link-preview-js - npm Package Compare versions

Comparing version 2.0.4 to 2.0.5

__tests__/index.spec.ts

@@ -5,3 +5,3 @@ import { getLinkPreview } from "../build/index";
   it(`should extract link info from just URL`, async () => {
-    const linkInfo = await getLinkPreview(
+    const linkInfo: any = await getLinkPreview(
       `https://www.youtube.com/watch?v=wuClZjOdT30`,
@@ -21,5 +21,3 @@ { headers: { "Accept-Language": `en-US` } },
     expect(linkInfo.videos.length).toEqual(0);
-    expect(linkInfo.favicons[0]).toEqual(
-      `https://www.youtube.com/yts/img/favicon_32-vflOogEID.png`,
-    );
+    expect(linkInfo.favicons[0]).not.toBe(``);
     expect(linkInfo.contentType.toLowerCase()).toEqual(
@@ -31,3 +29,3 @@ `text/html; charset=utf-8`,
   it(`should extract link info from a URL with a newline`, async () => {
-    const linkInfo = await getLinkPreview(
+    const linkInfo: any = await getLinkPreview(
       `
@@ -49,5 +47,3 @@ https://www.youtube.com/watch?v=wuClZjOdT30,
     expect(linkInfo.videos.length).toEqual(0);
-    expect(linkInfo.favicons[0]).toEqual(
-      `https://www.youtube.com/yts/img/favicon_32-vflOogEID.png`,
-    );
+    expect(linkInfo.favicons[0]).not.toBe(``);
     expect(linkInfo.contentType.toLowerCase()).toEqual(
@@ -59,3 +55,3 @@ `text/html; charset=utf-8`,
   it(`should extract link info from just text with a URL`, async () => {
-    const linkInfo = await getLinkPreview(
+    const linkInfo: any = await getLinkPreview(
       `This is some text blah blah https://www.youtube.com/watch?v=wuClZjOdT30 and more text`,
@@ -75,5 +71,3 @@ { headers: { "Accept-Language": `en-US` } },
     expect(linkInfo.videos.length).toEqual(0);
-    expect(linkInfo.favicons[0]).toEqual(
-      `https://www.youtube.com/yts/img/favicon_32-vflOogEID.png`,
-    );
+    expect(linkInfo.favicons[0]).toBeTruthy();
     expect(linkInfo.contentType.toLowerCase()).toEqual(
@@ -85,3 +79,3 @@ `text/html; charset=utf-8`,
   it(`should make request with different languages`, async () => {
-    let linkInfo = await getLinkPreview(`https://www.hsbc.ca/`, {
+    let linkInfo: any = await getLinkPreview(`https://www.hsbc.ca/`, {
       headers: { "Accept-Language": `fr` },
@@ -132,26 +126,22 @@ });
   it(`should handle unknown content type urls`, async () => {
+    const linkInfo = await getLinkPreview(`https://mjml.io/try-it-live`);
+
+    expect(linkInfo.url).toEqual(`https://mjml.io/try-it-live`);
+    expect(linkInfo.mediaType).toEqual(`website`);
+  });
+
+  // This site changed? it is not returning application any more but rather website
+  it.skip(`should handle application urls`, async () => {
     const linkInfo = await getLinkPreview(
-      `https://mjml.io/try-it-live`,
+      `https://assets.curtmfg.com/masterlibrary/56282/installsheet/CME_56282_INS.pdf`,
     );
 
     expect(linkInfo.url).toEqual(
-      `https://mjml.io/try-it-live`,
+      `https://assets.curtmfg.com/masterlibrary/56282/installsheet/CME_56282_INS.pdf`,
     );
-    expect(linkInfo.mediaType).toEqual(`website`);
+    expect(linkInfo.mediaType).toEqual(`application`);
+    expect(linkInfo.contentType.toLowerCase()).toEqual(`application/pdf`);
+    expect(linkInfo.favicons[0]).toBeTruthy();
   });
 
-  // This site changed? it is not returning application any more but rather website
-  // it(`should handle application urls`, async () => {
-  //   const linkInfo = await getLinkPreview(
-  //     `https://assets.curtmfg.com/masterlibrary/56282/installsheet/CME_56282_INS.pdf`,
-  //   );
-
-  //   expect(linkInfo.url).toEqual(
-  //     `https://assets.curtmfg.com/masterlibrary/56282/installsheet/CME_56282_INS.pdf`,
-  //   );
-  //   expect(linkInfo.mediaType).toEqual(`application`);
-  //   expect(linkInfo.contentType.toLowerCase()).toEqual(`application/pdf`);
-  //   expect(linkInfo.favicons[0]).toBeTruthy();
-  // });
-
   it(`no link in text should fail gracefully`, async () => {
@@ -171,5 +161,28 @@ await expect(
 
