@langchain/classic 1.0.0-alpha.1

package/dist/chains/question_answering/refine_prompts.cjs

@@ -0,0 +1,60 @@
+const require_rolldown_runtime = require('../../_virtual/rolldown_runtime.cjs');
+const __langchain_core_prompts = require_rolldown_runtime.__toESM(require("@langchain/core/prompts"));
+const __langchain_core_example_selectors = require_rolldown_runtime.__toESM(require("@langchain/core/example_selectors"));
+
+//#region src/chains/question_answering/refine_prompts.ts
+const DEFAULT_REFINE_PROMPT_TMPL = `The original question is as follows: {question}
+We have provided an existing answer: {existing_answer}
+We have the opportunity to refine the existing answer
+(only if needed) with some more context below.
+------------
+{context}
+------------
+Given the new context, refine the original answer to better answer the question. 
+If the context isn't useful, return the original answer.`;
+const DEFAULT_REFINE_PROMPT = /* @__PURE__ */ new __langchain_core_prompts.PromptTemplate({
+	inputVariables: [
+		"question",
+		"existing_answer",
+		"context"
+	],
+	template: DEFAULT_REFINE_PROMPT_TMPL
+});
+const refineTemplate = `The original question is as follows: {question}
+We have provided an existing answer: {existing_answer}
+We have the opportunity to refine the existing answer
+(only if needed) with some more context below.
+------------
+{context}
+------------
+Given the new context, refine the original answer to better answer the question. 
+If the context isn't useful, return the original answer.`;
+const messages = [
+	/* @__PURE__ */ __langchain_core_prompts.HumanMessagePromptTemplate.fromTemplate("{question}"),
+	/* @__PURE__ */ __langchain_core_prompts.AIMessagePromptTemplate.fromTemplate("{existing_answer}"),
+	/* @__PURE__ */ __langchain_core_prompts.HumanMessagePromptTemplate.fromTemplate(refineTemplate)
+];
+const CHAT_REFINE_PROMPT = /* @__PURE__ */ __langchain_core_prompts.ChatPromptTemplate.fromMessages(messages);
+const REFINE_PROMPT_SELECTOR = /* @__PURE__ */ new __langchain_core_example_selectors.ConditionalPromptSelector(DEFAULT_REFINE_PROMPT, [[__langchain_core_example_selectors.isChatModel, CHAT_REFINE_PROMPT]]);
+const DEFAULT_TEXT_QA_PROMPT_TMPL = `Context information is below. 
+---------------------
+{context}
+---------------------
+Given the context information and no prior knowledge, answer the question: {question}`;
+const DEFAULT_TEXT_QA_PROMPT = /* @__PURE__ */ new __langchain_core_prompts.PromptTemplate({
+	inputVariables: ["context", "question"],
+	template: DEFAULT_TEXT_QA_PROMPT_TMPL
+});
+const chat_qa_prompt_template = `Context information is below. 
+---------------------
+{context}
+---------------------
+Given the context information and no prior knowledge, answer any questions`;
+const chat_messages = [/* @__PURE__ */ __langchain_core_prompts.SystemMessagePromptTemplate.fromTemplate(chat_qa_prompt_template), /* @__PURE__ */ __langchain_core_prompts.HumanMessagePromptTemplate.fromTemplate("{question}")];
+const CHAT_QUESTION_PROMPT = /* @__PURE__ */ __langchain_core_prompts.ChatPromptTemplate.fromMessages(chat_messages);
+const QUESTION_PROMPT_SELECTOR = /* @__PURE__ */ new __langchain_core_example_selectors.ConditionalPromptSelector(DEFAULT_TEXT_QA_PROMPT, [[__langchain_core_example_selectors.isChatModel, CHAT_QUESTION_PROMPT]]);
+
+//#endregion
+exports.QUESTION_PROMPT_SELECTOR = QUESTION_PROMPT_SELECTOR;
+exports.REFINE_PROMPT_SELECTOR = REFINE_PROMPT_SELECTOR;
+//# sourceMappingURL=refine_prompts.cjs.map

package/dist/chains/question_answering/refine_prompts.cjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"refine_prompts.cjs","names":["PromptTemplate","HumanMessagePromptTemplate","AIMessagePromptTemplate","ChatPromptTemplate","ConditionalPromptSelector","isChatModel","SystemMessagePromptTemplate"],"sources":["../../../src/chains/question_answering/refine_prompts.ts"],"sourcesContent":["import {\n  PromptTemplate,\n  ChatPromptTemplate,\n  SystemMessagePromptTemplate,\n  HumanMessagePromptTemplate,\n  AIMessagePromptTemplate,\n} from \"@langchain/core/prompts\";\nimport {\n  ConditionalPromptSelector,\n  isChatModel,\n} from \"@langchain/core/example_selectors\";\n\nexport const DEFAULT_REFINE_PROMPT_TMPL = `The original question is as follows: {question}\nWe have provided an existing answer: {existing_answer}\nWe have the opportunity to refine the existing answer\n(only if needed) with some more context below.\n------------\n{context}\n------------\nGiven the new context, refine the original answer to better answer the question. \nIf the context isn't useful, return the original answer.`;\nexport const DEFAULT_REFINE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  inputVariables: [\"question\", \"existing_answer\", \"context\"],\n  template: DEFAULT_REFINE_PROMPT_TMPL,\n});\n\nconst refineTemplate = `The original question is as follows: {question}\nWe have provided an existing answer: {existing_answer}\nWe have the opportunity to refine the existing answer\n(only if needed) with some more context below.\n------------\n{context}\n------------\nGiven the new context, refine the original answer to better answer the question. \nIf the context isn't useful, return the original answer.`;\n\nconst messages = [\n  /*#__PURE__*/ HumanMessagePromptTemplate.fromTemplate(\"{question}\"),\n  /*#__PURE__*/ AIMessagePromptTemplate.fromTemplate(\"{existing_answer}\"),\n  /*#__PURE__*/ HumanMessagePromptTemplate.fromTemplate(refineTemplate),\n];\n\nexport const CHAT_REFINE_PROMPT =\n  /*#__PURE__*/ ChatPromptTemplate.fromMessages(messages);\n\nexport const REFINE_PROMPT_SELECTOR =\n  /*#__PURE__*/ new ConditionalPromptSelector(DEFAULT_REFINE_PROMPT, [\n    [isChatModel, CHAT_REFINE_PROMPT],\n  ]);\n\nexport const DEFAULT_TEXT_QA_PROMPT_TMPL = `Context information is below. \n---------------------\n{context}\n---------------------\nGiven the context information and no prior knowledge, answer the question: {question}`;\nexport const DEFAULT_TEXT_QA_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  inputVariables: [\"context\", \"question\"],\n  template: DEFAULT_TEXT_QA_PROMPT_TMPL,\n});\n\nconst chat_qa_prompt_template = `Context information is below. \n---------------------\n{context}\n---------------------\nGiven the context information and no prior knowledge, answer any questions`;\nconst chat_messages = [\n  /*#__PURE__*/ SystemMessagePromptTemplate.fromTemplate(\n    chat_qa_prompt_template\n  ),\n  /*#__PURE__*/ HumanMessagePromptTemplate.fromTemplate(\"{question}\"),\n];\nexport const CHAT_QUESTION_PROMPT =\n  /*#__PURE__*/ ChatPromptTemplate.fromMessages(chat_messages);\nexport const QUESTION_PROMPT_SELECTOR =\n  /*#__PURE__*/ new ConditionalPromptSelector(DEFAULT_TEXT_QA_PROMPT, [\n    [isChatModel, CHAT_QUESTION_PROMPT],\n  ]);\n"],"mappings":";;;;;AAYA,MAAa,6BAA6B,CAAC;;;;;;;;wDAQa,CAAC;AACzD,MAAa,wCAAsC,IAAIA,wCAAe;CACpE,gBAAgB;EAAC;EAAY;EAAmB;CAAU;CAC1D,UAAU;AACX;AAED,MAAM,iBAAiB,CAAC;;;;;;;;wDAQgC,CAAC;AAEzD,MAAM,WAAW;iBACDC,oDAA2B,aAAa,aAAa;iBACrDC,iDAAwB,aAAa,oBAAoB;iBACzDD,oDAA2B,aAAa,eAAe;AACtE;AAED,MAAa,qCACGE,4CAAmB,aAAa,SAAS;AAEzD,MAAa,yCACG,IAAIC,6DAA0B,uBAAuB,CACjE,CAACC,gDAAa,kBAAmB,CAClC;AAEH,MAAa,8BAA8B,CAAC;;;;qFAIyC,CAAC;AACtF,MAAa,yCAAuC,IAAIL,wCAAe;CACrE,gBAAgB,CAAC,WAAW,UAAW;CACvC,UAAU;AACX;AAED,MAAM,0BAA0B,CAAC;;;;0EAIyC,CAAC;AAC3E,MAAM,gBAAgB,iBACNM,qDAA4B,aACxC,wBACD,kBACaL,oDAA2B,aAAa,aAAa,AACpE;AACD,MAAa,uCACGE,4CAAmB,aAAa,cAAc;AAC9D,MAAa,2CACG,IAAIC,6DAA0B,wBAAwB,CAClE,CAACC,gDAAa,oBAAqB,CACpC"}

