEL_PROVIDER_CONFIG\n  ] as ModelProviderConfig;\n  if (!config) {\n    const supported = SUPPORTED_PROVIDERS.join(\", \");\n    throw new Error(\n      `Unsupported { modelProvider: ${modelProviderCopy} }.\\n\\nSupported model providers are: ${supported}`\n    );\n  }\n\n  // eslint-disable-next-line @typescript-eslint/no-unused-vars\n  const { modelProvider: _unused, ...passedParams } = params;\n  const ProviderClass = await getChatModelByClassName(config.className);\n  return new ProviderClass({ model, ...passedParams });\n}\n\n/**\n * Attempts to infer the model provider based on the given model name.\n *\n * @param {string} modelName - The name of the model to infer the provider for.\n * @returns {string | undefined} The inferred model provider name, or undefined if unable to infer.\n *\n * @example\n * _inferModelProvider(\"gpt-4\"); // returns \"openai\"\n * _inferModelProvider(\"claude-2\"); // returns \"anthropic\"\n * _inferModelProvider(\"unknown-model\"); // returns undefined\n */\nexport function _inferModelProvider(modelName: string): string | undefined {\n  if (\n    modelName.startsWith(\"gpt-3\") ||\n    modelName.startsWith(\"gpt-4\") ||\n    modelName.startsWith(\"o1\") ||\n    modelName.startsWith(\"o3\") ||\n    modelName.startsWith(\"o4\")\n  ) {\n    return \"openai\";\n  } else if (modelName.startsWith(\"claude\")) {\n    return \"anthropic\";\n  } else if (modelName.startsWith(\"command\")) {\n    return \"cohere\";\n  } else if (modelName.startsWith(\"accounts/fireworks\")) {\n    return \"fireworks\";\n  } else if (modelName.startsWith(\"gemini\")) {\n    return \"google-vertexai\";\n  } else if (modelName.startsWith(\"amazon.\")) {\n    return \"bedrock\";\n  } else {\n    return undefined;\n  }\n}\n\ninterface ConfigurableModelFields extends BaseChatModelParams {\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  defaultConfig?: Record<string, any>;\n  /**\n   * @default \"any\"\n   */\n  configurableFields?: string[] | \"any\";\n  /**\n   * @default \"\"\n   */\n  configPrefix?: string;\n  /**\n   * Methods which should be called after the model is initialized.\n   * The key will be the method name, and the value will be the arguments.\n   */\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  queuedMethodOperations?: Record<string, any>;\n}\n\n/**\n * Internal class used to create chat models.\n *\n * @internal\n */\nexport class ConfigurableModel<\n  RunInput extends BaseLanguageModelInput = BaseLanguageModelInput,\n  CallOptions extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions\n> extends BaseChatModel<CallOptions, AIMessageChunk> {\n  _llmType(): string {\n    return \"chat_model\";\n  }\n\n  lc_namespace = [\"langchain\", \"chat_models\"];\n\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  _defaultConfig?: Record<string, any> = {};\n\n  /**\n   * @default \"any\"\n   */\n  _configurableFields: string[] | \"any\" = \"any\";\n\n  /**\n   * @default \"\"\n   */\n  _configPrefix: string;\n\n  /**\n   * Methods which should be called after the model is initialized.\n   * The key will be the method name, and the value will be the arguments.\n   */\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  _queuedMethodOperations: Record<string, any> = {};\n\n  constructor(fields: ConfigurableModelFields) {\n    super(fields);\n    this._defaultConfig = fields.defaultConfig ?? {};\n\n    if (fields.configurableFields === \"any\") {\n      this._configurableFields = \"any\";\n    } else {\n      this._configurableFields = fields.configurableFields ?? [\n        \"model\",\n        \"modelProvider\",\n      ];\n    }\n\n    if (fields.configPrefix) {\n      this._configPrefix = fields.configPrefix.endsWith(\"_\")\n        ? fields.configPrefix\n        : `${fields.configPrefix}_`;\n    } else {\n      this._configPrefix = \"\";\n    }\n\n    this._queuedMethodOperations =\n      fields.queuedMethodOperations ?? this._queuedMethodOperations;\n  }\n\n  async _model(config?: RunnableConfig) {\n    const params = { ...this._defaultConfig, ...this._modelParams(config) };\n    let initializedModel = await _initChatModelHelper(\n      params.model,\n      params.modelProvider,\n      params\n    );\n\n    // Apply queued method operations\n    const queuedMethodOperationsEntries = Object.entries(\n      this._queuedMethodOperations\n    );\n    if (queuedMethodOperationsEntries.length > 0) {\n      for (const [method, args] of queuedMethodOperationsEntries) {\n        if (\n          method in initializedModel &&\n          // eslint-disable-next-line @typescript-eslint/no-explicit-any\n          typeof (initializedModel as any)[method] === \"function\"\n        ) {\n          // eslint-disable-next-line @typescript-eslint/no-explicit-any\n          initializedModel = await (initializedModel as any)[method](...args);\n        }\n      }\n    }\n\n    return initializedModel;\n  }\n\n  async _generate(\n    messages: BaseMessage[],\n    options?: this[\"ParsedCallOptions\"],\n    runManager?: CallbackManagerForLLMRun\n  ): Promise<ChatResult> {\n    const model = await this._model(options);\n    return model._generate(messages, options ?? {}, runManager);\n  }\n\n  override bindTools(\n    tools: BindToolsInput[],\n    // eslint-disable-next-line @typescript-eslint/no-explicit-any\n    params?: Record<string, any>\n  ): ConfigurableModel<RunInput, CallOptions> {\n    this._queuedMethodOperations.bindTools = [tools, params];\n    return new ConfigurableModel<RunInput, CallOptions>({\n      defaultConfig: this._defaultConfig,\n      configurableFields: this._configurableFields,\n      configPrefix: this._configPrefix,\n      queuedMethodOperations: this._queuedMethodOperations,\n    });\n  }\n\n  // Extract the input types from the `BaseModel` class.\n  withStructuredOutput: BaseChatModel[\"withStructuredOutput\"] = (\n    schema,\n    ...args\n  ): ReturnType<BaseChatModel[\"withStructuredOutput\"]> => {\n    this._queuedMethodOperations.withStructuredOutput = [schema, ...args];\n    return new ConfigurableModel<RunInput, CallOptions>({\n      defaultConfig: this._defaultConfig,\n      configurableFields: this._configurableFields,\n      configPrefix: this._configPrefix,\n      queuedMethodOperations: this._queuedMethodOperations,\n    }) as unknown as ReturnType<BaseChatModel[\"withStructuredOutput\"]>;\n  };\n\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  _modelParams(config?: RunnableConfig): Record<string, any> {\n    const configurable = config?.configurable ?? {};\n    // eslint-disable-next-line @typescript-eslint/no-explicit-any\n    let modelParams: Record<string, any> = {};\n\n    for (const [key, value] of Object.entries(configurable)) {\n      if (key.startsWith(this._configPrefix)) {\n        const strippedKey = this._removePrefix(key, this._configPrefix);\n        modelParams[strippedKey] = value;\n      }\n    }\n\n    if (this._configurableFields !== \"any\") {\n      modelParams = Object.fromEntries(\n        Object.entries(modelParams).filter(([key]) =>\n          this._configurableFields.includes(key)\n        )\n      );\n    }\n\n    return modelParams;\n  }\n\n  _removePrefix(str: string, prefix: string): string {\n    return str.startsWith(prefix) ? str.slice(prefix.length) : str;\n  }\n\n  /**\n   * Bind config to a Runnable, returning a new Runnable.\n   * @param {RunnableConfig | undefined} [config] - The config to bind.\n   * @returns {RunnableBinding<RunInput, RunOutput, CallOptions>} A new RunnableBinding with the bound config.\n   */\n  withConfig(\n    config?: RunnableConfig\n  ): RunnableBinding<RunInput, AIMessageChunk, CallOptions> {\n    const mergedConfig: RunnableConfig = { ...(config || {}) };\n    const modelParams = this._modelParams(mergedConfig);\n\n    const remainingConfig: RunnableConfig = Object.fromEntries(\n      Object.entries(mergedConfig).filter(([k]) => k !