pagerts 0.2.0 → 1.0.1

package/bin/package.json

@@ -1,40 +0,0 @@
-{
-    "name": "pagerts",
-    "description": "A tool for viewing external relations in a webpage",
-    "version": "0.1.9",
-    "main": "main.js",
-    "bin": {
-        "pagerts": "bin/main.js"
-    },
-    "scripts": {
-        "test": "jest",
-        "build": "esbuild src/main.ts --external:jsdom --bundle --outdir=bin --minify --sourcemap --platform=node",
-        "lint": "tsc",
-        "start": "node ./bin/main.js",
-        "dev": "npx tsx src/main.ts"
-    },
-    "keywords": [
-        "webpage",
-        "hierarchy",
-        "management"
-    ],
-    "author": "Kirill kn253 Nevzorov",
-    "license": "MIT",
-    "bugs": {
-        "url": "https://github.com/akinevz0/pagerts/issues"
-    },
-    "homepage": "https://github.com/akinevz0/pagerts",
-    "dependencies": {
-        "blessed": "^0.1.81",
-        "commander": "^12.1.0",
-        "dotenv": "^16.4.5",
-        "jsdom": "^26.0.0"
-    },
-    "devDependencies": {
-        "@types/blessed": "^0.1.25",
-        "@types/jsdom": "^21.1.7",
-        "@types/node": "^22.8.2",
-        "esbuild": "^0.25.1",
-        "ts-node": "^10.9.2"
-    }
-}

package/bin/src/extractors/AbstractExtractor.js

@@ -1,11 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.AbstractExtractor = void 0;
-class AbstractExtractor {
-    name;
-    constructor(name) {
-        this.name = name;
-    }
-}
-exports.AbstractExtractor = AbstractExtractor;
-//# sourceMappingURL=AbstractExtractor.js.map

package/bin/src/extractors/AbstractExtractor.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"AbstractExtractor.js","sourceRoot":"","sources":["../../../src/extractors/AbstractExtractor.ts"],"names":[],"mappings":";;;AACA,MAAsB,iBAAiB;IACd;IAArB,YAAqB,IAAW;QAAX,SAAI,GAAJ,IAAI,CAAO;IAAI,CAAC;CAExC;AAHD,8CAGC"}

package/bin/src/extractors/PageExtractor.js

@@ -1,13 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.PageExtractor = void 0;
-const AbstractExtractor_1 = require("./AbstractExtractor");
-class PageExtractor extends AbstractExtractor_1.AbstractExtractor {
-    constructor() { super("page-extractor"); }
-    async extract(value) {
-        const { window: { document: { title, location: { href: url } } } } = value;
-        return { title, url };
-    }
-}
-exports.PageExtractor = PageExtractor;
-//# sourceMappingURL=PageExtractor.js.map

package/bin/src/extractors/PageExtractor.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"PageExtractor.js","sourceRoot":"","sources":["../../../src/extractors/PageExtractor.ts"],"names":[],"mappings":";;;AAEA,2DAAwD;AAExD,MAAa,aAAc,SAAQ,qCAA8B;IAC7D,gBAAgB,KAAK,CAAC,gBAAgB,CAAC,CAAC,CAAC,CAAC;IAE1C,KAAK,CAAC,OAAO,CAAC,KAAY;QACtB,MAAM,EAAE,MAAM,EAAE,EAAE,QAAQ,EAAE,EAAE,KAAK,EAAE,QAAQ,EAAE,EAAE,IAAI,EAAE,GAAG,EAAE,EAAE,EAAE,EAAE,GAAG,KAAK,CAAA;QAC1E,OAAO,EAAE,KAAK,EAAE,GAAG,EAAE,CAAA;IACzB,CAAC;CACJ;AAPD,sCAOC"}

package/bin/src/extractors/ResourceExtractor.js

@@ -1,32 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.ResourceExtractor = void 0;
-const resource_1 = require("../resource");
-const AbstractExtractor_1 = require("./AbstractExtractor");
-class ResourceExtractor extends AbstractExtractor_1.AbstractExtractor {
-    tags;
-    constructor(tags) {
-        super("page-extractor");
-        this.tags = tags;
-    }
-    async extract(value) {
-        const { document } = value.window;
-        const externalResources = [];
-        for (const tag of this.tags) {
-            const selector = document.querySelectorAll(tag);
-            const elements = Array.from(selector);
-            for (const element of elements) {
-                const text = (0, resource_1.findResourceText)(element);
-                const link = (0, resource_1.findResourceLink)(element);
-                if (!text || !link)
-                    continue;
-                if (!link.url.startsWith("http"))
-                    continue;
-                externalResources.push({ text, link });
-            }
-        }
-        return externalResources;
-    }
-}
-exports.ResourceExtractor = ResourceExtractor;
-//# sourceMappingURL=ResourceExtractor.js.map

