2021-06-26 15:23:15 +00:00
|
|
|
|
<?php
|
|
|
|
|
|
|
|
|
|
namespace BookStack\Entities\Tools;
|
2019-10-05 11:55:01 +00:00
|
|
|
|
|
2020-11-22 00:17:45 +00:00
|
|
|
|
use BookStack\Entities\Models\Page;
|
2022-04-18 16:39:28 +00:00
|
|
|
|
use BookStack\Entities\Tools\Markdown\MarkdownToHtml;
|
2021-06-02 20:34:34 +00:00
|
|
|
|
use BookStack\Exceptions\ImageUploadException;
|
2022-09-05 15:40:42 +00:00
|
|
|
|
use BookStack\Facades\Theme;
|
|
|
|
|
use BookStack\Theming\ThemeEvents;
|
2021-04-20 23:41:21 +00:00
|
|
|
|
use BookStack\Uploads\ImageRepo;
|
2021-11-01 00:24:42 +00:00
|
|
|
|
use BookStack\Uploads\ImageService;
|
2021-06-26 15:23:15 +00:00
|
|
|
|
use BookStack\Util\HtmlContentFilter;
|
2019-10-05 11:55:01 +00:00
|
|
|
|
use DOMDocument;
|
2021-11-22 23:33:55 +00:00
|
|
|
|
use DOMElement;
|
|
|
|
|
use DOMNode;
|
2019-10-05 11:55:01 +00:00
|
|
|
|
use DOMNodeList;
|
|
|
|
|
use DOMXPath;
|
2021-06-02 20:34:34 +00:00
|
|
|
|
use Illuminate\Support\Str;
|
2019-10-05 11:55:01 +00:00
|
|
|
|
|
|
|
|
|
class PageContent
|
|
|
|
|
{
|
2023-02-22 14:32:40 +00:00
|
|
|
|
public function __construct(
|
|
|
|
|
protected Page $page
|
|
|
|
|
) {
|
2019-10-05 11:55:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* Update the content of the page with new provided HTML.
|
|
|
|
|
*/
|
2023-02-22 14:32:40 +00:00
|
|
|
|
public function setNewHTML(string $html): void
|
2019-10-05 11:55:01 +00:00
|
|
|
|
{
|
2021-10-18 10:42:50 +00:00
|
|
|
|
$html = $this->extractBase64ImagesFromHtml($html);
|
2019-10-05 11:55:01 +00:00
|
|
|
|
$this->page->html = $this->formatHtml($html);
|
2020-09-19 14:13:18 +00:00
|
|
|
|
$this->page->text = $this->toPlainText();
|
2020-11-22 14:56:19 +00:00
|
|
|
|
$this->page->markdown = '';
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* Update the content of the page with new provided Markdown content.
|
|
|
|
|
*/
|
2023-02-22 14:32:40 +00:00
|
|
|
|
public function setNewMarkdown(string $markdown): void
|
2020-11-22 14:56:19 +00:00
|
|
|
|
{
|
2021-10-18 10:42:50 +00:00
|
|
|
|
$markdown = $this->extractBase64ImagesFromMarkdown($markdown);
|
2020-11-22 14:56:19 +00:00
|
|
|
|
$this->page->markdown = $markdown;
|
2022-04-18 16:39:28 +00:00
|
|
|
|
$html = (new MarkdownToHtml($markdown))->convert();
|
2020-11-22 14:56:19 +00:00
|
|
|
|
$this->page->html = $this->formatHtml($html);
|
|
|
|
|
$this->page->text = $this->toPlainText();
|
|
|
|
|
}
|
|
|
|
|
|
2021-04-20 23:41:21 +00:00
|
|
|
|
/**
|
2021-06-26 15:23:15 +00:00
|
|
|
|
* Convert all base64 image data to saved images.