== \"configurable\")\n    );\n\n    remainingConfig.configurable = Object.fromEntries(\n      Object.entries(mergedConfig.configurable || {}).filter(\n        ([k]) =>\n          this._configPrefix &&\n          !Object.keys(modelParams).includes(\n            this._removePrefix(k, this._configPrefix)\n          )\n      )\n    );\n\n    const newConfigurableModel = new ConfigurableModel<RunInput, CallOptions>({\n      defaultConfig: { ...this._defaultConfig, ...modelParams },\n      configurableFields: Array.isArray(this._configurableFields)\n        ? [...this._configurableFields]\n        : this._configurableFields,\n      configPrefix: this._configPrefix,\n      queuedMethodOperations: this._queuedMethodOperations,\n    });\n\n    return new RunnableBinding<RunInput, AIMessageChunk, CallOptions>({\n      config: mergedConfig,\n      bound: newConfigurableModel,\n    });\n  }\n\n  async invoke(\n    input: RunInput,\n    options?: CallOptions\n  ): Promise<AIMessageChunk> {\n    const model = await this._model(options);\n    const config = ensureConfig(options);\n    return model.invoke(input, config);\n  }\n\n  async stream(\n    input: RunInput,\n    options?: CallOptions\n  ): Promise<IterableReadableStream<AIMessageChunk>> {\n    const model = await this._model(options);\n    const wrappedGenerator = new AsyncGeneratorWithSetup({\n      generator: await model.stream(input, options),\n      config: options,\n    });\n    await wrappedGenerator.setup;\n    return IterableReadableStream.fromAsyncGenerator(wrappedGenerator);\n  }\n\n  async batch(\n    inputs: RunInput[],\n    options?: Partial<CallOptions> | Partial<CallOptions>[],\n    batchOptions?: RunnableBatchOptions & { returnExceptions?: false }\n  ): Promise<AIMessageChunk[]>;\n\n  async batch(\n    inputs: RunInput[],\n    options?: Partial<CallOptions> | Partial<CallOptions>[],\n    batchOptions?: RunnableBatchOptions & { returnExceptions: true }\n  ): Promise<(AIMessageChunk | Error)[]>;\n\n  async batch(\n    inputs: RunInput[],\n    options?: Partial<CallOptions> | Partial<CallOptions>[],\n    batchOptions?: RunnableBatchOptions\n  ): Promise<(AIMessageChunk | Error)[]>;\n\n  async batch(\n    inputs: RunInput[],\n    options?: Partial<CallOptions> | Partial<CallOptions>[],\n    batchOptions?: RunnableBatchOptions\n  ): Promise<(AIMessageChunk | Error)[]> {\n    // We can super this since the base runnable implementation of\n    // `.batch` will call `.invoke` on each input.\n    return super.batch(inputs, options, batchOptions);\n  }\n\n  async *transform(\n    generator: AsyncGenerator<RunInput>,\n    options: CallOptions\n  ): AsyncGenerator<AIMessageChunk> {\n    const model = await this._model(options);\n    const config = ensureConfig(options);\n\n    yield* model.transform(generator, config);\n  }\n\n  async *streamLog(\n    input: RunInput,\n    options?: Partial<CallOptions>,\n    streamOptions?: Omit<LogStreamCallbackHandlerInput, \"autoClose\">\n  ): AsyncGenerator<RunLogPatch> {\n    const model = await this._model(options);\n    const config = ensureConfig(options);\n\n    yield* model.streamLog(input, config, {\n      ...streamOptions,\n      _schemaFormat: \"original\",\n      includeNames: streamOptions?.includeNames,\n      includeTypes: streamOptions?.includeTypes,\n      includeTags: streamOptions?.includeTags,\n      excludeNames: streamOptions?.excludeNames,\n      excludeTypes: streamOptions?.excludeTypes,\n      excludeTags: streamOptions?.excludeTags,\n    });\n  }\n\n  streamEvents(\n    input: RunInput,\n    options: Partial<CallOptions> & { version: \"v1\" | \"v2\" },\n    streamOptions?: Omit<EventStreamCallbackHandlerInput, \"autoClose\">\n  ): IterableReadableStream<StreamEvent>;\n\n  streamEvents(\n    input: RunInput,\n    options: Partial<CallOptions> & {\n      version: \"v1\" | \"v2\";\n      encoding: \"text/event-stream\";\n    },\n    streamOptions?: Omit<EventStreamCallbackHandlerInput, \"autoClose\">\n  ): IterableReadableStream<Uint8Array>;\n\n  streamEvents(\n    input: RunInput,\n    options: Partial<CallOptions> & {\n      version: \"v1\" | \"v2\";\n      encoding?: \"text/event-stream\" | undefined;\n    },\n    streamOptions?: Omit<EventStreamCallbackHandlerInput, \"autoClose\">\n  ): IterableReadableStream<StreamEvent | Uint8Array> {\n    // eslint-disable-next-line @typescript-eslint/no-this-alias\n    const outerThis = this;\n    async function* wrappedGenerator() {\n      const model = await outerThis._model(options);\n      const config = ensureConfig(options);\n      const eventStream = model.streamEvents(input, config, streamOptions);\n\n      for await (const chunk of eventStream) {\n        yield chunk;\n      }\n    }\n    return IterableReadableStream.fromAsyncGenerator(wrappedGenerator());\n  }\n}\n\n// eslint-disable-next-line @typescript-eslint/no-explicit-any\nexport interface InitChatModelFields extends Partial<Record<string, any>> {\n  modelProvider?: string;\n  configurableFields?: string[] | \"any\";\n  configPrefix?: string;\n}\n\nexport type ConfigurableFields = \"any\" | string[];\n\nexport async function initChatModel<\n  RunInput extends BaseLanguageModelInput = BaseLanguageModelInput,\n  CallOptions extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions\n>(\n  model: string,\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  fields?: Partial<Record<string, any>> & {\n    modelProvider?: string;\n    configurableFields?: never;\n    configPrefix?: string;\n  }\n): Promise<ConfigurableModel<RunInput, CallOptions>>;\n\nexport async function initChatModel<\n  RunInput extends BaseLanguageModelInput = BaseLanguageModelInput,\n  CallOptions extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions\n>(\n  model: never,\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  options?: Partial<Record<string, any>> & {\n    modelProvider?: string;\n    configurableFields?: never;\n    configPrefix?: string;\n  }\n): Promise<ConfigurableModel<RunInput, CallOptions>>;\n\nexport async function initChatModel<\n  RunInput extends BaseLanguageModelInput = BaseLanguageModelInput,\n  CallOptions extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions\n>(\n  model?: string,\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  options?: Partial<Record<string, any>> & {\n    modelProvider?: string;\n    configurableFields?: ConfigurableFields;\n    configPrefix?: string;\n  }\n): Promise<ConfigurableModel<RunInput, CallOptions>>;\n\n// ################################# FOR CONTRIBUTORS #################################\n//\n// If adding support for a new provider, please append the provider\n// name to the supported list in the docstring below.\n//\n// ####################################################################################\n\n/**\n * Initialize a ChatModel from the model name and provider.\n * Must have the integration package corresponding to the model provider installed.\n *\n * @template {extends BaseLanguageModelInput = BaseLanguageModelInput} RunInput - The input type for the model.\n * @template {extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions} CallOptions - Call options for the model.\n *\n * @param {string | ChatModelProvider} [model] - The name of the model, e.g. \"gpt-4\", \"claude-3-opus-20240229\".\n *   Can be prefixed with the model provider, e.g. \"openai:gpt-4\", \"anthropic:claude-3-opus-20240229\".\n * @param {Object} [fields] - Additional configuration options.\n * @param {string} [fields.modelProvider] - The model provider. Supported values include:\n *   - openai (@langchain/openai)\n *   - anthropic (@langchain/anthropic)\n *   - azure_openai (@langchain/openai)\n *   - google-vertexai (@langchain/google-vertexai)\n *   - google-vertexai-web (@langchain/google-vertexai-web)\n *   - google-genai (@langchain/google-genai)\n *   - bedrock (@langchain/aws)\n *   - cohere (@langchain/cohere)\n *   - fireworks (@langchain/community/chat_models/fireworks)\n *   - together (@langchain/community/chat_models/togetherai)\n *   - mistralai (@langchain/mistralai)\n *   - groq (@langchain/groq)\n *   - ollama (@langchain/ollama)\n *   - cerebras (@langchain/cerebras)\n *   - deepseek (@langchain/deepseek)\n *   - xai (@langchain/xai)\n * @param {string[] | \"any\"} [fields.configurableFields] - Which model parameters are configurable:\n *   - undefined: No configurable fields.\n *   - \"any\": All fields are configurable. (See Security Note in description)\n *   - string[]: Specified fields are configurable.\n * @param {string} [fields.configPrefix] - Prefix for configurable fields at runtime.\n * @param {Record<string, any>} [fields.params] - Additional keyword args to pass to the ChatModel constructor.\n * @returns {Promise<ConfigurableModel<RunInput, CallOptions>>} A class which extends BaseChatModel.\n * @throws {Error} If modelProvider cannot be inferred or isn't supported.\n * @throws {Error} If the model provider integration package is not installed.