Update to new YouTube API

2020-07-15 19:14:31 +02:00 · 2020-07-15 19:14:31 +02:00 · e7d87e3bbd
parent 8bd77e09e5
commit e7d87e3bbd
5 changed files with 150 additions and 171 deletions
--- a/explore/api.js
+++ b/explore/api.js
@ -5,20 +5,24 @@ const request = require('request');
 const cheerio = require('cheerio');

 const YOUTUBE_BASE = 'https://www.youtube.com/%s';
+const WATCH_BASE = util.format(YOUTUBE_BASE, 'watch?v=%s');
+
 const VIDEO_ID_REGEX = /watch\?v=([^&]*)/i;
-const END_SCREEN_BASE = util.format(YOUTUBE_BASE, 'get_endscreen?v=%s');
-const CARD_BASE = util.format(YOUTUBE_BASE, 'annotations_invideo?video_id=%s');
+// const END_SCREEN_BASE = util.format(YOUTUBE_BASE, 'get_endscreen?v=%s');
+// const CARD_BASE = util.format(YOUTUBE_BASE, 'annotations_invideo?video_id=%s');
+
+const playerRegex = /ytplayer\.config = (\{.*?\});/;

 /**
- * Récupère une liste des vidéos liées à la fin d’une autre vidéo.
+ * Fetch the `ytplayer.config` object for a YouTube video.
 *
- * @param videoId Identifiant de la vidéo source.
- * @return Liste des vidéos liées. Chaque vidéo est un objet contenant
- * son identifiant (videoId) et son titre (title).
+ * @async
+ * @param videoId Identifier of the video to fetch.
+ * @return The player configuration object.
 */
-const getEndScreenVideos = videoId =>
+const getPlayerConfig = videoId =>
 {
-    const url = util.format(END_SCREEN_BASE, videoId);
+    const url = util.format(WATCH_BASE, videoId);

    return new Promise((resolve, reject) =>
    {
@ -30,87 +34,84 @@ const getEndScreenVideos = videoId =>
                return;
            }

-            // Suppression des caractères initiaux inutiles si applicable
-            if (body.substr(0, 3) === ')]}')
+            // Look for the definition of ytplayer.config and unserialize it
+            // and the player_response subobject
+            try
            {
-                body = body.substr(3);
+                const playerConfig = JSON.parse(body.match(playerRegex)[1]);
+                playerConfig.args.player_response
+                    = JSON.parse(playerConfig.args.player_response);
+                resolve(playerConfig);
            }
-
-            // Interprétation du JSON
-            const data = JSON.parse(body);
-
-            // Aucun écran de fin
-            if (
-                typeof data !== 'object' || data === null ||
-                data.elements === undefined
-            )
+            catch (err)
            {
-                resolve([]);
-                return;
+                reject(err);
            }
-
-            // Filtrage des données pour extraire le titre et l’ID des vidéos
-            resolve(data.elements.map(elt =>
-            {
-                const data = elt.endscreenElementRenderer;
-                const videoIdResults = VIDEO_ID_REGEX.exec(
-                    data.endpoint.urlEndpoint.url
-                );
-
-                return {
-                    videoId: videoIdResults ? videoIdResults[1] : null,
-                    title: data.title.simpleText,
+        });
+    });
 };
-            }));
-        });
+
+exports.getPlayerConfig = getPlayerConfig;
+
+/**
+ * Get metadata about a YouTube video.
+ *
+ * @param playerConfig The `ytplayer.config` object corresponding to the source
+ * YouTube video, as obtained from `getPlayerConfig`.
+ * @return Object containing the video metadata.
+ */
+const getVideoMeta = playerConfig => ({
+    videoId: playerConfig.args.player_response.videoDetails.videoId,
+    title: playerConfig.args.player_response.videoDetails.title,
 });
+
+exports.getVideoMeta = getVideoMeta;
+
+/**
+ * Find videos linked from the endscreen of a YouTube video.
+ *
+ * @param playerConfig The `ytplayer.config` object corresponding to the source
+ * YouTube video, as obtained from `getPlayerConfig`.
+ * @return List of identifiers of linked videos.
+ */
+const getEndScreenVideos = playerConfig =>
+{
+    const response = playerConfig.args.player_response;
+
+    if (!('endscreen' in response))
+    {
+        return [];
+    }
+
+    return response.endscreen.endscreenRenderer.elements
+        .map(elt => elt.endscreenElementRenderer)
+        .filter(rdr => 'watchEndpoint' in rdr.endpoint)
+        .map(rdr => rdr.endpoint.watchEndpoint.videoId);
 };

 exports.getEndScreenVideos = getEndScreenVideos;