|
2021-04-20 23:41:21 +00:00
|
|
|
|
*/
|
2021-10-18 10:42:50 +00:00
|
|
|
|
protected function extractBase64ImagesFromHtml(string $htmlText): string
|
2021-04-20 23:41:21 +00:00
|
|
|
|
{
|
2023-02-22 14:32:40 +00:00
|
|
|
|
if (empty($htmlText) || !str_contains($htmlText, 'data:image')) {
|
2021-04-20 23:41:21 +00:00
|
|
|
|
return $htmlText;
|
|
|
|
|
}
|
|
|
|
|
|
2021-06-02 20:34:34 +00:00
|
|
|
|
$doc = $this->loadDocumentFromHtml($htmlText);
|
2021-04-20 23:41:21 +00:00
|
|
|
|
$container = $doc->documentElement;
|
|
|
|
|
$body = $container->childNodes->item(0);
|
|
|
|
|
$childNodes = $body->childNodes;
|
|
|
|
|
$xPath = new DOMXPath($doc);
|
|
|
|
|
|
|
|
|
|
// Get all img elements with image data blobs
|
|
|
|
|
$imageNodes = $xPath->query('//img[contains(@src, \'data:image\')]');
|
2021-06-02 20:34:34 +00:00
|
|
|
|
foreach ($imageNodes as $imageNode) {
|
2021-04-20 23:41:21 +00:00
|
|
|
|
$imageSrc = $imageNode->getAttribute('src');
|
2021-10-28 14:54:00 +00:00
|
|
|
|
$newUrl = $this->base64ImageUriToUploadedImageUrl($imageSrc);
|
|
|
|
|
$imageNode->setAttribute('src', $newUrl);
|
2021-04-20 23:41:21 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// Generate inner html as a string
|
|
|
|
|
$html = '';
|
|
|
|
|
foreach ($childNodes as $childNode) {
|
|
|
|
|
$html .= $doc->saveHTML($childNode);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return $html;
|
|
|
|
|
}
|
|
|
|
|
|
2021-10-18 10:42:50 +00:00
|
|
|
|
/**
|
|
|
|
|
* Convert all inline base64 content to uploaded image files.
|
2022-02-06 07:51:38 +00:00
|
|
|
|
* Regex is used to locate the start of data-uri definitions then
|
|
|
|
|
* manual looping over content is done to parse the whole data uri.
|
|
|
|
|
* Attempting to capture the whole data uri using regex can cause PHP
|
|
|
|
|
* PCRE limits to be hit with larger, multi-MB, files.
|
2021-10-18 10:42:50 +00:00
|
|
|
|
*/
|
2023-02-22 14:32:40 +00:00
|
|
|
|
protected function extractBase64ImagesFromMarkdown(string $markdown): string
|
2021-10-18 10:42:50 +00:00
|
|
|
|
{
|
|
|
|
|
$matches = [];
|
2022-02-06 07:51:38 +00:00
|
|
|
|
$contentLength = strlen($markdown);
|
|
|
|
|
$replacements = [];
|
|
|
|
|
preg_match_all('/!\[.*?]\(.*?(data:image\/.{1,6};base64,)/', $markdown, $matches, PREG_OFFSET_CAPTURE);
|
|
|
|
|
|
|
|
|
|
foreach ($matches[1] as $base64MatchPair) {
|
|
|
|
|
[$dataUri, $index] = $base64MatchPair;
|
|
|
|
|
|
|
|
|
|
for ($i = strlen($dataUri) + $index; $i < $contentLength; $i++) {
|
|
|
|
|
$char = $markdown[$i];
|
|
|
|
|
if ($char === ')' || $char === ' ' || $char === "\n" || $char === '"') {
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
$dataUri .= $char;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
$newUrl = $this->base64ImageUriToUploadedImageUrl($dataUri);
|
|
|
|
|
$replacements[] = [$dataUri, $newUrl];
|
|
|
|
|
}
|
2021-10-18 10:42:50 +00:00
|
|
|
|
|
2022-02-06 07:51:38 +00:00
|
|
|
|
foreach ($replacements as [$dataUri, $newUrl]) {
|
|
|
|
|
$markdown = str_replace($dataUri, $newUrl, $markdown);
|
2021-10-18 10:42:50 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return $markdown;
|
|
|
|
|
}
|
|
|
|
|
|
2021-10-28 14:54:00 +00:00
|
|
|
|
/**
|
|
|
|
|
* 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);
|
|
|
|
|
|
|
|
|
|
// Validate extension and content
|
2021-11-01 00:24:42 +00:00
|
|
|
|
if (empty($imageInfo['data']) || !ImageService::isExtensionSupported($imageInfo['extension'])) {
|
2021-10-28 14:54:00 +00:00
|
|
|
|
return '';
|
|
|
|
|
}
|
|
|
|
|
|
2021-11-14 22:03:22 +00:00
|
|
|
|
// Validate that the content is not over our upload limit
|
|
|
|
|
$uploadLimitBytes = (config('app.upload_limit') * 1000000);
|
|
|
|
|
if (strlen($imageInfo['data']) > $uploadLimitBytes) {
|
|
|
|
|
return '';
|
|
|
|
|
}
|
|
|
|
|
|
2021-10-28 14:54:00 +00:00
|
|
|
|
// 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.
