Feature : provides .epub files preview pictures

lib/AppInfo/Application.php

@@ -8,6 +8,7 @@
 use OCA\Epubviewer\Listener\BeforeTemplateRenderedListener;
 use OCA\Epubviewer\Listener\FilesLoadAdditionalScriptsListener;
 use OCA\Epubviewer\Listener\PublicShareBeforeTemplateRenderedListener;
+use OCA\Epubviewer\Preview\EPubPreview;
 use OCA\Files\Event\LoadAdditionalScriptsEvent;
 
 use OCP\AppFramework\App;
@@ -27,6 +28,8 @@ public function __construct() {
 	public function register(IRegistrationContext $context): void {
 		include_once __DIR__ . '/../../vendor/autoload.php';
 
+		$this->registerProvider($context);
+
 		// “Emitted before the rendering step of each TemplateResponse. The event holds a flag that specifies if a user is logged in.”
 		// See: https://docs.nextcloud.com/server/latest/developer_manual/basics/events.html#oca-settings-events-beforetemplaterenderedevent
 		$context->registerEventListener(BeforeTemplateRenderedEvent::class, BeforeTemplateRenderedListener::class);
@@ -44,6 +47,10 @@ public function register(IRegistrationContext $context): void {
 		//    Hooks::register();
 	}
 
+	private function registerProvider(IRegistrationContext $context): void {
+		$context->registerPreviewProvider(EPubPreview::class, '/^application\/epub\+zip$/');
+	}
+
 	public function boot(IBootContext $context): void {
 	}
 }

