@langchain/core 0.1.22 → 0.1.23

package/dist/output_parsers/index.cjs

@@ -20,3 +20,4 @@ __exportStar(require("./list.cjs"), exports);
 __exportStar(require("./string.cjs"), exports);
 __exportStar(require("./transform.cjs"), exports);
 __exportStar(require("./json.cjs"), exports);
+__exportStar(require("./xml.cjs"), exports);

package/dist/output_parsers/index.d.ts

@@ -4,3 +4,4 @@ export * from "./list.js";
 export * from "./string.js";
 export * from "./transform.js";
 export * from "./json.js";
+export * from "./xml.js";

package/dist/output_parsers/index.js

@@ -4,3 +4,4 @@ export * from "./list.js";
 export * from "./string.js";
 export * from "./transform.js";
 export * from "./json.js";
+export * from "./xml.js";

package/dist/output_parsers/json.cjs

@@ -57,7 +57,7 @@ function parseJsonMarkdown(s, parser = parsePartialJson) {
     }
 }
 exports.parseJsonMarkdown = parseJsonMarkdown;
-// Adapted from https://github.com/KillianLucas/open-interpreter/blob/main/interpreter/utils/parse_partial_json.py
+// Adapted from https://github.com/KillianLucas/open-interpreter/blob/main/interpreter/core/llm/utils/parse_partial_json.py
 // MIT License
 function parsePartialJson(s) {
     // If the input is undefined, return null to indicate failure.

package/dist/output_parsers/json.js

@@ -52,7 +52,7 @@ export function parseJsonMarkdown(s, parser = parsePartialJson) {
         return parser(match[2]);
     }
 }
-// Adapted from https://github.com/KillianLucas/open-interpreter/blob/main/interpreter/utils/parse_partial_json.py
+// Adapted from https://github.com/KillianLucas/open-interpreter/blob/main/interpreter/core/llm/utils/parse_partial_json.py
 // MIT License
 export function parsePartialJson(s) {
     // If the input is undefined, return null to indicate failure.