 /**
- * Récupère une liste des vidéos liées en tant que carte d’une autre vidéo.
+ * Find videos linked from as cards from a YouTube video.
 *
- * @param videoId Identifiant de la vidéo source.
- * @return Liste des vidéos liées. Chaque vidéo est un objet contenant
- * son identifiant (videoId) et son titre (title).
+ * @param playerConfig The `ytplayer.config` object corresponding to the source
+ * YouTube video, as obtained from `getPlayerConfig`.
+ * @return List of identifiers of linked videos.
 */
-const getCardVideos = videoId =>
+const getCardVideos = playerConfig =>
 {
-    const url = util.format(CARD_BASE, videoId);
+    const response = playerConfig.args.player_response;

-    return new Promise((resolve, reject) =>
+    if (!('cards' in response))
    {
-        request(url, (err, res, body) =>
-        {
-            if (err)
-            {
-                reject(err);
-                return;
+        return [];
    }

-            // Interprétation du XML externe et recherche des annotations
-            // de type carte
-            const nav = cheerio.load(body);
-            const cards = nav('annotation[type="card"][style="video"]');
-            const list = [];
-
-            cards.each((i, el) =>
-            {
-                // Interprétation du JSON de chaque carte, extraction
-                // du titre et de l’identifiant de la vidéo
-                const data = JSON.parse(nav(el).children('data').text());
-                const videoIdResults = VIDEO_ID_REGEX.exec(data.url);
-
-                list.push({
-                    title: data.title,
-                    videoId: videoIdResults ? videoIdResults[1] : null
-                });
-            });
-
-            resolve(list);
-        });
-    });
+    return response.cards.cardCollectionRenderer.cards
+        .map(card => card.cardRenderer.content)
+        .filter(content => 'videoInfoCardContentRenderer' in content)
+        .map(content => content.videoInfoCardContentRenderer)
+        .map(rdr => rdr.action.watchEndpoint.videoId);
 };

 exports.getCardVideos = getCardVideos;
--- a/explore/graph.js
+++ b/explore/graph.js
@ -7,19 +7,19 @@ const GRAPH_NODE_URL = '    "%s" [label="%s", URL="%s", fontcolor=blue]';
 const GRAPH_LINK = '    "%s" -> "%s"';

 /**
- * Convertit un graphe en format DOT.
+ * Convert a graph to the DOT format.
 *
- * @param graph Graphe à convertir.
- * @param [title=identity] Fonction donnant le titre de chaque nœud du graphe.
- * @param [url=none] Fonction donnant l’URL de chaque nœud du graphe, ou la
- * chaîne vide si un nœud n’a pas d’URL.
- * @return Représentation DOT du graphe.
+ * @param graph Graph to convert.
+ * @param [title=identity] Function giving the name of each node.
+ * @param [url=none] Function given the URL of each node, or an empty string
+ * if a node has no URL.
+ * @return DOT representation of the graph.
 */
 const graphToDOT = (graph, title = id => id, url = () => '') =>
 {
    const ser = graph.serialize();

-    // Conversion des nœuds de vidéo
+    // Convert nodes
    const nodes = ser.nodes.map(({id}) =>
    {
        const nodeTitle = title(id);
@ -36,7 +36,7 @@ const graphToDOT = (graph, title = id => id, url = () => '') =>
    }
    ).join('\n');

-    // Conversion des liens de vidéo
+    // Convert edges
    const links = ser.links.map(({source, target}) =>
        util.format(GRAPH_LINK, source, target)
    ).join('\n');
--- a/explore/index.js
+++ b/explore/index.js
@ -5,74 +5,59 @@ const fs = require('fs');
 const path = require('path');
 const util = require('util');