lib/Preview/EPubPreview.php

@@ -0,0 +1,248 @@
+<?php
+/**
+ *
+ * @author Sebastien Marinier <[email protected]>
+ *
+ * @license AGPL-3.0
+ *
+ * This code is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License, version 3,
+ * as published by the Free Software Foundation.
+ *
+ * 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, version 3,
+ * along with this program. If not, see <http://www.gnu.org/licenses/>
+ *
+ */
+
+namespace OCA\Epubviewer\Preview;
+
+//.epub
+use OC\Archive\ZIP;
+use OCP\Files\File;
+use OCP\Files\FileInfo;
+use OCP\IImage;
+use OCP\Preview\IProviderV2;
+use OCP\ITempManager;
+
+class EPubPreview implements IProviderV2 {
+	private ?ZIP $zip = null;
+
+	/**
+	 * {@inheritDoc}
+	 */
+	public function getMimeType(): string {
+		return '/application\/epub\+zip/';
+	}
+
+	/**
+	 * Check if a preview can be generated for $path
+	 *
+	 * {@inheritDoc}
+	 */
+	public function isAvailable(FileInfo $file): bool {
+		return true;
+	}
+
+	/**
+	 * @inheritDoc
+	 */
+	public function getThumbnail(File $file, int $maxX, int $maxY): ?IImage {
+		$image = $this->extractThumbnail($file, '');
+		if ($image && $image->valid()) {
+			return $image;
+		}
+		return null;
+	}
+
+	/**
+	 * extractThumbnail from complicated epub format
+	 */
+	private function extractThumbnail(File $file, string $path): ?IImage {
+		$tmpManager = \OC::$server->get(ITempManager::class);
+		$sourceTmp = $tmpManager->getTemporaryFile();
+
+		try {
+			$content = $file->fopen('r');
+			file_put_contents($sourceTmp, $content);
+
+			$this->zip = new ZIP($sourceTmp);
+
+			$img_data = null;
+			$contentPath = $this->getContentPath();
+			if ($contentPath) {
+				$package = $this->extractXML($contentPath);
+				if ($package) {
+					$path = $contentPath;
+					$img_src = $cover = null;
+					// Try first through <manifest>
+					$items = $package->manifest->children();
+					foreach($items as $item) {
+						if (($item['id'] == 'cover' || $item['id'] == 'cover-image') && preg_match('/image\//', (string) $item['media-type'])) {
+							$img_src = (string) $item['href'];
+							break;
+						}
+					}
+
+					// in references
+					if (!$img_src) {
+						$references = $package->guide->children();
+						foreach($references as $reference) {
+							if ($reference['type'] == 'cover' || $reference['type'] == 'title-page') {
+								$cover = (string) $reference['href'];
+								break;
+							}
+						}
+					}
+
+					// no cover ? no image ? take the first page
+					if (!$img_src && !$cover) {
+						$first_page_id = (string) $package->spine->itemref['idref'];
+						if ($first_page_id) {
+							foreach($items as $item) {
+								if ($item['id'] == $first_page_id) {
+									$cover = (string) $item['href'];
+									break;
+								}
+							}
+
+						}
+					}
+
+					// have we a "cover" file ?
+					if ($cover) {
+						// relative to container
+						$img_src = null;
+						$path = $this->resolvePath($path, $cover);
+						$dom = $this->extractHTML($path);
+						if ($dom) {
+							// search img
+							$images = $dom->getElementsByTagName('img');
+							if ($images->length) {
+								$img_src = $images[0]->getAttribute('src');
+							} else {
+								$images = $dom->getElementsByTagName('image');
+								if ($images->length) {
+									$img_src = $images[0]->getAttribute('xlink:href');
+								}
+							}
+						}
+					}// cover
+
+					// img ?
+					if ($img_src) {
+						$img_src = $this->resolvePath($path, $img_src);
+						$img_data = $this->extractFileData($img_src);
+					}
+				}
+			}
+
+			// Pfff. Make a pause
+			if ($img_data) {
+				$image = new \OC_Image();
+				$image->loadFromData($img_data);
+				return $image;
+			}
+			return null;
+		} catch (\Exception $e) {
+			return null;
+		}
+	}
+
+	/**
+	 * find the main content XML (usually "content.opf")
+	 */
+	private function getContentPath() : ?string {
+		$xml_container = $this->extractXML('META-INF/container.xml');
+		if (is_object($xml_container)) {
+			$full_path = $xml_container->rootfiles->rootfile['full-path'][0];
+			if ($full_path) {
+				return $full_path->__toString();
+			}
+		}
+		return null;
+	}
+
+	/**
+	 * extract HTML from Zip path
+	 * @param string $path
+	 * @return \DOMDocument|null
+	 */
+	protected function extractHTML(string $path): \DOMDocument|null {
+		$html = $this->extractFileData($path);
+		if (is_string($html)) {
+			$dom = new \DOMDocument('1.0', 'utf-8');
+			$dom->strictErrorChecking = false;
+			if (@$dom->loadHTML($html)) {
+				return $dom;
+			}
+		}
+		return null;
+	}
+
+	/**
+	 * extract XML from Zip path
+	 *
+	 * @psalm-param 'META-INF/container.xml' $path
+	 */
+	private function extractXML(string $path): \SimpleXMLElement|false|null {
+		$xml = $this->extractFileData($path);
+		if (is_string($xml)) {
+			return simplexml_load_string($xml);
+		}
+		return null;
+	}
+
+	/**
+	 * get unzipped data
+	 *
+	 * @param string $path file path in zip
+	 *
+	 * @psalm-param 'META-INF/container.xml' $path
+	 *
+	 * @return false|null|string
+	 */
+	private function extractFileData(string $path): string|false|null {
+		if ($this->zip === null) {
+			return null;
+		}
+		$fp = $this->zip->getStream($path, 'r');
+		if ($fp) {
+			$content = stream_get_contents($fp);
+			fclose($fp);
+			return $content;
+		}
+		return null;
+	}
+
+	/**
+	 * Resolve relative $relPath from $path (removes ./, ../)
+	 *
+	 * @param string $path reference path
+	 * @param string $relPath relative path
+	 * @return string
+	 */
+	private function resolvePath(string $path, string $relPath): string {
+		$path = dirname($path).'/'.$relPath;
+		$pieces = explode('/', $path);
+		$parents = [];
+		foreach($pieces as $dir) {
+			switch($dir) {
+				case '.':
+					// Don't need to do anything here
+					break;
+				case '..':
+					array_pop($parents);
+					break;
+				default:
+					$parents[] = $dir;
+					break;
+			}
+		}
+		return implode('/', $parents);
+	}
+}