|
2021-11-01 13:26:02 +00:00
|
|
|
|
*
|
2021-11-20 14:03:56 +00:00
|
|
|
|
* @return array{extension: string, data: string}
|
2021-10-28 14:54:00 +00:00
|
|
|
|
*/
|
|
|
|
|
protected function parseBase64ImageUri(string $uri): array
|
|
|
|
|
{
|
|
|
|
|
[$dataDefinition, $base64ImageData] = explode(',', $uri, 2);
|
|
|
|
|
$extension = strtolower(preg_split('/[\/;]/', $dataDefinition)[1] ?? '');
|
2021-11-01 13:26:02 +00:00
|
|
|
|
|
2021-10-28 14:54:00 +00:00
|
|
|
|
return [
|
|
|
|
|
'extension' => $extension,
|
2021-11-01 13:26:02 +00:00
|
|
|
|
'data' => base64_decode($base64ImageData) ?: '',
|
2021-10-28 14:54:00 +00:00
|
|
|
|
];
|
|
|
|
|
}
|
|
|
|
|
|
2019-10-05 11:55:01 +00:00
|
|
|
|
/**
|
|
|
|
|
* Formats a page's html to be tagged correctly within the system.
|
|
|
|
|
*/
|
|
|
|
|
protected function formatHtml(string $htmlText): string
|
|
|
|
|
{
|
2021-06-02 20:34:34 +00:00
|
|
|
|
if (empty($htmlText)) {
|
2019-10-05 11:55:01 +00:00
|
|
|
|
return $htmlText;
|
|
|
|
|
}
|
|
|
|
|
|
2021-06-02 20:34:34 +00:00
|
|
|
|
$doc = $this->loadDocumentFromHtml($htmlText);
|
2019-10-05 11:55:01 +00:00
|
|
|
|
$container = $doc->documentElement;
|
|
|
|
|
$body = $container->childNodes->item(0);
|
|
|
|
|
$childNodes = $body->childNodes;
|
2020-09-28 21:26:50 +00:00
|
|
|
|
$xPath = new DOMXPath($doc);
|
2019-10-05 11:55:01 +00:00
|
|
|
|
|
2023-02-22 14:32:40 +00:00
|
|
|
|
// Map to hold used ID references
|
2019-10-05 11:55:01 +00:00
|
|
|
|
$idMap = [];
|
2023-02-22 14:32:40 +00:00
|
|
|
|
// Map to hold changing ID references
|
|
|
|
|
$changeMap = [];
|
2019-10-05 11:55:01 +00:00
|
|
|
|
|
2023-02-22 14:32:40 +00:00
|
|
|
|
$this->updateIdsRecursively($body, 0, $idMap, $changeMap);
|
|
|
|
|
$this->updateLinks($xPath, $changeMap);
|
2019-10-05 11:55:01 +00:00
|
|
|
|
|
|
|
|
|
// Generate inner html as a string
|
|
|
|
|
$html = '';
|
|
|
|
|
foreach ($childNodes as $childNode) {
|
|
|
|
|
$html .= $doc->saveHTML($childNode);
|
|
|
|
|
}
|
|
|
|
|
|
2022-03-01 17:26:06 +00:00
|
|
|
|
// Perform required string-level tweaks
|
|
|
|
|
$html = str_replace(' ', ' ', $html);
|
|
|
|
|
|
2019-10-05 11:55:01 +00:00
|
|
|
|
return $html;
|
|
|
|
|
}
|
|
|
|
|
|
2020-09-28 21:26:50 +00:00
|
|
|
|
/**
|
2023-02-22 14:32:40 +00:00
|
|
|
|
* For the given DOMNode, traverse its children recursively and update IDs
|
|
|
|
|
* where required (Top-level, headers & elements with IDs).