package/dist/chains/question_answering/refine_prompts.js

@@ -0,0 +1,58 @@
+import { AIMessagePromptTemplate, ChatPromptTemplate, HumanMessagePromptTemplate, PromptTemplate, SystemMessagePromptTemplate } from "@langchain/core/prompts";
+import { ConditionalPromptSelector, isChatModel } from "@langchain/core/example_selectors";
+
+//#region src/chains/question_answering/refine_prompts.ts
+const DEFAULT_REFINE_PROMPT_TMPL = `The original question is as follows: {question}
+We have provided an existing answer: {existing_answer}
+We have the opportunity to refine the existing answer
+(only if needed) with some more context below.
+------------
+{context}
+------------
+Given the new context, refine the original answer to better answer the question. 
+If the context isn't useful, return the original answer.`;
+const DEFAULT_REFINE_PROMPT = /* @__PURE__ */ new PromptTemplate({
+	inputVariables: [
+		"question",
+		"existing_answer",
+		"context"
+	],
+	template: DEFAULT_REFINE_PROMPT_TMPL
+});
+const refineTemplate = `The original question is as follows: {question}
+We have provided an existing answer: {existing_answer}
+We have the opportunity to refine the existing answer
+(only if needed) with some more context below.
+------------
+{context}
+------------
+Given the new context, refine the original answer to better answer the question. 
+If the context isn't useful, return the original answer.`;
+const messages = [
+	/* @__PURE__ */ HumanMessagePromptTemplate.fromTemplate("{question}"),
+	/* @__PURE__ */ AIMessagePromptTemplate.fromTemplate("{existing_answer}"),
+	/* @__PURE__ */ HumanMessagePromptTemplate.fromTemplate(refineTemplate)
+];
+const CHAT_REFINE_PROMPT = /* @__PURE__ */ ChatPromptTemplate.fromMessages(messages);
+const REFINE_PROMPT_SELECTOR = /* @__PURE__ */ new ConditionalPromptSelector(DEFAULT_REFINE_PROMPT, [[isChatModel, CHAT_REFINE_PROMPT]]);
+const DEFAULT_TEXT_QA_PROMPT_TMPL = `Context information is below. 
+---------------------
+{context}
+---------------------
+Given the context information and no prior knowledge, answer the question: {question}`;
+const DEFAULT_TEXT_QA_PROMPT = /* @__PURE__ */ new PromptTemplate({
+	inputVariables: ["context", "question"],
+	template: DEFAULT_TEXT_QA_PROMPT_TMPL
+});
+const chat_qa_prompt_template = `Context information is below. 
+---------------------
+{context}
+---------------------
+Given the context information and no prior knowledge, answer any questions`;
+const chat_messages = [/* @__PURE__ */ SystemMessagePromptTemplate.fromTemplate(chat_qa_prompt_template), /* @__PURE__ */ HumanMessagePromptTemplate.fromTemplate("{question}")];
+const CHAT_QUESTION_PROMPT = /* @__PURE__ */ ChatPromptTemplate.fromMessages(chat_messages);
+const QUESTION_PROMPT_SELECTOR = /* @__PURE__ */ new ConditionalPromptSelector(DEFAULT_TEXT_QA_PROMPT, [[isChatModel, CHAT_QUESTION_PROMPT]]);
+
+//#endregion
+export { QUESTION_PROMPT_SELECTOR, REFINE_PROMPT_SELECTOR };
+//# sourceMappingURL=refine_prompts.js.map

package/dist/chains/question_answering/refine_prompts.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"refine_prompts.js","names":[],"sources":["../../../src/chains/question_answering/refine_prompts.ts"],"sourcesContent":["import {\n  PromptTemplate,\n  ChatPromptTemplate,\n  SystemMessagePromptTemplate,\n  HumanMessagePromptTemplate,\n  AIMessagePromptTemplate,\n} from \"@langchain/core/prompts\";\nimport {\n  ConditionalPromptSelector,\n  isChatModel,\n} from \"@langchain/core/example_selectors\";\n\nexport const DEFAULT_REFINE_PROMPT_TMPL = `The original question is as follows: {question}\nWe have provided an existing answer: {existing_answer}\nWe have the opportunity to refine the existing answer\n(only if needed) with some more context below.\n------------\n{context}\n------------\nGiven the new context, refine the original answer to better answer the question. \nIf the context isn't useful, return the original answer.`;\nexport const DEFAULT_REFINE_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  inputVariables: [\"question\", \"existing_answer\", \"context\"],\n  template: DEFAULT_REFINE_PROMPT_TMPL,\n});\n\nconst refineTemplate = `The original question is as follows: {question}\nWe have provided an existing answer: {existing_answer}\nWe have the opportunity to refine the existing answer\n(only if needed) with some more context below.\n------------\n{context}\n------------\nGiven the new context, refine the original answer to better answer the question. \nIf the context isn't useful, return the original answer.`;\n\nconst messages = [\n  /*#__PURE__*/ HumanMessagePromptTemplate.fromTemplate(\"{question}\"),\n  /*#__PURE__*/ AIMessagePromptTemplate.fromTemplate(\"{existing_answer}\"),\n  /*#__PURE__*/ HumanMessagePromptTemplate.fromTemplate(refineTemplate),\n];\n\nexport const CHAT_REFINE_PROMPT =\n  /*#__PURE__*/ ChatPromptTemplate.fromMessages(messages);\n\nexport const REFINE_PROMPT_SELECTOR =\n  /*#__PURE__*/ new ConditionalPromptSelector(DEFAULT_REFINE_PROMPT, [\n    [isChatModel, CHAT_REFINE_PROMPT],\n  ]);\n\nexport const DEFAULT_TEXT_QA_PROMPT_TMPL = `Context information is below. \n---------------------\n{context}\n---------------------\nGiven the context information and no prior knowledge, answer the question: {question}`;\nexport const DEFAULT_TEXT_QA_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  inputVariables: [\"context\", \"question\"],\n  template: DEFAULT_TEXT_QA_PROMPT_TMPL,\n});\n\nconst chat_qa_prompt_template = `Context information is below. \n---------------------\n{context}\n---------------------\nGiven the context information and no prior knowledge, answer any questions`;\nconst chat_messages = [\n  /*#__PURE__*/ SystemMessagePromptTemplate.fromTemplate(\n    chat_qa_prompt_template\n  ),\n  /*#__PURE__*/ HumanMessagePromptTemplate.fromTemplate(\"{question}\"),\n];\nexport const CHAT_QUESTION_PROMPT =\n  /*#__PURE__*/ ChatPromptTemplate.fromMessages(chat_messages);\nexport const QUESTION_PROMPT_SELECTOR =\n  /*#__PURE__*/ new ConditionalPromptSelector(DEFAULT_TEXT_QA_PROMPT, [\n    [isChatModel, CHAT_QUESTION_PROMPT],\n  ]);\n"],"mappings":";;;;AAYA,MAAa,6BAA6B,CAAC;;;;;;;;wDAQa,CAAC;AACzD,MAAa,wCAAsC,IAAI,eAAe;CACpE,gBAAgB;EAAC;EAAY;EAAmB;CAAU;CAC1D,UAAU;AACX;AAED,MAAM,iBAAiB,CAAC;;;;;;;;wDAQgC,CAAC;AAEzD,MAAM,WAAW;iBACD,2BAA2B,aAAa,aAAa;iBACrD,wBAAwB,aAAa,oBAAoB;iBACzD,2BAA2B,aAAa,eAAe;AACtE;AAED,MAAa,qCACG,mBAAmB,aAAa,SAAS;AAEzD,MAAa,yCACG,IAAI,0BAA0B,uBAAuB,CACjE,CAAC,aAAa,kBAAmB,CAClC;AAEH,MAAa,8BAA8B,CAAC;;;;qFAIyC,CAAC;AACtF,MAAa,yCAAuC,IAAI,eAAe;CACrE,gBAAgB,CAAC,WAAW,UAAW;CACvC,UAAU;AACX;AAED,MAAM,0BAA0B,CAAC;;;;0EAIyC,CAAC;AAC3E,MAAM,gBAAgB,iBACN,4BAA4B,aACxC,wBACD,kBACa,2BAA2B,aAAa,aAAa,AACpE;AACD,MAAa,uCACG,mBAAmB,aAAa,cAAc;AAC9D,MAAa,2CACG,IAAI,0BAA0B,wBAAwB,CAClE,CAAC,aAAa,oBAAqB,CACpC"}