package/dist/output_parsers/xml.cjs

@@ -0,0 +1,155 @@
+"use strict";
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.parseXMLMarkdown = exports.XMLOutputParser = exports.XML_FORMAT_INSTRUCTIONS = void 0;
+const sax_1 = __importDefault(require("sax"));
+const transform_js_1 = require("./transform.cjs");
+const json_patch_js_1 = require("../utils/json_patch.cjs");
+exports.XML_FORMAT_INSTRUCTIONS = `The output should be formatted as a XML file.
+1. Output should conform to the tags below. 
+2. If tags are not given, make them on your own.
+3. Remember to always open and close all the tags.
+
+As an example, for the tags ["foo", "bar", "baz"]:
+1. String "<foo>\n   <bar>\n      <baz></baz>\n   </bar>\n</foo>" is a well-formatted instance of the schema. 
+2. String "<foo>\n   <bar>\n   </foo>" is a badly-formatted instance.
+3. String "<foo>\n   <tag>\n   </tag>\n</foo>" is a badly-formatted instance.
+
+Here are the output tags:
+\`\`\`
+{tags}
+\`\`\``;
+class XMLOutputParser extends transform_js_1.BaseCumulativeTransformOutputParser {
+    constructor(fields) {
+        const f = fields ?? {};
+        const { tags, ...saxOptions } = f;
+        super(f);
+        Object.defineProperty(this, "tags", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: void 0
+        });
+        Object.defineProperty(this, "saxOptions", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: void 0
+        });
+        Object.defineProperty(this, "lc_namespace", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: ["langchain_core", "output_parsers"]
+        });
+        Object.defineProperty(this, "lc_serializable", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: true
+        });
+        this.tags = tags;
+        this.saxOptions = saxOptions;
+    }
+    static lc_name() {
+        return "XMLOutputParser";
+    }
+    _diff(prev, next) {
+        if (!next) {
+            return undefined;
+        }
+        if (!prev) {
+            return [{ op: "replace", path: "", value: next }];
+        }
+        return (0, json_patch_js_1.compare)(prev, next);
+    }
+    async parsePartialResult(generations) {
+        return parseXMLMarkdown(generations[0].text, this.saxOptions);
+    }
+    async parse(text) {
+        return parseXMLMarkdown(text, this.saxOptions);
+    }
+    getFormatInstructions() {
+        const withTags = !!(this.tags && this.tags.length > 0);
+        return withTags
+            ? exports.XML_FORMAT_INSTRUCTIONS.replace("{tags}", this.tags?.join(", ") ?? "")
+            : exports.XML_FORMAT_INSTRUCTIONS;
+    }
+}
+exports.XMLOutputParser = XMLOutputParser;
+const strip = (text) => text
+    .split("\n")
+    .map((line) => line.replace(/^\s+/, ""))
+    .join("\n")
+    .trim();
+const parseParsedResult = (input) => {
+    if (Object.keys(input).length === 0) {
+        return {};
+    }
+    const result = {};
+    if (input.children.length > 0) {
+        result[input.name] = input.children.map(parseParsedResult);
+        return result;
+    }
+    else {
+        result[input.name] = input.text ?? undefined;
+        return result;
+    }
+};
+function parseXMLMarkdown(s, saxOptions) {
+    const cleanedString = strip(s);
+    const parser = sax_1.default.parser(true, saxOptions);
+    let parsedResult = {};
+    const elementStack = [];
+    parser.onopentag = (node) => {
+        const element = {
+            name: node.name,
+            attributes: node.attributes,
+            children: [],
+            text: "",
+            isSelfClosing: node.isSelfClosing,
+        };
+        if (elementStack.length > 0) {
+            const parentElement = elementStack[elementStack.length - 1];
+            parentElement.children.push(element);
+        }
+        else {
+            parsedResult = element;
+        }
+        if (!node.isSelfClosing) {
+            elementStack.push(element);
+        }
+    };
+    parser.onclosetag = () => {
+        if (elementStack.length > 0) {
+            const lastElement = elementStack.pop();
+            if (elementStack.length === 0 && lastElement) {
+                parsedResult = lastElement;
+            }
+        }
+    };
+    parser.ontext = (text) => {
+        if (elementStack.length > 0) {
+            const currentElement = elementStack[elementStack.length - 1];
+            currentElement.text += text;
+        }
+    };
+    parser.onattribute = (attr) => {
+        if (elementStack.length > 0) {
+            const currentElement = elementStack[elementStack.length - 1];
+            currentElement.attributes[attr.name] = attr.value;
+        }
+    };
+    // Try to find XML string within triple backticks.
+    const match = /```(xml)?(.*)```/s.exec(cleanedString);
+    const xmlString = match ? match[2] : cleanedString;
+    parser.write(xmlString).close();
+    // Remove the XML declaration if present
+    if (parsedResult && parsedResult.name === "?xml") {
+        parsedResult = parsedResult.children[0];
+    }
+    return parseParsedResult(parsedResult);
+}
+exports.parseXMLMarkdown = parseXMLMarkdown;