-const {getEndScreenVideos, getCardVideos} = require('./api');
+const api = require('./api');
 const {graphToDOT} = require('./graph');

 const YOUTUBE_WATCH = 'https://youtu.be/%s';

-// Récupère le chemin vers le fichier de sortie depuis la ligne de commande
+// Fetch the output path from command line
 if (process.argv.length !== 3)
 {
-    console.error(`Utilisation : node src [sortie]`);
+    console.error(`Usage: node explore [output]`);
    process.exit(1);
 }

 const dest = process.argv[2];

-// Graphe des vidéos. Chaque nœud est une vidéo, et est liée à toutes
-// les vidéos vers lesquelles elle a un lien (par carte ou écran de fin)
+// Graph of visited videos. Each node is a video which is linked to all the
+// videos to which there is a link, either through a card or an endscreen item
 const videosGraph = Graph();

-// Stocke les métadonnées de chaque vidéo
+// Store metadata about each visited video
 const videosMeta = Object.create(null);

-// Se souvient des vidéos déja visitées
-const visitedVideos = Object.create(null);
-
 /**
- * Explore récursivement les liens d’une vidéo pour remplir le graphe
- * des vidéos.
+ * Recursively explore a video and the video linked from it to fill
+ * the video graph.
 *
- * @param videoId Identifiant de la vidéo source.
+ * @param videoId Source video identifier.
 */
-const exploreVideo = videoId =>
+const exploreVideo = async videoId =>
 {
-    // S’assure de ne pas visiter deux fois la même vidéo
-    if (visitedVideos[videoId] === true)
+    // Make sure we don’t explore the same video twice
+    if (videoId in videosMeta)
    {
        return Promise.resolve();
    }

-    visitedVideos[videoId] = true;
+    const playerConfig = await api.getPlayerConfig(videoId);
+    videosMeta[videoId] = api.getVideoMeta(playerConfig);

-    return Promise.all([
-        getEndScreenVideos(videoId),
-        getCardVideos(videoId)
-    ]).then(results =>
-    {
-        const videos = [].concat(...results);
-        const ids = videos.map(video => video.videoId);
+    const linkedVideos = [
+        ...api.getEndScreenVideos(playerConfig),
+        ...api.getCardVideos(playerConfig),
+    ];

-        // Ajout des vidéos liées dans les métadonnées
-        videos.forEach(video => videosMeta[video.videoId] = video);
+    // Add links between this video and the linked ones
+    linkedVideos.forEach(id => videosGraph.addEdge(videoId, id));

-        // Ajout des liens entre la vidéo et ses vidéos liées
-        ids.forEach(id => videosGraph.addEdge(videoId, id));
-
-        // Récursion sur les vidéos non-explorées
-        return Promise.all(ids.map(id => exploreVideo(id)));
-    });
+    // Recurse on linked videos
+    return Promise.all(linkedVideos.map(id => exploreVideo(id)));
 };

-// Métadonnées de la vidéo source
+// Metadata of the source video
 const rootVideoId = 'EZGra6O8ClQ';
-videosMeta[rootVideoId] = {
-    title: '1 avril 2017 : présentation des règles',
-    videoId: rootVideoId,
-    links: []
-};
-
-console.log('Démarrage de l’exploration !');
+console.log('Starting to explore!');

 exploreVideo(rootVideoId).then(() =>
 {
@ -85,5 +70,5 @@ exploreVideo(rootVideoId).then(() =>
        )
    );

