3 namespace BookStack\Entities\Tools;
5 use BookStack\Entities\Models\Page;
6 use BookStack\Entities\Tools\Markdown\CustomListItemRenderer;
7 use BookStack\Entities\Tools\Markdown\CustomStrikeThroughExtension;
8 use BookStack\Exceptions\ImageUploadException;
9 use BookStack\Facades\Theme;
10 use BookStack\Theming\ThemeEvents;
11 use BookStack\Uploads\ImageRepo;
12 use BookStack\Uploads\ImageService;
13 use BookStack\Util\HtmlContentFilter;
17 use Illuminate\Support\Str;
18 use League\CommonMark\Block\Element\ListItem;
19 use League\CommonMark\CommonMarkConverter;
20 use League\CommonMark\Environment;
21 use League\CommonMark\Extension\Table\TableExtension;
22 use League\CommonMark\Extension\TaskList\TaskListExtension;
29 * PageContent constructor.
31 public function __construct(Page $page)
37 * Update the content of the page with new provided HTML.
39 public function setNewHTML(string $html)
41 $html = $this->extractBase64ImagesFromHtml($html);
42 $this->page->html = $this->formatHtml($html);
43 $this->page->text = $this->toPlainText();
44 $this->page->markdown = '';
48 * Update the content of the page with new provided Markdown content.
50 public function setNewMarkdown(string $markdown)
52 $markdown = $this->extractBase64ImagesFromMarkdown($markdown);
53 $this->page->markdown = $markdown;
54 $html = $this->markdownToHtml($markdown);
55 $this->page->html = $this->formatHtml($html);
56 $this->page->text = $this->toPlainText();
60 * Convert the given Markdown content to a HTML string.
62 protected function markdownToHtml(string $markdown): string
64 $environment = Environment::createCommonMarkEnvironment();
65 $environment->addExtension(new TableExtension());
66 $environment->addExtension(new TaskListExtension());
67 $environment->addExtension(new CustomStrikeThroughExtension());
68 $environment = Theme::dispatch(ThemeEvents::COMMONMARK_ENVIRONMENT_CONFIGURE, $environment) ?? $environment;
69 $converter = new CommonMarkConverter([], $environment);
71 $environment->addBlockRenderer(ListItem::class, new CustomListItemRenderer(), 10);
73 return $converter->convertToHtml($markdown);
77 * Convert all base64 image data to saved images.
79 protected function extractBase64ImagesFromHtml(string $htmlText): string
81 if (empty($htmlText) || strpos($htmlText, 'data:image') === false) {
85 $doc = $this->loadDocumentFromHtml($htmlText);
86 $container = $doc->documentElement;
87 $body = $container->childNodes->item(0);
88 $childNodes = $body->childNodes;
89 $xPath = new DOMXPath($doc);
91 // Get all img elements with image data blobs
92 $imageNodes = $xPath->query('//img[contains(@src, \'data:image\')]');
93 foreach ($imageNodes as $imageNode) {
94 $imageSrc = $imageNode->getAttribute('src');
95 $newUrl = $this->base64ImageUriToUploadedImageUrl($imageSrc);
96 $imageNode->setAttribute('src', $newUrl);
99 // Generate inner html as a string
101 foreach ($childNodes as $childNode) {
102 $html .= $doc->saveHTML($childNode);
109 * Convert all inline base64 content to uploaded image files.
111 protected function extractBase64ImagesFromMarkdown(string $markdown)
114 preg_match_all('/!\[.*?]\(.*?(data:image\/.*?)[)"\s]/', $markdown, $matches);
116 foreach ($matches[1] as $base64Match) {
117 $newUrl = $this->base64ImageUriToUploadedImageUrl($base64Match);
118 $markdown = str_replace($base64Match, $newUrl, $markdown);
125 * Parse the given base64 image URI and return the URL to the created image instance.
126 * Returns an empty string if the parsed URI is invalid or causes an error upon upload.