|
|
|
|
|
* Will update the provided $changeMap array with changes made, where keys are the old
|
|
|
|
|
* ids and the corresponding values are the new ids.
|
|
|
|
|
*/
|
|
|
|
|
protected function updateIdsRecursively(DOMNode $element, int $depth, array &$idMap, array &$changeMap): void
|
|
|
|
|
{
|
|
|
|
|
/* @var DOMNode $child */
|
|
|
|
|
foreach ($element->childNodes as $child) {
|
|
|
|
|
if ($child instanceof DOMElement && ($depth === 0 || in_array($child->nodeName, ['h1', 'h2', 'h3', 'h4', 'h5', 'h6']) || $child->getAttribute('id'))) {
|
|
|
|
|
[$oldId, $newId] = $this->setUniqueId($child, $idMap);
|
|
|
|
|
if ($newId && $newId !== $oldId && !isset($idMap[$oldId])) {
|
|
|
|
|
$changeMap[$oldId] = $newId;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if ($child->hasChildNodes()) {
|
|
|
|
|
$this->updateIdsRecursively($child, $depth + 1, $idMap, $changeMap);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* Update the all links in the given xpath to apply requires changes within the
|
|
|
|
|
* given $changeMap array.
|
2020-09-28 21:26:50 +00:00
|
|
|
|
*/
|
2023-02-22 14:32:40 +00:00
|
|
|
|
protected function updateLinks(DOMXPath $xpath, array $changeMap): void
|
2020-09-28 21:26:50 +00:00
|
|
|
|
{
|
2023-02-22 14:32:40 +00:00
|
|
|
|
if (empty($changeMap)) {
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
$links = $xpath->query('//body//*//*[@href]');
|
|
|
|
|
/** @var DOMElement $domElem */
|
|
|
|
|
foreach ($links as $domElem) {
|
|
|
|
|
$href = ltrim($domElem->getAttribute('href'), '#');
|
|
|
|
|
$newHref = $changeMap[$href] ?? null;
|
|
|
|
|
if ($newHref) {
|
|
|
|
|
$domElem->setAttribute('href', '#' . $newHref);
|
|
|
|
|
}
|
2020-09-28 21:26:50 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2019-10-05 11:55:01 +00:00
|
|
|
|
/**
|
|
|
|
|
* Set a unique id on the given DOMElement.
|
2023-02-22 14:32:40 +00:00
|
|
|
|
* A map for existing ID's should be passed in to check for current existence,
|
|
|
|
|
* and this will be updated with any new IDs set upon elements.
|
2021-06-26 15:23:15 +00:00
|
|
|
|
* Returns a pair of strings in the format [old_id, new_id].
|
2019-10-05 11:55:01 +00:00
|
|
|
|
*/
|
2021-11-22 23:33:55 +00:00
|
|
|
|
protected function setUniqueId(DOMNode $element, array &$idMap): array
|
2019-10-05 11:55:01 +00:00
|
|
|
|
{
|
2021-11-22 23:33:55 +00:00
|
|
|
|
if (!$element instanceof DOMElement) {
|
2020-09-28 21:26:50 +00:00
|
|
|
|
return ['', ''];
|
2019-10-05 11:55:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
2020-09-28 21:26:50 +00:00
|
|
|
|
// Stop if there's an existing valid id that has not already been used.