package/bin/src/extractors/ResourceExtractor.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"ResourceExtractor.js","sourceRoot":"","sources":["../../../src/extractors/ResourceExtractor.ts"],"names":[],"mappings":";;;AACA,0CAAiH;AACjH,2DAAwD;AAExD,MAAa,iBAAkB,SAAQ,qCAA4C;IAClD;IAA7B,YAA6B,IAAW;QACpC,KAAK,CAAC,gBAAgB,CAAC,CAAC;QADC,SAAI,GAAJ,IAAI,CAAO;IAExC,CAAC;IACD,KAAK,CAAC,OAAO,CAAC,KAAY;QACtB,MAAM,EAAE,QAAQ,EAAE,GAAG,KAAK,CAAC,MAAM,CAAC;QAClC,MAAM,iBAAiB,GAAuB,EAAE,CAAC;QACjD,KAAK,MAAM,GAAG,IAAI,IAAI,CAAC,IAAI,EAAE,CAAC;YAC1B,MAAM,QAAQ,GAAG,QAAQ,CAAC,gBAAgB,CAAW,GAAG,CAAC,CAAA;YACzD,MAAM,QAAQ,GAAG,KAAK,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAA;YACrC,KAAK,MAAM,OAAO,IAAI,QAAQ,EAAE,CAAC;gBAC7B,MAAM,IAAI,GAAG,IAAA,2BAAgB,EAAC,OAAO,CAAC,CAAC;gBACvC,MAAM,IAAI,GAAG,IAAA,2BAAgB,EAAC,OAAO,CAAC,CAAC;gBACvC,IAAG,CAAC,IAAI,IAAI,CAAC,IAAI;oBAAE,SAAQ;gBAC3B,IAAI,CAAC,IAAI,CAAC,GAAG,CAAC,UAAU,CAAC,MAAM,CAAC;oBAAE,SAAQ;gBAC1C,iBAAiB,CAAC,IAAI,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,CAAC,CAAA;YAC1C,CAAC;QACL,CAAC;QACD,OAAO,iBAAiB,CAAC;IAC7B,CAAC;CACJ;AApBD,8CAoBC"}

package/bin/src/main.js

@@ -1,36 +0,0 @@
-#!/usr/bin/env node
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-const commander_1 = require("commander");
-const package_json_1 = require("../package.json");
-const PageExtractor_1 = require("./extractors/PageExtractor");
-const ResourceExtractor_1 = require("./extractors/ResourceExtractor");
-const PageFetcher_1 = require("./page/PageFetcher");
-const JSONStylePrinter_1 = require("./printers/JSONStylePrinter");
-const program = new commander_1.Command();
-const url = (0, commander_1.createArgument)("<url | file...>", "remote https://URL or local file://resource.html to extract from");
-(async () => {
-    await program
-        .name(package_json_1.name)
-        .version(package_json_1.version, "-v, --version")
-        .description(package_json_1.description)
-        .addArgument(url)
-        .action(async (urls) => {
-        const printer = new JSONStylePrinter_1.JSONStylePrinter();
-        // simple log style printer
-        // const printer = new LogStylePrinter();
-        const pageFetcher = new PageFetcher_1.PageFetcher();
-        const pageExtractor = new PageExtractor_1.PageExtractor();
-        const resourceExtractor = new ResourceExtractor_1.ResourceExtractor(["a", "meta", "link", "embed"]);
-        const pageResponses = await pageFetcher.fetchAll(urls);
-        const pageMetadatas = [];
-        for (const { content, url, error } of pageResponses) {
-            const resources = error in (content) ? [] : await resourceExtractor.extract(content);
-            const descriptor = error in content ? { url, error } : await pageExtractor.extract(content);
-            pageMetadatas.push({ ...descriptor, resources });
-        }
-        await printer.print(...pageMetadatas);
-    })
-        .parseAsync(process.argv);
-})();
-//# sourceMappingURL=main.js.map

