@xcrap/extractor 0.3.0

package/dist/index.js

@@ -0,0 +1,23 @@
+"use strict";
+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]; } };
+    }
+    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("./errors"), exports);
+__exportStar(require("./html"), exports);
+__exportStar(require("./json"), exports);
+__exportStar(require("./source-parser"), exports);
+__exportStar(require("./interfaces/extractor-model"), exports);
+__exportStar(require("./markdown"), exports);
+__exportStar(require("./query-builders"), exports);

package/dist/interfaces/extractor-model.d.ts

@@ -0,0 +1,3 @@
+export interface ExtractorModel<T = any> {
+    extract(source: string): Promise<T> | T;
+}

package/dist/interfaces/extractor-model.js

@@ -0,0 +1,2 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });

package/dist/json/extractor-model.d.ts

@@ -0,0 +1,22 @@
+import { ExtractorModel } from "../interfaces/extractor-model";
+export type JsonExtractorModelShapeValue = {
+    query: string;
+    default?: any;
+    model?: ExtractorModel;
+    multiple?: boolean;
+    limit?: number;
+};
+export type JsonExtractorModelShape = {
+    [key: string]: JsonExtractorModelShapeValue;
+};
+export type InferJsonValue<V extends JsonExtractorModelShapeValue> = V["model"] extends ExtractorModel<infer M> ? (V["multiple"] extends true ? M[] : M) : any;
+export type InferJsonShape<S extends JsonExtractorModelShape> = {
+    [K in keyof S]: InferJsonValue<S[K]>;
+};
+export declare class JsonExtractorModel<S extends JsonExtractorModelShape> implements ExtractorModel<InferJsonShape<S>> {
+    readonly shape: S;
+    constructor(shape: S);
+    extract(source: string): Promise<InferJsonShape<S>>;
+    extractValue(value: JsonExtractorModelShapeValue, root: any): any;
+    extractNestedValue(value: JsonExtractorModelShapeValue, root: any): Promise<any>;
+}

package/dist/json/extractor-model.js

@@ -0,0 +1,97 @@
+"use strict";
+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]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.JsonExtractorModel = void 0;
+const jmespath = __importStar(require("jmespath"));
+class JsonExtractorModel {
+    constructor(shape) {
+        this.shape = shape;
+    }
+    async extract(source) {
+        const root = JSON.parse(source);
+        const data = {};
+        for (const key in this.shape) {
+            const value = this.shape[key];
+            const isNestedValue = "model" in value;
+            if (isNestedValue) {
+                data[key] = await this.extractNestedValue(value, root);
+            }
+            else {
+                data[key] = this.extractValue(value, root);
+            }
+        }
+        return data;
+    }
+    extractValue(value, root) {
+        const extractedData = jmespath.search(root, value.query);
+        if (extractedData === null && value.default !== undefined) {
+            return value.default;
+        }
+        return extractedData;
+    }
+    async extractNestedValue(value, root) {
+        const extractedData = jmespath.search(root, value.query);
+        const model = value.model;
+        const modelIsJsonExtractor = model.constructor.name === JsonExtractorModel.name;
+        if (extractedData === null && value.default !== undefined) {
+            return value.default;
+        }
+        if (value.multiple) {
+            if (!Array.isArray(extractedData)) {
+                throw new Error(`Expected an array for multiple values, but got ${typeof extractedData}`);
+            }
+            if (value.limit !== undefined) {
+                extractedData.splice(value.limit);
+            }
+            if (!modelIsJsonExtractor) {
+                if (extractedData.some((item) => typeof item !== "string")) {
+                    throw new Error(`Expected an array of strings for model parsing, but got ${typeof extractedData[0]}`);
+                }
+                return await Promise.all(extractedData.map((item) => model.extract(JSON.stringify(item))));
+            }
+        }
+        else {
+            if (!modelIsJsonExtractor && typeof extractedData !== "string") {
+                throw new Error(`Expected a string for model parsing, but got ${typeof extractedData}`);
+            }
+            if (!modelIsJsonExtractor) {
+                return await model.extract(extractedData);
+            }
+            return await model.extract(JSON.stringify(extractedData));
+        }
+        return extractedData;
+    }
+}
+exports.JsonExtractorModel = JsonExtractorModel;