-  it(`hould handle empty strings gracefully`, async () => {
+  it(`should handle empty strings gracefully`, async () => {
     await expect(getLinkPreview(``)).rejects.toThrowErrorMatchingSnapshot();
   });
+
+  it(`should handle a proxy url option`, async () => {
+    // origin header is required by cors-anywhere
+    const linkInfo: any = await getLinkPreview(
+      `https://www.youtube.com/watch?v=wuClZjOdT30`,
+      { proxyUrl: `https://cors-anywhere.herokuapp.com/`, headers: { Origin: `http://localhost:8000`, "Accept-Language": `en-US` } },
+    );
+
+    expect(linkInfo.url).toEqual(`https://www.youtube.com/watch?v=wuClZjOdT30`);
+    expect(linkInfo.siteName).toEqual(`YouTube`);
+    expect(linkInfo.title).toEqual(`Geography Now! Germany`);
+    expect(linkInfo.description).toBeTruthy();
+    expect(linkInfo.mediaType).toEqual(`video.other`);
+    expect(linkInfo.images.length).toEqual(1);
+    expect(linkInfo.images[0]).toEqual(
+      `https://i.ytimg.com/vi/wuClZjOdT30/maxresdefault.jpg`,
+    );
+    expect(linkInfo.videos.length).toEqual(0);
+    expect(linkInfo.favicons[0]).not.toBe(``);
+    expect(linkInfo.contentType.toLowerCase()).toEqual(
+      `text/html; charset=utf-8`,
+    );
+  });
 });

build/index.d.ts

 interface ILinkPreviewOptions {
     headers?: Record<string, string>;
     imagesPropertyType?: string;
+    proxyUrl?: string;
 }
@@ -5,0 +6,0 @@ export declare function getLinkPreview(text: string, options?: ILinkPreviewOptions): Promise<{