package/bin/src/main.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"main.js","sourceRoot":"","sources":["../../src/main.ts"],"names":[],"mappings":";;;AACA,yCAAoD;AAEpD,kDAA6D;AAC7D,8DAA2D;AAC3D,sEAAmE;AACnE,oDAAiD;AAEjD,kEAA+D;AAG/D,MAAM,OAAO,GAAG,IAAI,mBAAO,EAAE,CAAC;AAE9B,MAAM,GAAG,GAAG,IAAA,0BAAc,EAAC,iBAAiB,EAAE,kEAAkE,CAAC,CAAC;AAElH,CAAC,KAAK,IAAI,EAAE;IACV,MAAM,OAAO;SACV,IAAI,CAAC,mBAAI,CAAC;SACV,OAAO,CAAC,sBAAO,EAAE,eAAe,CAAC;SACjC,WAAW,CAAC,0BAAW,CAAC;SACxB,WAAW,CAAC,GAAG,CAAC;SAChB,MAAM,CAAC,KAAK,EAAE,IAAc,EAAE,EAAE;QAC/B,MAAM,OAAO,GAAG,IAAI,mCAAgB,EAAE,CAAC;QACvC,2BAA2B;QAC3B,yCAAyC;QAEzC,MAAM,WAAW,GAAG,IAAI,yBAAW,EAAE,CAAA;QACrC,MAAM,aAAa,GAAG,IAAI,6BAAa,EAAE,CAAA;QACzC,MAAM,iBAAiB,GAAG,IAAI,qCAAiB,CAAC,CAAC,GAAG,EAAE,MAAM,EAAE,MAAM,EAAE,OAAO,CAAC,CAAC,CAAA;QAE/E,MAAM,aAAa,GAAG,MAAM,WAAW,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC;QACvD,MAAM,aAAa,GAAmB,EAAE,CAAC;QAEzC,KAAK,MAAM,EAAE,OAAO,EAAE,GAAG,EAAE,KAAK,EAAE,IAAI,aAAa,EAAE,CAAC;YACpD,MAAM,SAAS,GAAG,KAAK,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,MAAM,iBAAiB,CAAC,OAAO,CAAC,OAAO,CAAC,CAAC;YACrF,MAAM,UAAU,GAAG,KAAK,IAAI,OAAO,CAAC,CAAC,CAAC,EAAE,GAAG,EAAE,KAAK,EAAE,CAAC,CAAC,CAAC,MAAM,aAAa,CAAC,OAAO,CAAC,OAAO,CAAC,CAAC;YAC5F,aAAa,CAAC,IAAI,CAAC,EAAE,GAAG,UAAU,EAAE,SAAS,EAAE,CAAC,CAAC;QACnD,CAAC;QAED,MAAM,OAAO,CAAC,KAAK,CAAC,GAAG,aAAa,CAAC,CAAC;IACxC,CAAC,CAAC;SACD,UAAU,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC;AAC9B,CAAC,CAAC,EAAE,CAAC"}

package/bin/src/page/Page.js

@@ -1,8 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.isPage = exports.isError = void 0;
-const isError = (page) => 'error' in page;
-exports.isError = isError;
-const isPage = (page) => "resources" in page && Array.isArray(page.resources);
-exports.isPage = isPage;
-//# sourceMappingURL=Page.js.map

package/bin/src/page/Page.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"Page.js","sourceRoot":"","sources":["../../../src/page/Page.ts"],"names":[],"mappings":";;;AAgBO,MAAM,OAAO,GAAG,CAAC,IAAkB,EAA6B,EAAE,CAAC,OAAO,IAAI,IAAI,CAAC;AAA7E,QAAA,OAAO,WAAsE;AACnF,MAAM,MAAM,GAAG,CAAC,IAAS,EAAgB,EAAE,CAC9C,WAAW,IAAI,IAAI,IAAI,KAAK,CAAC,OAAO,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC;AAD5C,QAAA,MAAM,UACsC"}

package/bin/src/page/PageFetcher.js

@@ -1,26 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.PageFetcher = void 0;
-const jsdom_1 = require("jsdom");
-class PageFetcher {
-    async fetchPage(url) {
-        let dom;
-        const virtualConsole = new jsdom_1.VirtualConsole().on('jsdomError', (error) => {
-            process.stderr.write(`Error parsing ${url}:${error.message}\n`);
-        });
-        if (url.startsWith("file://")) {
-            dom = jsdom_1.JSDOM.fromFile(url, { virtualConsole });
-        }
-        else {
-            dom = jsdom_1.JSDOM.fromURL(url, { virtualConsole });
-        }
-        return dom.then(content => ({ url, content }))
-            .catch(({ message }) => ({ url, error: `JSDOM failed to parse: ${message}` }));
-    }
-    async fetchAll(urls) {
-        const responses = await Promise.all(urls.map(url => this.fetchPage(url)));
-        return responses.filter(response => response.content !== undefined);
-    }
-}
-exports.PageFetcher = PageFetcher;
-//# sourceMappingURL=PageFetcher.js.map