package/dist/json/index.d.ts

@@ -0,0 +1,2 @@
+export * from "./extractor-model";
+export * from "./parser";

package/dist/json/index.js

@@ -0,0 +1,18 @@
+"use strict";
+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]; } };
+    }
+    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("./extractor-model"), exports);
+__exportStar(require("./parser"), exports);

package/dist/json/parser.d.ts

@@ -0,0 +1,6 @@
+import { SourceParser } from "../source-parser";
+export declare class JsonParser extends SourceParser {
+    readonly data: any;
+    constructor(source: string);
+    extract(query: string): any;
+}

package/dist/json/parser.js

@@ -0,0 +1,48 @@
+"use strict";
+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]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.JsonParser = void 0;
+const jmespath = __importStar(require("jmespath"));
+const source_parser_1 = require("../source-parser");
+class JsonParser extends source_parser_1.SourceParser {
+    constructor(source) {
+        super(source);
+        this.data = JSON.parse(source);
+    }
+    extract(query) {
+        return jmespath.search(this.data, query);
+    }
+}
+exports.JsonParser = JsonParser;

package/dist/markdown/constants.d.ts

@@ -0,0 +1,2 @@
+import MarkdownIt from "markdown-it";
+export declare const markdownIt: MarkdownIt;

package/dist/markdown/constants.js

@@ -0,0 +1,11 @@
+"use strict";
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.markdownIt = void 0;
+const markdown_it_anchor_1 = __importDefault(require("markdown-it-anchor"));
+const markdown_it_1 = __importDefault(require("markdown-it"));
+exports.markdownIt = new markdown_it_1.default({
+    html: true,
+}).use(markdown_it_anchor_1.default);

package/dist/markdown/extractor-model.d.ts

@@ -0,0 +1,4 @@
+import { HtmlExtrctorModel, HtmlExtrctorModelShape, InferHtmlShape } from "../html";
+export declare class MarkdownExtractorModel<S extends HtmlExtrctorModelShape> extends HtmlExtrctorModel<S> {
+    extract(source: string): Promise<InferHtmlShape<S>>;
+}

package/dist/markdown/extractor-model.js

@@ -0,0 +1,17 @@
+"use strict";
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.MarkdownExtractorModel = void 0;
+const he_1 = __importDefault(require("he"));
+const html_1 = require("../html");
+const constants_1 = require("./constants");
+class MarkdownExtractorModel extends html_1.HtmlExtrctorModel {
+    async extract(source) {
+        const htmlSource = constants_1.markdownIt.render(source);
+        const decodedSource = he_1.default.decode(htmlSource).replace(/\u00A0/g, " ");
+        return await super.extract(`<!DOCTYPE html><html lang="en"><head><title>Markdown Document</title></head><body>${decodedSource}</body></html>`);
+    }
+}
+exports.MarkdownExtractorModel = MarkdownExtractorModel;

package/dist/markdown/index.d.ts

@@ -0,0 +1,2 @@
+export * from "./parser";
+export * from "./extractor-model";

package/dist/markdown/index.js

@@ -0,0 +1,18 @@
+"use strict";
+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]; } };
+    }
+    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("./parser"), exports);
+__exportStar(require("./extractor-model"), exports);

package/dist/markdown/parser.d.ts

@@ -0,0 +1,5 @@
+import { HtmlParser } from "../html";
+export declare class MarkdownParser extends HtmlParser {
+    readonly markdownSource: string;
+    constructor(source: string);
+}

