langchain 0.1.35 → 0.1.37

package/dist/chains/conversational_retrieval_chain.cjs

@@ -13,37 +13,79 @@ Chat History:
 Follow Up Input: {question}
 Standalone question:`;
 /**
+ * @deprecated This class will be removed in 0.3.0. See below for an example implementation using
+ * `createRetrievalChain`.
+ *
  * Class for conducting conversational question-answering tasks with a
  * retrieval component. Extends the BaseChain class and implements the
  * ConversationalRetrievalQAChainInput interface.
  * @example
  * ```typescript
- * const model = new ChatAnthropic({});
- *
- * const text = fs.readFileSync("state_of_the_union.txt", "utf8");
- *
- * const textSplitter = new RecursiveCharacterTextSplitter({ chunkSize: 1000 });
- * const docs = await textSplitter.createDocuments([text]);
+ * import { ChatAnthropic } from "@langchain/anthropic";
+ * import {
+ *   ChatPromptTemplate,
+ *   MessagesPlaceholder,
+ * } from "@langchain/core/prompts";
+ * import { BaseMessage } from "@langchain/core/messages";
+ * import { createStuffDocumentsChain } from "langchain/chains/combine_documents";
+ * import { createHistoryAwareRetriever } from "langchain/chains/history_aware_retriever";
+ * import { createRetrievalChain } from "langchain/chains/retrieval";
  *
- * const vectorStore = await HNSWLib.fromDocuments(docs, new OpenAIEmbeddings());
+ * const retriever = ...your retriever;
+ * const llm = new ChatAnthropic();
  *
- * const chain = ConversationalRetrievalQAChain.fromLLM(
- *   model,
- *   vectorStore.asRetriever(),
- * );
+ * // Contextualize question
+ * const contextualizeQSystemPrompt = `
+ * Given a chat history and the latest user question
+ * which might reference context in the chat history,
+ * formulate a standalone question which can be understood
+ * without the chat history. Do NOT answer the question, just
+ * reformulate it if needed and otherwise return it as is.`;
+ * const contextualizeQPrompt = ChatPromptTemplate.fromMessages([
+ *   ["system", contextualizeQSystemPrompt],
+ *   new MessagesPlaceholder("chat_history"),
+ *   ["human", "{input}"],
+ * ]);
+ * const historyAwareRetriever = await createHistoryAwareRetriever({
+ *   llm,
+ *   retriever,
+ *   rephrasePrompt: contextualizeQPrompt,
+ * });
  *
- * const question = "What did the president say about Justice Breyer?";
+ * // Answer question
+ * const qaSystemPrompt = `
+ * You are an assistant for question-answering tasks. Use
+ * the following pieces of retrieved context to answer the
+ * question. If you don't know the answer, just say that you
+ * don't know. Use three sentences maximum and keep the answer
+ * concise.
+ * \n\n
+ * {context}`;
+ * const qaPrompt = ChatPromptTemplate.fromMessages([
+ *   ["system", qaSystemPrompt],
+ *   new MessagesPlaceholder("chat_history"),
+ *   ["human", "{input}"],
+ * ]);
  *
- * const res = await chain.call({ question, chat_history: "" });
- * console.log(res);
+ * // Below we use createStuffDocuments_chain to feed all retrieved context
+ * // into the LLM. Note that we can also use StuffDocumentsChain and other
+ * // instances of BaseCombineDocumentsChain.
+ * const questionAnswerChain = await createStuffDocumentsChain({
+ *   llm,
+ *   prompt: qaPrompt,
+ * });
  *
- * const chatHistory = `${question}\n${res.text}`;
- * const followUpRes = await chain.call({
- *   question: "Was that nice?",
- *   chat_history: chatHistory,
+ * const ragChain = await createRetrievalChain({
+ *   retriever: historyAwareRetriever,
+ *   combineDocsChain: questionAnswerChain,
  * });
- * console.log(followUpRes);
  *
+ * // Usage:
+ * const chat_history: BaseMessage[] = [];
+ * const response = await ragChain.invoke({
+ *   chat_history,
+ *   input: "...",
+ * });
  * ```
  */
 class ConversationalRetrievalQAChain extends base_js_1.BaseChain {