package/bin/src/page/PageFetcher.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"PageFetcher.js","sourceRoot":"","sources":["../../../src/page/PageFetcher.ts"],"names":[],"mappings":";;;AAAA,iCAA8C;AAS9C,MAAa,WAAW;IACZ,KAAK,CAAC,SAAS,CAAC,GAAW;QAC/B,IAAI,GAAmB,CAAC;QACxB,MAAM,cAAc,GAAG,IAAI,sBAAc,EAAE,CAAC,EAAE,CAAC,YAAY,EAAE,CAAC,KAAK,EAAE,EAAE;YACnE,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,iBAAiB,GAAG,IAAI,KAAK,CAAC,OAAO,IAAI,CAAC,CAAC;QACpE,CAAC,CAAC,CAAC;QACH,IAAI,GAAG,CAAC,UAAU,CAAC,SAAS,CAAC,EAAE,CAAC;YAC5B,GAAG,GAAG,aAAK,CAAC,QAAQ,CAAC,GAAG,EAAE,EAAE,cAAc,EAAE,CAAC,CAAC;QAClD,CAAC;aAAM,CAAC;YACJ,GAAG,GAAG,aAAK,CAAC,OAAO,CAAC,GAAG,EAAE,EAAE,cAAc,EAAE,CAAC,CAAC;QACjD,CAAC;QAED,OAAO,GAAG,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC,CAAC,EAAE,GAAG,EAAE,OAAO,EAAE,CAAC,CAAC;aACzC,KAAK,CAAC,CAAC,EAAE,OAAO,EAAE,EAAE,EAAE,CAAC,CAAC,EAAE,GAAG,EAAE,KAAK,EAAE,0BAA0B,OAAO,EAAE,EAAE,CAAC,CAAC,CAAC;IACvF,CAAC;IACD,KAAK,CAAC,QAAQ,CAAC,IAAc;QACzB,MAAM,SAAS,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC,EAAE,CAAC,IAAI,CAAC,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC;QAC1E,OAAO,SAAS,CAAC,MAAM,CAAC,QAAQ,CAAC,EAAE,CAAC,QAAQ,CAAC,OAAO,KAAK,SAAS,CAAC,CAAC;IACxE,CAAC;CAEJ;AApBD,kCAoBC"}

package/bin/src/printers/AbstractResourcePrinter.js

@@ -1,8 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.AbstractResourcePrinter = void 0;
-class AbstractResourcePrinter {
-    constructor() { }
-}
-exports.AbstractResourcePrinter = AbstractResourcePrinter;
-//# sourceMappingURL=AbstractResourcePrinter.js.map

package/bin/src/printers/AbstractResourcePrinter.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"AbstractResourcePrinter.js","sourceRoot":"","sources":["../../../src/printers/AbstractResourcePrinter.ts"],"names":[],"mappings":";;;AAEA,MAAsB,uBAAuB;IACzC,gBAAiB,CAAC;CAErB;AAHD,0DAGC"}

package/bin/src/printers/JSONStylePrinter.js

@@ -1,12 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.JSONStylePrinter = void 0;
-const AbstractResourcePrinter_1 = require("./AbstractResourcePrinter");
-class JSONStylePrinter extends AbstractResourcePrinter_1.AbstractResourcePrinter {
-    print(...pages) {
-        const json = JSON.stringify(pages);
-        process.stdout.write(json + "\n");
-    }
-}
-exports.JSONStylePrinter = JSONStylePrinter;
-//# sourceMappingURL=JSONStylePrinter.js.map

package/bin/src/printers/JSONStylePrinter.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"JSONStylePrinter.js","sourceRoot":"","sources":["../../../src/printers/JSONStylePrinter.ts"],"names":[],"mappings":";;;AACA,uEAAoE;AAGpE,MAAa,gBAAiB,SAAQ,iDAAuB;IACzD,KAAK,CAAC,GAAG,KAAqB;QAC1B,MAAM,IAAI,GAAG,IAAI,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC;QACnC,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,IAAI,GAAG,IAAI,CAAC,CAAA;IACrC,CAAC;CAGJ;AAPD,4CAOC"}

package/bin/src/printers/LogStylePrinter.js

@@ -1,27 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.LogStylePrinter = void 0;
-const Page_1 = require("../page/Page");
-const AbstractResourcePrinter_1 = require("./AbstractResourcePrinter");
-class LogStylePrinter extends AbstractResourcePrinter_1.AbstractResourcePrinter {
-    write(str) {
-        process.stdout.write(str);
-    }
-    async print(...pages) {
-        for (const page of pages) {
-            if (!(0, Page_1.isPage)(page)) {
-                this.write(page.error);
-                continue;
-            }
-            const { resources, title, url } = page;
-            this.write(`Title: ${title}\n`);
-            this.write(`URL: ${url}\n\n`);
-            for (const resource of resources) {
-                const { link: { url }, text: { value } } = resource;
-                this.write(`${value}: ${url}\n`);
-            }
-        }
-    }
-}
-exports.LogStylePrinter = LogStylePrinter;
-//# sourceMappingURL=LogStylePrinter.js.map