package/dist/output_parsers/xml.d.ts

@@ -0,0 +1,31 @@
+import { SAXOptions } from "sax";
+import { BaseCumulativeTransformOutputParser, BaseCumulativeTransformOutputParserInput } from "./transform.js";
+import { Operation } from "../utils/json_patch.js";
+import { ChatGeneration, Generation } from "../outputs.js";
+export declare const XML_FORMAT_INSTRUCTIONS = "The output should be formatted as a XML file.\n1. Output should conform to the tags below. \n2. If tags are not given, make them on your own.\n3. Remember to always open and close all the tags.\n\nAs an example, for the tags [\"foo\", \"bar\", \"baz\"]:\n1. String \"<foo>\n   <bar>\n      <baz></baz>\n   </bar>\n</foo>\" is a well-formatted instance of the schema. \n2. String \"<foo>\n   <bar>\n   </foo>\" is a badly-formatted instance.\n3. String \"<foo>\n   <tag>\n   </tag>\n</foo>\" is a badly-formatted instance.\n\nHere are the output tags:\n```\n{tags}\n```";
+export interface XMLOutputParserFields extends SAXOptions, BaseCumulativeTransformOutputParserInput {
+    /**
+     * Optional list of tags that the output should conform to.
+     * Only used in formatting of the prompt.
+     */
+    tags?: string[];
+}
+export type Content = string | undefined | Array<{
+    [key: string]: Content;
+}>;
+export type XMLResult = {
+    [key: string]: Content;
+};
+export declare class XMLOutputParser extends BaseCumulativeTransformOutputParser<XMLResult> {
+    tags?: string[];
+    saxOptions?: SAXOptions;
+    constructor(fields?: XMLOutputParserFields);
+    static lc_name(): string;
+    lc_namespace: string[];
+    lc_serializable: boolean;
+    protected _diff(prev: unknown | undefined, next: unknown): Operation[] | undefined;
+    parsePartialResult(generations: ChatGeneration[] | Generation[]): Promise<XMLResult | undefined>;
+    parse(text: string): Promise<XMLResult>;
+    getFormatInstructions(): string;
+}
+export declare function parseXMLMarkdown(s: string, saxOptions?: SAXOptions): XMLResult;