package/dist/chains/conversational_retrieval_chain.d.ts

@@ -21,37 +21,79 @@ export interface ConversationalRetrievalQAChainInput extends ChainInputs {
     inputKey?: string;
 }
 /**
+ * @deprecated This class will be removed in 0.3.0. See below for an example implementation using
+ * `createRetrievalChain`.
+ *
  * Class for conducting conversational question-answering tasks with a
  * retrieval component. Extends the BaseChain class and implements the
  * ConversationalRetrievalQAChainInput interface.
  * @example
  * ```typescript
- * const model = new ChatAnthropic({});
- *
- * const text = fs.readFileSync("state_of_the_union.txt", "utf8");
- *
- * const textSplitter = new RecursiveCharacterTextSplitter({ chunkSize: 1000 });
- * const docs = await textSplitter.createDocuments([text]);
+ * import { ChatAnthropic } from "@langchain/anthropic";
+ * import {
+ *   ChatPromptTemplate,
+ *   MessagesPlaceholder,
+ * } from "@langchain/core/prompts";
+ * import { BaseMessage } from "@langchain/core/messages";
+ * import { createStuffDocumentsChain } from "langchain/chains/combine_documents";
+ * import { createHistoryAwareRetriever } from "langchain/chains/history_aware_retriever";
+ * import { createRetrievalChain } from "langchain/chains/retrieval";
  *
- * const vectorStore = await HNSWLib.fromDocuments(docs, new OpenAIEmbeddings());
+ * const retriever = ...your retriever;
+ * const llm = new ChatAnthropic();
  *
- * const chain = ConversationalRetrievalQAChain.fromLLM(
- *   model,
- *   vectorStore.asRetriever(),
- * );
+ * // Contextualize question
+ * const contextualizeQSystemPrompt = `
+ * Given a chat history and the latest user question
+ * which might reference context in the chat history,
+ * formulate a standalone question which can be understood
+ * without the chat history. Do NOT answer the question, just
+ * reformulate it if needed and otherwise return it as is.`;
+ * const contextualizeQPrompt = ChatPromptTemplate.fromMessages([
+ *   ["system", contextualizeQSystemPrompt],
+ *   new MessagesPlaceholder("chat_history"),
+ *   ["human", "{input}"],
+ * ]);
+ * const historyAwareRetriever = await createHistoryAwareRetriever({
+ *   llm,
+ *   retriever,
+ *   rephrasePrompt: contextualizeQPrompt,
+ * });
  *
- * const question = "What did the president say about Justice Breyer?";
+ * // Answer question
+ * const qaSystemPrompt = `
+ * You are an assistant for question-answering tasks. Use
+ * the following pieces of retrieved context to answer the
+ * question. If you don't know the answer, just say that you
+ * don't know. Use three sentences maximum and keep the answer
+ * concise.
+ * \n\n
+ * {context}`;
+ * const qaPrompt = ChatPromptTemplate.fromMessages([
+ *   ["system", qaSystemPrompt],
+ *   new MessagesPlaceholder("chat_history"),
+ *   ["human", "{input}"],
+ * ]);
  *
- * const res = await chain.call({ question, chat_history: "" });
- * console.log(res);
+ * // Below we use createStuffDocuments_chain to feed all retrieved context
+ * // into the LLM. Note that we can also use StuffDocumentsChain and other
+ * // instances of BaseCombineDocumentsChain.
+ * const questionAnswerChain = await createStuffDocumentsChain({
+ *   llm,
+ *   prompt: qaPrompt,
+ * });
  *
- * const chatHistory = `${question}\n${res.text}`;
- * const followUpRes = await chain.call({
- *   question: "Was that nice?",
- *   chat_history: chatHistory,
+ * const ragChain = await createRetrievalChain({
+ *   retriever: historyAwareRetriever,
+ *   combineDocsChain: questionAnswerChain,
  * });
- * console.log(followUpRes);
  *
+ * // Usage:
+ * const chat_history: BaseMessage[] = [];
+ * const response = await ragChain.invoke({
+ *   chat_history,
+ *   input: "...",
+ * });
  * ```
  */
 export declare class ConversationalRetrievalQAChain extends BaseChain implements ConversationalRetrievalQAChainInput {