package/bin/src/printers/LogStylePrinter.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"LogStylePrinter.js","sourceRoot":"","sources":["../../../src/printers/LogStylePrinter.ts"],"names":[],"mappings":";;;AAAA,uCAAoE;AACpE,uEAAoE;AAEpE,MAAa,eAAgB,SAAQ,iDAAuB;IAExD,KAAK,CAAC,GAAW;QACb,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,GAAG,CAAC,CAAA;IAC7B,CAAC;IAED,KAAK,CAAC,KAAK,CAAC,GAAG,KAAqB;QAChC,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;YACvB,IAAI,CAAC,IAAA,aAAM,EAAC,IAAI,CAAC,EAAE,CAAC;gBAChB,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;gBACtB,SAAQ;YACZ,CAAC;YAED,MAAM,EAAC,SAAS,EAAE,KAAK,EAAE,GAAG,EAAE,GAAG,IAAI,CAAA;YAErC,IAAI,CAAC,KAAK,CAAC,UAAU,KAAK,IAAI,CAAC,CAAA;YAC/B,IAAI,CAAC,KAAK,CAAC,QAAQ,GAAG,MAAM,CAAC,CAAA;YAE7B,KAAK,MAAM,QAAQ,IAAI,SAAS,EAAE,CAAC;gBAC/B,MAAM,EAAE,IAAI,EAAE,EAAE,GAAG,EAAE,EAAE,IAAI,EAAE,EAAE,KAAK,EAAE,EAAE,GAAG,QAAQ,CAAA;gBACnD,IAAI,CAAC,KAAK,CAAC,GAAG,KAAK,KAAK,GAAG,IAAI,CAAC,CAAA;YACpC,CAAC;QACL,CAAC;IACL,CAAC;CACJ;AAxBD,0CAwBC"}

package/bin/src/resource.js

@@ -1,56 +0,0 @@
-"use strict";
-/**
- * @license MIT
- * We are interested in visualising a page as a collection of tags.
- *
- * We wish to work with tags that can be compactly previewed on a webpage.
- * Here we must declare all of the element types that can be used to represent
- * a resource that can be hyperlinked off a webpage.
- */
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.isKeyDefined = exports.isResourceKey = exports.RESOURCE_LINK_KEYS = exports.RESOURCE_DISPLAYABLE_KEYS = void 0;
-exports.findResourceText = findResourceText;
-exports.findResourceLink = findResourceLink;
-function findDefinedKey(element, keys) {
-    for (const key of keys) {
-        if ((0, exports.isKeyDefined)(key, element)) {
-            return key;
-        }
-    }
-}
-exports.RESOURCE_DISPLAYABLE_KEYS = [
-    'id',
-    'innerText',
-    'textContent',
-    'class',
-    'ariaLabel',
-    'ariaDescription',
-    'alt',
-    'rel'
-];
-exports.RESOURCE_LINK_KEYS = [
-    "href",
-    "data-src",
-    "target",
-    "action",
-    "src",
-    "url"
-];
-function findResourceText(element) {
-    for (const key of exports.RESOURCE_DISPLAYABLE_KEYS) {
-        const value = element[key];
-        if (value && typeof value === 'string' && value.trim() !== '')
-            return { key, value };
-    }
-}
-function findResourceLink(element) {
-    const key = findDefinedKey(element, [...exports.RESOURCE_LINK_KEYS]);
-    const url = element[key];
-    if (url && typeof url === 'string' && url.trim() !== '')
-        return { key, url };
-}
-const isResourceKey = (key) => key in exports.RESOURCE_LINK_KEYS;
-exports.isResourceKey = isResourceKey;
-const isKeyDefined = (key, element) => key in element && element[key] !== undefined;
-exports.isKeyDefined = isKeyDefined;
-//# sourceMappingURL=resource.js.map