\n *\n * @example Initialize non-configurable models\n * ```typescript\n * import { initChatModel } from \"langchain/chat_models/universal\";\n *\n * const gpt4 = await initChatModel(\"openai:gpt-4\", {\n *   temperature: 0.25,\n * });\n * const gpt4Result = await gpt4.invoke(\"what's your name\");\n *\n * const claude = await initChatModel(\"anthropic:claude-3-opus-20240229\", {\n *   temperature: 0.25,\n * });\n * const claudeResult = await claude.invoke(\"what's your name\");\n *\n * const gemini = await initChatModel(\"gemini-1.5-pro\", {\n *   modelProvider: \"google-vertexai\",\n *   temperature: 0.25,\n * });\n * const geminiResult = await gemini.invoke(\"what's your name\");\n * ```\n *\n * @example Create a partially configurable model with no default model\n * ```typescript\n * import { initChatModel } from \"langchain/chat_models/universal\";\n *\n * const configurableModel = await initChatModel(undefined, {\n *   temperature: 0,\n *   configurableFields: [\"model\", \"apiKey\"],\n * });\n *\n * const gpt4Result = await configurableModel.invoke(\"what's your name\", {\n *   configurable: {\n *     model: \"gpt-4\",\n *   },\n * });\n *\n * const claudeResult = await configurableModel.invoke(\"what's your name\", {\n *   configurable: {\n *     model: \"claude-3-5-sonnet-20240620\",\n *   },\n * });\n * ```\n *\n * @example Create a fully configurable model with a default model and a config prefix\n * ```typescript\n * import { initChatModel } from \"langchain/chat_models/universal\";\n *\n * const configurableModelWithDefault = await initChatModel(\"gpt-4\", {\n *   modelProvider: \"openai\",\n *   configurableFields: \"any\",\n *   configPrefix: \"foo\",\n *   temperature: 0,\n * });\n *\n * const openaiResult = await configurableModelWithDefault.invoke(\n *   \"what's your name\",\n *   {\n *     configurable: {\n *       foo_apiKey: process.env.OPENAI_API_KEY,\n *     },\n *   }\n * );\n *\n * const claudeResult = await configurableModelWithDefault.invoke(\n *   \"what's your name\",\n *   {\n *     configurable: {\n *       foo_model: \"claude-3-5-sonnet-20240620\",\n *       foo_modelProvider: \"anthropic\",\n *       foo_temperature: 0.6,\n *       foo_apiKey: process.env.ANTHROPIC_API_KEY,\n *     },\n *   }\n * );\n * ```\n *\n * @example Bind tools to a configurable model:\n * ```typescript\n * import { initChatModel } from \"langchain/chat_models/universal\";\n * import { z } from \"zod\";\n * import { tool } from \"@langchain/core/tools\";\n *\n * const getWeatherTool = tool(\n *   (input) => {\n *     // Do something with the input\n *     return JSON.stringify(input);\n *   },\n *   {\n *     schema: z\n *       .object({\n *         location: z\n *           .string()\n *           .describe(\"The city and state, e.g. San Francisco, CA\"),\n *       })\n *       .describe(\"Get the current weather in a given location\"),\n *     name: \"GetWeather\",\n *     description: \"Get the current weather in a given location\",\n *   }\n * );\n *\n * const getPopulationTool = tool(\n *   (input) => {\n *     // Do something with the input\n *     return JSON.stringify(input);\n *   },\n *   {\n *     schema: z\n *       .object({\n *         location: z\n *           .string()\n *           .describe(\"The city and state, e.g. San Francisco, CA\"),\n *       })\n *       .describe(\"Get the current population in a given location\"),\n *     name: \"GetPopulation\",\n *     description: \"Get the current population in a given location\",\n *   }\n * );\n *\n * const configurableModel = await initChatModel(\"gpt-4\", {\n *   configurableFields: [\"model\", \"modelProvider\", \"apiKey\"],\n *   temperature: 0,\n * });\n *\n * const configurableModelWithTools = configurableModel.bindTools([\n *   getWeatherTool,\n *   getPopulationTool,\n * ]);\n *\n * const configurableToolResult = await configurableModelWithTools.invoke(\n *   \"Which city is hotter today and which is bigger: LA or NY?\",\n *   {\n *     configurable: {\n *       apiKey: process.env.OPENAI_API_KEY,\n *     },\n *   }\n * );\n *\n * const configurableToolResult2 = await configurableModelWithTools.invoke(\n *   \"Which city is hotter today and which is bigger: LA or NY?\",\n *   {\n *     configurable: {\n *       model: \"claude-3-5-sonnet-20240620\",\n *       apiKey: process.env.ANTHROPIC_API_KEY,\n *     },\n *   }\n * );\n * ```\n *\n * @description\n * This function initializes a ChatModel based on the provided model name and provider.\n * It supports various model providers and allows for runtime configuration of model parameters.\n *\n * Security Note: Setting `configurableFields` to \"any\" means fields like apiKey, baseUrl, etc.\n * can be altered at runtime, potentially redirecting model requests to a different service/user.\n * Make sure that if you're accepting untrusted configurations, you enumerate the\n * `configurableFields` explicitly.\n *\n * The function will attempt to infer the model provider from the model name if not specified.\n * Certain model name prefixes are associated with specific providers:\n * - gpt-3... or gpt-4... -> openai\n * - claude... -> anthropic\n * - amazon.... -> bedrock\n * - gemini... -> google-vertexai\n * - command... -> cohere\n * - accounts/fireworks... -> fireworks\n *\n * @since 0.2.11\n * @version 0.2.11\n */\nexport async function initChatModel<\n  RunInput extends BaseLanguageModelInput = BaseLanguageModelInput,\n  CallOptions extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions\n>(\n  model?: string,\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  fields?: Partial<Record<string, any>> & {\n    modelProvider?: string;\n    configurableFields?: string[] | \"any\";\n    configPrefix?: string;\n  }\n): Promise<ConfigurableModel<RunInput, CallOptions>> {\n  // eslint-disable-next-line prefer-const\n  let { configurableFields, configPrefix, modelProvider, ...params } = {\n    configPrefix: \"\",\n    ...(fields ?? {}),\n  };\n  if (modelProvider === undefined && model?.includes(\":\")) {\n    const modelComponents = model.split(\":\", 2);\n    if (SUPPORTED_PROVIDERS.includes(modelComponents[0] as ChatModelProvider)) {\n      // eslint-disable-next-line no-param-reassign\n      [modelProvider, model] = modelComponents;\n    }\n  }\n  let configurableFieldsCopy = Array.isArray(configurableFields)\n    ? [...configurableFields]\n    : configurableFields;\n\n  if (!model && configurableFieldsCopy === undefined) {\n    configurableFieldsCopy = [\"model\", \"modelProvider\"];\n  }\n  if (configPrefix && configurableFieldsCopy === undefined) {\n    console.warn(\n      `{ configPrefix: ${configPrefix} } has been set but no fields are configurable. Set ` +\n        `{ configurableFields: [...] } to specify the model params that are ` +\n        `configurable.`\n    );\n  }\n\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  const paramsCopy: Record<string, any> = { ...params };\n\n  if (configurableFieldsCopy === undefined) {\n    return new ConfigurableModel<RunInput, CallOptions>({\n      defaultConfig: {\n        ...paramsCopy,\n        model,\n        modelProvider,\n      },\n      configPrefix,\n    });\n  } else {\n    if (model) {\n      paramsCopy.model = model;\n    }\n    if (modelProvider) {\n      paramsCopy.modelProvider = modelProvider;\n    }\n    return new ConfigurableModel<RunInput, CallOptions>({\n      defaultConfig: paramsCopy,\n      configPrefix,\n      configurableFields: configurableFieldsCopy,\n    });\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;AA+CA,MAAa,wBAAwB;CACnC,QAAQ;EACN,SAAS;EACT,WAAW;CACZ;CACD,WAAW;EACT,SAAS;EACT,WAAW;CACZ;CACD,cAAc;EACZ,SAAS;EACT,WAAW;CACZ;CACD,QAAQ;EACN,SAAS;EACT,WAAW;CACZ;CACD,mBAAmB;EACjB,SAAS;EACT,WAAW;CACZ;CACD,uBAAuB;EACrB,SAAS;EACT,WAAW;CACZ;CACD,gBAAgB;EACd,SAAS;EACT,WAAW;CACZ;CACD,QAAQ;EACN,SAAS;EACT,WAAW;CACZ;CACD,WAAW;EACT,SAAS;EACT,WAAW;CACZ;CACD,MAAM;EACJ,SAAS;EACT,WAAW;CACZ;CACD,UAAU;EACR,SAAS;EACT,WAAW;CACZ;CACD,SAAS;EACP,SAAS;EACT,WAAW;CACZ;CACD,UAAU;EACR,SAAS;EACT,WAAW;CACZ;CACD,KAAK;EACH,SAAS;EACT,WAAW;CACZ;CACD,WAAW;EACT,SAAS;EACT,WAAW;EACX,uBAAuB;CACxB;CACD,UAAU;EACR,SAAS;EACT,WAAW;EACX,uBAAuB;CACxB;AACF;AAED,MAAM,sBAAsB,OAAO,KACjC,sBACD;;;;;;AAaD,eAAsB,wBAAwBA,WAAmB;CAE/D,MAAM,gBAAgB,OAAO,QAAQ,sBAAsB,CAAC,KAC1D,CAAC,GAAGC,SAAO,KAAKA,SAAO,cAAc,UACtC;AAED,KAAI,CAAC,cACH,QAAO;CAGT,MAAM,GAAG,OAAO,GAAG;AACnB,KAAI;EACF,MAAMC,WAAS,MAAM,OAAO,OAAO;AACnC,SAAOA,SAAO,OAAO;CACtB,SAAQC,GAAY;EACnB,MAAM,MAAM;AACZ,MACE,UAAU,OACV,IAAI,MAAM,UAAU,CAAC,SAAS,uBAAuB,EACrD;GACA,MAAM,mBAAmB,IAAI,QAC1B,MAAM,+BAA+B,CAAC,GACtC,MAAM,IAAI,CAAC;AACd,SAAM,IAAI,MACR,CAAC,iBAAiB,EAAE,iBAAiB,oCAAsB,EACxC,iBAAiB,qBAAqB,EAAE,iBAAiB,EAAE,CAAC;EAElF;AACD,QAAM;CACP;AACF;AAED,eAAe,qBACbC,OACAC,eAEAC,SAA8B,CAAE,GACR;CACxB,MAAM,oBAAoB,iBAAiB,oBAAoB,MAAM;AACrE,KAAI,CAAC,kBACH,OAAM,IAAI,MACR,CAAC,4CAA4C,EAAE,MAAM,0CAA0C,CAAC;CAIpG,MAAM,SAAS,sBACb;AAEF,KAAI,CAAC,QAAQ;EACX,MAAM,YAAY,oBAAoB,KAAK,KAAK;AAChD,QAAM,IAAI,MACR,CAAC,6BAA6B,EAAE,kBAAkB,sCAAsC,EAAE,WAAW;CAExG;CAGD,MAAM,EAAE,eAAe,QAAS,GAAG,cAAc,GAAG;CACpD,MAAM,gBAAgB,MAAM,wBAAwB,OAAO,UAAU;AACrE,QAAO,IAAI,cAAc;EAAE;EAAO,GAAG;CAAc;AACpD;;;;;;;;;;;;AAaD,SAAgB,oBAAoBC,WAAuC;AACzE,KACE,UAAU,WAAW,QAAQ,IAC7B,UAAU,WAAW,QAAQ,IAC7B,UAAU,WAAW,KAAK,IAC1B,UAAU,WAAW,KAAK,IAC1B,UAAU,WAAW,KAAK,CAE1B,QAAO;UACE,UAAU,WAAW,SAAS,CACvC,QAAO;UACE,UAAU,WAAW,UAAU,CACxC,QAAO;UACE,UAAU,WAAW,qBAAqB,CACnD,QAAO;UACE,UAAU,WAAW,SAAS,CACvC,QAAO;UACE,UAAU,WAAW,UAAU,CACxC,QAAO;KAEP,QAAO;AAEV;;;;;;AA0BD,IAAa,oBAAb,MAAa,0BAGHC,2DAA2C;CACnD,WAAmB;AACjB,SAAO;CACR;CAED,eAAe,CAAC,aAAa,aAAc;CAG3C,iBAAuC,CAAE;;;;CAKzC,sBAAwC;;;;CAKxC;;;;;CAOA,0BAA+C,CAAE;CAEjD,YAAYC,QAAiC;EAC3C,MAAM,OAAO;EACb,KAAK,iBAAiB,OAAO,iBAAiB,CAAE;AAEhD,MAAI,OAAO,uBAAuB,OAChC,KAAK,sBAAsB;OAE3B,KAAK,sBAAsB,OAAO,sBAAsB,CACtD,SACA,eACD;AAGH,MAAI,OAAO,cACT,KAAK,gBAAgB,OAAO,aAAa,SAAS,IAAI,GAClD,OAAO,eACP,GAAG,OAAO,aAAa,CAAC,CAAC;OAE7B,KAAK,gBAAgB;EAGvB,KAAK,0BACH,OAAO,0BAA0B,KAAK;CACzC;CAED,MAAM,OAAOC,QAAyB;EACpC,MAAM,SAAS;GAAE,GAAG,KAAK;GAAgB,GAAG,KAAK,aAAa,OAAO;EAAE;EACvE,IAAI,mBAAmB,MAAM,qBAC3B,OAAO,OACP,OAAO,eACP,OACD;EAGD,MAAM,gCAAgC,OAAO,QAC3C,KAAK,wBACN;AACD,MAAI,8BAA8B,SAAS,GACzC;QAAK,MAAM,CAAC,QAAQ,KAAK,IAAI,8BAC3B,KACE,UAAU,oBAEV,OAAQ,iBAAyB,YAAY,YAG7C,mBAAmB,MAAO,iBAAyB,QAAQ,GAAG,KAAK;EAEtE;AAGH,SAAO;CACR;CAED,MAAM,UACJC,UACAC,SACAC,YACqB;EACrB,MAAM,QAAQ,MAAM,KAAK,OAAO,QAAQ;AACxC,SAAO,MAAM,UAAU,UAAU,WAAW,CAAE,GAAE,WAAW;CAC5D;CAED,AAAS,UACPC,OAEAC,QAC0C;EAC1C,KAAK,wBAAwB,YAAY,CAAC,OAAO,MAAO;AACxD,SAAO,IAAI,kBAAyC;GAClD,eAAe,KAAK;GACpB,oBAAoB,KAAK;GACzB,cAAc,KAAK;GACnB,wBAAwB,KAAK;EAC9B;CACF;CAGD,uBAA8D,CAC5D,QACA,GAAG,SACmD;EACtD,KAAK,wBAAwB,uBAAuB,CAAC,QAAQ,GAAG,IAAK;AACrE,SAAO,IAAI,kBAAyC;GAClD,eAAe,KAAK;GACpB,oBAAoB,KAAK;GACzB,cAAc,KAAK;GACnB,wBAAwB,KAAK;EAC9B;CACF;CAGD,aAAaL,QAA8C;EACzD,MAAM,eAAe,QAAQ,gBAAgB,CAAE;EAE/C,IAAIM,cAAmC,CAAE;AAEzC,OAAK,MAAM,CAAC,KAAK,MAAM,IAAI,OAAO,QAAQ,aAAa,CACrD,KAAI,IAAI,WAAW,KAAK,cAAc,EAAE;GACtC,MAAM,cAAc,KAAK,cAAc,KAAK,KAAK,cAAc;GAC/D,YAAY,eAAe;EAC5B;AAGH,MAAI,KAAK,wBAAwB,OAC/B,cAAc,OAAO,YACnB,OAAO,QAAQ,YAAY,CAAC,OAAO,CAAC,CAAC,IAAI,KACvC,KAAK,oBAAoB,SAAS,IAAI,CACvC,CACF;AAGH,SAAO;CACR;CAED,cAAcC,KAAaC,QAAwB;AACjD,SAAO,IAAI,WAAW,OAAO,GAAG,IAAI,MAAM,OAAO,OAAO,GAAG;CAC5D;;;;;;CAOD,WACER,QACwD;EACxD,MAAMS,eAA+B,EAAE,GAAI,UAAU,CAAE,EAAG;EAC1D,MAAM,cAAc,KAAK,aAAa,aAAa;EAEnD,MAAMC,kBAAkC,OAAO,YAC7C,OAAO,QAAQ,aAAa,CAAC,OAAO,CAAC,CAAC,EAAE,KAAK,MAAM,eAAe,CACnE;EAED,gBAAgB,eAAe,OAAO,YACpC,OAAO,QAAQ,aAAa,gBAAgB,CAAE,EAAC,CAAC,OAC9C,CAAC,CAAC,EAAE,KACF,KAAK,iBACL,CAAC,OAAO,KAAK,YAAY,CAAC,SACxB,KAAK,cAAc,GAAG,KAAK,cAAc,CAC1C,CACJ,CACF;EAED,MAAM,uBAAuB,IAAI,kBAAyC;GACxE,eAAe;IAAE,GAAG,KAAK;IAAgB,GAAG;GAAa;GACzD,oBAAoB,MAAM,QAAQ,KAAK,oBAAoB,GACvD,CAAC,GAAG,KAAK,mBAAoB,IAC7B,KAAK;GACT,cAAc,KAAK;GACnB,wBAAwB,KAAK;EAC9B;AAED,SAAO,IAAIC,2CAAuD;GAChE,QAAQ;GACR,OAAO;EACR;CACF;CAED,MAAM,OACJC,OACAC,SACyB;EACzB,MAAM,QAAQ,MAAM,KAAK,OAAO,QAAQ;EACxC,MAAM,sDAAsB,QAAQ;AACpC,SAAO,MAAM,OAAO,OAAO,OAAO;CACnC;CAED,MAAM,OACJD,OACAC,SACiD;EACjD,MAAM,QAAQ,MAAM,KAAK,OAAO,QAAQ;EACxC,MAAM,mBAAmB,IAAIC,sDAAwB;GACnD,WAAW,MAAM,MAAM,OAAO,OAAO,QAAQ;GAC7C,QAAQ;EACT;EACD,MAAM,iBAAiB;AACvB,SAAOC,qDAAuB,mBAAmB,iBAAiB;CACnE;CAoBD,MAAM,MACJC,QACAC,SACAC,cACqC;AAGrC,SAAO,MAAM,MAAM,QAAQ,SAAS,aAAa;CAClD;CAED,OAAO,UACLC,WACAC,SACgC;EAChC,MAAM,QAAQ,MAAM,KAAK,OAAO,QAAQ;EACxC,MAAM,sDAAsB,QAAQ;EAEpC,OAAO,MAAM,UAAU,WAAW,OAAO;CAC1C;CAED,OAAO,UACLR,OACAS,SACAC,eAC6B;EAC7B,MAAM,QAAQ,MAAM,KAAK,OAAO,QAAQ;EACxC,MAAM,sDAAsB,QAAQ;EAEpC,OAAO,MAAM,UAAU,OAAO,QAAQ;GACpC,GAAG;GACH,eAAe;GACf,cAAc,eAAe;GAC7B,cAAc,eAAe;GAC7B,aAAa,eAAe;GAC5B,cAAc,eAAe;GAC7B,cAAc,eAAe;GAC7B,aAAa,eAAe;EAC7B,EAAC;CACH;CAiBD,aACEV,OACAW,SAIAC,eACkD;EAElD,MAAM,YAAY;EAClB,gBAAgB,mBAAmB;GACjC,MAAM,QAAQ,MAAM,UAAU,OAAO,QAAQ;GAC7C,MAAM,sDAAsB,QAAQ;GACpC,MAAM,cAAc,MAAM,aAAa,OAAO,QAAQ,cAAc;AAEpE,cAAW,MAAM,SAAS,aACxB,MAAM;EAET;AACD,SAAOT,qDAAuB,mBAAmB,kBAAkB,CAAC;CACrE;AACF;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAuQD,eAAsB,cAIpBU,OAEAC,QAKmD;CAEnD,IAAI,EAAE,oBAAoB,cAAc,cAAe,GAAG,QAAQ,GAAG;EACnE,cAAc;EACd,GAAI,UAAU,CAAE;CACjB;AACD,KAAI,kBAAkB,UAAa,OAAO,SAAS,IAAI,EAAE;EACvD,MAAM,kBAAkB,MAAM,MAAM,KAAK,EAAE;AAC3C,MAAI,oBAAoB,SAAS,gBAAgB,GAAwB,EAEvE,CAAC,eAAe,MAAM,GAAG;CAE5B;CACD,IAAI,yBAAyB,MAAM,QAAQ,mBAAmB,GAC1D,CAAC,GAAG,kBAAmB,IACvB;AAEJ,KAAI,CAAC,SAAS,2BAA2B,QACvC,yBAAyB,CAAC,SAAS,eAAgB;AAErD,KAAI,gBAAgB,2BAA2B,QAC7C,QAAQ,KACN,CAAC,gBAAgB,EAAE,aAAa,oIAAoD,CAEnE,CAClB;CAIH,MAAMC,aAAkC,EAAE,GAAG,OAAQ;AAErD,KAAI,2BAA2B,OAC7B,QAAO,IAAI,kBAAyC;EAClD,eAAe;GACb,GAAG;GACH;GACA;EACD;EACD;CACD;MACI;AACL,MAAI,OACF,WAAW,QAAQ;AAErB,MAAI,eACF,WAAW,gBAAgB;AAE7B,SAAO,IAAI,kBAAyC;GAClD,eAAe;GACf;GACA,oBAAoB;EACrB;CACF;AACF"}
+{"version":3,"file":"universal.cjs","names":["className: string","config","module","e: unknown","model: string","modelProvider?: string","params: Record<string, any>","modelName: string","BaseChatModel","fields: ConfigurableModelFields","config?: RunnableConfig","messages: BaseMessage[]","options?: this[\"ParsedCallOptions\"]","runManager?: CallbackManagerForLLMRun","tools: BindToolsInput[]","params?: Record<string, any>","modelParams: Record<string, any>","str: string","prefix: string","mergedConfig: RunnableConfig","remainingConfig: RunnableConfig","RunnableBinding","input: RunInput","options?: CallOptions","AsyncGeneratorWithSetup","IterableReadableStream","inputs: RunInput[]","options?: Partial<CallOptions> | Partial<CallOptions>[]","batchOptions?: RunnableBatchOptions","generator: AsyncGenerator<RunInput>","options: CallOptions","options?: Partial<CallOptions>","streamOptions?: Omit<LogStreamCallbackHandlerInput, \"autoClose\">","options: Partial<CallOptions> & {\n      version: \"v1\" | \"v2\";\n      encoding?: \"text/event-stream\" | undefined;\n    }","streamOptions?: Omit<EventStreamCallbackHandlerInput, \"autoClose\">","model?: string","fields?: Partial<Record<string, any>> & {\n    modelProvider?: string;\n    configurableFields?: string[] | \"any\";\n    configPrefix?: string;\n  }","paramsCopy: Record<string, any>"],"sources":["../../src/chat_models/universal.ts"],"sourcesContent":["import {\n  BaseLanguageModelInput,\n  ToolDefinition,\n} from \"@langchain/core/language_models/base\";\nimport {\n  BaseChatModel,\n  BaseChatModelParams,\n  BindToolsInput,\n  type BaseChatModelCallOptions,\n} from \"@langchain/core/language_models/chat_models\";\nimport {\n  BaseMessage,\n  type AIMessageChunk,\n  MessageStructure,\n} from \"@langchain/core/messages\";\nimport {\n  type RunnableBatchOptions,\n  RunnableBinding,\n  type RunnableConfig,\n  type RunnableToolLike,\n  ensureConfig,\n} from \"@langchain/core/runnables\";\nimport {\n  AsyncGeneratorWithSetup,\n  IterableReadableStream,\n} from \"@langchain/core/utils/stream\";\nimport {\n  type LogStreamCallbackHandlerInput,\n  type RunLogPatch,\n  type StreamEvent,\n} from \"@langchain/core/tracers/log_stream\";\nimport { type StructuredToolInterface } from \"@langchain/core/tools\";\nimport { CallbackManagerForLLMRun } from \"@langchain/core/callbacks/manager\";\nimport { ChatResult } from \"@langchain/core/outputs\";\n\n// TODO: remove once `EventStreamCallbackHandlerInput` is exposed in core\ninterface EventStreamCallbackHandlerInput\n  extends Omit<LogStreamCallbackHandlerInput, \"_schemaFormat\"> {}\n\nexport interface ConfigurableChatModelCallOptions\n  extends BaseChatModelCallOptions {\n  tools?: (\n    | StructuredToolInterface\n    | Record<string, unknown>\n    | ToolDefinition\n    | RunnableToolLike\n  )[];\n}\n\n// Configuration map for model providers\nexport const MODEL_PROVIDER_CONFIG = {\n  openai: {\n    package: \"@langchain/openai\",\n    className: \"ChatOpenAI\",\n  },\n  anthropic: {\n    package: \"@langchain/anthropic\",\n    className: \"ChatAnthropic\",\n  },\n  azure_openai: {\n    package: \"@langchain/openai\",\n    className: \"AzureChatOpenAI\",\n  },\n  cohere: {\n    package: \"@langchain/cohere\",\n    className: \"ChatCohere\",\n  },\n  \"google-vertexai\": {\n    package: \"@langchain/google-vertexai\",\n    className: \"ChatVertexAI\",\n  },\n  \"google-vertexai-web\": {\n    