package/dist/chains/conversational_retrieval_chain.js

@@ -10,37 +10,79 @@ Chat History:
 Follow Up Input: {question}
 Standalone question:`;
 /**
+ * @deprecated This class will be removed in 0.3.0. See below for an example implementation using
+ * `createRetrievalChain`.
+ *
  * Class for conducting conversational question-answering tasks with a
  * retrieval component. Extends the BaseChain class and implements the
  * ConversationalRetrievalQAChainInput interface.
  * @example
  * ```typescript
- * const model = new ChatAnthropic({});
- *
- * const text = fs.readFileSync("state_of_the_union.txt", "utf8");
- *
- * const textSplitter = new RecursiveCharacterTextSplitter({ chunkSize: 1000 });
- * const docs = await textSplitter.createDocuments([text]);
+ * import { ChatAnthropic } from "@langchain/anthropic";
+ * import {
+ *   ChatPromptTemplate,
+ *   MessagesPlaceholder,
+ * } from "@langchain/core/prompts";
+ * import { BaseMessage } from "@langchain/core/messages";
+ * import { createStuffDocumentsChain } from "langchain/chains/combine_documents";
+ * import { createHistoryAwareRetriever } from "langchain/chains/history_aware_retriever";
+ * import { createRetrievalChain } from "langchain/chains/retrieval";
  *
- * const vectorStore = await HNSWLib.fromDocuments(docs, new OpenAIEmbeddings());
+ * const retriever = ...your retriever;
+ * const llm = new ChatAnthropic();
  *
- * const chain = ConversationalRetrievalQAChain.fromLLM(
- *   model,
- *   vectorStore.asRetriever(),
- * );
+ * // Contextualize question
+ * const contextualizeQSystemPrompt = `
+ * Given a chat history and the latest user question
+ * which might reference context in the chat history,
+ * formulate a standalone question which can be understood
+ * without the chat history. Do NOT answer the question, just
+ * reformulate it if needed and otherwise return it as is.`;
+ * const contextualizeQPrompt = ChatPromptTemplate.fromMessages([
+ *   ["system", contextualizeQSystemPrompt],
+ *   new MessagesPlaceholder("chat_history"),
+ *   ["human", "{input}"],
+ * ]);
+ * const historyAwareRetriever = await createHistoryAwareRetriever({
+ *   llm,
+ *   retriever,
+ *   rephrasePrompt: contextualizeQPrompt,
+ * });
  *
- * const question = "What did the president say about Justice Breyer?";
+ * // Answer question
+ * const qaSystemPrompt = `
+ * You are an assistant for question-answering tasks. Use
+ * the following pieces of retrieved context to answer the
+ * question. If you don't know the answer, just say that you
+ * don't know. Use three sentences maximum and keep the answer
+ * concise.
+ * \n\n
+ * {context}`;
+ * const qaPrompt = ChatPromptTemplate.fromMessages([
+ *   ["system", qaSystemPrompt],
+ *   new MessagesPlaceholder("chat_history"),
+ *   ["human", "{input}"],
+ * ]);
  *
- * const res = await chain.call({ question, chat_history: "" });
- * console.log(res);
+ * // Below we use createStuffDocuments_chain to feed all retrieved context
+ * // into the LLM. Note that we can also use StuffDocumentsChain and other
+ * // instances of BaseCombineDocumentsChain.
+ * const questionAnswerChain = await createStuffDocumentsChain({
+ *   llm,
+ *   prompt: qaPrompt,
+ * });
  *
- * const chatHistory = `${question}\n${res.text}`;
- * const followUpRes = await chain.call({
- *   question: "Was that nice?",
- *   chat_history: chatHistory,
+ * const ragChain = await createRetrievalChain({
+ *   retriever: historyAwareRetriever,
+ *   combineDocsChain: questionAnswerChain,
  * });
- * console.log(followUpRes);
  *
+ * // Usage:
+ * const chat_history: BaseMessage[] = [];
+ * const response = await ragChain.invoke({
+ *   chat_history,
+ *   input: "...",
+ * });
  * ```
  */
 export class ConversationalRetrievalQAChain extends BaseChain {