package/dist/chains/question_answering/stuff_prompts.cjs

@@ -0,0 +1,20 @@
+const require_rolldown_runtime = require('../../_virtual/rolldown_runtime.cjs');
+const __langchain_core_prompts = require_rolldown_runtime.__toESM(require("@langchain/core/prompts"));
+const __langchain_core_example_selectors = require_rolldown_runtime.__toESM(require("@langchain/core/example_selectors"));
+
+//#region src/chains/question_answering/stuff_prompts.ts
+const DEFAULT_QA_PROMPT = /* @__PURE__ */ new __langchain_core_prompts.PromptTemplate({
+	template: "Use the following pieces of context to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer.\n\n{context}\n\nQuestion: {question}\nHelpful Answer:",
+	inputVariables: ["context", "question"]
+});
+const system_template = `Use the following pieces of context to answer the users question. 
+If you don't know the answer, just say that you don't know, don't try to make up an answer.
+----------------
+{context}`;
+const messages = [/* @__PURE__ */ __langchain_core_prompts.SystemMessagePromptTemplate.fromTemplate(system_template), /* @__PURE__ */ __langchain_core_prompts.HumanMessagePromptTemplate.fromTemplate("{question}")];
+const CHAT_PROMPT = /* @__PURE__ */ __langchain_core_prompts.ChatPromptTemplate.fromMessages(messages);
+const QA_PROMPT_SELECTOR = /* @__PURE__ */ new __langchain_core_example_selectors.ConditionalPromptSelector(DEFAULT_QA_PROMPT, [[__langchain_core_example_selectors.isChatModel, CHAT_PROMPT]]);
+
+//#endregion
+exports.QA_PROMPT_SELECTOR = QA_PROMPT_SELECTOR;
+//# sourceMappingURL=stuff_prompts.cjs.map

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

@@ -0,0 +1 @@
+{"version":3,"file":"stuff_prompts.cjs","names":["PromptTemplate","SystemMessagePromptTemplate","HumanMessagePromptTemplate","ChatPromptTemplate","ConditionalPromptSelector","isChatModel"],"sources":["../../../src/chains/question_answering/stuff_prompts.ts"],"sourcesContent":["import {\n  ChatPromptTemplate,\n  SystemMessagePromptTemplate,\n  HumanMessagePromptTemplate,\n  PromptTemplate,\n} from \"@langchain/core/prompts\";\nimport {\n  ConditionalPromptSelector,\n  isChatModel,\n} from \"@langchain/core/example_selectors\";\n\nexport const DEFAULT_QA_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template:\n    \"Use the following pieces of context to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer.\\n\\n{context}\\n\\nQuestion: {question}\\nHelpful Answer:\",\n  inputVariables: [\"context\", \"question\"],\n});\n\nconst system_template = `Use the following pieces of context to answer the users question. \nIf you don't know the answer, just say that you don't know, don't try to make up an answer.\n----------------\n{context}`;\nconst messages = [\n  /*#__PURE__*/ SystemMessagePromptTemplate.fromTemplate(system_template),\n  /*#__PURE__*/ HumanMessagePromptTemplate.fromTemplate(\"{question}\"),\n];\nconst CHAT_PROMPT = /*#__PURE__*/ ChatPromptTemplate.fromMessages(messages);\n\nexport const QA_PROMPT_SELECTOR = /*#__PURE__*/ new ConditionalPromptSelector(\n  DEFAULT_QA_PROMPT,\n  [[isChatModel, CHAT_PROMPT]]\n);\n"],"mappings":";;;;;AAWA,MAAa,oCAAkC,IAAIA,wCAAe;CAChE,UACE;CACF,gBAAgB,CAAC,WAAW,UAAW;AACxC;AAED,MAAM,kBAAkB,CAAC;;;SAGhB,CAAC;AACV,MAAM,WAAW,iBACDC,qDAA4B,aAAa,gBAAgB,kBACzDC,oDAA2B,aAAa,aAAa,AACpE;AACD,MAAM,8BAA4BC,4CAAmB,aAAa,SAAS;AAE3E,MAAa,qCAAmC,IAAIC,6DAClD,mBACA,CAAC,CAACC,gDAAa,WAAY,CAAC"}

package/dist/chains/question_answering/stuff_prompts.js

@@ -0,0 +1,19 @@
+import { ChatPromptTemplate, HumanMessagePromptTemplate, PromptTemplate, SystemMessagePromptTemplate } from "@langchain/core/prompts";
+import { ConditionalPromptSelector, isChatModel } from "@langchain/core/example_selectors";
+
+//#region src/chains/question_answering/stuff_prompts.ts
+const DEFAULT_QA_PROMPT = /* @__PURE__ */ new PromptTemplate({
+	template: "Use the following pieces of context to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer.\n\n{context}\n\nQuestion: {question}\nHelpful Answer:",
+	inputVariables: ["context", "question"]
+});
+const system_template = `Use the following pieces of context to answer the users question. 
+If you don't know the answer, just say that you don't know, don't try to make up an answer.
+----------------
+{context}`;
+const messages = [/* @__PURE__ */ SystemMessagePromptTemplate.fromTemplate(system_template), /* @__PURE__ */ HumanMessagePromptTemplate.fromTemplate("{question}")];
+const CHAT_PROMPT = /* @__PURE__ */ ChatPromptTemplate.fromMessages(messages);
+const QA_PROMPT_SELECTOR = /* @__PURE__ */ new ConditionalPromptSelector(DEFAULT_QA_PROMPT, [[isChatModel, CHAT_PROMPT]]);
+
+//#endregion
+export { QA_PROMPT_SELECTOR };
+//# sourceMappingURL=stuff_prompts.js.map

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

@@ -0,0 +1 @@
+{"version":3,"file":"stuff_prompts.js","names":[],"sources":["../../../src/chains/question_answering/stuff_prompts.ts"],"sourcesContent":["import {\n  ChatPromptTemplate,\n  SystemMessagePromptTemplate,\n  HumanMessagePromptTemplate,\n  PromptTemplate,\n} from \"@langchain/core/prompts\";\nimport {\n  ConditionalPromptSelector,\n  isChatModel,\n} from \"@langchain/core/example_selectors\";\n\nexport const DEFAULT_QA_PROMPT = /*#__PURE__*/ new PromptTemplate({\n  template:\n    \"Use the following pieces of context to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer.\\n\\n{context}\\n\\nQuestion: {question}\\nHelpful Answer:\",\n  inputVariables: [\"context\", \"question\"],\n});\n\nconst system_template = `Use the following pieces of context to answer the users question. \nIf you don't know the answer, just say that you don't know, don't try to make up an answer.\n----------------\n{context}`;\nconst messages = [\n  /*#__PURE__*/ SystemMessagePromptTemplate.fromTemplate(system_template),\n  /*#__PURE__*/ HumanMessagePromptTemplate.fromTemplate(\"{question}\"),\n];\nconst CHAT_PROMPT = /*#__PURE__*/ ChatPromptTemplate.fromMessages(messages);\n\nexport const QA_PROMPT_SELECTOR = /*#__PURE__*/ new ConditionalPromptSelector(\n  DEFAULT_QA_PROMPT,\n  [[isChatModel, CHAT_PROMPT]]\n);\n"],"mappings":";;;;AAWA,MAAa,oCAAkC,IAAI,eAAe;CAChE,UACE;CACF,gBAAgB,CAAC,WAAW,UAAW;AACxC;AAED,MAAM,kBAAkB,CAAC;;;SAGhB,CAAC;AACV,MAAM,WAAW,iBACD,4BAA4B,aAAa,gBAAgB,kBACzD,2BAA2B,aAAa,aAAa,AACpE;AACD,MAAM,8BAA4B,mBAAmB,aAAa,SAAS;AAE3E,MAAa,qCAAmC,IAAI,0BAClD,mBACA,CAAC,CAAC,aAAa,WAAY,CAAC"}

