Mangarr/src/Domain/Scraping/Application/CommandHandler/ScrapeChapterHandler.php

<?php

namespace App\Domain\Scraping\Application\CommandHandler;

use App\Domain\Scraping\Application\Command\ScrapeChapter;
use App\Domain\Scraping\Domain\Contract\Repository\ChapterRepositoryInterface;
use App\Domain\Scraping\Domain\Contract\Repository\MangaRepositoryInterface;
use App\Domain\Scraping\Domain\Contract\Repository\SourceRepositoryInterface;
use App\Domain\Scraping\Domain\Contract\Service\CbzGeneratorInterface;
use App\Domain\Scraping\Domain\Contract\Service\ImageDownloaderInterface;
use App\Domain\Scraping\Domain\Contract\Service\ScraperInterface;
use App\Domain\Scraping\Domain\Event\ChapterScraped;
use App\Domain\Scraping\Domain\Event\ChapterScrapingFailed;
use App\Domain\Scraping\Domain\Model\ScrapingJob;
use App\Domain\Scraping\Domain\Model\Source;
use App\Domain\Scraping\Domain\Model\ValueObject\CbzGenerationRequest;
use App\Domain\Scraping\Domain\Model\ValueObject\ScrapingRequest;
use App\Domain\Scraping\Domain\Model\ValueObject\TempDirectory;
use App\Domain\Shared\Domain\Contract\JobRepositoryInterface;
use Ramsey\Uuid\Uuid;
use Symfony\Component\Messenger\MessageBusInterface;
use Doctrine\ORM\EntityManagerInterface;

readonly class ScrapeChapterHandler
{
    public function __construct(
        private ScraperInterface $scraper,
        private ImageDownloaderInterface $imageDownloader,
        private CbzGeneratorInterface $cbzGenerator,
        private JobRepositoryInterface $jobRepository,
        private ChapterRepositoryInterface $chapterRepository,
        private MangaRepositoryInterface $mangaRepository,
        private SourceRepositoryInterface $sourceRepository,
        private MessageBusInterface $eventBus,
        private EntityManagerInterface $entityManager
    ) {
    }

    public function handle(ScrapeChapter $command): void
    {
        $job = null;
        try {
            // 1. Récupération du chapitre
            $chapter = $this->chapterRepository->getById($command->chapterId);
            if (!$chapter) {
                throw new \InvalidArgumentException("Chapter not found with ID: {$command->chapterId}");
            }

            // 2. Récupération du manga
            $manga = $this->mangaRepository->getById($chapter->mangaId);
            if (!$manga) {
                throw new \InvalidArgumentException("Manga not found with ID: {$chapter->mangaId}");
            }

            // 3. Détermination des sources à utiliser
            $sources = $this->getSourcesToTry($manga);
            if (empty($sources)) {
                throw new \InvalidArgumentException("No sources available for scraping");
            }

            // 4. Essai de scraping sur chaque source jusqu'à succès
            $success = false;
            $lastException = null;

            foreach ($sources as $source) {
                $job = new ScrapingJob(
                    Uuid::uuid4()->toString(),
                    $chapter->mangaId,
                    $chapter->chapterNumber,
                    $source->getId()->getValue()
                );

                // Ajout de l'ID du chapitre dans le contexte du job
                $job->context['chapterId'] = $command->chapterId;

                $job->start();
                $this->jobRepository->save($job);

                try {
                    $this->entityManager->beginTransaction();

                    // 5. Scraping des URLs
                    $scrapingRequest = new ScrapingRequest(
                        'html',
                        $source->buildChapterUrl($manga->getSlug(), $chapter->chapterNumber),
                        $source->getScrappingParameters()
                    );

                    $scrapingResult = $this->scraper->scrape($scrapingRequest);

                    // 6. Téléchargement des images
                    $tempDir = new TempDirectory();
                    $downloadResults = $this->imageDownloader->downloadBatch(
                        $scrapingResult->getImageUrls(),
                        $tempDir,
                        $job->id
                    );

                    // 7. Génération du CBZ
                    $cbzRequest = new CbzGenerationRequest(
                        $manga->getTitle(),
                        $manga->getPublicationYear(),
                        $chapter->volumeNumber,
                        $chapter->chapterNumber,
                        $tempDir,
                        array_map(fn($r) => $r->getLocalPath(), $downloadResults)
                    );

                    $cbzPath = $this->cbzGenerator->generate($cbzRequest);

                    // 8. Mise à jour et sauvegarde
                    $chapter->cbzPath = $cbzPath->getPath();
                    $this->chapterRepository->save($chapter);

                    $job->complete();
                    $this->jobRepository->save($job);

                    $this->entityManager->commit();

                    $this->eventBus->dispatch(new ChapterScraped($job->id));

                    // 9. Nettoyage
                    $tempDir->cleanup();

                    // Scraping réussi, pas besoin d'essayer d'autres sources
                    $success = true;
                    break;

                } catch (\Exception $e) {
                    $this->entityManager->rollback();

                    if (isset($job)) {
                        $job->fail($e->getMessage());
                        $this->jobRepository->save($job);
                    }

                    $lastException = $e;

                    // Continuer avec la source suivante
                }
            }

            // Si toutes les sources ont échoué
            if (!$success) {
                $errorMessage = $lastException ? $lastException->getMessage() : "Failed to scrape chapter from all available sources";
                $this->eventBus->dispatch(new ChapterScrapingFailed($chapter->mangaId, $chapter->chapterNumber, $errorMessage));
            }

        } catch (\Exception $e) {
            if (isset($job)) {
                $job->fail($e->getMessage());
                $this->jobRepository->save($job);
            }
            $this->eventBus->dispatch(new ChapterScrapingFailed($chapter->mangaId ?? 'unknown', $chapter->chapterNumber ?? 'unknown', $e->getMessage()));
        }
    }

    /**
     * Détermine les sources à utiliser pour le scraping en fonction des préférences du manga
     *
     * @param \App\Domain\Scraping\Domain\Model\Manga $manga
     * @return Source[]
     */
    private function getSourcesToTry(\App\Domain\Scraping\Domain\Model\Manga $manga): array
    {
        // Si le manga a des sources préférées, les utiliser
        if ($manga->hasPreferredSources()) {
            $preferredSources = [];
            foreach ($manga->getPreferredSources() as $sourceId) {
                $source = $this->sourceRepository->getById($sourceId);
                if ($source) {
                    $preferredSources[] = $source;
                }
            }

            if (!empty($preferredSources)) {
                return $preferredSources;
            }
        }

        // Sinon, utiliser toutes les sources disponibles
        return $this->sourceRepository->getAll();
    }
}