metadata-extractor

#!/usr/bin/env php
<?php

declare(strict_types=1);

ini_set('memory_limit', -1);

use Symfony\Component\DomCrawler\Crawler;
use Symfony\Component\Console\Command\Command;
use Symfony\Component\Console\Helper\ProgressBar;
use Symfony\Component\Console\Input\InputInterface;
use Symfony\Component\Console\Input\InputOption;
use Symfony\Component\Console\Output\OutputInterface;
use Symfony\Component\Console\SingleCommandApplication;
use Symfony\Component\Console\Style\SymfonyStyle;

require_once __DIR__.DIRECTORY_SEPARATOR.'vendor'.DIRECTORY_SEPARATOR.'autoload.php';

(new SingleCommandApplication())
    ->setVersion('0.1.0')
    ->addOption('html-file', null, InputOption::VALUE_REQUIRED, 'Path to the HTML file', '.'.DIRECTORY_SEPARATOR.'YouTube Music.html')
    ->addOption('metadata-file', null, InputOption::VALUE_REQUIRED, 'Path to the metadata.json file', '.'.DIRECTORY_SEPARATOR.'metadata.json')
    ->setCode(function (InputInterface $input, OutputInterface $output) {
        $io = new SymfonyStyle($input, $output);
        $io->title('YouTube Music Upload Metadata Extractor');

        $htmlFile = $input->getOption('html-file');
        $metadataFile = $input->getOption('metadata-file');

        if (!file_exists($htmlFile)) {
            $io->error('html-file not found');

            return Command::INVALID;
        }

        $io->text(sprintf('Will extract %s from %s:', $metadataFile, $htmlFile));

        $crawler = (new Crawler(file_get_contents($htmlFile)))->filter('ytmusic-responsive-list-item-renderer');
        
        $progressBar = new ProgressBar($output);
        $progressBar->setFormat('verbose');
        $progressBar->start($crawler->count());

        $data = $crawler->each(function (Crawler $node) use ($progressBar) {
            if (!$node || !\count($node->filter('.title a'))) {
                $progressBar->advance();
                return;
            }

            $qs = [];
            // Parses the URL to get the ID
            parse_str(parse_url($node->filter('.title a')->attr('href'), PHP_URL_QUERY), $qs);

            $progressBar->advance();
            return [
                'id' => $qs['v'],
                'album' => $node->filter('.secondary-flex-columns')->children()->eq(1)->text() ?: null,
                'artist' => $node->filter('.secondary-flex-columns')->children()->eq(0)->text() ?: null,
                'duration' => $node->filter('.fixed-columns.style-scope.ytmusic-responsive-list-item-renderer')->text(),
                'href' => str_replace('&list=MLPT', '', $node->filter('.title a')->attr('href')),
                'is_liked' => match ($node->filter('ytmusic-like-button-renderer')->attr('like-status')) {
                    'LIKE' => true,
                    'DISLIKE' => false,
                    default => null
                },
                'title' => $node->filter('.title a')->text(),
            ];
        });

        file_put_contents(
            $metadataFile,
            json_encode(
                array_values(
                    array_filter($data)
                )
            )
        );

        $progressBar->finish();
        $io->newLine();
        $io->success('done');

        return Command::SUCCESS;
    })
    ->run();