package/dist/chains/retrieval.cjs

@@ -0,0 +1,55 @@
+const require_rolldown_runtime = require('../_virtual/rolldown_runtime.cjs');
+const __langchain_core_runnables = require_rolldown_runtime.__toESM(require("@langchain/core/runnables"));
+
+//#region src/chains/retrieval.ts
+var retrieval_exports = {};
+require_rolldown_runtime.__export(retrieval_exports, { createRetrievalChain: () => createRetrievalChain });
+function isBaseRetriever(x) {
+	return !!x && typeof x.invoke === "function";
+}
+/**
+* Create a retrieval chain that retrieves documents and then passes them on.
+* @param {CreateRetrievalChainParams} params A params object
+*     containing a retriever and a combineDocsChain.
+* @returns An LCEL Runnable which returns a an object
+*     containing at least `context` and `answer` keys.
+* @example
+* ```typescript
+* // pnpm add langchain @langchain/openai
+*
+* import { ChatOpenAI } from "@langchain/openai";
+* import { pull } from "langchain/hub";
+* import { createRetrievalChain } from "@langchain/classic/chains/retrieval";
+* import { createStuffDocumentsChain } from "@langchain/classic/chains/combine_documents";
+*
+* const retrievalQAChatPrompt = await pull("langchain-ai/retrieval-qa-chat");
+* const llm = new ChatOpenAI({ model: "gpt-4o-mini" });
+* const retriever = ...
+* const combineDocsChain = await createStuffDocumentsChain(...);
+* const retrievalChain = await createRetrievalChain({
+*   retriever,
+*   combineDocsChain,
+* });
+* const response = await chain.invoke({ input: "..." });
+* ```
+*/
+async function createRetrievalChain({ retriever, combineDocsChain }) {
+	let retrieveDocumentsChain;
+	if (isBaseRetriever(retriever)) retrieveDocumentsChain = __langchain_core_runnables.RunnableSequence.from([(input) => input.input, retriever]);
+	else retrieveDocumentsChain = retriever;
+	const retrievalChain = __langchain_core_runnables.RunnableSequence.from([__langchain_core_runnables.RunnablePassthrough.assign({
+		context: retrieveDocumentsChain.withConfig({ runName: "retrieve_documents" }),
+		chat_history: (input) => input.chat_history ?? []
+	}), __langchain_core_runnables.RunnablePassthrough.assign({ answer: combineDocsChain })]).withConfig({ runName: "retrieval_chain" });
+	return retrievalChain;
+}
+
+//#endregion
+exports.createRetrievalChain = createRetrievalChain;
+Object.defineProperty(exports, 'retrieval_exports', {
+  enumerable: true,
+  get: function () {
+    return retrieval_exports;
+  }
+});
+//# sourceMappingURL=retrieval.cjs.map

package/dist/chains/retrieval.cjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"retrieval.cjs","names":["x: unknown","retrieveDocumentsChain: Runnable<{ input: string }, DocumentInterface[]>","RunnableSequence","RunnablePassthrough"],"sources":["../../src/chains/retrieval.ts"],"sourcesContent":["import type { BaseRetrieverInterface } from \"@langchain/core/retrievers\";\nimport {\n  type Runnable,\n  RunnableSequence,\n  type RunnableInterface,\n  RunnablePassthrough,\n} from \"@langchain/core/runnables\";\nimport type { BaseMessage } from \"@langchain/core/messages\";\nimport type { DocumentInterface, Document } from \"@langchain/core/documents\";\n\n/**\n * Parameters for the createRetrievalChain method.\n */\nexport type CreateRetrievalChainParams<RunOutput> = {\n  /**\n   * Retriever-like object that returns list of documents. Should\n   * either be a subclass of BaseRetriever or a Runnable that returns\n   * a list of documents. If a subclass of BaseRetriever, then it\n   * is expected that an `input` key be passed in - this is what\n   * is will be used to pass into the retriever. If this is NOT a\n   * subclass of BaseRetriever, then all the inputs will be passed\n   * into this runnable, meaning that runnable should take a object\n   * as input.\n   */\n  retriever:\n    | BaseRetrieverInterface\n    // eslint-disable-next-line @typescript-eslint/no-explicit-any\n    | RunnableInterface<Record<string, any>, DocumentInterface[]>;\n  /**\n   * Runnable that takes inputs and produces a string output.\n   * The inputs to this will be any original inputs to this chain, a new\n   * context key with the retrieved documents, and chat_history (if not present\n   * in the inputs) with a value of `[]` (to easily enable conversational\n   * retrieval).\n   */\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  combineDocsChain: RunnableInterface<Record<string, any>, RunOutput>;\n};\n\nfunction isBaseRetriever(x: unknown): x is BaseRetrieverInterface {\n  return !!x && typeof (x as BaseRetrieverInterface).invoke === \"function\";\n}\n\n/**\n * Create a retrieval chain that retrieves documents and then passes them on.\n * @param {CreateRetrievalChainParams} params A params object\n *     containing a retriever and a combineDocsChain.\n * @returns An LCEL Runnable which returns a an object\n *     containing at least `context` and `answer` keys.\n * @example\n * ```typescript\n * // pnpm add langchain @langchain/openai\n *\n * import { ChatOpenAI } from \"@langchain/openai\";\n * import { pull } from \"langchain/hub\";\n * import { createRetrievalChain } from \"@langchain/classic/chains/retrieval\";\n * import { createStuffDocumentsChain } from \"@langchain/classic/chains/combine_documents\";\n *\n * const retrievalQAChatPrompt = await pull(\"langchain-ai/retrieval-qa-chat\");\n * const llm = new ChatOpenAI({ model: \"gpt-4o-mini\" });\n * const retriever = ...\n * const combineDocsChain = await createStuffDocumentsChain(...);\n * const retrievalChain = await createRetrievalChain({\n *   retriever,\n *   combineDocsChain,\n * });\n * const response = await chain.invoke({ input: \"...\" });\n * ```\n */\nexport async function createRetrievalChain<RunOutput>({\n  retriever,\n  combineDocsChain,\n}: CreateRetrievalChainParams<RunOutput>): Promise<\n  Runnable<\n    { input: string; chat_history?: BaseMessage[] | string } & {\n      [key: string]: unknown;\n    },\n    { context: Document[]; answer: RunOutput } & { [key: string]: unknown }\n  >\n> {\n  let retrieveDocumentsChain: Runnable<{ input: string }, DocumentInterface[]>;\n  if (isBaseRetriever(retriever)) {\n    retrieveDocumentsChain = RunnableSequence.from([\n      (input) => input.input,\n      retriever,\n    ]);\n  } else {\n    // TODO: Fix typing by adding withConfig to core RunnableInterface\n    retrieveDocumentsChain = retriever as Runnable;\n  }\n  const retrievalChain = RunnableSequence.from<{\n    input: string;\n    chat_history?: BaseMessage[] | string;\n  }>([\n    RunnablePassthrough.assign({\n      context: retrieveDocumentsChain.withConfig({\n        runName: \"retrieve_documents\",\n      }),\n      chat_history: (input) => input.chat_history ?? [],\n    }),\n    RunnablePassthrough.assign({\n      answer: combineDocsChain,\n    }),\n  ]).withConfig({ runName: \"retrieval_chain\" });\n  return retrievalChain;\n}\n"],"mappings":";;;;;;AAuCA,SAAS,gBAAgBA,GAAyC;AAChE,QAAO,CAAC,CAAC,KAAK,OAAQ,EAA6B,WAAW;AAC/D;;;;;;;;;;;;;;;;;;;;;;;;;;;AA4BD,eAAsB,qBAAgC,EACpD,WACA,kBACsC,EAOtC;CACA,IAAIC;AACJ,KAAI,gBAAgB,UAAU,EAC5B,yBAAyBC,4CAAiB,KAAK,CAC7C,CAAC,UAAU,MAAM,OACjB,SACD,EAAC;MAGF,yBAAyB;CAE3B,MAAM,iBAAiBA,4CAAiB,KAGrC,CACDC,+CAAoB,OAAO;EACzB,SAAS,uBAAuB,WAAW,EACzC,SAAS,qBACV,EAAC;EACF,cAAc,CAAC,UAAU,MAAM,gBAAgB,CAAE;CAClD,EAAC,EACFA,+CAAoB,OAAO,EACzB,QAAQ,iBACT,EAAC,AACH,EAAC,CAAC,WAAW,EAAE,SAAS,kBAAmB,EAAC;AAC7C,QAAO;AACR"}