package: \"@langchain/google-vertexai-web\",\n    className: \"ChatVertexAI\",\n  },\n  \"google-genai\": {\n    package: \"@langchain/google-genai\",\n    className: \"ChatGoogleGenerativeAI\",\n  },\n  ollama: {\n    package: \"@langchain/ollama\",\n    className: \"ChatOllama\",\n  },\n  mistralai: {\n    package: \"@langchain/mistralai\",\n    className: \"ChatMistralAI\",\n  },\n  groq: {\n    package: \"@langchain/groq\",\n    className: \"ChatGroq\",\n  },\n  cerebras: {\n    package: \"@langchain/cerebras\",\n    className: \"ChatCerebras\",\n  },\n  bedrock: {\n    package: \"@langchain/aws\",\n    className: \"ChatBedrockConverse\",\n  },\n  deepseek: {\n    package: \"@langchain/deepseek\",\n    className: \"ChatDeepSeek\",\n  },\n  xai: {\n    package: \"@langchain/xai\",\n    className: \"ChatXAI\",\n  },\n  fireworks: {\n    package: \"@langchain/community/chat_models/fireworks\",\n    className: \"ChatFireworks\",\n    hasCircularDependency: true,\n  },\n  together: {\n    package: \"@langchain/community/chat_models/togetherai\",\n    className: \"ChatTogetherAI\",\n    hasCircularDependency: true,\n  },\n} as const;\n\nconst SUPPORTED_PROVIDERS = Object.keys(\n  MODEL_PROVIDER_CONFIG\n) as (keyof typeof MODEL_PROVIDER_CONFIG)[];\nexport type ChatModelProvider = keyof typeof MODEL_PROVIDER_CONFIG;\ntype ModelProviderConfig = {\n  package: string;\n  className: string;\n  hasCircularDependency?: boolean;\n};\n\n/**\n * Helper function to get a chat model class by its class name\n * @param className The class name (e.g., \"ChatOpenAI\", \"ChatAnthropic\")\n * @returns The imported model class or undefined if not found\n */\nexport async function getChatModelByClassName(className: string) {\n  // Find the provider config that matches the class name\n  const providerEntry = Object.entries(MODEL_PROVIDER_CONFIG).find(\n    ([, config]) => config.className === className\n  );\n\n  if (!providerEntry) {\n    return undefined;\n  }\n\n  const [, config] = providerEntry;\n  try {\n    const module = await import(config.package);\n    return module[config.className];\n  } catch (e: unknown) {\n    const err = e as Error;\n    if (\n      \"code\" in err &&\n      err.code?.toString().includes(\"ERR_MODULE_NOT_FOUND\")\n    ) {\n      const attemptedPackage = err.message\n        .split(\"Error: Cannot find package '\")[1]\n        .split(\"'\")[0];\n      throw new Error(\n        `Unable to import ${attemptedPackage}. Please install with ` +\n          `\\`npm install ${attemptedPackage}\\` or \\`pnpm install ${attemptedPackage}\\``\n      );\n    }\n    throw e;\n  }\n}\n\nasync function _initChatModelHelper(\n  model: string,\n  modelProvider?: string,\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  params: Record<string, any> = {}\n): Promise<BaseChatModel> {\n  const modelProviderCopy = modelProvider || _inferModelProvider(model);\n  if (!modelProviderCopy) {\n    throw new Error(\n      `Unable to infer model provider for { model: ${model} }, please specify modelProvider directly.`\n    );\n  }\n\n  const config = MODEL_PROVIDER_CONFIG[\n    modelProviderCopy as keyof typeof MODEL_PROVIDER_CONFIG\n  ] as ModelProviderConfig;\n  if (!config) {\n    const supported = SUPPORTED_PROVIDERS.join(\", \");\n    throw new Error(\n      `Unsupported { modelProvider: ${modelProviderCopy} }.\\n\\nSupported model providers are: ${supported}`\n    );\n  }\n\n  const { modelProvider: _unused, ...passedParams } = params;\n  const ProviderClass = await getChatModelByClassName(config.className);\n  return new ProviderClass({ model, ...passedParams });\n}\n\n/**\n * Attempts to infer the model provider based on the given model name.\n *\n * @param {string} modelName - The name of the model to infer the provider for.\n * @returns {string | undefined} The inferred model provider name, or undefined if unable to infer.\n *\n * @example\n * _inferModelProvider(\"gpt-4\"); // returns \"openai\"\n * _inferModelProvider(\"claude-2\"); // returns \"anthropic\"\n * _inferModelProvider(\"unknown-model\"); // returns undefined\n */\nexport function _inferModelProvider(modelName: string): string | undefined {\n  if (\n    modelName.startsWith(\"gpt-3\") ||\n    modelName.startsWith(\"gpt-4\") ||\n    modelName.startsWith(\"o1\") ||\n    modelName.startsWith(\"o3\") ||\n    modelName.startsWith(\"o4\")\n  ) {\n    return \"openai\";\n  } else if (modelName.startsWith(\"claude\")) {\n    return \"anthropic\";\n  } else if (modelName.startsWith(\"command\")) {\n    return \"cohere\";\n  } else if (modelName.startsWith(\"accounts/fireworks\")) {\n    return \"fireworks\";\n  } else if (modelName.startsWith(\"gemini\")) {\n    return \"google-vertexai\";\n  } else if (modelName.startsWith(\"amazon.\")) {\n    return \"bedrock\";\n  } else if (modelName.startsWith(\"mistral\")) {\n    return \"mistralai\";\n  } else {\n    return undefined;\n  }\n}\n\ninterface ConfigurableModelFields extends BaseChatModelParams {\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  defaultConfig?: Record<string, any>;\n  /**\n   * @default \"any\"\n   */\n  configurableFields?: string[] | \"any\";\n  /**\n   * @default \"\"\n   */\n  configPrefix?: string;\n  /**\n   * Methods which should be called after the model is initialized.\n   * The key will be the method name, and the value will be the arguments.\n   */\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  queuedMethodOperations?: Record<string, any>;\n}\n\n/**\n * Internal class used to create chat models.\n *\n * @internal\n */\nexport class ConfigurableModel<\n  RunInput extends BaseLanguageModelInput = BaseLanguageModelInput,\n  CallOptions extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions\n> extends BaseChatModel<CallOptions, AIMessageChunk> {\n  _llmType(): string {\n    return \"chat_model\";\n  }\n\n  lc_namespace = [\"langchain\", \"chat_models\"];\n\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  _defaultConfig?: Record<string, any> = {};\n\n  /**\n   * @default \"any\"\n   */\n  _configurableFields: string[] | \"any\" = \"any\";\n\n  /**\n   * @default \"\"\n   */\n  _configPrefix: string;\n\n  /**\n   * Methods which should be called after the model is initialized.\n   * The key will be the method name, and the value will be the arguments.\n   */\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  _queuedMethodOperations: Record<string, any> = {};\n\n  constructor(fields: ConfigurableModelFields) {\n    super(fields);\n    this._defaultConfig = fields.defaultConfig ?? {};\n\n    if (fields.configurableFields === \"any\") {\n      this._configurableFields = \"any\";\n    } else {\n      this._configurableFields = fields.configurableFields ?? [\n        \"model\",\n        \"modelProvider\",\n      ];\n    }\n\n    if (fields.configPrefix) {\n      this._configPrefix = fields.configPrefix.endsWith(\"_\")\n        ? fields.configPrefix\n        : `${fields.configPrefix}_`;\n    } else {\n      this._configPrefix = \"\";\n    }\n\n    this._queuedMethodOperations =\n      fields.queuedMethodOperations ?? this._queuedMethodOperations;\n  }\n\n  async _model(\n    config?: RunnableConfig\n  ): Promise<\n    BaseChatModel<BaseChatModelCallOptions, AIMessageChunk<MessageStructure>>\n  > {\n    const params = { ...this._defaultConfig, ...this._modelParams(config) };\n    let initializedModel = await _initChatModelHelper(\n      params.model,\n      params.modelProvider,\n      params\n    );\n\n    // Apply queued method operations\n    const queuedMethodOperationsEntries = Object.entries(\n      this._queuedMethodOperations\n    );\n    if (queuedMethodOperationsEntries.length > 0) {\n      for (const [method, args] of queuedMethodOperationsEntries) {\n        if (\n          method in initializedModel &&\n          // eslint-disable-next-line @typescript-eslint/no-explicit-any\n          typeof (initializedModel as any)[method] === \"function\"\n        ) {\n          // eslint-disable-next-line @typescript-eslint/no-explicit-any\n          initializedModel = await (initializedModel as any)[method](...args);\n        }\n      }\n    }\n\n    return initializedModel;\n  }\n\n  async _generate(\n    messages: BaseMessage[],\n    options?: this[\"ParsedCallOptions\"],\n    runManager?: CallbackManagerForLLMRun\n  ): Promise<ChatResult> {\n    const model = await this._model(options);\n    return model._generate(messages, options ?? {}, runManager);\n  }\n\n  override bindTools(\n    tools: BindToolsInput[],\n    // eslint-disable-next-line @typescript-eslint/no-explicit-any\n    params?: Record<string, any>\n  ): ConfigurableModel<RunInput, CallOptions> {\n    const newQueuedOperations = { ...this._queuedMethodOperations };\n    newQueuedOperations.bindTools = [tools, params];\n    return new ConfigurableModel<RunInput, CallOptions>({\n      defaultConfig: this._defaultConfig,\n      configurableFields: this._configurableFields,\n      configPrefix: this._configPrefix,\n      queuedMethodOperations: newQueuedOperations,\n    });\n  }\n\n  // Extract the input types from the `BaseModel` class.