package/dist/chains/llm_chain.cjs

@@ -29,16 +29,21 @@ function _getLanguageModel(llmLike) {
     }
 }
 /**
+ * @deprecated This class will be removed in 0.3.0. Use the LangChain Expression Language (LCEL) instead.
+ * See the example below for how to use LCEL with the LLMChain class:
+ *
  * Chain to run queries against LLMs.
  *
  * @example
  * ```ts
- * import { LLMChain } from "langchain/chains";
- * import { OpenAI } from "langchain/llms/openai";
- * import { PromptTemplate } from "langchain/prompts";
+ * import { ChatPromptTemplate } from "@langchain/core/prompts";
+ * import { ChatOpenAI } from "@langchain/openai";
+ *
+ * const prompt = ChatPromptTemplate.fromTemplate("Tell me a {adjective} joke");
+ * const llm = new ChatOpenAI();
+ * const chain = prompt.pipe(llm);
  *
- * const prompt = PromptTemplate.fromTemplate("Tell me a {adjective} joke");
- * const llm = new LLMChain({ llm: new OpenAI(), prompt });
+ * const response = await chain.invoke({ adjective: "funny" });
  * ```
  */
 class LLMChain extends base_js_1.BaseChain {

package/dist/chains/llm_chain.d.ts

@@ -29,16 +29,21 @@ export interface LLMChainInput<T extends string | object = string, Model extends
     outputKey?: string;
 }
 /**
+ * @deprecated This class will be removed in 0.3.0. Use the LangChain Expression Language (LCEL) instead.
+ * See the example below for how to use LCEL with the LLMChain class:
+ *
  * Chain to run queries against LLMs.
  *
  * @example
  * ```ts
- * import { LLMChain } from "langchain/chains";
- * import { OpenAI } from "langchain/llms/openai";
- * import { PromptTemplate } from "langchain/prompts";
+ * import { ChatPromptTemplate } from "@langchain/core/prompts";
+ * import { ChatOpenAI } from "@langchain/openai";
+ *
+ * const prompt = ChatPromptTemplate.fromTemplate("Tell me a {adjective} joke");
+ * const llm = new ChatOpenAI();
+ * const chain = prompt.pipe(llm);
  *
- * const prompt = PromptTemplate.fromTemplate("Tell me a {adjective} joke");
- * const llm = new LLMChain({ llm: new OpenAI(), prompt });
+ * const response = await chain.invoke({ adjective: "funny" });
  * ```
  */
 export declare class LLMChain<T extends string | object = string, Model extends LLMType = LLMType> extends BaseChain implements LLMChainInput<T> {