package/dist/chains/retrieval.d.cts

@@ -0,0 +1,77 @@
+import { BaseMessage } from "@langchain/core/messages";
+import { Runnable, RunnableInterface } from "@langchain/core/runnables";
+import { BaseRetrieverInterface } from "@langchain/core/retrievers";
+import { Document, DocumentInterface } from "@langchain/core/documents";
+
+//#region src/chains/retrieval.d.ts
+
+/**
+ * Parameters for the createRetrievalChain method.
+ */
+type CreateRetrievalChainParams<RunOutput> = {
+  /**
+   * Retriever-like object that returns list of documents. Should
+   * either be a subclass of BaseRetriever or a Runnable that returns
+   * a list of documents. If a subclass of BaseRetriever, then it
+   * is expected that an `input` key be passed in - this is what
+   * is will be used to pass into the retriever. If this is NOT a
+   * subclass of BaseRetriever, then all the inputs will be passed
+   * into this runnable, meaning that runnable should take a object
+   * as input.
+   */
+  retriever: BaseRetrieverInterface
+  // eslint-disable-next-line @typescript-eslint/no-explicit-any
+  | RunnableInterface<Record<string, any>, DocumentInterface[]>;
+  /**
+   * Runnable that takes inputs and produces a string output.
+   * The inputs to this will be any original inputs to this chain, a new
+   * context key with the retrieved documents, and chat_history (if not present
+   * in the inputs) with a value of `[]` (to easily enable conversational
+   * retrieval).
+   */
+  // eslint-disable-next-line @typescript-eslint/no-explicit-any
+  combineDocsChain: RunnableInterface<Record<string, any>, RunOutput>;
+};
+/**
+ * Create a retrieval chain that retrieves documents and then passes them on.
+ * @param {CreateRetrievalChainParams} params A params object
+ *     containing a retriever and a combineDocsChain.
+ * @returns An LCEL Runnable which returns a an object
+ *     containing at least `context` and `answer` keys.
+ * @example
+ * ```typescript
+ * // pnpm add langchain @langchain/openai
+ *
+ * import { ChatOpenAI } from "@langchain/openai";
+ * import { pull } from "langchain/hub";
+ * import { createRetrievalChain } from "@langchain/classic/chains/retrieval";
+ * import { createStuffDocumentsChain } from "@langchain/classic/chains/combine_documents";
+ *
+ * const retrievalQAChatPrompt = await pull("langchain-ai/retrieval-qa-chat");
+ * const llm = new ChatOpenAI({ model: "gpt-4o-mini" });
+ * const retriever = ...
+ * const combineDocsChain = await createStuffDocumentsChain(...);
+ * const retrievalChain = await createRetrievalChain({
+ *   retriever,
+ *   combineDocsChain,
+ * });
+ * const response = await chain.invoke({ input: "..." });
+ * ```
+ */
+declare function createRetrievalChain<RunOutput>({
+  retriever,
+  combineDocsChain
+}: CreateRetrievalChainParams<RunOutput>): Promise<Runnable<{
+  input: string;
+  chat_history?: BaseMessage[] | string;
+} & {
+  [key: string]: unknown;
+}, {
+  context: Document[];
+  answer: RunOutput;
+} & {
+  [key: string]: unknown;
+}>>;
+//#endregion
+export { CreateRetrievalChainParams, createRetrievalChain };
+//# sourceMappingURL=retrieval.d.cts.map

package/dist/chains/retrieval.d.cts.map

@@ -0,0 +1 @@
+{"version":3,"file":"retrieval.d.cts","names":["BaseRetrieverInterface","Runnable","RunnableInterface","BaseMessage","DocumentInterface","Document","CreateRetrievalChainParams","Record","RunOutput","createRetrievalChain","retriever","combineDocsChain","Promise"],"sources":["../../src/chains/retrieval.d.ts"],"sourcesContent":["import type { BaseRetrieverInterface } from \"@langchain/core/retrievers\";\nimport { type Runnable, type RunnableInterface } from \"@langchain/core/runnables\";\nimport type { BaseMessage } from \"@langchain/core/messages\";\nimport type { DocumentInterface, Document } from \"@langchain/core/documents\";\n/**\n * Parameters for the createRetrievalChain method.\n */\nexport type CreateRetrievalChainParams<RunOutput> = {\n    /**\n     * Retriever-like object that returns list of documents. Should\n     * either be a subclass of BaseRetriever or a Runnable that returns\n     * a list of documents. If a subclass of BaseRetriever, then it\n     * is expected that an `input` key be passed in - this is what\n     * is will be used to pass into the retriever. If this is NOT a\n     * subclass of BaseRetriever, then all the inputs will be passed\n     * into this runnable, meaning that runnable should take a object\n     * as input.\n     */\n    retriever: BaseRetrieverInterface\n    // eslint-disable-next-line @typescript-eslint/no-explicit-any\n     | RunnableInterface<Record<string, any>, DocumentInterface[]>;\n    /**\n     * Runnable that takes inputs and produces a string output.\n     * The inputs to this will be any original inputs to this chain, a new\n     * context key with the retrieved documents, and chat_history (if not present\n     * in the inputs) with a value of `[]` (to easily enable conversational\n     * retrieval).\n     */\n    // eslint-disable-next-line @typescript-eslint/no-explicit-any\n    combineDocsChain: RunnableInterface<Record<string, any>, RunOutput>;\n};\n/**\n * Create a retrieval chain that retrieves documents and then passes them on.\n * @param {CreateRetrievalChainParams} params A params object\n *     containing a retriever and a combineDocsChain.\n * @returns An LCEL Runnable which returns a an object\n *     containing at least `context` and `answer` keys.\n * @example\n * ```typescript\n * // pnpm add langchain @langchain/openai\n *\n * import { ChatOpenAI } from \"@langchain/openai\";\n * import { pull } from \"langchain/hub\";\n * import { createRetrievalChain } from \"@langchain/classic/chains/retrieval\";\n * import { createStuffDocumentsChain } from \"@langchain/classic/chains/combine_documents\";\n *\n * const retrievalQAChatPrompt = await pull(\"langchain-ai/retrieval-qa-chat\");\n * const llm = new ChatOpenAI({ model: \"gpt-4o-mini\" });\n * const retriever = ...\n * const combineDocsChain = await createStuffDocumentsChain(...);\n * const retrievalChain = await createRetrievalChain({\n *   retriever,\n *   combineDocsChain,\n * });\n * const response = await chain.invoke({ input: \"...\" });\n * ```\n */\nexport declare function createRetrievalChain<RunOutput>({ retriever, combineDocsChain }: CreateRetrievalChainParams<RunOutput>): Promise<Runnable<{\n    input: string;\n    chat_history?: BaseMessage[] | string;\n} & {\n    [key: string]: unknown;\n}, {\n    context: Document[];\n    answer: RunOutput;\n} & {\n    [key: string]: unknown;\n}>>;\n"],"mappings":";;;;;;;;;AAOA;AAAsC,KAA1BM,0BAA0B,CAAA,SAAA,CAAA,GAAA;EAAA;;;;;;;AAsBC;AA4BvC;;EAA4C,SAAcI,EAvC3CV;EAuCoD;EAAA,EArC5DE,iBAqC6GM,CArC3FD,MAqC2FC,CAAAA,MAAAA,EAAAA,GAAAA,CAAAA,EArCtEJ,iBAqCsEI,EAAAA,CAAAA;EAAS;;;;;;AAAW;;oBA5BlHN,kBAAkBK,qBAAqBC;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBA4BrCC;;;GAAiEH,2BAA2BE,aAAaI,QAAQX;;iBAEtHE;;;;WAINE;UACDG"}

package/dist/chains/retrieval.d.ts

@@ -0,0 +1,77 @@
+import { Document, DocumentInterface } from "@langchain/core/documents";
+import { Runnable, RunnableInterface } from "@langchain/core/runnables";
+import { BaseMessage } from "@langchain/core/messages";
+import { BaseRetrieverInterface } from "@langchain/core/retrievers";
+
+//#region src/chains/retrieval.d.ts
+
+/**
+ * Parameters for the createRetrievalChain method.
+ */
+type CreateRetrievalChainParams<RunOutput> = {
+  /**
+   * Retriever-like object that returns list of documents. Should
+   * either be a subclass of BaseRetriever or a Runnable that returns
+   * a list of documents. If a subclass of BaseRetriever, then it
+   * is expected that an `input` key be passed in - this is what
+   * is will be used to pass into the retriever. If this is NOT a
+   * subclass of BaseRetriever, then all the inputs will be passed
+   * into this runnable, meaning that runnable should take a object
+   * as input.
+   */
+  retriever: BaseRetrieverInterface
+  // eslint-disable-next-line @typescript-eslint/no-explicit-any
+  | RunnableInterface<Record<string, any>, DocumentInterface[]>;
+  /**
+   * Runnable that takes inputs and produces a string output.
+   * The inputs to this will be any original inputs to this chain, a new
+   * context key with the retrieved documents, and chat_history (if not present
+   * in the inputs) with a value of `[]` (to easily enable conversational
+   * retrieval).
+   */
+  // eslint-disable-next-line @typescript-eslint/no-explicit-any
+  combineDocsChain: RunnableInterface<Record<string, any>, RunOutput>;
+};
+/**
+ * Create a retrieval chain that retrieves documents and then passes them on.
+ * @param {CreateRetrievalChainParams} params A params object
+ *     containing a retriever and a combineDocsChain.
+ * @returns An LCEL Runnable which returns a an object
+ *     containing at least `context` and `answer` keys.
+ * @example
+ * ```typescript
+ * // pnpm add langchain @langchain/openai
+ *
+ * import { ChatOpenAI } from "@langchain/openai";
+ * import { pull } from "langchain/hub";
+ * import { createRetrievalChain } from "@langchain/classic/chains/retrieval";
+ * import { createStuffDocumentsChain } from "@langchain/classic/chains/combine_documents";
+ *
+ * const retrievalQAChatPrompt = await pull("langchain-ai/retrieval-qa-chat");
+ * const llm = new ChatOpenAI({ model: "gpt-4o-mini" });
+ * const retriever = ...
+ * const combineDocsChain = await createStuffDocumentsChain(...);
+ * const retrievalChain = await createRetrievalChain({
+ *   retriever,
+ *   combineDocsChain,
+ * });
+ * const response = await chain.invoke({ input: "..." });
+ * ```
+ */
+declare function createRetrievalChain<RunOutput>({
+  retriever,
+  combineDocsChain
+}: CreateRetrievalChainParams<RunOutput>): Promise<Runnable<{
+  input: string;
+  chat_history?: BaseMessage[] | string;
+} & {
+  [key: string]: unknown;
+}, {
+  context: Document[];
+  answer: RunOutput;
+} & {
+  [key: string]: unknown;
+}>>;
+//#endregion
+export { CreateRetrievalChainParams, createRetrievalChain };
+//# sourceMappingURL=retrieval.d.ts.map