\n  withStructuredOutput: BaseChatModel[\"withStructuredOutput\"] = (\n    schema,\n    ...args\n  ): ReturnType<BaseChatModel[\"withStructuredOutput\"]> => {\n    const newQueuedOperations = { ...this._queuedMethodOperations };\n    newQueuedOperations.withStructuredOutput = [schema, ...args];\n    return new ConfigurableModel<RunInput, CallOptions>({\n      defaultConfig: this._defaultConfig,\n      configurableFields: this._configurableFields,\n      configPrefix: this._configPrefix,\n      queuedMethodOperations: newQueuedOperations,\n    }) as unknown as ReturnType<BaseChatModel[\"withStructuredOutput\"]>;\n  };\n\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  _modelParams(config?: RunnableConfig): Record<string, any> {\n    const configurable = config?.configurable ?? {};\n    // eslint-disable-next-line @typescript-eslint/no-explicit-any\n    let modelParams: Record<string, any> = {};\n\n    for (const [key, value] of Object.entries(configurable)) {\n      if (key.startsWith(this._configPrefix)) {\n        const strippedKey = this._removePrefix(key, this._configPrefix);\n        modelParams[strippedKey] = value;\n      }\n    }\n\n    if (this._configurableFields !== \"any\") {\n      modelParams = Object.fromEntries(\n        Object.entries(modelParams).filter(([key]) =>\n          this._configurableFields.includes(key)\n        )\n      );\n    }\n\n    return modelParams;\n  }\n\n  _removePrefix(str: string, prefix: string): string {\n    return str.startsWith(prefix) ? str.slice(prefix.length) : str;\n  }\n\n  /**\n   * Bind config to a Runnable, returning a new Runnable.\n   * @param {RunnableConfig | undefined} [config] - The config to bind.\n   * @returns {RunnableBinding<RunInput, RunOutput, CallOptions>} A new RunnableBinding with the bound config.\n   */\n  withConfig(\n    config?: RunnableConfig\n  ): RunnableBinding<RunInput, AIMessageChunk, CallOptions> {\n    const mergedConfig: RunnableConfig = { ...(config || {}) };\n    const modelParams = this._modelParams(mergedConfig);\n\n    const remainingConfig: RunnableConfig = Object.fromEntries(\n      Object.entries(mergedConfig).filter(([k]) => k !== \"configurable\")\n    );\n\n    remainingConfig.configurable = Object.fromEntries(\n      Object.entries(mergedConfig.configurable || {}).filter(\n        ([k]) =>\n          this._configPrefix &&\n          !Object.keys(modelParams).includes(\n            this._removePrefix(k, this._configPrefix)\n          )\n      )\n    );\n\n    const newConfigurableModel = new ConfigurableModel<RunInput, CallOptions>({\n      defaultConfig: { ...this._defaultConfig, ...modelParams },\n      configurableFields: Array.isArray(this._configurableFields)\n        ? [...this._configurableFields]\n        : this._configurableFields,\n      configPrefix: this._configPrefix,\n      queuedMethodOperations: this._queuedMethodOperations,\n    });\n\n    return new RunnableBinding<RunInput, AIMessageChunk, CallOptions>({\n      config: mergedConfig,\n      bound: newConfigurableModel,\n    });\n  }\n\n  async invoke(\n    input: RunInput,\n    options?: CallOptions\n  ): Promise<AIMessageChunk> {\n    const model = await this._model(options);\n    const config = ensureConfig(options);\n    return model.invoke(input, config);\n  }\n\n  async stream(\n    input: RunInput,\n    options?: CallOptions\n  ): Promise<IterableReadableStream<AIMessageChunk>> {\n    const model = await this._model(options);\n    const wrappedGenerator = new AsyncGeneratorWithSetup({\n      generator: await model.stream(input, options),\n      config: options,\n    });\n    await wrappedGenerator.setup;\n    return IterableReadableStream.fromAsyncGenerator(wrappedGenerator);\n  }\n\n  async batch(\n    inputs: RunInput[],\n    options?: Partial<CallOptions> | Partial<CallOptions>[],\n    batchOptions?: RunnableBatchOptions & { returnExceptions?: false }\n  ): Promise<AIMessageChunk[]>;\n\n  async batch(\n    inputs: RunInput[],\n    options?: Partial<CallOptions> | Partial<CallOptions>[],\n    batchOptions?: RunnableBatchOptions & { returnExceptions: true }\n  ): Promise<(AIMessageChunk | Error)[]>;\n\n  async batch(\n    inputs: RunInput[],\n    options?: Partial<CallOptions> | Partial<CallOptions>[],\n    batchOptions?: RunnableBatchOptions\n  ): Promise<(AIMessageChunk | Error)[]>;\n\n  async batch(\n    inputs: RunInput[],\n    options?: Partial<CallOptions> | Partial<CallOptions>[],\n    batchOptions?: RunnableBatchOptions\n  ): Promise<(AIMessageChunk | Error)[]> {\n    // We can super this since the base runnable implementation of\n    // `.batch` will call `.invoke` on each input.\n    return super.batch(inputs, options, batchOptions);\n  }\n\n  async *transform(\n    generator: AsyncGenerator<RunInput>,\n    options: CallOptions\n  ): AsyncGenerator<AIMessageChunk> {\n    const model = await this._model(options);\n    const config = ensureConfig(options);\n\n    yield* model.transform(generator, config);\n  }\n\n  async *streamLog(\n    input: RunInput,\n    options?: Partial<CallOptions>,\n    streamOptions?: Omit<LogStreamCallbackHandlerInput, \"autoClose\">\n  ): AsyncGenerator<RunLogPatch> {\n    const model = await this._model(options);\n    const config = ensureConfig(options);\n\n    yield* model.streamLog(input, config, {\n      ...streamOptions,\n      _schemaFormat: \"original\",\n      includeNames: streamOptions?.includeNames,\n      includeTypes: streamOptions?.includeTypes,\n      includeTags: streamOptions?.includeTags,\n      excludeNames: streamOptions?.excludeNames,\n      excludeTypes: streamOptions?.excludeTypes,\n      excludeTags: streamOptions?.excludeTags,\n    });\n  }\n\n  streamEvents(\n    input: RunInput,\n    options: Partial<CallOptions> & { version: \"v1\" | \"v2\" },\n    streamOptions?: Omit<EventStreamCallbackHandlerInput, \"autoClose\">\n  ): IterableReadableStream<StreamEvent>;\n\n  streamEvents(\n    input: RunInput,\n    options: Partial<CallOptions> & {\n      version: \"v1\" | \"v2\";\n      encoding: \"text/event-stream\";\n    },\n    streamOptions?: Omit<EventStreamCallbackHandlerInput, \"autoClose\">\n  ): IterableReadableStream<Uint8Array>;\n\n  streamEvents(\n    input: RunInput,\n    options: Partial<CallOptions> & {\n      version: \"v1\" | \"v2\";\n      encoding?: \"text/event-stream\" | undefined;\n    },\n    streamOptions?: Omit<EventStreamCallbackHandlerInput, \"autoClose\">\n  ): IterableReadableStream<StreamEvent | Uint8Array> {\n    const outerThis = this;\n    async function* wrappedGenerator() {\n      const model = await outerThis._model(options);\n      const config = ensureConfig(options);\n      const eventStream = model.streamEvents(input, config, streamOptions);\n\n      for await (const chunk of eventStream) {\n        yield chunk;\n      }\n    }\n    return IterableReadableStream.fromAsyncGenerator(wrappedGenerator());\n  }\n}\n\n// eslint-disable-next-line @typescript-eslint/no-explicit-any\nexport interface InitChatModelFields extends Partial<Record<string, any>> {\n  modelProvider?: string;\n  configurableFields?: string[] | \"any\";\n  configPrefix?: string;\n}\n\nexport type ConfigurableFields = \"any\" | string[];\n\nexport async function initChatModel<\n  RunInput extends BaseLanguageModelInput = BaseLanguageModelInput,\n  CallOptions extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions\n>(\n  model: string,\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  fields?: Partial<Record<string, any>> & {\n    modelProvider?: string;\n    configurableFields?: never;\n    configPrefix?: string;\n  }\n): Promise<ConfigurableModel<RunInput, CallOptions>>;\n\nexport async function initChatModel<\n  RunInput extends BaseLanguageModelInput = BaseLanguageModelInput,\n  CallOptions extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions\n>(\n  model: never,\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  options?: Partial<Record<string, any>> & {\n    modelProvider?: string;\n    configurableFields?: never;\n    configPrefix?: string;\n  }\n): Promise<ConfigurableModel<RunInput, CallOptions>>;\n\nexport async function initChatModel<\n  RunInput extends BaseLanguageModelInput = BaseLanguageModelInput,\n  CallOptions extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions\n>(\n  model?: string,\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  options?: Partial<Record<string, any>> & {\n    modelProvider?: string;\n    configurableFields?: ConfigurableFields;\n    configPrefix?: string;\n  }\n): Promise<ConfigurableModel<RunInput, CallOptions>>;\n\n// ################################# FOR CONTRIBUTORS #################################\n//\n// If adding support for a new provider, please append the provider\n// name to the supported list in the docstring below.