package/dist/chains/llm_chain.js

@@ -26,16 +26,21 @@ function _getLanguageModel(llmLike) {
     }
 }
 /**
+ * @deprecated This class will be removed in 0.3.0. Use the LangChain Expression Language (LCEL) instead.
+ * See the example below for how to use LCEL with the LLMChain class:
+ *
  * Chain to run queries against LLMs.
  *
  * @example
  * ```ts
- * import { LLMChain } from "langchain/chains";
- * import { OpenAI } from "langchain/llms/openai";
- * import { PromptTemplate } from "langchain/prompts";
+ * import { ChatPromptTemplate } from "@langchain/core/prompts";
+ * import { ChatOpenAI } from "@langchain/openai";
+ *
+ * const prompt = ChatPromptTemplate.fromTemplate("Tell me a {adjective} joke");
+ * const llm = new ChatOpenAI();
+ * const chain = prompt.pipe(llm);
  *
- * const prompt = PromptTemplate.fromTemplate("Tell me a {adjective} joke");
- * const llm = new LLMChain({ llm: new OpenAI(), prompt });
+ * const response = await chain.invoke({ adjective: "funny" });
  * ```
  */
 export class LLMChain extends BaseChain {

package/dist/chains/openai_functions/base.cjs

@@ -71,6 +71,8 @@ function isZodSchema(schema) {
     return typeof schema.safeParse === "function";
 }
 /**
+ * @deprecated Prefer the `.withStructuredOutput` method on chat model classes.
+ *
  * Create a runnable that uses an OpenAI function to get a structured output.
  * @param config Params required to create the runnable.
  * @returns A runnable sequence that will pass the given function to the model when run.

package/dist/chains/openai_functions/base.d.ts

@@ -98,6 +98,8 @@ export type CreateStructuredOutputRunnableConfig<RunInput extends Record<string,
     outputParser?: BaseOutputParser<RunOutput>;
 };
 /**
+ * @deprecated Prefer the `.withStructuredOutput` method on chat model classes.
+ *
  * Create a runnable that uses an OpenAI function to get a structured output.
  * @param config Params required to create the runnable.
  * @returns A runnable sequence that will pass the given function to the model when run.

package/dist/chains/openai_functions/base.js

@@ -67,6 +67,8 @@ function isZodSchema(schema) {
     return typeof schema.safeParse === "function";
 }
 /**
+ * @deprecated Prefer the `.withStructuredOutput` method on chat model classes.
+ *
  * Create a runnable that uses an OpenAI function to get a structured output.
  * @param config Params required to create the runnable.
  * @returns A runnable sequence that will pass the given function to the model when run.

package/dist/chains/query_constructor/index.cjs

@@ -1,6 +1,6 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.loadQueryConstructorChain = exports.formatAttributeInfo = exports.StructuredQueryOutputParser = exports.EXAMPLE_PROMPT = exports.DEFAULT_SUFFIX = exports.DEFAULT_SCHEMA = exports.DEFAULT_PREFIX = exports.DEFAULT_EXAMPLES = exports.QueryTransformer = void 0;
+exports.loadQueryConstructorRunnable = exports.formatAttributeInfo = exports.StructuredQueryOutputParser = exports.EXAMPLE_PROMPT = exports.DEFAULT_SUFFIX = exports.DEFAULT_SCHEMA = exports.DEFAULT_PREFIX = exports.DEFAULT_EXAMPLES = exports.QueryTransformer = void 0;
 const zod_1 = require("zod");
 const prompts_1 = require("@langchain/core/prompts");
 const parser_js_1 = require("./parser.cjs");
@@ -12,7 +12,6 @@ Object.defineProperty(exports, "DEFAULT_PREFIX", { enumerable: true, get: functi
 Object.defineProperty(exports, "DEFAULT_SCHEMA", { enumerable: true, get: function () { return prompt_js_1.DEFAULT_SCHEMA; } });
 Object.defineProperty(exports, "DEFAULT_SUFFIX", { enumerable: true, get: function () { return prompt_js_1.DEFAULT_SUFFIX; } });
 Object.defineProperty(exports, "EXAMPLE_PROMPT", { enumerable: true, get: function () { return prompt_js_1.EXAMPLE_PROMPT; } });
-const llm_chain_js_1 = require("../llm_chain.cjs");
 const structured_js_1 = require("../../output_parsers/structured.cjs");
 const queryInputSchema = /* #__PURE__ */ zod_1.z.object({
     query: /* #__PURE__ */ zod_1.z
@@ -119,11 +118,9 @@ function _getPrompt(documentContents, attributeInfo, allowedComparators, allowed
         outputParser,
     });
 }
-function loadQueryConstructorChain(opts) {
+function loadQueryConstructorRunnable(opts) {
     const prompt = _getPrompt(opts.documentContents, opts.attributeInfo, opts.allowedComparators, opts.allowedOperators, opts.examples);
-    return new llm_chain_js_1.LLMChain({
-        llm: opts.llm,
-        prompt,
-    });
+    const outputParser = StructuredQueryOutputParser.fromComponents(opts.allowedComparators, opts.allowedOperators);
+    return prompt.pipe(opts.llm).pipe(outputParser);
 }
-exports.loadQueryConstructorChain = loadQueryConstructorChain;
+exports.loadQueryConstructorRunnable = loadQueryConstructorRunnable;