package/dist/chains/retrieval.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"retrieval.d.ts","names":["BaseRetrieverInterface","Runnable","RunnableInterface","BaseMessage","DocumentInterface","Document","CreateRetrievalChainParams","Record","RunOutput","createRetrievalChain","retriever","combineDocsChain","Promise"],"sources":["../../src/chains/retrieval.d.ts"],"sourcesContent":["import type { BaseRetrieverInterface } from \"@langchain/core/retrievers\";\nimport { type Runnable, type RunnableInterface } from \"@langchain/core/runnables\";\nimport type { BaseMessage } from \"@langchain/core/messages\";\nimport type { DocumentInterface, Document } from \"@langchain/core/documents\";\n/**\n * Parameters for the createRetrievalChain method.\n */\nexport type CreateRetrievalChainParams<RunOutput> = {\n    /**\n     * Retriever-like object that returns list of documents. Should\n     * either be a subclass of BaseRetriever or a Runnable that returns\n     * a list of documents. If a subclass of BaseRetriever, then it\n     * is expected that an `input` key be passed in - this is what\n     * is will be used to pass into the retriever. If this is NOT a\n     * subclass of BaseRetriever, then all the inputs will be passed\n     * into this runnable, meaning that runnable should take a object\n     * as input.\n     */\n    retriever: BaseRetrieverInterface\n    // eslint-disable-next-line @typescript-eslint/no-explicit-any\n     | RunnableInterface<Record<string, any>, DocumentInterface[]>;\n    /**\n     * Runnable that takes inputs and produces a string output.\n     * The inputs to this will be any original inputs to this chain, a new\n     * context key with the retrieved documents, and chat_history (if not present\n     * in the inputs) with a value of `[]` (to easily enable conversational\n     * retrieval).\n     */\n    // eslint-disable-next-line @typescript-eslint/no-explicit-any\n    combineDocsChain: RunnableInterface<Record<string, any>, RunOutput>;\n};\n/**\n * Create a retrieval chain that retrieves documents and then passes them on.\n * @param {CreateRetrievalChainParams} params A params object\n *     containing a retriever and a combineDocsChain.\n * @returns An LCEL Runnable which returns a an object\n *     containing at least `context` and `answer` keys.\n * @example\n * ```typescript\n * // pnpm add langchain @langchain/openai\n *\n * import { ChatOpenAI } from \"@langchain/openai\";\n * import { pull } from \"langchain/hub\";\n * import { createRetrievalChain } from \"@langchain/classic/chains/retrieval\";\n * import { createStuffDocumentsChain } from \"@langchain/classic/chains/combine_documents\";\n *\n * const retrievalQAChatPrompt = await pull(\"langchain-ai/retrieval-qa-chat\");\n * const llm = new ChatOpenAI({ model: \"gpt-4o-mini\" });\n * const retriever = ...\n * const combineDocsChain = await createStuffDocumentsChain(...);\n * const retrievalChain = await createRetrievalChain({\n *   retriever,\n *   combineDocsChain,\n * });\n * const response = await chain.invoke({ input: \"...\" });\n * ```\n */\nexport declare function createRetrievalChain<RunOutput>({ retriever, combineDocsChain }: CreateRetrievalChainParams<RunOutput>): Promise<Runnable<{\n    input: string;\n    chat_history?: BaseMessage[] | string;\n} & {\n    [key: string]: unknown;\n}, {\n    context: Document[];\n    answer: RunOutput;\n} & {\n    [key: string]: unknown;\n}>>;\n"],"mappings":";;;;;;;;;AAOA;AAAsC,KAA1BM,0BAA0B,CAAA,SAAA,CAAA,GAAA;EAAA;;;;;;;AAsBC;AA4BvC;;EAA4C,SAAcI,EAvC3CV;EAuCoD;EAAA,EArC5DE,iBAqC6GM,CArC3FD,MAqC2FC,CAAAA,MAAAA,EAAAA,GAAAA,CAAAA,EArCtEJ,iBAqCsEI,EAAAA,CAAAA;EAAS;;;;;;AAAW;;oBA5BlHN,kBAAkBK,qBAAqBC;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBA4BrCC;;;GAAiEH,2BAA2BE,aAAaI,QAAQX;;iBAEtHE;;;;WAINE;UACDG"}

package/dist/chains/retrieval.js

@@ -0,0 +1,49 @@
+import { __export } from "../_virtual/rolldown_runtime.js";
+import { RunnablePassthrough, RunnableSequence } from "@langchain/core/runnables";
+
+//#region src/chains/retrieval.ts
+var retrieval_exports = {};
+__export(retrieval_exports, { createRetrievalChain: () => createRetrievalChain });
+function isBaseRetriever(x) {
+	return !!x && typeof x.invoke === "function";
+}
+/**
+* Create a retrieval chain that retrieves documents and then passes them on.
+* @param {CreateRetrievalChainParams} params A params object
+*     containing a retriever and a combineDocsChain.
+* @returns An LCEL Runnable which returns a an object
+*     containing at least `context` and `answer` keys.
+* @example
+* ```typescript
+* // pnpm add langchain @langchain/openai
+*
+* import { ChatOpenAI } from "@langchain/openai";
+* import { pull } from "langchain/hub";
+* import { createRetrievalChain } from "@langchain/classic/chains/retrieval";
+* import { createStuffDocumentsChain } from "@langchain/classic/chains/combine_documents";
+*
+* const retrievalQAChatPrompt = await pull("langchain-ai/retrieval-qa-chat");
+* const llm = new ChatOpenAI({ model: "gpt-4o-mini" });
+* const retriever = ...
+* const combineDocsChain = await createStuffDocumentsChain(...);
+* const retrievalChain = await createRetrievalChain({
+*   retriever,
+*   combineDocsChain,
+* });
+* const response = await chain.invoke({ input: "..." });
+* ```
+*/
+async function createRetrievalChain({ retriever, combineDocsChain }) {
+	let retrieveDocumentsChain;
+	if (isBaseRetriever(retriever)) retrieveDocumentsChain = RunnableSequence.from([(input) => input.input, retriever]);
+	else retrieveDocumentsChain = retriever;
+	const retrievalChain = RunnableSequence.from([RunnablePassthrough.assign({
+		context: retrieveDocumentsChain.withConfig({ runName: "retrieve_documents" }),
+		chat_history: (input) => input.chat_history ?? []
+	}), RunnablePassthrough.assign({ answer: combineDocsChain })]).withConfig({ runName: "retrieval_chain" });
+	return retrievalChain;
+}
+
+//#endregion
+export { createRetrievalChain, retrieval_exports };
+//# sourceMappingURL=retrieval.js.map