build/index.js

@@ -46,4 +46,9 @@ "use strict";
 var constants_1 = require("./constants");
+var metaTag = function (doc, type, attr) {
+    var nodes = doc("meta[" + attr + "='" + type + "']");
+    return nodes.length ? nodes : null;
+};
+var metaTagContent = function (doc, type, attr) { return doc("meta[" + attr + "='" + type + "']").attr("content"); };
 function getTitle(doc) {
-    var title = doc("meta[property='og:title']").attr("content");
+    var title = metaTagContent(doc, "og:title", "property") || metaTagContent(doc, "og:title", "name");
     if (!title) {
@@ -55,24 +60,19 @@ title = doc("title").text();
 function getSiteName(doc) {
-    var siteName = doc("meta[property='og:site_name']").attr("content");
+    var siteName = metaTagContent(doc, "og:site_name", "property") || metaTagContent(doc, "og:site_name", "name");
     return siteName;
 }
 function getDescription(doc) {
-    var description = doc("meta[name=description]").attr("content");
-    if (description === undefined) {
-        description = doc("meta[name=Description]").attr("content");
-    }
-    if (description === undefined) {
-        description = doc("meta[property='og:description']").attr("content");
-    }
+    var description = metaTagContent(doc, "description", "name") || metaTagContent(doc, "Description", "name") || metaTagContent(doc, "og:description", "property");
     return description;
 }
 function getMediaType(doc) {
-    var node = doc("meta[name=medium]");
-    if (node.length) {
+    var node = metaTag(doc, "medium", "name");
+    if (node) {
         var content = node.attr("content");
         return content === "image" ? "photo" : content;
     }
-    return doc("meta[property='og:type']").attr("content");
+    return (metaTagContent(doc, "og:type", "property") || metaTagContent(doc, "og:type", "name"));
 }
 function getImages(doc, rootUrl, imagesPropertyType) {
+    var _a;
     var images = [];
@@ -83,4 +83,4 @@ var nodes;
     var imagePropertyType = (imagesPropertyType !== null && imagesPropertyType !== void 0 ? imagesPropertyType : "og");
-    nodes = doc("meta[property='" + imagePropertyType + ":image']");
-    if (nodes.length) {
+    nodes = metaTag(doc, imagePropertyType + ":image", "property") || metaTag(doc, imagePropertyType + ":image", "name");
+    if (nodes) {
         nodes.each(function (_, node) {
@@ -102,3 +102,3 @@ src = node.attribs.content;
             nodes = doc("img");
-            if (nodes.length) {
+            if ((_a = nodes) === null || _a === void 0 ? void 0 : _a.length) {
                 dic = {};
@@ -121,2 +121,3 @@ images = [];
 function getVideos(doc) {
+    var _a;
     var videos = [];
@@ -134,10 +135,9 @@ var nodeTypes;
     var index;
-    var nodes = doc("meta[property='og:video']");
-    var length = nodes.length;
-    if (length) {
-        nodeTypes = doc("meta[property='og:video:type']");
-        nodeSecureUrls = doc("meta[property='og:video:secure_url']");
-        width = doc("meta[property='og:video:width']").attr("content");
-        height = doc("meta[property='og:video:height']").attr("content");
-        for (index = 0; index < length; index += 1) {
+    var nodes = metaTag(doc, "og:video", "property") || metaTag(doc, "og:video", "name");
+    if ((_a = nodes) === null || _a === void 0 ? void 0 : _a.length) {
+        nodeTypes = metaTag(doc, "og:video:type", "property") || metaTag(doc, "og:video:type", "name");
+        nodeSecureUrls = metaTag(doc, "og:video:secure_url", "property") || metaTag(doc, "og:video:secure_url", "name");
+        width = metaTagContent(doc, "og:video:width", "property") || metaTagContent(doc, "og:video:width", "name");
+        height = metaTagContent(doc, "og:video:height", "property") || metaTagContent(doc, "og:video:height", "name");
+        for (index = 0; index < nodes.length; index += 1) {
             video = nodes[index].attribs.content;
@@ -251,7 +251,7 @@ nodeType = nodeTypes[index];
 function getLinkPreview(text, options) {
-    var _a, _b;
+    var _a, _b, _c, _d;
     return __awaiter(this, void 0, void 0, function () {
-        var detectedUrl, fetchOptions, response, finalUrl, contentType, htmlString_1, htmlString_2, htmlString, e_1;
-        return __generator(this, function (_c) {
-            switch (_c.label) {
+        var detectedUrl, fetchOptions, fetchUrl, response, finalUrl, contentType, htmlString_1, htmlString_2, htmlString, e_1;
+        return __generator(this, function (_e) {
+            switch (_e.label) {
                 case 0:
@@ -261,6 +261,3 @@ if (!text || typeof text !== "string") {
                     }
-                    detectedUrl = text
-                        .replace(/\n/g, " ")
-                        .split(" ")
-                        .find(function (token) { return constants_1.CONSTANTS.REGEX_VALID_URL.test(token); });
+                    detectedUrl = text.replace(/\n/g, " ").split(" ").find(function (token) { return constants_1.CONSTANTS.REGEX_VALID_URL.test(token); });
                     if (!detectedUrl) {
@@ -270,9 +267,11 @@ throw new Error("link-preview-js did not receive a valid a url or text");
                     fetchOptions = { headers: (_b = (_a = options) === null || _a === void 0 ? void 0 : _a.headers, (_b !== null && _b !== void 0 ? _b : {})) };
-                    _c.label = 1;
+                    fetchUrl = ((_c = options) === null || _c === void 0 ? void 0 : _c.proxyUrl) ? options.proxyUrl.concat(detectedUrl) : detectedUrl;
+                    _e.label = 1;
                 case 1:
-                    _c.trys.push([1, 8, , 9]);
-                    return [4 /*yield*/, cross_fetch_1.fetch(detectedUrl, fetchOptions)];
+                    _e.trys.push([1, 8, , 9]);
+                    return [4 /*yield*/, cross_fetch_1.fetch(fetchUrl, fetchOptions)];
                 case 2:
-                    response = _c.sent();
-                    finalUrl = response.url;
+                    response = _e.sent();
+                    finalUrl = ((_d = options) === null || _d === void 0 ? void 0 : _d.proxyUrl) ? response.url.replace(options.proxyUrl, "")
+                        : response.url;
                     contentType = response.headers.get("content-type");
@@ -282,3 +281,3 @@ if (!!contentType) return [3 /*break*/, 4];
                 case 3:
-                    htmlString_1 = _c.sent();
+                    htmlString_1 = _e.sent();
                     return [2 /*return*/, parseUnknownResponse(htmlString_1, finalUrl, options)];
@@ -303,3 +302,3 @@ case 4:
                 case 5:
-                    htmlString_2 = _c.sent();
+                    htmlString_2 = _e.sent();
                     return [2 /*return*/, parseTextResponse(htmlString_2, finalUrl, options, contentType)];
@@ -312,6 +311,6 @@ case 6:
                 case 7:
-                    htmlString = _c.sent();
+                    htmlString = _e.sent();
                     return [2 /*return*/, parseUnknownResponse(htmlString, finalUrl, options)];
                 case 8:
-                    e_1 = _c.sent();
+                    e_1 = _e.sent();
                     throw new Error("link-preview-js could not fetch link information " + e_1.toString());
@@ -318,0 +317,0 @@ case 9: return [2 /*return*/];

index.ts

@@ -9,11 +9,17 @@ import cheerio from "cheerio-without-node-native";
   imagesPropertyType?: string;
+  proxyUrl?: string;
 }
 