package/dist/chains/query_constructor/index.d.ts

@@ -1,10 +1,9 @@
-import type { BaseLanguageModelInterface } from "@langchain/core/language_models/base";
 import { z } from "zod";
+import type { BaseLanguageModelInterface } from "@langchain/core/language_models/base";
 import { InputValues } from "@langchain/core/utils/types";
 import { QueryTransformer, TraverseType } from "./parser.js";
 import { Comparator, Operator, StructuredQuery } from "./ir.js";
 import { DEFAULT_EXAMPLES, DEFAULT_PREFIX, DEFAULT_SCHEMA, DEFAULT_SUFFIX, EXAMPLE_PROMPT } from "./prompt.js";
-import { LLMChain } from "../llm_chain.js";
 import { AsymmetricStructuredOutputParser } from "../../output_parsers/structured.js";
 import { AttributeInfo } from "../../schema/query_constructor.js";
 export { QueryTransformer, type TraverseType };
@@ -50,7 +49,7 @@ export declare function formatAttributeInfo(info: AttributeInfo[]): string;
 /**
  * A type that represents options for the query constructor chain.
  */
-export type QueryConstructorChainOptions = {
+export type QueryConstructorRunnableOptions = {
     llm: BaseLanguageModelInterface;
     documentContents: string;
     attributeInfo: AttributeInfo[];
@@ -58,4 +57,6 @@ export type QueryConstructorChainOptions = {
     allowedComparators?: Comparator[];
     allowedOperators?: Operator[];
 };
-export declare function loadQueryConstructorChain(opts: QueryConstructorChainOptions): LLMChain<string, BaseLanguageModelInterface<any, import("@langchain/core/language_models/base").BaseLanguageModelCallOptions>>;
+/** @deprecated */
+export type QueryConstructorChainOptions = QueryConstructorRunnableOptions;
+export declare function loadQueryConstructorRunnable(opts: QueryConstructorRunnableOptions): import("@langchain/core/runnables").Runnable<any, StructuredQuery, import("@langchain/core/runnables").RunnableConfig>;

package/dist/chains/query_constructor/index.js

@@ -3,7 +3,6 @@ import { interpolateFString, FewShotPromptTemplate, } from "@langchain/core/prom
 import { QueryTransformer } from "./parser.js";
 import { Comparators, Operators, StructuredQuery, } from "./ir.js";
 import { DEFAULT_EXAMPLES, DEFAULT_PREFIX, DEFAULT_SCHEMA, DEFAULT_SUFFIX, EXAMPLE_PROMPT, } from "./prompt.js";
-import { LLMChain } from "../llm_chain.js";
 import { AsymmetricStructuredOutputParser } from "../../output_parsers/structured.js";
 export { QueryTransformer };
 export { DEFAULT_EXAMPLES, DEFAULT_PREFIX, DEFAULT_SCHEMA, DEFAULT_SUFFIX, EXAMPLE_PROMPT, };
@@ -110,10 +109,8 @@ function _getPrompt(documentContents, attributeInfo, allowedComparators, allowed
         outputParser,
     });
 }
-export function loadQueryConstructorChain(opts) {
+export function loadQueryConstructorRunnable(opts) {
     const prompt = _getPrompt(opts.documentContents, opts.attributeInfo, opts.allowedComparators, opts.allowedOperators, opts.examples);
-    return new LLMChain({
-        llm: opts.llm,
-        prompt,
-    });
+    const outputParser = StructuredQueryOutputParser.fromComponents(opts.allowedComparators, opts.allowedOperators);
+    return prompt.pipe(opts.llm).pipe(outputParser);
 }