package/dist/output_parsers/xml.js

@@ -0,0 +1,147 @@
+import sax from "sax";
+import { BaseCumulativeTransformOutputParser, } from "./transform.js";
+import { compare } from "../utils/json_patch.js";
+export const XML_FORMAT_INSTRUCTIONS = `The output should be formatted as a XML file.
+1. Output should conform to the tags below. 
+2. If tags are not given, make them on your own.
+3. Remember to always open and close all the tags.
+
+As an example, for the tags ["foo", "bar", "baz"]:
+1. String "<foo>\n   <bar>\n      <baz></baz>\n   </bar>\n</foo>" is a well-formatted instance of the schema. 
+2. String "<foo>\n   <bar>\n   </foo>" is a badly-formatted instance.
+3. String "<foo>\n   <tag>\n   </tag>\n</foo>" is a badly-formatted instance.
+
+Here are the output tags:
+\`\`\`
+{tags}
+\`\`\``;
+export class XMLOutputParser extends BaseCumulativeTransformOutputParser {
+    constructor(fields) {
+        const f = fields ?? {};
+        const { tags, ...saxOptions } = f;
+        super(f);
+        Object.defineProperty(this, "tags", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: void 0
+        });
+        Object.defineProperty(this, "saxOptions", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: void 0
+        });
+        Object.defineProperty(this, "lc_namespace", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: ["langchain_core", "output_parsers"]
+        });
+        Object.defineProperty(this, "lc_serializable", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: true
+        });
+        this.tags = tags;
+        this.saxOptions = saxOptions;
+    }
+    static lc_name() {
+        return "XMLOutputParser";
+    }
+    _diff(prev, next) {
+        if (!next) {
+            return undefined;
+        }
+        if (!prev) {
+            return [{ op: "replace", path: "", value: next }];
+        }
+        return compare(prev, next);
+    }
+    async parsePartialResult(generations) {
+        return parseXMLMarkdown(generations[0].text, this.saxOptions);
+    }
+    async parse(text) {
+        return parseXMLMarkdown(text, this.saxOptions);
+    }
+    getFormatInstructions() {
+        const withTags = !!(this.tags && this.tags.length > 0);
+        return withTags
+            ? XML_FORMAT_INSTRUCTIONS.replace("{tags}", this.tags?.join(", ") ?? "")
+            : XML_FORMAT_INSTRUCTIONS;
+    }
+}
+const strip = (text) => text
+    .split("\n")
+    .map((line) => line.replace(/^\s+/, ""))
+    .join("\n")
+    .trim();
+const parseParsedResult = (input) => {
+    if (Object.keys(input).length === 0) {
+        return {};
+    }
+    const result = {};
+    if (input.children.length > 0) {
+        result[input.name] = input.children.map(parseParsedResult);
+        return result;
+    }
+    else {
+        result[input.name] = input.text ?? undefined;
+        return result;
+    }
+};
+export function parseXMLMarkdown(s, saxOptions) {
+    const cleanedString = strip(s);
+    const parser = sax.parser(true, saxOptions);
+    let parsedResult = {};
+    const elementStack = [];
+    parser.onopentag = (node) => {
+        const element = {
+            name: node.name,
+            attributes: node.attributes,
+            children: [],
+            text: "",
+            isSelfClosing: node.isSelfClosing,
+        };
+        if (elementStack.length > 0) {
+            const parentElement = elementStack[elementStack.length - 1];
+            parentElement.children.push(element);
+        }
+        else {
+            parsedResult = element;
+        }
+        if (!node.isSelfClosing) {
+            elementStack.push(element);
+        }
+    };
+    parser.onclosetag = () => {
+        if (elementStack.length > 0) {
+            const lastElement = elementStack.pop();
+            if (elementStack.length === 0 && lastElement) {
+                parsedResult = lastElement;
+            }
+        }
+    };
+    parser.ontext = (text) => {
+        if (elementStack.length > 0) {
+            const currentElement = elementStack[elementStack.length - 1];
+            currentElement.text += text;
+        }
+    };
+    parser.onattribute = (attr) => {
+        if (elementStack.length > 0) {
+            const currentElement = elementStack[elementStack.length - 1];
+            currentElement.attributes[attr.name] = attr.value;
+        }
+    };
+    // Try to find XML string within triple backticks.
+    const match = /```(xml)?(.*)```/s.exec(cleanedString);
+    const xmlString = match ? match[2] : cleanedString;
+    parser.write(xmlString).close();
+    // Remove the XML declaration if present
+    if (parsedResult && parsedResult.name === "?xml") {
+        parsedResult = parsedResult.children[0];
+    }
+    return parseParsedResult(parsedResult);
+}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@langchain/core",
-  "version": "0.1.22",
+  "version": "0.1.23",
   "description": "Core LangChain.js abstractions and schemas",
   "type": "module",
   "engines": {
@@ -23,7 +23,7 @@
     "lint:dpdm": "dpdm --exit-code circular:1 --no-warning --no-tree src/*.ts src/**/*.ts",
     "lint": "yarn lint:eslint && yarn lint:dpdm",
     "lint:fix": "yarn lint:eslint --fix && yarn lint:dpdm",
-    "clean": "rimraf .turbo/ dist/ && NODE_OPTIONS=--max-old-space-size=4096 yarn create-entrypoints -- --pre",
+    "clean": "rimraf .turbo/ dist/ && NODE_OPTIONS=--max-old-space-size=4096 yarn lc-build --config ./langchain.config.js --create-entrypoints --pre",
     "prepack": "yarn build",
     "release": "release-it --only-version --config .release-it.json",
     "test": "NODE_OPTIONS=--experimental-vm-modules jest --testPathIgnorePatterns=\\.int\\.test.ts --testTimeout 30000 --maxWorkers=50%",
@@ -46,6 +46,7 @@
     "ml-distance": "^4.0.0",
     "p-queue": "^6.6.2",
     "p-retry": "4",
+    "sax": "^1.3.0",
     "uuid": "^9.0.0",
     "zod": "^3.22.4",
     "zod-to-json-schema": "^3.22.3"
@@ -55,6 +56,7 @@
     "@langchain/scripts": "~0.0",
     "@swc/core": "^1.3.90",
     "@swc/jest": "^0.2.29",
+    "@types/sax": "^1",
     "dpdm": "^3.12.0",
     "eslint": "^8.33.0",
     "eslint-config-airbnb-base": "^15.0.0",