|
2019-10-05 11:55:01 +00:00
|
|
|
|
$existingId = $element->getAttribute('id');
|
2023-02-22 14:32:40 +00:00
|
|
|
|
if (str_starts_with($existingId, 'bkmrk') && !isset($idMap[$existingId])) {
|
2019-10-05 11:55:01 +00:00
|
|
|
|
$idMap[$existingId] = true;
|
2021-06-26 15:23:15 +00:00
|
|
|
|
|
2020-09-28 21:26:50 +00:00
|
|
|
|
return [$existingId, $existingId];
|
2019-10-05 11:55:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
2021-11-20 14:03:56 +00:00
|
|
|
|
// Create a unique id for the element
|
2019-10-05 11:55:01 +00:00
|
|
|
|
// 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);
|
|
|
|
|
$newId = urlencode($contentId);
|
2023-02-22 14:32:40 +00:00
|
|
|
|
$loopIndex = 1;
|
2019-10-05 11:55:01 +00:00
|
|
|
|
|
|
|
|
|
while (isset($idMap[$newId])) {
|
|
|
|
|
$newId = urlencode($contentId . '-' . $loopIndex);
|
|
|
|
|
$loopIndex++;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
$element->setAttribute('id', $newId);
|
|
|
|
|
$idMap[$newId] = true;
|
2021-06-26 15:23:15 +00:00
|
|
|
|
|
2020-09-28 21:26:50 +00:00
|
|
|
|
return [$existingId, $newId];
|
2019-10-05 11:55:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* Get a plain-text visualisation of this page.
|
|
|
|
|
*/
|
|
|
|
|
protected function toPlainText(): string
|
|
|
|
|
{
|
|
|
|
|
$html = $this->render(true);
|
2021-06-26 15:23:15 +00:00
|
|
|
|
|
2020-09-19 14:13:18 +00:00
|
|
|
|
return html_entity_decode(strip_tags($html));
|
2019-10-05 11:55:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
2021-06-26 15:23:15 +00:00
|
|
|
|
* Render the page for viewing.
|
2019-10-05 11:55:01 +00:00
|
|
|
|
*/
|
2021-06-02 20:34:34 +00:00
|
|
|
|
public function render(bool $blankIncludes = false): string
|
2019-10-05 11:55:01 +00:00
|
|
|
|
{
|
2021-08-22 12:15:58 +00:00
|
|
|
|
$content = $this->page->html ?? '';
|
2019-10-05 11:55:01 +00:00
|
|
|
|
|
|
|
|
|
if (!config('app.allow_content_scripts')) {
|
2021-05-03 22:59:52 +00:00
|
|
|
|
$content = HtmlContentFilter::removeScripts($content);
|
2019-10-05 11:55:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if ($blankIncludes) {
|
|
|
|
|
$content = $this->blankPageIncludes($content);
|
2023-04-27 14:52:16 +00:00
|
|
|
|
} else {
|
2023-05-02 11:44:55 +00:00
|
|
|
|
for ($includeDepth = 0; $includeDepth < 3; $includeDepth++) {
|
2023-04-27 14:52:16 +00:00
|
|
|
|
$content = $this->parsePageIncludes($content);
|
|
|
|
|
}
|
2019-10-05 11:55:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return $content;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
2021-06-26 15:23:15 +00:00
|
|
|
|
* Parse the headers on the page to get a navigation menu.