package/dist/markdown/parser.js

@@ -0,0 +1,18 @@
+"use strict";
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.MarkdownParser = void 0;
+const he_1 = __importDefault(require("he"));
+const constants_1 = require("./constants");
+const html_1 = require("../html");
+class MarkdownParser extends html_1.HtmlParser {
+    constructor(source) {
+        const htmlSource = constants_1.markdownIt.render(source);
+        const decodedSource = he_1.default.decode(htmlSource).replace(/\u00A0/g, " ");
+        super(`<!DOCTYPE html><html lang="en"><head><title>Markdown Document</title></head><body>${decodedSource}</body></html>`);
+        this.markdownSource = source;
+    }
+}
+exports.MarkdownParser = MarkdownParser;

package/dist/query-builders/html.d.ts

@@ -0,0 +1,6 @@
+export type BuildedQuery = {
+    value: string;
+    type: "css" | "xpath";
+};
+export declare function css(query: string): BuildedQuery;
+export declare function xpath(query: string): BuildedQuery;

package/dist/query-builders/html.js

@@ -0,0 +1,16 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.css = css;
+exports.xpath = xpath;
+function css(query) {
+    return {
+        value: query,
+        type: "css",
+    };
+}
+function xpath(query) {
+    return {
+        value: query,
+        type: "xpath",
+    };
+}

package/dist/query-builders/index.d.ts

@@ -0,0 +1 @@
+export * from "./html";

package/dist/query-builders/index.js

@@ -0,0 +1,17 @@
+"use strict";
+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]; } };
+    }
+    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("./html"), exports);

package/dist/source-parser.d.ts

@@ -0,0 +1,13 @@
+import { Abortable } from "node:events";
+import { OpenMode } from "node:fs";
+import { ExtractorModel } from "./interfaces/extractor-model";
+export type SourceParserLoadFileOptions = ({
+    encoding?: BufferEncoding | null;
+    flag?: OpenMode;
+} & Abortable) | null;
+export declare class SourceParser {
+    readonly source: string;
+    constructor(source: string);
+    extractWithModel<T>(extractorModel: ExtractorModel<T>): Promise<T>;
+    static loadFile<T extends typeof SourceParser>(this: T, path: string, options?: SourceParserLoadFileOptions): Promise<InstanceType<T>>;
+}

package/dist/source-parser.js

@@ -0,0 +1,22 @@
+"use strict";
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.SourceParser = void 0;
+const node_fs_1 = __importDefault(require("node:fs"));
+class SourceParser {
+    constructor(source) {
+        this.source = source;
+    }
+    async extractWithModel(extractorModel) {
+        return await extractorModel.extract(this.source);
+    }
+    static async loadFile(path, options = {
+        encoding: "utf-8",
+    }) {
+        const fileContent = await node_fs_1.default.promises.readFile(path, options);
+        return Reflect.construct(this, [fileContent.toString()]);
+    }
+}
+exports.SourceParser = SourceParser;

package/dist/utils/index.d.ts

@@ -0,0 +1,3 @@
+export * from "./select-first-element";
+export * from "./select-many-elements";
+export * from "./slugify";

package/dist/utils/index.js

@@ -0,0 +1,19 @@
+"use strict";
+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]; } };
+    }
+    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("./select-first-element"), exports);
+__exportStar(require("./select-many-elements"), exports);
+__exportStar(require("./slugify"), exports);

package/dist/utils/select-first-element.d.ts

@@ -0,0 +1,3 @@
+import { HTMLElement } from "node-html-parser";
+import { BuildedQuery } from "../query-builders";
+export declare function selectFirstElement(query: BuildedQuery, root: HTMLElement): HTMLElement | null;

package/dist/utils/select-first-element.js

@@ -0,0 +1,60 @@
+"use strict";
+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]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.selectFirstElement = selectFirstElement;
+const xmldom_1 = require("xmldom");
+const node_html_parser_1 = __importDefault(require("node-html-parser"));
+const xpathLib = __importStar(require("xpath"));
+function selectFirstElement(query, root) {
+    if (query.type === "css") {
+        return root.querySelector(query.value);
+    }
+    else {
+        const doc = new xmldom_1.DOMParser({
+            locator: {},
+            errorHandler: {
+                warning: () => { },
+            },
+        }).parseFromString(root.toString());
+        const elements = xpathLib.select(query.value, doc);
+        if (!elements || elements.length === 0) {
+            return null;
+        }
+        return node_html_parser_1.default.parse(elements[0].toString()).childNodes[0];
+    }
+}