128 protected function base64ImageUriToUploadedImageUrl(string $uri): string
130 $imageRepo = app()->make(ImageRepo::class);
131 $imageInfo = $this->parseBase64ImageUri($uri);
133 // Validate extension and content
134 if (empty($imageInfo['data']) || !ImageService::isExtensionSupported($imageInfo['extension'])) {
138 // Save image from data with a random name
139 $imageName = 'embedded-image-' . Str::random(8) . '.' . $imageInfo['extension'];
142 $image = $imageRepo->saveNewFromData($imageName, $imageInfo['data'], 'gallery', $this->page->id);
143 } catch (ImageUploadException $exception) {
151 * Parse a base64 image URI into the data and extension.
153 * @return array{extension: array, data: string}
155 protected function parseBase64ImageUri(string $uri): array
157 [$dataDefinition, $base64ImageData] = explode(',', $uri, 2);
158 $extension = strtolower(preg_split('/[\/;]/', $dataDefinition)[1] ?? '');
161 'extension' => $extension,
162 'data' => base64_decode($base64ImageData) ?: '',
167 * Formats a page's html to be tagged correctly within the system.
169 protected function formatHtml(string $htmlText): string
171 if (empty($htmlText)) {
175 $doc = $this->loadDocumentFromHtml($htmlText);
176 $container = $doc->documentElement;
177 $body = $container->childNodes->item(0);
178 $childNodes = $body->childNodes;
179 $xPath = new DOMXPath($doc);
181 // Set ids on top-level nodes
183 foreach ($childNodes as $index => $childNode) {
184 [$oldId, $newId] = $this->setUniqueId($childNode, $idMap);
185 if ($newId && $newId !== $oldId) {
186 $this->updateLinks($xPath, '#' . $oldId, '#' . $newId);
190 // Ensure no duplicate ids within child items
191 $idElems = $xPath->query('//body//*//*[@id]');
192 foreach ($idElems as $domElem) {
193 [$oldId, $newId] = $this->setUniqueId($domElem, $idMap);
194 if ($newId && $newId !== $oldId) {
195 $this->updateLinks($xPath, '#' . $oldId, '#' . $newId);
199 // Generate inner html as a string
201 foreach ($childNodes as $childNode) {
202 $html .= $doc->saveHTML($childNode);
209 * Update the all links to the $old location to instead point to $new.
211 protected function updateLinks(DOMXPath $xpath, string $old, string $new)
213 $old = str_replace('"', '', $old);
214 $matchingLinks = $xpath->query('//body//*//*[@href="' . $old . '"]');
215 foreach ($matchingLinks as $domElem) {
216 $domElem->setAttribute('href', $new);
221 * Set a unique id on the given DOMElement.
222 * A map for existing ID's should be passed in to check for current existence.
223 * Returns a pair of strings in the format [old_id, new_id].
225 protected function setUniqueId(\DOMNode $element, array &$idMap): array
227 if (get_class($element) !== 'DOMElement') {
231 // Stop if there's an existing valid id that has not already been used.
232 $existingId = $element->getAttribute('id');
233 if (strpos($existingId, 'bkmrk') === 0 && !isset($idMap[$existingId])) {
234 $idMap[$existingId] = true;
236 return [$existingId, $existingId];
239 // Create an unique id for the element
240 // Uses the content as a basis to ensure output is the same every time
241 // the same content is passed through.
242 $contentId = 'bkmrk-' . mb_substr(strtolower(preg_replace('/\s+/', '-', trim($element->nodeValue))), 0, 20);
243 $newId = urlencode($contentId);
246 while (isset($idMap[$newId])) {
247 $newId = urlencode($contentId . '-' . $loopIndex);
251 $element->setAttribute('id', $newId);
252 $idMap[$newId] = true;
254 return [$existingId, $newId];
258 * Get a plain-text visualisation of this page.
260 protected function toPlainText(): string
262 $html = $this->render(true);
264 return html_entity_decode(strip_tags($html));
268 * Render the page for viewing.