package/dist/chains/retrieval.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"retrieval.js","names":["x: unknown","retrieveDocumentsChain: Runnable<{ input: string }, DocumentInterface[]>"],"sources":["../../src/chains/retrieval.ts"],"sourcesContent":["import type { BaseRetrieverInterface } from \"@langchain/core/retrievers\";\nimport {\n  type Runnable,\n  RunnableSequence,\n  type RunnableInterface,\n  RunnablePassthrough,\n} from \"@langchain/core/runnables\";\nimport type { BaseMessage } from \"@langchain/core/messages\";\nimport type { DocumentInterface, Document } from \"@langchain/core/documents\";\n\n/**\n * Parameters for the createRetrievalChain method.\n */\nexport type CreateRetrievalChainParams<RunOutput> = {\n  /**\n   * Retriever-like object that returns list of documents. Should\n   * either be a subclass of BaseRetriever or a Runnable that returns\n   * a list of documents. If a subclass of BaseRetriever, then it\n   * is expected that an `input` key be passed in - this is what\n   * is will be used to pass into the retriever. If this is NOT a\n   * subclass of BaseRetriever, then all the inputs will be passed\n   * into this runnable, meaning that runnable should take a object\n   * as input.\n   */\n  retriever:\n    | BaseRetrieverInterface\n    // eslint-disable-next-line @typescript-eslint/no-explicit-any\n    | RunnableInterface<Record<string, any>, DocumentInterface[]>;\n  /**\n   * Runnable that takes inputs and produces a string output.\n   * The inputs to this will be any original inputs to this chain, a new\n   * context key with the retrieved documents, and chat_history (if not present\n   * in the inputs) with a value of `[]` (to easily enable conversational\n   * retrieval).\n   */\n  // eslint-disable-next-line @typescript-eslint/no-explicit-any\n  combineDocsChain: RunnableInterface<Record<string, any>, RunOutput>;\n};\n\nfunction isBaseRetriever(x: unknown): x is BaseRetrieverInterface {\n  return !!x && typeof (x as BaseRetrieverInterface).invoke === \"function\";\n}\n\n/**\n * Create a retrieval chain that retrieves documents and then passes them on.\n * @param {CreateRetrievalChainParams} params A params object\n *     containing a retriever and a combineDocsChain.\n * @returns An LCEL Runnable which returns a an object\n *     containing at least `context` and `answer` keys.\n * @example\n * ```typescript\n * // pnpm add langchain @langchain/openai\n *\n * import { ChatOpenAI } from \"@langchain/openai\";\n * import { pull } from \"langchain/hub\";\n * import { createRetrievalChain } from \"@langchain/classic/chains/retrieval\";\n * import { createStuffDocumentsChain } from \"@langchain/classic/chains/combine_documents\";\n *\n * const retrievalQAChatPrompt = await pull(\"langchain-ai/retrieval-qa-chat\");\n * const llm = new ChatOpenAI({ model: \"gpt-4o-mini\" });\n * const retriever = ...\n * const combineDocsChain = await createStuffDocumentsChain(...);\n * const retrievalChain = await createRetrievalChain({\n *   retriever,\n *   combineDocsChain,\n * });\n * const response = await chain.invoke({ input: \"...\" });\n * ```\n */\nexport async function createRetrievalChain<RunOutput>({\n  retriever,\n  combineDocsChain,\n}: CreateRetrievalChainParams<RunOutput>): Promise<\n  Runnable<\n    { input: string; chat_history?: BaseMessage[] | string } & {\n      [key: string]: unknown;\n    },\n    { context: Document[]; answer: RunOutput } & { [key: string]: unknown }\n  >\n> {\n  let retrieveDocumentsChain: Runnable<{ input: string }, DocumentInterface[]>;\n  if (isBaseRetriever(retriever)) {\n    retrieveDocumentsChain = RunnableSequence.from([\n      (input) => input.input,\n      retriever,\n    ]);\n  } else {\n    // TODO: Fix typing by adding withConfig to core RunnableInterface\n    retrieveDocumentsChain = retriever as Runnable;\n  }\n  const retrievalChain = RunnableSequence.from<{\n    input: string;\n    chat_history?: BaseMessage[] | string;\n  }>([\n    RunnablePassthrough.assign({\n      context: retrieveDocumentsChain.withConfig({\n        runName: \"retrieve_documents\",\n      }),\n      chat_history: (input) => input.chat_history ?? [],\n    }),\n    RunnablePassthrough.assign({\n      answer: combineDocsChain,\n    }),\n  ]).withConfig({ runName: \"retrieval_chain\" });\n  return retrievalChain;\n}\n"],"mappings":";;;;;;AAuCA,SAAS,gBAAgBA,GAAyC;AAChE,QAAO,CAAC,CAAC,KAAK,OAAQ,EAA6B,WAAW;AAC/D;;;;;;;;;;;;;;;;;;;;;;;;;;;AA4BD,eAAsB,qBAAgC,EACpD,WACA,kBACsC,EAOtC;CACA,IAAIC;AACJ,KAAI,gBAAgB,UAAU,EAC5B,yBAAyB,iBAAiB,KAAK,CAC7C,CAAC,UAAU,MAAM,OACjB,SACD,EAAC;MAGF,yBAAyB;CAE3B,MAAM,iBAAiB,iBAAiB,KAGrC,CACD,oBAAoB,OAAO;EACzB,SAAS,uBAAuB,WAAW,EACzC,SAAS,qBACV,EAAC;EACF,cAAc,CAAC,UAAU,MAAM,gBAAgB,CAAE;CAClD,EAAC,EACF,oBAAoB,OAAO,EACzB,QAAQ,iBACT,EAAC,AACH,EAAC,CAAC,WAAW,EAAE,SAAS,kBAAmB,EAAC;AAC7C,QAAO;AACR"}

package/dist/chains/retrieval_qa.cjs

@@ -0,0 +1,104 @@
+const require_base = require('./base.cjs');
+const require_load = require('./question_answering/load.cjs');
+
+//#region src/chains/retrieval_qa.ts
+/**
+* Class representing a chain for performing question-answering tasks with
+* a retrieval component.
+* @example
+* ```typescript
+* import { createStuffDocumentsChain } from "@langchain/classic/chains/combine_documents";
+* import { ChatPromptTemplate } from "@langchain/core/prompts";
+* import { createRetrievalChain } from "@langchain/classic/chains/retrieval";
+* import { MemoryVectorStore } from "@langchain/classic/vectorstores/memory";
+*
+* const documents = [...your documents here];
+* const embeddings = ...your embeddings model;
+* const llm = ...your LLM model;
+*
+* const vectorstore = await MemoryVectorStore.fromDocuments(
+*   documents,
+*   embeddings
+* );
+* const prompt = ChatPromptTemplate.fromTemplate(`Answer the user's question: {input} based on the following context {context}`);
+*
+* const combineDocsChain = await createStuffDocumentsChain({
+*   llm,
+*   prompt,
+* });
+* const retriever = vectorstore.asRetriever();
+*
+* const retrievalChain = await createRetrievalChain({
+*   combineDocsChain,
+*   retriever,
+* });
+* ```
+*/
+var RetrievalQAChain = class extends require_base.BaseChain {
+	static lc_name() {
+		return "RetrievalQAChain";
+	}
+	inputKey = "query";
+	get inputKeys() {
+		return [this.inputKey];
+	}
+	get outputKeys() {
+		return this.combineDocumentsChain.outputKeys.concat(this.returnSourceDocuments ? ["sourceDocuments"] : []);
+	}
+	retriever;
+	combineDocumentsChain;
+	returnSourceDocuments = false;
+	constructor(fields) {
+		super(fields);
+		this.retriever = fields.retriever;
+		this.combineDocumentsChain = fields.combineDocumentsChain;
+		this.inputKey = fields.inputKey ?? this.inputKey;
+		this.returnSourceDocuments = fields.returnSourceDocuments ?? this.returnSourceDocuments;
+	}
+	/** @ignore */
+	async _call(values, runManager) {
+		if (!(this.inputKey in values)) throw new Error(`Question key "${this.inputKey}" not found.`);
+		const question = values[this.inputKey];
+		const docs = await this.retriever.invoke(question, runManager?.getChild("retriever"));
+		const inputs = {
+			question,
+			input_documents: docs,
+			...values
+		};
+		const result = await this.combineDocumentsChain.call(inputs, runManager?.getChild("combine_documents"));
+		if (this.returnSourceDocuments) return {
+			...result,
+			sourceDocuments: docs
+		};
+		return result;
+	}
+	_chainType() {
+		return "retrieval_qa";
+	}
+	static async deserialize(_data, _values) {
+		throw new Error("Not implemented");
+	}
+	serialize() {
+		throw new Error("Not implemented");
+	}
+	/**
+	* Creates a new instance of RetrievalQAChain using a BaseLanguageModel
+	* and a BaseRetriever.
+	* @param llm The BaseLanguageModel used to generate a new question.
+	* @param retriever The BaseRetriever used to retrieve relevant documents.
+	* @param options Optional parameters for the RetrievalQAChain.
+	* @returns A new instance of RetrievalQAChain.
+	*/
+	static fromLLM(llm, retriever, options) {
+		const qaChain = require_load.loadQAStuffChain(llm, { prompt: options?.prompt });
+		return new this({
+			...options,
+			retriever,
+			combineDocumentsChain: qaChain
+		});
+	}
+};
+
+//#endregion
+exports.RetrievalQAChain = RetrievalQAChain;
+//# sourceMappingURL=retrieval_qa.cjs.map

