Refactored media link parsing, started work on YT-DLP integration.

config.example.json

@@ -5,6 +5,7 @@
     "proxied": false,
     "protocol": "http",
     "domain": "localhost",
+    "ytdlpPath": "/home/canopy/.local/pipx/venvs/yt-dlp/bin/yt-dlp",
     "sessionSecret": "CHANGE_ME",
     "altchaSecret": "CHANGE_ME",
     "ipSecret": "CHANGE_ME",

config.example.jsonc

@@ -13,6 +13,9 @@
     "protocol": "http",
     //Domain the server is available at, used for server-side link generation
     "domain": "localhost",
+    //Path to YT-DLP Executable for scraping youtube, dailymotion, and vimeo
+    //Dailymotion and Vimeo could work using official apis w/o keys, but you wouldn't have any raw file playback options :P
+    "ytdlpPath": "/home/canopy/.local/pipx/venvs/yt-dlp/bin/yt-dlp",
     //Be careful with what you keep in secrets, you should use special chars, but test your deployment, as some chars may break account registration
     //An update to either kill the server and bitch it's planned so it's not so confusing for new admins
     //Session secret used to secure session keys

package.json

@@ -16,7 +16,8 @@
     "mongoose": "^8.4.3",
     "node-cron": "^3.0.3",
     "nodemailer": "^6.9.16",
-    "socket.io": "^4.8.1"
+    "socket.io": "^4.8.1",
+    "youtube-dl-exec": "^3.0.20"
   },
   "scripts": {
     "start": "node ./src/server.js",

src/app/channel/media/media.js

@@ -14,16 +14,14 @@ GNU Affero General Public License for more details.
 You should have received a copy of the GNU Affero General Public License
 along with this program.  If not, see <https://www.gnu.org/licenses/>.*/
 
-//Node imports
-const crypto = require('node:crypto');
-
 module.exports = class{
-    constructor(title, fileName, url, id, type, duration){
+    constructor(title, fileName, url, id, type, duration, rawLink = url){
         this.title = title;
         this.fileName = fileName
         this.url = url;
         this.id = id;
         this.type = type;
         this.duration = duration;
+        this.rawLink = rawLink;
     }
 }

src/utils/media/internetArchiveUtils.js

@@ -23,17 +23,9 @@ const media = require('../../app/channel/media/media.js');
 const regexUtils = require('../regexUtils.js');
 const loggerUtils = require('../loggerUtils.js')
 
-module.exports.fetchMetadata = async function(link, title){
-    //Parse link
-    const parsedLink = new url.URL(link);
-    //Split link path
-    const splitPath = parsedLink.pathname.split('/');
-    //Get ItemID from link path
-    const itemID = splitPath[2]
-    //Splice the empty string, request type, and item ID out from link path
-    splitPath.splice(0,3)
-    //Join remaining link path back together to get requested file path within the given archive.org upload
-    const requestedPath = decodeURIComponent(splitPath.join('/'));
+module.exports.fetchMetadata = async function(fullID, title){
+    //Split fullID by first slash
+    const [itemID, requestedPath] = decodeURIComponent(fullID).split(/\/(.*)/);
     //Create empty list to hold media objects
     const mediaList = [];
     //Create empty variable to hold return data object
@@ -64,7 +56,7 @@ module.exports.fetchMetadata = async function(link, title){
 
 
     //If we're requesting an empty path
-    if(requestedPath == ''){
+    if(requestedPath == '' || requestedPath == null){
         //Return item metadata and compatible files
         data = {
                 files: compatibleFiles,

src/utils/media/yanker.js

@@ -19,37 +19,45 @@ const validator = require('validator');//No express here, so regular validator i
 
 //local import
 const iaUtil = require('./internetArchiveUtils');
+const ytdlpUtil = require('./ytdlpUtils');
 
 module.exports.yankMedia = async function(url, title){
     //Get pull type
     const pullType = await this.getMediaType(url);
 