package/bin/src/resource.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"resource.js","sourceRoot":"","sources":["../../src/resource.ts"],"names":[],"mappings":";AAAA;;;;;;;GAOG;;;AA8CH,4CAMC;AAED,4CAKC;AAvDD,SAAS,cAAc,CAAC,OAAiB,EAAE,IAAe;IACtD,KAAK,MAAM,GAAG,IAAI,IAAI,EAAE,CAAC;QACrB,IAAI,IAAA,oBAAY,EAAC,GAAG,EAAE,OAAO,CAAC,EAAE,CAAC;YAC7B,OAAO,GAAG,CAAC;QACf,CAAC;IACL,CAAC;AACL,CAAC;AAEY,QAAA,yBAAyB,GAAG;IACrC,IAAI;IACJ,WAAW;IACX,aAAa;IACb,OAAO;IACP,WAAW;IACX,iBAAiB;IACjB,KAAK;IACL,KAAK;CACC,CAAC;AASE,QAAA,kBAAkB,GAAG;IAC9B,MAAM;IACN,UAAU;IACV,QAAQ;IACR,QAAQ;IACR,KAAK;IACL,KAAK;CACC,CAAC;AASX,SAAgB,gBAAgB,CAAC,OAAiB;IAC9C,KAAK,MAAM,GAAG,IAAI,iCAAyB,EAAE,CAAC;QAC1C,MAAM,KAAK,GAAG,OAAO,CAAC,GAAG,CAAC,CAAA;QAC1B,IAAI,KAAK,IAAI,OAAO,KAAK,KAAK,QAAQ,IAAI,KAAK,CAAC,IAAI,EAAE,KAAK,EAAE;YACzD,OAAO,EAAE,GAAG,EAAE,KAAK,EAAE,CAAC;IAC9B,CAAC;AACL,CAAC;AAED,SAAgB,gBAAgB,CAAC,OAAiB;IAC9C,MAAM,GAAG,GAAG,cAAc,CAAC,OAAO,EAAE,CAAC,GAAG,0BAAkB,CAAC,CAAC,CAAC;IAC7D,MAAM,GAAG,GAAG,OAAO,CAAC,GAAG,CAAC,CAAC;IACzB,IAAI,GAAG,IAAI,OAAO,GAAG,KAAK,QAAQ,IAAI,GAAG,CAAC,IAAI,EAAE,KAAK,EAAE;QACnD,OAAO,EAAE,GAAG,EAAE,GAAG,EAAE,CAAC;AAC5B,CAAC;AAOM,MAAM,aAAa,GAAG,CAAC,GAAW,EAAkB,EAAE,CAAC,GAAG,IAAI,0BAAkB,CAAC;AAA3E,QAAA,aAAa,iBAA8D;AAEjF,MAAM,YAAY,GAAG,CAA6B,GAAW,EAAE,OAAU,EAAW,EAAE,CACzF,GAAG,IAAI,OAAO,IAAI,OAAO,CAAC,GAAG,CAAC,KAAK,SAAS,CAAC;AADpC,QAAA,YAAY,gBACwB"}