|
2019-10-05 11:55:01 +00:00
|
|
|
|
*/
|
|
|
|
|
public function getNavigation(string $htmlContent): array
|
|
|
|
|
{
|
|
|
|
|
if (empty($htmlContent)) {
|
|
|
|
|
return [];
|
|
|
|
|
}
|
|
|
|
|
|
2021-06-02 20:34:34 +00:00
|
|
|
|
$doc = $this->loadDocumentFromHtml($htmlContent);
|
2019-10-05 11:55:01 +00:00
|
|
|
|
$xPath = new DOMXPath($doc);
|
2021-06-26 15:23:15 +00:00
|
|
|
|
$headers = $xPath->query('//h1|//h2|//h3|//h4|//h5|//h6');
|
2019-10-05 11:55:01 +00:00
|
|
|
|
|
|
|
|
|
return $headers ? $this->headerNodesToLevelList($headers) : [];
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* Convert a DOMNodeList into an array of readable header attributes
|
|
|
|
|
* with levels normalised to the lower header level.
|
|
|
|
|
*/
|
|
|
|
|
protected function headerNodesToLevelList(DOMNodeList $nodeList): array
|
|
|
|
|
{
|
2021-11-22 23:33:55 +00:00
|
|
|
|
$tree = collect($nodeList)->map(function (DOMElement $header) {
|
2019-10-05 11:55:01 +00:00
|
|
|
|
$text = trim(str_replace("\xc2\xa0", '', $header->nodeValue));
|
|
|
|
|
$text = mb_substr($text, 0, 100);
|
|
|
|
|
|
|
|
|
|
return [
|
|
|
|
|
'nodeName' => strtolower($header->nodeName),
|
2021-06-26 15:23:15 +00:00
|
|
|
|
'level' => intval(str_replace('h', '', $header->nodeName)),
|
|
|
|
|
'link' => '#' . $header->getAttribute('id'),
|
|
|
|
|
'text' => $text,
|
2019-10-05 11:55:01 +00:00
|
|
|
|
];
|
|
|
|
|
})->filter(function ($header) {
|
|
|
|
|
return mb_strlen($header['text']) > 0;
|
|
|
|
|
});
|
|
|
|
|
|
|
|
|
|
// Shift headers if only smaller headers have been used
|
|
|
|
|
$levelChange = ($tree->pluck('level')->min() - 1);
|
|
|
|
|
$tree = $tree->map(function ($header) use ($levelChange) {
|
|
|
|
|
$header['level'] -= ($levelChange);
|
2021-06-26 15:23:15 +00:00
|
|
|
|
|
2019-10-05 11:55:01 +00:00
|
|
|
|
return $header;
|
|
|
|
|
});
|
|
|
|
|
|
|
|
|
|
return $tree->toArray();
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* Remove any page include tags within the given HTML.
|
|
|
|
|
*/
|
2021-06-02 20:34:34 +00:00
|
|
|
|
protected function blankPageIncludes(string $html): string
|
2019-10-05 11:55:01 +00:00
|
|
|
|
{
|
|
|
|
|
return preg_replace("/{{@\s?([0-9].*?)}}/", '', $html);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* Parse any include tags "{{@<page_id>#section}}" to be part of the page.
|
|
|
|
|
*/
|
2021-06-02 20:34:34 +00:00
|
|
|
|
protected function parsePageIncludes(string $html): string
|
2019-10-05 11:55:01 +00:00
|
|
|
|
{
|
|
|
|
|
$matches = [];
|
|
|
|
|
preg_match_all("/{{@\s?([0-9].*?)}}/", $html, $matches);
|
|
|
|
|
|
|
|
|
|
foreach ($matches[1] as $index => $includeId) {
|
|
|
|
|
$fullMatch = $matches[0][$index];
|
|
|
|
|
$splitInclude = explode('#', $includeId, 2);
|
|
|
|
|
|
|
|
|
|
// Get page id from reference
|
|
|
|
|
$pageId = intval($splitInclude[0]);
|
|
|
|
|
if (is_nan($pageId)) {
|
|
|
|
|
continue;
|
|
|
|
|
}
|
|
|
|
|
|
2022-09-05 15:40:42 +00:00
|
|
|
|
// Find page to use, and default replacement to empty string for non-matches.