package/dist/chains/retrieval_qa.cjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"retrieval_qa.cjs","names":["BaseChain","fields: RetrievalQAChainInput","values: ChainValues","runManager?: CallbackManagerForChainRun","question: string","_data: SerializedVectorDBQAChain","_values: LoadValues","llm: BaseLanguageModelInterface","retriever: BaseRetrieverInterface","options?: Partial<\n      Omit<\n        RetrievalQAChainInput,\n        \"retriever\" | \"combineDocumentsChain\" | \"index\"\n      >\n    > &\n      StuffQAChainParams","loadQAStuffChain"],"sources":["../../src/chains/retrieval_qa.ts"],"sourcesContent":["import type { BaseLanguageModelInterface } from \"@langchain/core/language_models/base\";\nimport type { BaseRetrieverInterface } from \"@langchain/core/retrievers\";\nimport { ChainValues } from \"@langchain/core/utils/types\";\nimport { CallbackManagerForChainRun } from \"@langchain/core/callbacks/manager\";\nimport { BaseChain, ChainInputs } from \"./base.js\";\nimport { SerializedVectorDBQAChain } from \"./serde.js\";\nimport {\n  StuffQAChainParams,\n  loadQAStuffChain,\n} from \"./question_answering/load.js\";\n\n// eslint-disable-next-line @typescript-eslint/no-explicit-any\nexport type LoadValues = Record<string, any>;\n\n/**\n * Interface for the input parameters of the RetrievalQAChain class.\n */\nexport interface RetrievalQAChainInput extends Omit<ChainInputs, \"memory\"> {\n  retriever: BaseRetrieverInterface;\n  combineDocumentsChain: BaseChain;\n  inputKey?: string;\n  returnSourceDocuments?: boolean;\n}\n\n/**\n * Class representing a chain for performing question-answering tasks with\n * a retrieval component.\n * @example\n * ```typescript\n * import { createStuffDocumentsChain } from \"@langchain/classic/chains/combine_documents\";\n * import { ChatPromptTemplate } from \"@langchain/core/prompts\";\n * import { createRetrievalChain } from \"@langchain/classic/chains/retrieval\";\n * import { MemoryVectorStore } from \"@langchain/classic/vectorstores/memory\";\n *\n * const documents = [...your documents here];\n * const embeddings = ...your embeddings model;\n * const llm = ...your LLM model;\n *\n * const vectorstore = await MemoryVectorStore.fromDocuments(\n *   documents,\n *   embeddings\n * );\n * const prompt = ChatPromptTemplate.fromTemplate(`Answer the user's question: {input} based on the following context {context}`);\n *\n * const combineDocsChain = await createStuffDocumentsChain({\n *   llm,\n *   prompt,\n * });\n * const retriever = vectorstore.asRetriever();\n *\n * const retrievalChain = await createRetrievalChain({\n *   combineDocsChain,\n *   retriever,\n * });\n * ```\n */\nexport class RetrievalQAChain\n  extends BaseChain\n  implements RetrievalQAChainInput\n{\n  static lc_name() {\n    return \"RetrievalQAChain\";\n  }\n\n  inputKey = \"query\";\n\n  get inputKeys() {\n    return [this.inputKey];\n  }\n\n  get outputKeys() {\n    return this.combineDocumentsChain.outputKeys.concat(\n      this.returnSourceDocuments ? [\"sourceDocuments\"] : []\n    );\n  }\n\n  retriever: BaseRetrieverInterface;\n\n  combineDocumentsChain: BaseChain;\n\n  returnSourceDocuments = false;\n\n  constructor(fields: RetrievalQAChainInput) {\n    super(fields);\n    this.retriever = fields.retriever;\n    this.combineDocumentsChain = fields.combineDocumentsChain;\n    this.inputKey = fields.inputKey ?? this.inputKey;\n    this.returnSourceDocuments =\n      fields.returnSourceDocuments ?? this.returnSourceDocuments;\n  }\n\n  /** @ignore */\n  async _call(\n    values: ChainValues,\n    runManager?: CallbackManagerForChainRun\n  ): Promise<ChainValues> {\n    if (!(this.inputKey in values)) {\n      throw new Error(`Question key \"${this.inputKey}\" not found.`);\n    }\n    const question: string = values[this.inputKey];\n    const docs = await this.retriever.invoke(\n      question,\n      runManager?.getChild(\"retriever\")\n    );\n    const inputs = { question, input_documents: docs, ...values };\n    const result = await this.combineDocumentsChain.call(\n      inputs,\n      runManager?.getChild(\"combine_documents\")\n    );\n    if (this.returnSourceDocuments) {\n      return {\n        ...result,\n        sourceDocuments: docs,\n      };\n    }\n    return result;\n  }\n\n  _chainType() {\n    return \"retrieval_qa\" as const;\n  }\n\n  static async deserialize(\n    _data: SerializedVectorDBQAChain,\n    _values: LoadValues\n  ): Promise<RetrievalQAChain> {\n    throw new Error(\"Not implemented\");\n  }\n\n  serialize(): SerializedVectorDBQAChain {\n    throw new Error(\"Not implemented\");\n  }\n\n  /**\n   * Creates a new instance of RetrievalQAChain using a BaseLanguageModel\n   * and a BaseRetriever.\n   * @param llm The BaseLanguageModel used to generate a new question.\n   * @param retriever The BaseRetriever used to retrieve relevant documents.\n   * @param options Optional parameters for the RetrievalQAChain.\n   * @returns A new instance of RetrievalQAChain.\n   */\n  static fromLLM(\n    llm: BaseLanguageModelInterface,\n    retriever: BaseRetrieverInterface,\n    options?: Partial<\n      Omit<\n        RetrievalQAChainInput,\n        \"retriever\" | \"combineDocumentsChain\" | \"index\"\n      >\n    > &\n      StuffQAChainParams\n  ): RetrievalQAChain {\n    const qaChain = loadQAStuffChain(llm, {\n      prompt: options?.prompt,\n    });\n    return new this({\n      ...options,\n      retriever,\n      combineDocumentsChain: qaChain,\n    });\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAwDA,IAAa,mBAAb,cACUA,uBAEV;CACE,OAAO,UAAU;AACf,SAAO;CACR;CAED,WAAW;CAEX,IAAI,YAAY;AACd,SAAO,CAAC,KAAK,QAAS;CACvB;CAED,IAAI,aAAa;AACf,SAAO,KAAK,sBAAsB,WAAW,OAC3C,KAAK,wBAAwB,CAAC,iBAAkB,IAAG,CAAE,EACtD;CACF;CAED;CAEA;CAEA,wBAAwB;CAExB,YAAYC,QAA+B;EACzC,MAAM,OAAO;EACb,KAAK,YAAY,OAAO;EACxB,KAAK,wBAAwB,OAAO;EACpC,KAAK,WAAW,OAAO,YAAY,KAAK;EACxC,KAAK,wBACH,OAAO,yBAAyB,KAAK;CACxC;;CAGD,MAAM,MACJC,QACAC,YACsB;AACtB,MAAI,EAAE,KAAK,YAAY,QACrB,OAAM,IAAI,MAAM,CAAC,cAAc,EAAE,KAAK,SAAS,YAAY,CAAC;EAE9D,MAAMC,WAAmB,OAAO,KAAK;EACrC,MAAM,OAAO,MAAM,KAAK,UAAU,OAChC,UACA,YAAY,SAAS,YAAY,CAClC;EACD,MAAM,SAAS;GAAE;GAAU,iBAAiB;GAAM,GAAG;EAAQ;EAC7D,MAAM,SAAS,MAAM,KAAK,sBAAsB,KAC9C,QACA,YAAY,SAAS,oBAAoB,CAC1C;AACD,MAAI,KAAK,sBACP,QAAO;GACL,GAAG;GACH,iBAAiB;EAClB;AAEH,SAAO;CACR;CAED,aAAa;AACX,SAAO;CACR;CAED,aAAa,YACXC,OACAC,SAC2B;AAC3B,QAAM,IAAI,MAAM;CACjB;CAED,YAAuC;AACrC,QAAM,IAAI,MAAM;CACjB;;;;;;;;;CAUD,OAAO,QACLC,KACAC,WACAC,SAOkB;EAClB,MAAM,UAAUC,8BAAiB,KAAK,EACpC,QAAQ,SAAS,OAClB,EAAC;AACF,SAAO,IAAI,KAAK;GACd,GAAG;GACH;GACA,uBAAuB;EACxB;CACF;AACF"}