-    console.log('Terminé. Résultat dans ' + dest);
+    console.log('Finished. Result in ' + dest);
 }).catch(console.error);
--- a/package-lock.json
+++ b/package-lock.json
@ -10,11 +10,11 @@
      "integrity": "sha512-Uy0PN4R5vgBUXFoJrKryf5aTk3kJ8Rv3PdlHjl6UaX+Cqp1QE0yPQ68MPXGrZOfG7gZVNDIJZYyot0B9ubXUrQ=="
    },
    "ajv": {
-      "version": "6.10.1",
-      "resolved": "https://registry.npmjs.org/ajv/-/ajv-6.10.1.tgz",
-      "integrity": "sha512-w1YQaVGNC6t2UCPjEawK/vo/dG8OOrVtUmhBT1uJJYxbl5kU2Tj3v6LGqBcsysN1yhuCStJCCA3GqdvKY8sqXQ==",
+      "version": "6.12.3",
+      "resolved": "https://registry.npmjs.org/ajv/-/ajv-6.12.3.tgz",
+      "integrity": "sha512-4K0cK3L1hsqk9xIb2z9vs/XU+PGJZ9PNpJRDS9YLzmNdX6jmVPfamLvTJr0aDAusnHyCHO6MjzlkAsgtqp9teA==",
      "requires": {
-        "fast-deep-equal": "^2.0.1",
+        "fast-deep-equal": "^3.1.1",
        "fast-json-stable-stringify": "^2.0.0",
        "json-schema-traverse": "^0.4.1",
        "uri-js": "^4.2.2"
@ -44,9 +44,9 @@
      "integrity": "sha1-tG6JCTSpWR8tL2+G1+ap8bP+dqg="
    },
    "aws4": {
-      "version": "1.8.0",
-      "resolved": "https://registry.npmjs.org/aws4/-/aws4-1.8.0.tgz",
-      "integrity": "sha512-ReZxvNHIOv88FlT7rxcXIIC0fPt4KZqZbOlivyWtXLt8ESx84zd3kMC6iK5jVeS2qt+g7ftS7ye4fi06X5rtRQ=="
+      "version": "1.10.0",
+      "resolved": "https://registry.npmjs.org/aws4/-/aws4-1.10.0.tgz",
+      "integrity": "sha512-3YDiu347mtVtjpyV3u5kVqQLP242c06zwDOgpeRnybmXlYYsLbtTrUBUm8i8srONt+FWobl5aibnU1030PeeuA=="
    },
    "bcrypt-pbkdf": {
      "version": "1.0.2",
@ -177,14 +177,14 @@
      "integrity": "sha1-lpGEQOMEGnpBT4xS48V06zw+HgU="
    },
    "fast-deep-equal": {
-      "version": "2.0.1",
-      "resolved": "https://registry.npmjs.org/fast-deep-equal/-/fast-deep-equal-2.0.1.tgz",
-      "integrity": "sha1-ewUhjd+WZ79/Nwv3/bLLFf3Qqkk="
+      "version": "3.1.3",
+      "resolved": "https://registry.npmjs.org/fast-deep-equal/-/fast-deep-equal-3.1.3.tgz",
+      "integrity": "sha512-f3qQ9oQy9j2AhBe/H9VC91wLmKBCCU/gDOnKNAYG5hswO7BLKj09Hc5HYNz9cGI++xlpDCIgDaitVs03ATR84Q=="
    },
    "fast-json-stable-stringify": {
-      "version": "2.0.0",
-      "resolved": "https://registry.npmjs.org/fast-json-stable-stringify/-/fast-json-stable-stringify-2.0.0.tgz",
-      "integrity": "sha1-1RQsDK7msRifh9OnYREGT4bIu/I="
+      "version": "2.1.0",
+      "resolved": "https://registry.npmjs.org/fast-json-stable-stringify/-/fast-json-stable-stringify-2.1.0.tgz",
+      "integrity": "sha512-lhd/wF+Lk98HZoTCtlVraHtfh5XYijIjalXck7saUtuanSDyLMxnHhSXEDJqHxD7msR8D0uCmqlkwjCV8xvwHw=="
    },
    "forever-agent": {
      "version": "0.6.1",
@ -210,9 +210,9 @@
      }
    },
    "graph-data-structure": {
-      "version": "1.8.0",