     //Check pull type
-    switch(pullType){
+    switch(pullType.type){
         case "ia":
             //return media object list from IA module
-            return await iaUtil.fetchMetadata(url, title); 
+            return await iaUtil.fetchMetadata(pullType.id, title); 
         default:
             //return null to signify a bad url
             return null;
     }
 }
 
+//I'd be lying if this didn't take at least some inspiration/regex patterns from extractQueryParam() in cytube/forest's browser-side 'util.js'
+//Still this has some improvements like url pre-checks and the fact that it's handled serverside, recuing possibility of bad requests
 module.exports.getMediaType = async function(url){
-    //Encode URI in-case we where handed something a little too humie friendly
-    url = encodeURI(url);
-
-    //Check if we have a valid url
-    if(!validator.isURL(url)){
+    //Check if we have a valid url, encode it on the fly in case it's too humie-friendly
+    if(!validator.isURL(encodeURI(url))){
         //If not toss the fucker out
-        return null;
+        return {
+            type: null,
+            id: url
         }
-
     //If we have link to a resource from archive.org
-    if(url.match(/^https\:\/\/archive.org\//g)){
+    }else if(match = url.match(/archive\.org\/(?:details|download)\/([a-zA-Z0-9\/._-\s\%]+)/)){
         //return internet archive code
-        return "ia";
+        return {
+            type: "ia",
+            id: match[1]
+        }
     }
 
-    return null;
+    //If we fell through all of our media types without a match
+    return{
+        type: null,
+        id: url
+    }
 }

src/utils/media/ytdlpUtils.js

@@ -0,0 +1,65 @@
+/*Canopy - The next generation of stoner streaming software
+Copyright (C) 2024-2025 Rainbownapkin and the TTN Community
+
+This program is free software: you can redistribute it and/or modify
+it under the terms of the GNU Affero General Public License as
+published by the Free Software Foundation, either version 3 of the
+License, or (at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+GNU Affero General Public License for more details.
+
+You should have received a copy of the GNU Affero General Public License
+along with this program.  If not, see <https://www.gnu.org/licenses/>.*/
+
+//Config
+const config = require('../../../config.json');
+
+//Node Imports
+const { create: ytdlpMaker } = require('youtube-dl-exec');
+//Import ytdlp w/ custom path from config so we can force the newest build of yt-dlp from pip
+const ytdlp = ytdlpMaker(config.ytdlpPath);
+const url = require("node:url");
+const validator = require('validator');
+
+//Local Imports
+const media = require('../../app/channel/media/media.js');
+const regexUtils = require('../regexUtils.js');
+const loggerUtils = require('../loggerUtils.js')
+
+module.exports.fetchYoutubeVideoMetadata = async function(id, title){
+    const media = await fetchMetadata(`youtu.be/${id}`, title,'yt');
+
+    return media;
+}
+
+//Generic YTDLP function meant to be used by service-sepecific fetchers which will then be used to fetch video metadata
+async function fetchMetadata(link, title, type){
+    //Create media list
+    const mediaList = [];    
+
+    //Pull raw metadata
+    const rawMetadata = await ytdlp(link, {
+        dumpSingleJson: true,
+        format: 'b'
+    });
+
+    //Pull data from rawMetadata, sanatizing title to prevent XSS
+    const name = validator.escape(validator.trim(rawMetadata.title));
+    const rawLink = rawMetadata.requested_downloads[0].url;
+    const id = rawMetadata.id;
+
+    //if we where handed a null title
+    if(title == null || title == ''){
+        //Create new media object from file info substituting filename for title
+        mediaList.push(new media(name, name, link, id, type, Number(rawMetadata.duration), rawLink));
+    }else{
+        //Create new media object from file info
+        mediaList.push(new media(title, name, link, id, type, Number(rawMetadata.duration), rawLink));
+    }
+
+    //Return list of media
+    return mediaList;
+}