package/jest.config.js

@@ -1,198 +0,0 @@
-/**
- * For a detailed explanation regarding each configuration property, visit:
- * https://jestjs.io/docs/configuration
- */
-
-/** @type {import('jest').Config} */
-const config = {
-  // All imported modules in your tests should be mocked automatically
-  // automock: false,
-
-  // Stop running tests after `n` failures
-  // bail: 0,
-
-  // The directory where Jest should store its cached dependency information
-  // cacheDirectory: "/tmp/jest_rs",
-
-  // Automatically clear mock calls, instances, contexts and results before every test
-  // clearMocks: false,
-
-  // Indicates whether the coverage information should be collected while executing the test
-  collectCoverage: true,
-
-  // An array of glob patterns indicating a set of files for which coverage information should be collected
-  // collectCoverageFrom: undefined,
-
-  // The directory where Jest should output its coverage files
-  coverageDirectory: "coverage",
-
-  // An array of regexp pattern strings used to skip coverage collection
-  // coveragePathIgnorePatterns: [
-  //   "/node_modules/"
-  // ],
-
-  // Indicates which provider should be used to instrument code for coverage
-  coverageProvider: "v8",
-
-  // A list of reporter names that Jest uses when writing coverage reports
-  // coverageReporters: [
-  //   "json",
-  //   "text",
-  //   "lcov",
-  //   "clover"
-  // ],
-
-  // An object that configures minimum threshold enforcement for coverage results
-  // coverageThreshold: undefined,
-
-  // A path to a custom dependency extractor
-  // dependencyExtractor: undefined,
-
-  // Make calling deprecated APIs throw helpful error messages
-  // errorOnDeprecated: false,
-
-  // The default configuration for fake timers
-  // fakeTimers: {
-  //   "enableGlobally": false
-  // },
-
-  // Force coverage collection from ignored files using an array of glob patterns
-  // forceCoverageMatch: [],
-
-  // A path to a module which exports an async function that is triggered once before all test suites
-  // globalSetup: undefined,
-
-  // A path to a module which exports an async function that is triggered once after all test suites
-  // globalTeardown: undefined,
-
-  // A set of global variables that need to be available in all test environments
-  // globals: {},
-
-  // The maximum amount of workers used to run your tests. Can be specified as % or a number. E.g. maxWorkers: 10% will use 10% of your CPU amount + 1 as the maximum worker number. maxWorkers: 2 will use a maximum of 2 workers.
-  // maxWorkers: "50%",
-
-  // An array of directory names to be searched recursively up from the requiring module's location
-  // moduleDirectories: [
-  //   "node_modules"
-  // ],
-
-  // An array of file extensions your modules use
-  // moduleFileExtensions: [
-  //   "js",
-  //   "mjs",
-  //   "cjs",
-  //   "jsx",
-  //   "ts",
-  //   "tsx",
-  //   "json",
-  //   "node"
-  // ],
-
-  // A map from regular expressions to module names or to arrays of module names that allow to stub out resources with a single module
-  // moduleNameMapper: {},
-
-  // An array of regexp pattern strings, matched against all module paths before considered 'visible' to the module loader
-  // modulePathIgnorePatterns: [],
-
-  // Activates notifications for test results
-  // notify: false,
-
-  // An enum that specifies notification mode. Requires { notify: true }
-  // notifyMode: "failure-change",
-
-  // A preset that is used as a base for Jest's configuration
-  // preset: undefined,
-
-  // Run tests from one or more projects
-  // projects: undefined,
-
-  // Use this configuration option to add custom reporters to Jest
-  // reporters: undefined,
-
-  // Automatically reset mock state before every test
-  // resetMocks: false,
-
-  // Reset the module registry before running each individual test
-  // resetModules: false,
-
-  // A path to a custom resolver
-  // resolver: undefined,
-
-  // Automatically restore mock state and implementation before every test
-  // restoreMocks: false,
-
-  // The root directory that Jest should scan for tests and modules within
-  // rootDir: undefined,
-
-  // A list of paths to directories that Jest should use to search for files in
-  // roots: [
-  //   "<rootDir>"
-  // ],
-
-  // Allows you to use a custom runner instead of Jest's default test runner
-  // runner: "jest-runner",
-
-  // The paths to modules that run some code to configure or set up the testing environment before each test
-  // setupFiles: [],
-
-  // A list of paths to modules that run some code to configure or set up the testing framework before each test
-  // setupFilesAfterEnv: [],
-
-  // The number of seconds after which a test is considered as slow and reported as such in the results.
-  // slowTestThreshold: 5,
-
-  // A list of paths to snapshot serializer modules Jest should use for snapshot testing
-  // snapshotSerializers: [],
-
-  // The test environment that will be used for testing
-  // testEnvironment: "jest-environment-node",
-
-  // Options that will be passed to the testEnvironment
-  // testEnvironmentOptions: {},
-
-  // Adds a location field to test results
-  // testLocationInResults: false,
-
-  // The glob patterns Jest uses to detect test files
-  // testMatch: [
-  //   "**/__tests__/**/*.[jt]s?(x)",
-  //   "**/?(*.)+(spec|test).[tj]s?(x)"
-  // ],
-
-  // An array of regexp pattern strings that are matched against all test paths, matched tests are skipped
-  // testPathIgnorePatterns: [
-  //   "/node_modules/"
-  // ],
-
-  // The regexp pattern or array of patterns that Jest uses to detect test files
-  // testRegex: [],
-
-  // This option allows the use of a custom results processor
-  // testResultsProcessor: undefined,
-
-  // This option allows use of a custom test runner
-  // testRunner: "jest-circus/runner",
-
-  // A map from regular expressions to paths to transformers
-  // transform: undefined,
-
-  // An array of regexp pattern strings that are matched against all source file paths, matched files will skip transformation
-  // transformIgnorePatterns: [
-  //   "/node_modules/",
-  //   "\\.pnp\\.[^\\/]+$"
-  // ],
-
-  // An array of regexp pattern strings that are matched against all modules before the module loader will automatically return a mock for them
-  // unmockedModulePathPatterns: undefined,
-
-  // Indicates whether each individual test should be reported during the run
-  // verbose: undefined,
-
-  // An array of regexp patterns that are matched against all source file paths before re-running tests in watch mode
-  // watchPathIgnorePatterns: [],
-
-  // Whether to use watchman for file crawling
-  // watchman: true,
-};
-
-module.exports = config;

package/src/extractors/AbstractExtractor.ts

@@ -1,5 +0,0 @@
-
-export abstract class AbstractExtractor<V, R> {
-    constructor(readonly name:string) { }
-    abstract extract(value: V): Promise<R>;
-}