+const metaTag = (doc: any, type: string, attr: string) => {
+  const nodes = doc(`meta[${attr}='${type}']`);
+  return nodes.length ? nodes : null;
+};
+
+const metaTagContent = (doc: any, type: string, attr: string) => doc(`meta[${attr}='${type}']`).attr(`content`);
+
 function getTitle(doc: any) {
-  let title = doc(`meta[property='og:title']`).attr(`content`);
-
+  let title = metaTagContent(doc, `og:title`, `property`) || metaTagContent(doc, `og:title`, `name`);
   if (!title) {
     title = doc(`title`).text();
   }
-
   return title;
@@ -23,4 +29,3 @@ }
 function getSiteName(doc: any) {
-  const siteName = doc(`meta[property='og:site_name']`).attr(`content`);
-
+  const siteName = metaTagContent(doc, `og:site_name`, `property`) || metaTagContent(doc, `og:site_name`, `name`);
   return siteName;
@@ -30,12 +35,3 @@ }
 function getDescription(doc: any) {
-  let description = doc(`meta[name=description]`).attr(`content`);
-
-  if (description === undefined) {
-    description = doc(`meta[name=Description]`).attr(`content`);
-  }
-
-  if (description === undefined) {
-    description = doc(`meta[property='og:description']`).attr(`content`);
-  }
-
+  const description = metaTagContent(doc, `description`, `name`) || metaTagContent(doc, `Description`, `name`) || metaTagContent(doc, `og:description`, `property`);
   return description;
@@ -45,9 +41,8 @@ }
 function getMediaType(doc: any) {
-  const node = doc(`meta[name=medium]`);
-
-  if (node.length) {
+  const node = metaTag(doc, `medium`, `name`);
+  if (node) {
     const content = node.attr(`content`);
     return content === `image` ? `photo` : content;
   }
-  return doc(`meta[property='og:type']`).attr(`content`);
+  return (metaTagContent(doc, `og:type`, `property`) || metaTagContent(doc, `og:type`, `name`));
 }
@@ -62,5 +57,5 @@
   const imagePropertyType = imagesPropertyType ?? `og`;
-  nodes = doc(`meta[property='${imagePropertyType}:image']`);
+  nodes = metaTag(doc, `${imagePropertyType}:image`, `property`) || metaTag(doc, `${imagePropertyType}:image`, `name`);
 
-  if (nodes.length) {
+  if (nodes) {
     nodes.each((_: number, node: any) => {
@@ -83,3 +78,3 @@ src = node.attribs.content;
 
-      if (nodes.length) {
+      if (nodes?.length) {
         dic = {};
@@ -117,12 +112,11 @@ images = [];
 
-  const nodes = doc(`meta[property='og:video']`);
-  const { length } = nodes;
+  const nodes = metaTag(doc, `og:video`, `property`) || metaTag(doc, `og:video`, `name`);
 
-  if (length) {
-    nodeTypes = doc(`meta[property='og:video:type']`);
-    nodeSecureUrls = doc(`meta[property='og:video:secure_url']`);
-    width = doc(`meta[property='og:video:width']`).attr(`content`);
-    height = doc(`meta[property='og:video:height']`).attr(`content`);
+  if (nodes?.length) {
+    nodeTypes = metaTag(doc, `og:video:type`, `property`) || metaTag(doc, `og:video:type`, `name`);
+    nodeSecureUrls = metaTag(doc, `og:video:secure_url`, `property`) || metaTag(doc, `og:video:secure_url`, `name`);
+    width = metaTagContent(doc, `og:video:width`, `property`) || metaTagContent(doc, `og:video:width`, `name`);
+    height = metaTagContent(doc, `og:video:height`, `property`) || metaTagContent(doc, `og:video:height`, `name`);
 
-    for (index = 0; index < length; index += 1) {
+    for (index = 0; index < nodes.length; index += 1) {
       video = nodes[index].attribs.content;
@@ -159,3 +153,2 @@
 