-      "resolved": "https://registry.npmjs.org/graph-data-structure/-/graph-data-structure-1.8.0.tgz",
-      "integrity": "sha512-G2Jl7JLGsq0FGBvFMfKkQEwjnk+1//ssIirR5GTQxXBbKqYZkmjDFEVcJ8H87dVpw4D8lZDJj9v9ggnHWW8M+g=="
+      "version": "1.12.1",
+      "resolved": "https://registry.npmjs.org/graph-data-structure/-/graph-data-structure-1.12.1.tgz",
+      "integrity": "sha512-0DHxFEUk2EHO19PQrcOckz91WZPk7Itl2mmNpGdpSIZUtBUHRVJPtuuZCJAXB69YRL4fKDCRv2cX3ly8aZZ0QQ=="
    },
    "har-schema": {
      "version": "2.0.0",
@ -298,21 +298,21 @@
      }
    },
    "lodash": {
-      "version": "4.17.11",
-      "resolved": "https://registry.npmjs.org/lodash/-/lodash-4.17.11.tgz",
-      "integrity": "sha512-cQKh8igo5QUhZ7lg38DYWAxMvjSAKG0A8wGSVimP07SIUEK2UO+arSRKbRZWtelMtN5V0Hkwh5ryOto/SshYIg=="
+      "version": "4.17.19",
+      "resolved": "https://registry.npmjs.org/lodash/-/lodash-4.17.19.tgz",
+      "integrity": "sha512-JNvd8XER9GQX0v2qJgsaN/mzFCNA5BRe/j8JN9d+tWyGLSodKQHKFicdwNYzWwI3wjRnaKPsGj1XkBjx/F96DQ=="
    },
    "mime-db": {
-      "version": "1.40.0",
-      "resolved": "https://registry.npmjs.org/mime-db/-/mime-db-1.40.0.tgz",
-      "integrity": "sha512-jYdeOMPy9vnxEqFRRo6ZvTZ8d9oPb+k18PKoYNYUe2stVEBPPwsln/qWzdbmaIvnhZ9v2P+CuecK+fpUfsV2mA=="
+      "version": "1.44.0",
+      "resolved": "https://registry.npmjs.org/mime-db/-/mime-db-1.44.0.tgz",
+      "integrity": "sha512-/NOTfLrsPBVeH7YtFPgsVWveuL+4SjjYxaQ1xtM1KMFj7HdxlBlxeyNLzhyJVx7r4rZGJAZ/6lkKCitSc/Nmpg=="
    },
    "mime-types": {
-      "version": "2.1.24",
-      "resolved": "https://registry.npmjs.org/mime-types/-/mime-types-2.1.24.tgz",
-      "integrity": "sha512-WaFHS3MCl5fapm3oLxU4eYDw77IQM2ACcxQ9RIxfaC3ooc6PFuBMGZZsYpvoXS5D5QTWPieo1jjLdAm3TBP3cQ==",
+      "version": "2.1.27",
+      "resolved": "https://registry.npmjs.org/mime-types/-/mime-types-2.1.27.tgz",
+      "integrity": "sha512-JIhqnCasI9yD+SsmkquHBxTSEuZdQX5BuQnS2Vc7puQQQ+8yiP5AY5uWhpdv4YL4VM5c6iliiYWPgJ/nJQLp7w==",
      "requires": {
-        "mime-db": "1.40.0"
+        "mime-db": "1.44.0"
      }
    },
    "nth-check": {
@ -342,9 +342,9 @@
      "integrity": "sha1-Ywn04OX6kT7BxpMHrjZLSzd8nns="
    },
    "psl": {
-      "version": "1.2.0",
-      "resolved": "https://registry.npmjs.org/psl/-/psl-1.2.0.tgz",
-      "integrity": "sha512-GEn74ZffufCmkDDLNcl3uuyF/aSD6exEyh1v/ZSdAomB82t6G9hzJVRx0jBmLDW+VfZqks3aScmMw9DszwUalA=="
+      "version": "1.8.0",
+      "resolved": "https://registry.npmjs.org/psl/-/psl-1.8.0.tgz",
+      "integrity": "sha512-RIdOzyoavK+hA18OGGWDqUTsCLhtA7IcZ/6NCs4fFJaHBDab+pDDmDIByWFRQJq2Cd7r1OoQxBGKOaztq+hjIQ=="
    },
    "punycode": {
      "version": "2.1.1",
@ -367,9 +367,9 @@
      }
    },
    "request": {
-      "version": "2.88.0",