package/src/extractors/PageExtractor.ts

@@ -1,12 +0,0 @@
-import { isError, type Page } from '../page/Page';
-import { JSDOM } from 'jsdom';
-import { AbstractExtractor } from './AbstractExtractor';
-
-export class PageExtractor extends AbstractExtractor<JSDOM, Page> {
-    constructor() { super("page-extractor"); }
-
-    async extract(value: JSDOM): Promise<Page> {
-        const { window: { document: { title, location: { href: url } } } } = value
-        return { title, url }
-    }
-}

package/src/extractors/ResourceExtractor.ts

@@ -1,25 +0,0 @@
-import type { JSDOM } from "jsdom";
-import { findResourceLink, findResourceText, type ExternalResource, type Resource, type Tag } from "../resource";
-import { AbstractExtractor } from './AbstractExtractor';
-
-export class ResourceExtractor extends AbstractExtractor<JSDOM, ExternalResource[]> {
-    constructor(private readonly tags: Tag[]) {
-        super("page-extractor");
-    }
-    async extract(value: JSDOM): Promise<ExternalResource[]> {
-        const { document } = value.window;
-        const externalResources: ExternalResource[] = [];
-        for (const tag of this.tags) {
-            const selector = document.querySelectorAll<Resource>(tag)
-            const elements = Array.from(selector)
-            for (const element of elements) {
-                const text = findResourceText(element);
-                const link = findResourceLink(element);
-                if(!text || !link) continue
-                if (!link.url.startsWith("http")) continue
-                externalResources.push({ text, link })
-            }
-        }
-        return externalResources;
-    }
-}

package/src/extractors/TagExtractor.ts

@@ -1,14 +0,0 @@
-import { JSDOM } from 'jsdom';
-import type { Resource, Tag } from '../resource';
-import { AbstractExtractor } from './AbstractExtractor';
-
-export class TagExtractor<T extends Tag> extends AbstractExtractor<JSDOM, Resource[]> {
-    extract(value: JSDOM): Promise<Resource[]> {
-        const linkNodes = value.window.document.querySelectorAll<Resource>(this.tagName);
-        return Promise.resolve(Array.from(linkNodes));
-    }
-    constructor(private readonly tagName: T) {
-        super(`extract <${tagName}>`)
-    };
-
-}

package/src/main.ts

@@ -1,43 +0,0 @@
-#!/usr/bin/env node
-import { Command, createArgument } from "commander";
-
-import { description, name, version } from '../package.json';
-import { PageExtractor } from "./extractors/PageExtractor";
-import { ResourceExtractor } from "./extractors/ResourceExtractor";
-import { PageFetcher } from "./page/PageFetcher";
-import type { Page, PageMetadata } from "./page/Page";
-import { JSONStylePrinter } from "./printers/JSONStylePrinter";
-import { LogStylePrinter } from "./printers/LogStylePrinter";
-
-const program = new Command();
-
-const url = createArgument("<url | file...>", "remote https://URL or local file://resource.html to extract from");
-
-(async () => {
-  await program
-    .name(name)
-    .version(version, "-v, --version")
-    .description(description)
-    .addArgument(url)
-    .action(async (urls: string[]) => {
-      const printer = new JSONStylePrinter();
-      // simple log style printer
-      // const printer = new LogStylePrinter();
-
-      const pageFetcher = new PageFetcher()
-      const pageExtractor = new PageExtractor()
-      const resourceExtractor = new ResourceExtractor(["a", "meta", "link", "embed"])
-
-      const pageResponses = await pageFetcher.fetchAll(urls);
-      const pageMetadatas: PageMetadata[] = [];
-
-      for (const { content, url, error } of pageResponses) {
-        const resources = error in (content) ? [] : await resourceExtractor.extract(content);
-        const descriptor = error in content ? { url, error } : await pageExtractor.extract(content);
-        pageMetadatas.push({ ...descriptor, resources });
-      }
-
-      await printer.print(...pageMetadatas);
-    })
-    .parseAsync(process.argv);
-})();

package/src/page/Page.ts

@@ -1,19 +0,0 @@
-import type { ExternalResource } from "../resource";
-
-type hasTitle = {
-    title: string;
-};
-
-type hasUrl = {
-    url: string;
-};
-
-type hasResources = {
-    resources: ExternalResource[];
-};
-
-export type Page = hasTitle & hasUrl
-export type PageMetadata = (Page & hasResources) | { error: string }
-export const isError = (page: PageMetadata): page is { error: string } => 'error' in page;
-export const isPage = (page: any): page is Page =>
-    "resources" in page && Array.isArray(page.resources);