X-Git-Url: http://source.bookstackapp.com/bookstack/blobdiff_plain/2e9ac21b383cad6f120a05130cda9d32e54695a5..refs/pull/3193/head:/app/Entities/Tools/PageContent.php diff --git a/app/Entities/Tools/PageContent.php b/app/Entities/Tools/PageContent.php index 9f4ac2893..b95131fce 100644 --- a/app/Entities/Tools/PageContent.php +++ b/app/Entities/Tools/PageContent.php @@ -9,8 +9,11 @@ use BookStack\Exceptions\ImageUploadException; use BookStack\Facades\Theme; use BookStack\Theming\ThemeEvents; use BookStack\Uploads\ImageRepo; +use BookStack\Uploads\ImageService; use BookStack\Util\HtmlContentFilter; use DOMDocument; +use DOMElement; +use DOMNode; use DOMNodeList; use DOMXPath; use Illuminate\Support\Str; @@ -86,30 +89,13 @@ class PageContent $body = $container->childNodes->item(0); $childNodes = $body->childNodes; $xPath = new DOMXPath($doc); - $imageRepo = app()->make(ImageRepo::class); // Get all img elements with image data blobs $imageNodes = $xPath->query('//img[contains(@src, \'data:image\')]'); foreach ($imageNodes as $imageNode) { $imageSrc = $imageNode->getAttribute('src'); - [$dataDefinition, $base64ImageData] = explode(',', $imageSrc, 2); - $extension = strtolower(preg_split('/[\/;]/', $dataDefinition)[1] ?? 'png'); - - // Validate extension - if (!$imageRepo->imageExtensionSupported($extension)) { - $imageNode->setAttribute('src', ''); - continue; - } - - // Save image from data with a random name - $imageName = 'embedded-image-' . Str::random(8) . '.' . $extension; - - try { - $image = $imageRepo->saveNewFromData($imageName, base64_decode($base64ImageData), 'gallery', $this->page->id); - $imageNode->setAttribute('src', $image->url); - } catch (ImageUploadException $exception) { - $imageNode->setAttribute('src', ''); - } + $newUrl = $this->base64ImageUriToUploadedImageUrl($imageSrc); + $imageNode->setAttribute('src', $newUrl); } // Generate inner html as a string @@ -126,32 +112,63 @@ class PageContent */ protected function extractBase64ImagesFromMarkdown(string $markdown) { - $imageRepo = app()->make(ImageRepo::class); $matches = []; preg_match_all('/!\[.*?]\(.*?(data:image\/.*?)[)"\s]/', $markdown, $matches); foreach ($matches[1] as $base64Match) { - [$dataDefinition, $base64ImageData] = explode(',', $base64Match, 2); - $extension = strtolower(preg_split('/[\/;]/', $dataDefinition)[1] ?? 'png'); + $newUrl = $this->base64ImageUriToUploadedImageUrl($base64Match); + $markdown = str_replace($base64Match, $newUrl, $markdown); + } - // Validate extension - if (!$imageRepo->imageExtensionSupported($extension)) { - $markdown = str_replace($base64Match, '', $markdown); - continue; - } + return $markdown; + } - // Save image from data with a random name - $imageName = 'embedded-image-' . Str::random(8) . '.' . $extension; + /** + * Parse the given base64 image URI and return the URL to the created image instance. + * Returns an empty string if the parsed URI is invalid or causes an error upon upload. + */ + protected function base64ImageUriToUploadedImageUrl(string $uri): string + { + $imageRepo = app()->make(ImageRepo::class); + $imageInfo = $this->parseBase64ImageUri($uri); - try { - $image = $imageRepo->saveNewFromData($imageName, base64_decode($base64ImageData), 'gallery', $this->page->id); - $markdown = str_replace($base64Match, $image->url, $markdown); - } catch (ImageUploadException $exception) { - $markdown = str_replace($base64Match, '', $markdown); - } + // Validate extension and content + if (empty($imageInfo['data']) || !ImageService::isExtensionSupported($imageInfo['extension'])) { + return ''; } - return $markdown; + // Validate that the content is not over our upload limit + $uploadLimitBytes = (config('app.upload_limit') * 1000000); + if (strlen($imageInfo['data']) > $uploadLimitBytes) { + return ''; + } + + // Save image from data with a random name + $imageName = 'embedded-image-' . Str::random(8) . '.' . $imageInfo['extension']; + + try { + $image = $imageRepo->saveNewFromData($imageName, $imageInfo['data'], 'gallery', $this->page->id); + } catch (ImageUploadException $exception) { + return ''; + } + + return $image->url; + } + + /** + * Parse a base64 image URI into the data and extension. + * + * @return array{extension: string, data: string} + */ + protected function parseBase64ImageUri(string $uri): array + { + [$dataDefinition, $base64ImageData] = explode(',', $uri, 2); + $extension = strtolower(preg_split('/[\/;]/', $dataDefinition)[1] ?? ''); + + return [ + 'extension' => $extension, + 'data' => base64_decode($base64ImageData) ?: '', + ]; } /** @@ -178,6 +195,15 @@ class PageContent } } + // Set ids on nested header nodes + $nestedHeaders = $xPath->query('//body//*//h1|//body//*//h2|//body//*//h3|//body//*//h4|//body//*//h5|//body//*//h6'); + foreach ($nestedHeaders as $nestedHeader) { + [$oldId, $newId] = $this->setUniqueId($nestedHeader, $idMap); + if ($newId && $newId !== $oldId) { + $this->updateLinks($xPath, '#' . $oldId, '#' . $newId); + } + } + // Ensure no duplicate ids within child items $idElems = $xPath->query('//body//*//*[@id]'); foreach ($idElems as $domElem) { @@ -213,9 +239,9 @@ class PageContent * A map for existing ID's should be passed in to check for current existence. * Returns a pair of strings in the format [old_id, new_id]. */ - protected function setUniqueId(\DOMNode $element, array &$idMap): array + protected function setUniqueId(DOMNode $element, array &$idMap): array { - if (get_class($element) !== 'DOMElement') { + if (!$element instanceof DOMElement) { return ['', '']; } @@ -227,7 +253,7 @@ class PageContent return [$existingId, $existingId]; } - // Create an unique id for the element + // Create a unique id for the element // Uses the content as a basis to ensure output is the same every time // the same content is passed through. $contentId = 'bkmrk-' . mb_substr(strtolower(preg_replace('/\s+/', '-', trim($element->nodeValue))), 0, 20); @@ -297,7 +323,7 @@ class PageContent */ protected function headerNodesToLevelList(DOMNodeList $nodeList): array { - $tree = collect($nodeList)->map(function ($header) { + $tree = collect($nodeList)->map(function (DOMElement $header) { $text = trim(str_replace("\xc2\xa0", '', $header->nodeValue)); $text = mb_substr($text, 0, 100); @@ -375,7 +401,7 @@ class PageContent */ protected function fetchSectionOfPage(Page $page, string $sectionId): string { - $topLevelTags = ['table', 'ul', 'ol']; + $topLevelTags = ['table', 'ul', 'ol', 'pre']; $doc = $this->loadDocumentFromHtml($page->html); // Search included content for the id given and blank out if not exists.