package/dist/utils/select-many-elements.d.ts

@@ -0,0 +1,3 @@
+import { HTMLElement } from "node-html-parser";
+import { BuildedQuery } from "../query-builders";
+export declare function selectManyElements(query: BuildedQuery, root: HTMLElement): HTMLElement[];

package/dist/utils/select-many-elements.js

@@ -0,0 +1,57 @@
+"use strict";
+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]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.selectManyElements = selectManyElements;
+const xmldom_1 = require("xmldom");
+const node_html_parser_1 = __importDefault(require("node-html-parser"));
+const xpathLib = __importStar(require("xpath"));
+function selectManyElements(query, root) {
+    if (query.type === "css") {
+        return root.querySelectorAll(query.value);
+    }
+    else {
+        const doc = new xmldom_1.DOMParser({
+            locator: {},
+            errorHandler: {
+                warning: () => { },
+            },
+        }).parseFromString(root.toString());
+        const elements = xpathLib.select(query.value, doc);
+        return elements.map((element) => node_html_parser_1.default.parse(element.toString()).childNodes[0]);
+    }
+}

package/dist/utils/slugify.d.ts

@@ -0,0 +1 @@
+export declare function slugify(text: string): string;

package/dist/utils/slugify.js

@@ -0,0 +1,10 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.slugify = slugify;
+function slugify(text) {
+    return text
+        .toLowerCase()
+        .replace(/[^\w]+/g, "-")
+        .replace(/(^-+|-+$)/g, "")
+        .trim();
+}

package/package.json

@@ -0,0 +1,64 @@
+{
+  "name": "@xcrap/extractor",
+  "version": "0.3.0",
+  "description": "Xcrap Extractor is a package of the Xcrap framework, it was developed to take care of the data extraction part of text files (currently supporting only HTML, JSON and Markdown) using declarative models.",
+  "keywords": [
+    "xcrap",
+    "extractor",
+    "parser",
+    "web scraping",
+    "node-html-parser",
+    "parsel",
+    "scrapy"
+  ],
+  "homepage": "https://github.com/Xcrap-Cloud/parser#readme",
+  "bugs": {
+    "url": "https://github.com/Xcrap-Cloud/parser/issues"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/Xcrap-Cloud/parser.git"
+  },
+  "license": "MIT",
+  "author": "Marcuth",
+  "main": "./dist/index.js",
+  "module": "./dist/index.js",
+  "types": "./dist/index.d.ts",
+  "files": [
+    "dist/*",
+    "!/**/__tests__"
+  ],
+  "publishConfig": {
+    "access": "public"
+  },
+  "scripts": {
+    "build": "tsc",
+    "test": "jest",
+    "format": "prettier --write \"src/**/*.ts\" \"__tests__/**/*.ts\""
+  },
+  "devDependencies": {
+    "@types/he": "^1.2.3",
+    "@types/jest": "^29.5.14",
+    "@types/jmespath": "^0.15.2",
+    "@types/node": "^22.13.14",
+    "@types/xmldom": "^0.1.34",
+    "eslint": "^10.0.0",
+    "eslint-config-prettier": "^10.1.8",
+    "eslint-plugin-prettier": "^5.5.5",
+    "jest": "^29.7.0",
+    "prettier": "^3.8.1",
+    "prettier-plugin-sort-imports": "^1.8.10",
+    "ts-jest": "^29.3.0",
+    "ts-node": "^10.9.2",
+    "typescript": "^5.8.2"
+  },
+  "dependencies": {
+    "he": "^1.2.0",
+    "jmespath": "^0.16.0",
+    "markdown-it": "^14.1.0",
+    "markdown-it-anchor": "^9.2.0",
+    "node-html-parser": "^7.0.1",
+    "xmldom": "^0.6.0",
+    "xpath": "^0.0.34"
+  }
+}