270 public function render(bool $blankIncludes = false): string
272 $content = $this->page->html ?? '';
274 if (!config('app.allow_content_scripts')) {
275 $content = HtmlContentFilter::removeScripts($content);
278 if ($blankIncludes) {
279 $content = $this->blankPageIncludes($content);
281 $content = $this->parsePageIncludes($content);
288 * Parse the headers on the page to get a navigation menu.
290 public function getNavigation(string $htmlContent): array
292 if (empty($htmlContent)) {
296 $doc = $this->loadDocumentFromHtml($htmlContent);
297 $xPath = new DOMXPath($doc);
298 $headers = $xPath->query('//h1|//h2|//h3|//h4|//h5|//h6');
300 return $headers ? $this->headerNodesToLevelList($headers) : [];
304 * Convert a DOMNodeList into an array of readable header attributes
305 * with levels normalised to the lower header level.
307 protected function headerNodesToLevelList(DOMNodeList $nodeList): array
309 $tree = collect($nodeList)->map(function ($header) {
310 $text = trim(str_replace("\xc2\xa0", '', $header->nodeValue));
311 $text = mb_substr($text, 0, 100);
314 'nodeName' => strtolower($header->nodeName),
315 'level' => intval(str_replace('h', '', $header->nodeName)),
316 'link' => '#' . $header->getAttribute('id'),
319 })->filter(function ($header) {
320 return mb_strlen($header['text']) > 0;
323 // Shift headers if only smaller headers have been used
324 $levelChange = ($tree->pluck('level')->min() - 1);
325 $tree = $tree->map(function ($header) use ($levelChange) {
326 $header['level'] -= ($levelChange);
331 return $tree->toArray();
335 * Remove any page include tags within the given HTML.
337 protected function blankPageIncludes(string $html): string
339 return preg_replace("/{{@\s?([0-9].*?)}}/", '', $html);
343 * Parse any include tags "{{@<page_id>#section}}" to be part of the page.
345 protected function parsePageIncludes(string $html): string
348 preg_match_all("/{{@\s?([0-9].*?)}}/", $html, $matches);
350 foreach ($matches[1] as $index => $includeId) {
351 $fullMatch = $matches[0][$index];
352 $splitInclude = explode('#', $includeId, 2);
354 // Get page id from reference
355 $pageId = intval($splitInclude[0]);
356 if (is_nan($pageId)) {
360 // Find page and skip this if page not found
361 /** @var ?Page $matchedPage */
362 $matchedPage = Page::visible()->find($pageId);
363 if ($matchedPage === null) {
364 $html = str_replace($fullMatch, '', $html);
368 // If we only have page id, just insert all page html and continue.
369 if (count($splitInclude) === 1) {
370 $html = str_replace($fullMatch, $matchedPage->html, $html);
374 // Create and load HTML into a document
375 $innerContent = $this->fetchSectionOfPage($matchedPage, $splitInclude[1]);
376 $html = str_replace($fullMatch, trim($innerContent), $html);
383 * Fetch the content from a specific section of the given page.
385 protected function fetchSectionOfPage(Page $page, string $sectionId): string
387 $topLevelTags = ['table', 'ul', 'ol'];
388 $doc = $this->loadDocumentFromHtml($page->html);
390 // Search included content for the id given and blank out if not exists.
391 $matchingElem = $doc->getElementById($sectionId);
392 if ($matchingElem === null) {
396 // Otherwise replace the content with the found content
397 // Checks if the top-level wrapper should be included by matching on tag types
399 $isTopLevel = in_array(strtolower($matchingElem->nodeName), $topLevelTags);
401 $innerContent .= $doc->saveHTML($matchingElem);
403 foreach ($matchingElem->childNodes as $childNode) {
404 $innerContent .= $doc->saveHTML($childNode);
407 libxml_clear_errors();
409 return $innerContent;
413 * Create and load a DOMDocument from the given html content.
415 protected function loadDocumentFromHtml(string $html): DOMDocument
417 libxml_use_internal_errors(true);
418 $doc = new DOMDocument();
419 $html = '<body>' . $html . '</body>';
420 $doc->loadHTML(mb_convert_encoding($html, 'HTML-ENTITIES', 'UTF-8'));