\n//\n// ####################################################################################\n\n/**\n * Initialize a ChatModel from the model name and provider.\n * Must have the integration package corresponding to the model provider installed.\n *\n * @template {extends BaseLanguageModelInput = BaseLanguageModelInput} RunInput - The input type for the model.\n * @template {extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions} CallOptions - Call options for the model.\n *\n * @param {string | ChatModelProvider} [model] - The name of the model, e.g. \"gpt-4\", \"claude-3-opus-20240229\".\n *   Can be prefixed with the model provider, e.g. \"openai:gpt-4\", \"anthropic:claude-3-opus-20240229\".\n * @param {Object} [fields] - Additional configuration options.\n * @param {string} [fields.modelProvider] - The model provider. Supported values include:\n *   - openai (@langchain/openai)\n *   - anthropic (@langchain/anthropic)\n *   - azure_openai (@langchain/openai)\n *   - google-vertexai (@langchain/google-vertexai)\n *   - google-vertexai-web (@langchain/google-vertexai-web)\n *   - google-genai (@langchain/google-genai)\n *   - bedrock (@langchain/aws)\n *   - cohere (@langchain/cohere)\n *   - fireworks (@langchain/community/chat_models/fireworks)\n *   - together (@langchain/community/chat_models/togetherai)\n *   - mistralai (@langchain/mistralai)\n *   - groq (@langchain/groq)\n *   - ollama (@langchain/ollama)\n *   - cerebras (@langchain/cerebras)\n *   - deepseek (@langchain/deepseek)\n *   - xai (@langchain/xai)\n * @param {string[] | \"any\"} [fields.configurableFields] - Which model parameters are configurable:\n *   - undefined: No configurable fields.\n *   - \"any\": All fields are configurable. (See Security Note in description)\n *   - string[]: Specified fields are configurable.\n * @param {string} [fields.configPrefix] - Prefix for configurable fields at runtime.\n * @param {Record<string, any>} [fields.params] - Additional keyword args to pass to the ChatModel constructor.\n * @returns {Promise<ConfigurableModel<RunInput, CallOptions>>} A class which extends BaseChatModel.\n * @throws {Error} If modelProvider cannot be inferred or isn't supported.\n * @throws {Error} If the model provider integration package is not installed.\n *\n * @example Initialize non-configurable models\n * ```typescript\n * import { initChatModel } from \"langchain/chat_models/universal\";\n *\n * const gpt4 = await initChatModel(\"openai:gpt-4\", {\n *   temperature: 0.25,\n * });\n * const gpt4Result = await gpt4.invoke(\"what's your name\");\n *\n * const claude = await initChatModel(\"anthropic:claude-3-opus-20240229\", {\n *   temperature: 0.25,\n * });\n * const claudeResult = await claude.invoke(\"what's your name\");\n *\n * const gemini = await initChatModel(\"gemini-1.5-pro\", {\n *   modelProvider: \"google-vertexai\",\n *   temperature: 0.25,\n * });\n * const geminiResult = await gemini.invoke(\"what's your name\");\n * ```\n *\n * @example Create a partially configurable model with no default model\n * ```typescript\n * import { initChatModel } from \"langchain/chat_models/universal\";\n *\n * const configurableModel = await initChatModel(undefined, {\n *   temperature: 0,\n *   configurableFields: [\"model\", \"apiKey\"],\n * });\n *\n * const gpt4Result = await configurableModel.invoke(\"what's your name\", {\n *   configurable: {\n *     model: \"gpt-4\",\n *   },\n * });\n *\n * const claudeResult = await configurableModel.invoke(\"what's your name\", {\n *   configurable: {\n *     model: \"claude-3-5-sonnet-20240620\",\n *   },\n * });\n * ```\n *\n * @example Create a fully configurable model with a default model and a config prefix\n * ```typescript\n * import { initChatModel } from \"langchain/chat_models/universal\";\n *\n * const configurableModelWithDefault = await initChatModel(\"gpt-4\", {\n *   modelProvider: \"openai\",\n *   configurableFields: \"any\",\n *   configPrefix: \"foo\",\n *   temperature: 0,\n * });\n *\n * const openaiResult = await configurableModelWithDefault.invoke(\n *   \"what's your name\",\n *   {\n *     configurable: {\n *       foo_apiKey: process.env.OPENAI_API_KEY,\n *     },\n *   }\n * );\n *\n * const claudeResult = await configurableModelWithDefault.invoke(\n *   \"what's your name\",\n *   {\n *     configurable: {\n *       foo_model: \"claude-3-5-sonnet-20240620\",\n *       foo_modelProvider: \"anthropic\",\n *       foo_temperature: 0.6,\n *       foo_apiKey: process.env.ANTHROPIC_API_KEY,\n *     },\n *   }\n * );\n * ```\n *\n * @example Bind tools to a configurable model:\n * ```typescript\n * import { initChatModel } from \"langchain/chat_models/universal\";\n * import { z } from \"zod/v3\";\n * import { tool } from \"@langchain/core/tools\";\n *\n * const getWeatherTool = tool(\n *   (input) => {\n *     // Do something with the input\n *     return JSON.stringify(input);\n *   },\n *   {\n *     schema: z\n *       .object({\n *         location: z\n *           .string()\n *           .describe(\"The city and state, e.g. San Francisco, CA\"),\n *       })\n *       .describe(\"Get the current weather in a given location\"),\n *     name: \"GetWeather\",\n *     description: \"Get the current weather in a given location\",\n *   }\n * );\n *\n * const getPopulationTool = tool(\n *   (input) => {\n *     // Do something with the input\n *     return JSON.stringify(input);\n *   },\n *   {\n *     schema: z\n *       .object({\n *         location: z\n *           .string()\n *           .describe(\"The city and state, e.g. San Francisco, CA\"),\n *       })\n *       .describe(\"Get the current population in a given location\"),\n *     name: \"GetPopulation\",\n *     description: \"Get the current population in a given location\",\n *   }\n * );\n *\n * const configurableModel = await initChatModel(\"gpt-4\", {\n *   configurableFields: [\"model\", \"modelProvider\", \"apiKey\"],\n *   temperature: 0,\n * });\n *\n * const configurableModelWithTools = configurableModel.bindTools([\n *   getWeatherTool,\n *   getPopulationTool,\n * ]);\n *\n * const configurableToolResult = await configurableModelWithTools.invoke(\n *   \"Which city is hotter today and which is bigger: LA or NY?\",\n *   {\n *     configurable: {\n *       apiKey: process.env.OPENAI_API_KEY,\n *     },\n *   }\n * );\n *\n * const configurableToolResult2 = await configurableModelWithTools.invoke(\n *   \"Which city is hotter today and which is bigger: LA or NY?\",\n *   {\n *     configurable: {\n *       model: \"claude-3-5-sonnet-20240620\",\n *       apiKey: process.env.ANTHROPIC_API_KEY,\n *     },\n *   }\n * );\n * ```\n *\n * @description\n * This function initializes a ChatModel based on the provided model name and provider.\n * It supports various model providers and allows for runtime configuration of model parameters.\n *\n * Security Note: Setting `configurableFields` to \"any\" means fields like apiKey, baseUrl, etc.\n * can be altered at runtime, potentially redirecting model requests to a different service/user.\n * Make sure that if you're accepting untrusted configurations, you enumerate the\n * `configurableFields` explicitly.\n *\n * The function will attempt to infer the model provider from the model name if not specified.