-
 // returns an array of URL's to favicon images
@@ -271,6 +264,3 @@ function getFavicons(doc: any, rootUrl: string) {
 
-  const detectedUrl = text
-    .replace(/\n/g, ` `)
-    .split(` `)
-    .find((token) => CONSTANTS.REGEX_VALID_URL.test(token));
+  const detectedUrl = text.replace(/\n/g, ` `).split(` `).find((token) => CONSTANTS.REGEX_VALID_URL.test(token));
 
@@ -283,7 +273,11 @@ if (!detectedUrl) {
 
+  const fetchUrl = options?.proxyUrl ? options.proxyUrl.concat(detectedUrl) : detectedUrl;
+
   try {
-    const response = await fetch(detectedUrl, fetchOptions);
+    const response = await fetch(fetchUrl, fetchOptions);
 
-    // get final URL (after any redirects)
-    const finalUrl = response.url;
+    // get final URL (after any redirects, strip out proxy url from response url)
+    const finalUrl = options?.proxyUrl
+      ? response.url.replace(options.proxyUrl, ``)
+      : response.url;
 
@@ -290,0 +284,0 @@ // get content type of response

package.json

 {
   "name": "link-preview-js",
-  "version": "2.0.4",
+  "version": "2.0.5",
   "description": "Javascript module to extract and fetch HTTP link information from blocks of text. ",
@@ -5,0 +5,0 @@ "main": "build/index.js",

README.md

 # link-preview-js
 
-[![CircleCI](https://circleci.com/gh/ospfranco/link-preview-js/tree/master.svg?style=svg)](https://circleci.com/gh/ospfranco/link-preview-js/tree/master) [![npm version](https://badge.fury.io/js/link-preview-js.svg)](https://badge.fury.io/js/link-preview-js)
-
 Typescript library that allows you to extract information from a URL or parse text and retrieve information from the first available link.
 
-## This library does not work on CORS protected environments, i.e: all the major browsers
+## Does not work on CORS protected environments, means: **all the browsers**
 
-Chrome, Firefox, Safari, etc DO NOT ALLOW YOU TO DO CROSS SITE REQUESTS therefore you cannot request another domain from your web application, read more about [CORS](https://developer.mozilla.org/en-US/docs/Web/HTTP/CORS).
+A (respectable) browser **DOES NOT ALLOW YOU TO DO CROSS ORIGIN REQUESTS**, you cannot do a request to a different domain from your web application, if do not know how *same-origin-policy* works you can read [this](https://dev.to/lydiahallie/cs-visualized-cors-5b8h) fantastic piece written by @lydiahallie
 
-## This library uses [cheerio-without-native](https://github.com/oyyd/cheerio-without-node-native), github is now warning me that there are security vunerabilities because the package has been abandonded, I'm not responsible for any security implications this might carry, I could use cheerio but that means loosing compatibility with RN, which actually might not be a bad idea...
 
-As of 23 of April of 2020: Do not use https://google.com it does not return the appropiate tags to be parsed
+## Security disclaimer
+This library uses [cheerio-without-native](https://github.com/oyyd/cheerio-without-node-native) which unfortunately has been abandoned and now has some security vunerabilities (according to github) I'm not responsible for any security implications this might carry
 
-# Migration to 2.X.X
+As of 23 of April of 2020: Do not use https://google.com it does not return the appropiate meta-tags to be parsed, so test with another domain, don't waste your time
 
-The api for version 2.X.X changed slightly, there is no longer a default unnamed export, only a named method export `getLinkPreview`
-
 ## Install
 
-`$ yarn add link-preview-js`
+```
+yarn add link-preview-js
+```
 
@@ -32,5 +30,9 @@ ## Usage
 
+// pass the link directly
 getLinkPreview('https://www.youtube.com/watch?v=MejbOFk7H6c')
   .then((data) => console.debug(data));
 
+// OR
+
+// a chunk of text
 getLinkPreview('This is a text supposed to be parsed and the first link displayed https://www.youtube.com/watch?v=MejbOFk7H6c')
@@ -135,5 +137,1 @@ .then((data) => console.debug(data));
 MIT license
-
-## Sponsor
-
-If you find this package useful, please considering [sponsoring](https://github.com/sponsors/ospfranco), buying a coffee is enough, thanks!

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

47521

increased by6.51%

Number of package files

24

increased by4.35%

Lines of code

937

increased by1.08%

Worsened metrics

Number of lines in readme file

136

decreased by-1.45%

No dependency changes