package/dist/chains/query_constructor/ir.cjs

@@ -1,141 +1,17 @@
 "use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.StructuredQuery = exports.Operation = exports.Comparison = exports.FilterDirective = exports.Expression = exports.Visitor = exports.Comparators = exports.Operators = void 0;
-exports.Operators = {
-    and: "and",
-    or: "or",
-    not: "not",
-};
-exports.Comparators = {
-    eq: "eq",
-    ne: "ne",
-    lt: "lt",
-    gt: "gt",
-    lte: "lte",
-    gte: "gte",
-};
-/**
- * Abstract class for visiting expressions. Subclasses must implement
- * visitOperation, visitComparison, and visitStructuredQuery methods.
- */
-class Visitor {
-}
-exports.Visitor = Visitor;
-/**
- * Abstract class representing an expression. Subclasses must implement
- * the exprName property and the accept method.
- */
-class Expression {
-    accept(visitor) {
-        if (this.exprName === "Operation") {
-            return visitor.visitOperation(this);
-        }
-        else if (this.exprName === "Comparison") {
-            return visitor.visitComparison(this);
-        }
-        else if (this.exprName === "StructuredQuery") {
-            return visitor.visitStructuredQuery(this);
-        }
-        else {
-            throw new Error("Unknown Expression type");
-        }
-    }
-}
-exports.Expression = Expression;
-/**
- * Abstract class representing a filter directive. It extends the
- * Expression class.
- */
-class FilterDirective extends Expression {
-}
-exports.FilterDirective = FilterDirective;
-/**
- * Class representing a comparison filter directive. It extends the
- * FilterDirective class.
- */
-class Comparison extends FilterDirective {
-    constructor(comparator, attribute, value) {
-        super();
-        Object.defineProperty(this, "comparator", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: comparator
-        });
-        Object.defineProperty(this, "attribute", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: attribute
-        });
-        Object.defineProperty(this, "value", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: value
-        });
-        Object.defineProperty(this, "exprName", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: "Comparison"
-        });
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
     }
-}
-exports.Comparison = Comparison;
-/**
- * Class representing an operation filter directive. It extends the
- * FilterDirective class.
- */
-class Operation extends FilterDirective {
-    constructor(operator, args) {
-        super();
-        Object.defineProperty(this, "operator", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: operator
-        });
-        Object.defineProperty(this, "args", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: args
-        });
-        Object.defineProperty(this, "exprName", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: "Operation"
-        });
-    }
-}
-exports.Operation = Operation;
-/**
- * Class representing a structured query expression. It extends the
- * Expression class.
- */
-class StructuredQuery extends Expression {
-    constructor(query, filter) {
-        super();
-        Object.defineProperty(this, "query", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: query
-        });
-        Object.defineProperty(this, "filter", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: filter
-        });
-        Object.defineProperty(this, "exprName", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: "StructuredQuery"
-        });
-    }
-}
-exports.StructuredQuery = StructuredQuery;
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __exportStar = (this && this.__exportStar) || function(m, exports) {
+    for (var p in m) if (p !== "default" && !Object.prototype.hasOwnProperty.call(exports, p)) __createBinding(exports, m, p);
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+__exportStar(require("@langchain/core/structured_query"), exports);