-      "resolved": "https://registry.npmjs.org/request/-/request-2.88.0.tgz",
-      "integrity": "sha512-NAqBSrijGLZdM0WZNsInLJpkJokL72XYjUpnB0iwsRgxh7dB6COrHnTBNwN0E+lHDAJzu7kLAkDeY08z2/A0hg==",
+      "version": "2.88.2",
+      "resolved": "https://registry.npmjs.org/request/-/request-2.88.2.tgz",
+      "integrity": "sha512-MsvtOrfG9ZcrOwAW+Qi+F6HbD0CWXEh9ou77uOb7FM2WPhwT7smM833PzanhJLsgXjN89Ir6V2PczXNnMpwKhw==",
      "requires": {
        "aws-sign2": "~0.7.0",
        "aws4": "^1.8.0",
@ -378,7 +378,7 @@
        "extend": "~3.0.2",
        "forever-agent": "~0.6.1",
        "form-data": "~2.3.2",
-        "har-validator": "~5.1.0",
+        "har-validator": "~5.1.3",
        "http-signature": "~1.2.0",
        "is-typedarray": "~1.0.0",
        "isstream": "~0.1.2",
@ -388,7 +388,7 @@
        "performance-now": "^2.1.0",
        "qs": "~6.5.2",
        "safe-buffer": "^5.1.2",
-        "tough-cookie": "~2.4.3",
+        "tough-cookie": "~2.5.0",
        "tunnel-agent": "^0.6.0",
        "uuid": "^3.3.2"
      }
@ -428,19 +428,12 @@
      }
    },
    "tough-cookie": {
-      "version": "2.4.3",
-      "resolved": "https://registry.npmjs.org/tough-cookie/-/tough-cookie-2.4.3.tgz",
-      "integrity": "sha512-Q5srk/4vDM54WJsJio3XNn6K2sCG+CQ8G5Wz6bZhRZoAe/+TxjWB/GlFAnYEbkYVlON9FMk/fE3h2RLpPXo4lQ==",
+      "version": "2.5.0",
+      "resolved": "https://registry.npmjs.org/tough-cookie/-/tough-cookie-2.5.0.tgz",
+      "integrity": "sha512-nlLsUzgm1kfLXSXfRZMc1KLAugd4hqJHDTvc2hDIwS3mZAfMEuMbc03SujMF+GEcpaX/qboeycw6iO8JwVv2+g==",
      "requires": {
-        "psl": "^1.1.24",
-        "punycode": "^1.4.1"
-      },
-      "dependencies": {
-        "punycode": {
-          "version": "1.4.1",
-          "resolved": "https://registry.npmjs.org/punycode/-/punycode-1.4.1.tgz",
-          "integrity": "sha1-wNWmOycYgArY4esPpSachN1BhF4="
-        }
+        "psl": "^1.1.28",
+        "punycode": "^2.1.1"
      }
    },
    "tunnel-agent": {
@ -470,9 +463,9 @@
      "integrity": "sha1-RQ1Nyfpw3nMnYvvS1KKJgUGaDM8="
    },
    "uuid": {
-      "version": "3.3.2",
-      "resolved": "https://registry.npmjs.org/uuid/-/uuid-3.3.2.tgz",
-      "integrity": "sha512-yXJmeNaw3DnnKAOKJE51sL/ZaYfWJRl1pK9dr19YFCu0ObS231AB1/LbqTKRAQ5kw8A90rA6fr4riOUpTZvQZA=="
+      "version": "3.4.0",
+      "resolved": "https://registry.npmjs.org/uuid/-/uuid-3.4.0.tgz",
+      "integrity": "sha512-HjSDRw6gZE5JMggctHBcjVak08+KEVhSIiDzFnT9S9aegmp85S/bReBVTb4QTFaRNptJ9kuYaNhnbNEOkbKb/A=="
    },
    "verror": {
      "version": "1.10.0",
--- a/package.json
+++ b/package.json
@ -9,7 +9,7 @@
  },
  "dependencies": {
    "cheerio": "1.0.0-rc.3",
-    "graph-data-structure": "^1.8.0",
-    "request": "^2.88.0"
+    "graph-data-structure": "^1.12.1",
+    "request": "^2.88.2"
  }
 }