|
2021-10-04 10:26:26 +00:00
|
|
|
|
/** @var ?Page $matchedPage */
|
2019-10-05 11:55:01 +00:00
|
|
|
|
$matchedPage = Page::visible()->find($pageId);
|
2022-09-05 15:40:42 +00:00
|
|
|
|
$replacement = '';
|
|
|
|
|
|
|
|
|
|
if ($matchedPage && count($splitInclude) === 1) {
|
|
|
|
|
// If we only have page id, just insert all page html and continue.
|
|
|
|
|
$replacement = $matchedPage->html;
|
2022-09-06 16:41:32 +00:00
|
|
|
|
} elseif ($matchedPage && count($splitInclude) > 1) {
|
2022-09-05 15:40:42 +00:00
|
|
|
|
// Otherwise, if our include tag defines a section, load that specific content
|
|
|
|
|
$innerContent = $this->fetchSectionOfPage($matchedPage, $splitInclude[1]);
|
|
|
|
|
$replacement = trim($innerContent);
|
2019-10-05 11:55:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
2022-09-05 15:40:42 +00:00
|
|
|
|
$themeReplacement = Theme::dispatch(
|
|
|
|
|
ThemeEvents::PAGE_INCLUDE_PARSE,
|
|
|
|
|
$includeId,
|
|
|
|
|
$replacement,
|
|
|
|
|
clone $this->page,
|
|
|
|
|
$matchedPage ? (clone $matchedPage) : null,
|
|
|
|
|
);
|
2019-10-05 11:55:01 +00:00
|
|
|
|
|
2022-09-05 15:40:42 +00:00
|
|
|
|
// Perform the content replacement
|
|
|
|
|
$html = str_replace($fullMatch, $themeReplacement ?? $replacement, $html);
|
2019-10-05 11:55:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return $html;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* Fetch the content from a specific section of the given page.
|
|
|
|
|
*/
|
|
|
|
|
protected function fetchSectionOfPage(Page $page, string $sectionId): string
|
|
|
|
|
{
|
2021-11-15 00:48:05 +00:00
|
|
|
|
$topLevelTags = ['table', 'ul', 'ol', 'pre'];
|
2021-06-13 11:53:04 +00:00
|
|
|
|
$doc = $this->loadDocumentFromHtml($page->html);
|
2019-10-05 11:55:01 +00:00
|
|
|
|
|
|
|
|
|
// Search included content for the id given and blank out if not exists.
|
|
|
|
|
$matchingElem = $doc->getElementById($sectionId);
|
|
|
|
|
if ($matchingElem === null) {
|
|
|
|
|
return '';
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// Otherwise replace the content with the found content
|
|
|
|
|
// Checks if the top-level wrapper should be included by matching on tag types
|
|
|
|
|
$innerContent = '';
|
|
|
|
|
$isTopLevel = in_array(strtolower($matchingElem->nodeName), $topLevelTags);
|
|
|
|
|
if ($isTopLevel) {
|
|
|
|
|
$innerContent .= $doc->saveHTML($matchingElem);
|
|
|
|
|
} else {
|
|
|
|
|
foreach ($matchingElem->childNodes as $childNode) {
|
|
|
|
|
$innerContent .= $doc->saveHTML($childNode);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
libxml_clear_errors();
|
|
|
|
|
|
|
|
|
|
return $innerContent;
|
|
|
|
|
}
|
2021-06-02 20:34:34 +00:00
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* Create and load a DOMDocument from the given html content.
|
|
|
|
|
*/
|
|
|
|
|
protected function loadDocumentFromHtml(string $html): DOMDocument
|
|
|
|
|
{
|
|
|
|
|
libxml_use_internal_errors(true);
|
|
|
|
|
$doc = new DOMDocument();
|
2023-02-23 22:59:26 +00:00
|
|
|
|
$html = '<?xml encoding="utf-8" ?><body>' . $html . '</body>';
|
|
|
|
|
$doc->loadHTML($html);
|
2021-06-26 15:23:15 +00:00
|
|
|
|
|
2021-06-02 20:34:34 +00:00
|
|
|
|
return $doc;
|
|
|
|
|
}
|
2019-10-05 11:55:01 +00:00
|
|
|
|
}
|