\n * Certain model name prefixes are associated with specific providers:\n * - gpt-3... or gpt-4... -> openai\n * - claude... -> anthropic\n * - amazon.... -> bedrock\n * - gemini... -> google-vertexai\n * - command... -> cohere\n * - accounts/fireworks... -> fireworks\n *\n * @since 0.2.11\n * @version 0.2.11\n */\nexport async function initChatModel<\n  RunInput extends BaseLanguageModelInput = BaseLanguageModelInput,\n  CallOptions extends ConfigurableChatModelCallOptions = ConfigurableChatModelCallOptions\n>(\n  model?: string,\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  fields?: Partial<Record<string, any>> & {\n    modelProvider?: string;\n    configurableFields?: string[] | \"any\";\n    configPrefix?: string;\n  }\n): Promise<ConfigurableModel<RunInput, CallOptions>> {\n  // eslint-disable-next-line prefer-const\n  let { configurableFields, configPrefix, modelProvider, ...params } = {\n    configPrefix: \"\",\n    ...(fields ?? {}),\n  };\n  if (modelProvider === undefined && model?.includes(\":\")) {\n    const modelComponents = model.split(\":\", 2);\n    if (SUPPORTED_PROVIDERS.includes(modelComponents[0] as ChatModelProvider)) {\n      // eslint-disable-next-line no-param-reassign\n      [modelProvider, model] = modelComponents;\n    }\n  }\n  let configurableFieldsCopy = Array.isArray(configurableFields)\n    ? [...configurableFields]\n    : configurableFields;\n\n  if (!model && configurableFieldsCopy === undefined) {\n    configurableFieldsCopy = [\"model\", \"modelProvider\"];\n  }\n  if (configPrefix && configurableFieldsCopy === undefined) {\n    console.warn(\n      `{ configPrefix: ${configPrefix} } has been set but no fields are configurable. Set ` +\n        `{ configurableFields: [...] } to specify the model params that are ` +\n        `configurable.`\n    );\n  }\n\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  const paramsCopy: Record<string, any> = { ...params };\n\n  if (configurableFieldsCopy === undefined) {\n    return new ConfigurableModel<RunInput, CallOptions>({\n      defaultConfig: {\n        ...paramsCopy,\n        model,\n        modelProvider,\n      },\n      configPrefix,\n    });\n  } else {\n    if (model) {\n      paramsCopy.model = model;\n    }\n    if (modelProvider) {\n      paramsCopy.modelProvider = modelProvider;\n    }\n    return new ConfigurableModel<RunInput, CallOptions>({\n      defaultConfig: paramsCopy,\n      configPrefix,\n      configurableFields: configurableFieldsCopy,\n    });\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;AAkDA,MAAa,wBAAwB;CACnC,QAAQ;EACN,SAAS;EACT,WAAW;CACZ;CACD,WAAW;EACT,SAAS;EACT,WAAW;CACZ;CACD,cAAc;EACZ,SAAS;EACT,WAAW;CACZ;CACD,QAAQ;EACN,SAAS;EACT,WAAW;CACZ;CACD,mBAAmB;EACjB,SAAS;EACT,WAAW;CACZ;CACD,uBAAuB;EACrB,SAAS;EACT,WAAW;CACZ;CACD,gBAAgB;EACd,SAAS;EACT,WAAW;CACZ;CACD,QAAQ;EACN,SAAS;EACT,WAAW;CACZ;CACD,WAAW;EACT,SAAS;EACT,WAAW;CACZ;CACD,MAAM;EACJ,SAAS;EACT,WAAW;CACZ;CACD,UAAU;EACR,SAAS;EACT,WAAW;CACZ;CACD,SAAS;EACP,SAAS;EACT,WAAW;CACZ;CACD,UAAU;EACR,SAAS;EACT,WAAW;CACZ;CACD,KAAK;EACH,SAAS;EACT,WAAW;CACZ;CACD,WAAW;EACT,SAAS;EACT,WAAW;EACX,uBAAuB;CACxB;CACD,UAAU;EACR,SAAS;EACT,WAAW;EACX,uBAAuB;CACxB;AACF;AAED,MAAM,sBAAsB,OAAO,KACjC,sBACD;;;;;;AAaD,eAAsB,wBAAwBA,WAAmB;CAE/D,MAAM,gBAAgB,OAAO,QAAQ,sBAAsB,CAAC,KAC1D,CAAC,GAAGC,SAAO,KAAKA,SAAO,cAAc,UACtC;AAED,KAAI,CAAC,cACH,QAAO;CAGT,MAAM,GAAG,OAAO,GAAG;AACnB,KAAI;EACF,MAAMC,WAAS,MAAM,OAAO,OAAO;AACnC,SAAOA,SAAO,OAAO;CACtB,SAAQC,GAAY;EACnB,MAAM,MAAM;AACZ,MACE,UAAU,OACV,IAAI,MAAM,UAAU,CAAC,SAAS,uBAAuB,EACrD;GACA,MAAM,mBAAmB,IAAI,QAC1B,MAAM,+BAA+B,CAAC,GACtC,MAAM,IAAI,CAAC;AACd,SAAM,IAAI,MACR,CAAC,iBAAiB,EAAE,iBAAiB,oCAAsB,EACxC,iBAAiB,qBAAqB,EAAE,iBAAiB,EAAE,CAAC;EAElF;AACD,QAAM;CACP;AACF;AAED,eAAe,qBACbC,OACAC,eAEAC,SAA8B,CAAE,GACR;CACxB,MAAM,oBAAoB,iBAAiB,oBAAoB,MAAM;AACrE,KAAI,CAAC,kBACH,OAAM,IAAI,MACR,CAAC,4CAA4C,EAAE,MAAM,0CAA0C,CAAC;CAIpG,MAAM,SAAS,sBACb;AAEF,KAAI,CAAC,QAAQ;EACX,MAAM,YAAY,oBAAoB,KAAK,KAAK;AAChD,QAAM,IAAI,MACR,CAAC,6BAA6B,EAAE,kBAAkB,sCAAsC,EAAE,WAAW;CAExG;CAED,MAAM,EAAE,eAAe,QAAS,GAAG,cAAc,GAAG;CACpD,MAAM,gBAAgB,MAAM,wBAAwB,OAAO,UAAU;AACrE,QAAO,IAAI,cAAc;EAAE;EAAO,GAAG;CAAc;AACpD;;;;;;;;;;;;AAaD,SAAgB,oBAAoBC,WAAuC;AACzE,KACE,UAAU,WAAW,QAAQ,IAC7B,UAAU,WAAW,QAAQ,IAC7B,UAAU,WAAW,KAAK,IAC1B,UAAU,WAAW,KAAK,IAC1B,UAAU,WAAW,KAAK,CAE1B,QAAO;UACE,UAAU,WAAW,SAAS,CACvC,QAAO;UACE,UAAU,WAAW,UAAU,CACxC,QAAO;UACE,UAAU,WAAW,qBAAqB,CACnD,QAAO;UACE,UAAU,WAAW,SAAS,CACvC,QAAO;UACE,UAAU,WAAW,UAAU,CACxC,QAAO;UACE,UAAU,WAAW,UAAU,CACxC,QAAO;KAEP,QAAO;AAEV;;;;;;AA0BD,IAAa,oBAAb,MAAa,0BAGHC,2DAA2C;CACnD,WAAmB;AACjB,SAAO;CACR;CAED,eAAe,CAAC,aAAa,aAAc;CAG3C,iBAAuC,CAAE;;;;CAKzC,sBAAwC;;;;CAKxC;;;;;CAOA,0BAA+C,CAAE;CAEjD,YAAYC,QAAiC;EAC3C,MAAM,OAAO;EACb,KAAK,iBAAiB,OAAO,iBAAiB,CAAE;AAEhD,MAAI,OAAO,uBAAuB,OAChC,KAAK,sBAAsB;OAE3B,KAAK,sBAAsB,OAAO,sBAAsB,CACtD,SACA,eACD;AAGH,MAAI,OAAO,cACT,KAAK,gBAAgB,OAAO,aAAa,SAAS,IAAI,GAClD,OAAO,eACP,GAAG,OAAO,aAAa,CAAC,CAAC;OAE7B,KAAK,gBAAgB;EAGvB,KAAK,0BACH,OAAO,0BAA0B,KAAK;CACzC;CAED,MAAM,OACJC,QAGA;EACA,MAAM,SAAS;GAAE,GAAG,KAAK;GAAgB,GAAG,KAAK,aAAa,OAAO;EAAE;EACvE,IAAI,mBAAmB,MAAM,qBAC3B,OAAO,OACP,OAAO,eACP,OACD;EAGD,MAAM,gCAAgC,OAAO,QAC3C,KAAK,wBACN;AACD,MAAI,8BAA8B,SAAS,GACzC;QAAK,MAAM,CAAC,QAAQ,KAAK,IAAI,8BAC3B,KACE,UAAU,oBAEV,OAAQ,iBAAyB,YAAY,YAG7C,mBAAmB,MAAO,iBAAyB,QAAQ,GAAG,KAAK;EAEtE;AAGH,SAAO;CACR;CAED,MAAM,UACJC,UACAC,SACAC,YACqB;EACrB,MAAM,QAAQ,MAAM,KAAK,OAAO,QAAQ;AACxC,SAAO,MAAM,UAAU,UAAU,WAAW,CAAE,GAAE,WAAW;CAC5D;CAED,AAAS,UACPC,OAEAC,QAC0C;EAC1C,MAAM,sBAAsB,EAAE,GAAG,KAAK,wBAAyB;EAC/D,oBAAoB,YAAY,CAAC,OAAO,MAAO;AAC/C,SAAO,IAAI,kBAAyC;GAClD,eAAe,KAAK;GACpB,oBAAoB,KAAK;GACzB,cAAc,KAAK;GACnB,wBAAwB;EACzB;CACF;CAGD,uBAA8D,CAC5D,QACA,GAAG,SACmD;EACtD,MAAM,sBAAsB,EAAE,GAAG,KAAK,wBAAyB;EAC/D,oBAAoB,uBAAuB,CAAC,QAAQ,GAAG,IAAK;AAC5D,SAAO,IAAI,kBAAyC;GAClD,eAAe,KAAK;GACpB,oBAAoB,KAAK;GACzB,cAAc,KAAK;GACnB,wBAAwB;EACzB;CACF;CAGD,aAAaL,QAA8C;EACzD,MAAM,eAAe,QAAQ,gBAAgB,CAAE;EAE/C,IAAIM,cAAmC,CAAE;AAEzC,OAAK,MAAM,CAAC,KAAK,MAAM,IAAI,OAAO,QAAQ,aAAa,CACrD,KAAI,IAAI,WAAW,KAAK,cAAc,EAAE;GACtC,MAAM,cAAc,KAAK,cAAc,KAAK,KAAK,cAAc;GAC/D,YAAY,eAAe;EAC5B;AAGH,MAAI,KAAK,wBAAwB,OAC/B,cAAc,OAAO,YACnB,OAAO,QAAQ,YAAY,CAAC,OAAO,CAAC,CAAC,IAAI,KACvC,KAAK,oBAAoB,SAAS,IAAI,CACvC,CACF;AAGH,SAAO;CACR;CAED,cAAcC,KAAaC,QAAwB;AACjD,SAAO,IAAI,WAAW,OAAO,GAAG,IAAI,MAAM,OAAO,OAAO,GAAG;CAC5D;;;;;;CAOD,WACER,QACwD;EACxD,MAAMS,eAA+B,EAAE,GAAI,UAAU,CAAE,EAAG;EAC1D,MAAM,cAAc,KAAK,aAAa,aAAa;EAEnD,MAAMC,kBAAkC,OAAO,YAC7C,OAAO,QAAQ,aAAa,CAAC,OAAO,CAAC,CAAC,EAAE,KAAK,MAAM,eAAe,CACnE;EAED,gBAAgB,eAAe,OAAO,YACpC,OAAO,QAAQ,aAAa,gBAAgB,CAAE,EAAC,CAAC,OAC9C,CAAC,CAAC,EAAE,KACF,KAAK,iBACL,CAAC,OAAO,KAAK,YAAY,CAAC,SACxB,KAAK,cAAc,GAAG,KAAK,cAAc,CAC1C,CACJ,CACF;EAED,MAAM,uBAAuB,IAAI,kBAAyC;GACxE,eAAe;IAAE,GAAG,KAAK;IAAgB,GAAG;GAAa;GACzD,oBAAoB,MAAM,QAAQ,KAAK,oBAAoB,GACvD,CAAC,GAAG,KAAK,mBAAoB,IAC7B,KAAK;GACT,cAAc,KAAK;GACnB,wBAAwB,KAAK;EAC9B;AAED,SAAO,IAAIC,2CAAuD;GAChE,QAAQ;GACR,OAAO;EACR;CACF;CAED,MAAM,OACJC,OACAC,SACyB;EACzB,MAAM,QAAQ,MAAM,KAAK,OAAO,QAAQ;EACxC,MAAM,sDAAsB,QAAQ;AACpC,SAAO,MAAM,OAAO,OAAO,OAAO;CACnC;CAED,MAAM,OACJD,OACAC,SACiD;EACjD,MAAM,QAAQ,MAAM,KAAK,OAAO,QAAQ;EACxC,MAAM,mBAAmB,IAAIC,sDAAwB;GACnD,WAAW,MAAM,MAAM,OAAO,OAAO,QAAQ;GAC7C,QAAQ;EACT;EACD,MAAM,iBAAiB;AACvB,SAAOC,qDAAuB,mBAAmB,iBAAiB;CACnE;CAoBD,MAAM,MACJC,QACAC,SACAC,cACqC;AAGrC,SAAO,MAAM,MAAM,QAAQ,SAAS,aAAa;CAClD;CAED,OAAO,UACLC,WACAC,SACgC;EAChC,MAAM,QAAQ,MAAM,KAAK,OAAO,QAAQ;EACxC,MAAM,sDAAsB,QAAQ;EAEpC,OAAO,MAAM,UAAU,WAAW,OAAO;CAC1C;CAED,OAAO,UACLR,OACAS,SACAC,eAC6B;EAC7B,MAAM,QAAQ,MAAM,KAAK,OAAO,QAAQ;EACxC,MAAM,sDAAsB,QAAQ;EAEpC,OAAO,MAAM,UAAU,OAAO,QAAQ;GACpC,GAAG;GACH,eAAe;GACf,cAAc,eAAe;GAC7B,cAAc,eAAe;GAC7B,aAAa,eAAe;GAC5B,cAAc,eAAe;GAC7B,cAAc,eAAe;GAC7B,aAAa,eAAe;EAC7B,EAAC;CACH;CAiBD,aACEV,OACAW,SAIAC,eACkD;EAClD,MAAM,YAAY;EAClB,gBAAgB,mBAAmB;GACjC,MAAM,QAAQ,MAAM,UAAU,OAAO,QAAQ;GAC7C,MAAM,sDAAsB,QAAQ;GACpC,MAAM,cAAc,MAAM,aAAa,OAAO,QAAQ,cAAc;AAEpE,cAAW,MAAM,SAAS,aACxB,MAAM;EAET;AACD,SAAOT,qDAAuB,mBAAmB,kBAAkB,CAAC;CACrE;AACF;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAuQD,eAAsB,cAIpBU,OAEAC,QAKmD;CAEnD,IAAI,EAAE,oBAAoB,cAAc,cAAe,GAAG,QAAQ,GAAG;EACnE,cAAc;EACd,GAAI,UAAU,CAAE;CACjB;AACD,KAAI,kBAAkB,UAAa,OAAO,SAAS,IAAI,EAAE;EACvD,MAAM,kBAAkB,MAAM,MAAM,KAAK,EAAE;AAC3C,MAAI,oBAAoB,SAAS,gBAAgB,GAAwB,EAEvE,CAAC,eAAe,MAAM,GAAG;CAE5B;CACD,IAAI,yBAAyB,MAAM,QAAQ,mBAAmB,GAC1D,CAAC,GAAG,kBAAmB,IACvB;AAEJ,KAAI,CAAC,SAAS,2BAA2B,QACvC,yBAAyB,CAAC,SAAS,eAAgB;AAErD,KAAI,gBAAgB,2BAA2B,QAC7C,QAAQ,KACN,CAAC,gBAAgB,EAAE,aAAa,oIAAoD,CAEnE,CAClB;CAIH,MAAMC,aAAkC,EAAE,GAAG,OAAQ;AAErD,KAAI,2BAA2B,OAC7B,QAAO,IAAI,kBAAyC;EAClD,eAAe;GACb,GAAG;GACH;GACA;EACD;EACD;CACD;MACI;AACL,MAAI,OACF,WAAW,QAAQ;AAErB,MAAI,eACF,WAAW,gBAAgB;AAE7B,SAAO,IAAI,kBAAyC;GAClD,eAAe;GACf;GACA,oBAAoB;EACrB;CACF;AACF"}