1
0
mirror of https://github.com/BookStackApp/BookStack.git synced 2024-10-29 23:22:34 +01:00

Includes: Switched page to new system

- Added mulit-level depth parsing.
- Updating usage of HTML doc in page content to be efficient.
- Removed now redundant PageContentTest cases.
- Made some include system fixes based upon testing.
This commit is contained in:
Dan Brown 2023-11-27 19:54:47 +00:00
parent 4874dc1304
commit 71c93c8878
No known key found for this signature in database
GPG Key ID: 46D9F943C24A2EF9
7 changed files with 60 additions and 148 deletions

View File

@ -275,21 +275,33 @@ class PageContent
*/ */
public function render(bool $blankIncludes = false): string public function render(bool $blankIncludes = false): string
{ {
$content = $this->page->html ?? ''; $html = $this->page->html ?? '';
if (empty($html)) {
return $html;
}
$doc = new HtmlDocument($html);
$contentProvider = function (int $id) use ($blankIncludes) {
if ($blankIncludes) {
return '';
}
return Page::visible()->find($id)->html ?? '';
};
$parser = new PageIncludeParser($doc, $contentProvider);
$nodesAdded = 1;
for ($includeDepth = 0; $includeDepth < 3 && $nodesAdded !== 0; $includeDepth++) {
$nodesAdded = $parser->parse();
}
if (!config('app.allow_content_scripts')) { if (!config('app.allow_content_scripts')) {
$content = HtmlContentFilter::removeScripts($content); HtmlContentFilter::removeScriptsFromDocument($doc);
} }
if ($blankIncludes) { return $doc->getBodyInnerHtml();
$content = $this->blankPageIncludes($content);
} else {
for ($includeDepth = 0; $includeDepth < 3; $includeDepth++) {
$content = $this->parsePageIncludes($content);
}
}
return $content;
} }
/** /**
@ -337,83 +349,4 @@ class PageContent
return $tree->toArray(); return $tree->toArray();
} }
/**
* Remove any page include tags within the given HTML.
*/
protected function blankPageIncludes(string $html): string
{
return preg_replace("/{{@\s?([0-9].*?)}}/", '', $html);
}
/**
* Parse any include tags "{{@<page_id>#section}}" to be part of the page.
*/
protected function parsePageIncludes(string $html): string
{
$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;
}
// Find page to use, and default replacement to empty string for non-matches.
/** @var ?Page $matchedPage */
$matchedPage = Page::visible()->find($pageId);
$replacement = '';
if ($matchedPage && count($splitInclude) === 1) {
// If we only have page id, just insert all page html and continue.
$replacement = $matchedPage->html;
} elseif ($matchedPage && count($splitInclude) > 1) {
// Otherwise, if our include tag defines a section, load that specific content
$innerContent = $this->fetchSectionOfPage($matchedPage, $splitInclude[1]);
$replacement = trim($innerContent);
}
$themeReplacement = Theme::dispatch(
ThemeEvents::PAGE_INCLUDE_PARSE,
$includeId,
$replacement,
clone $this->page,
$matchedPage ? (clone $matchedPage) : null,
);
// Perform the content replacement
$html = str_replace($fullMatch, $themeReplacement ?? $replacement, $html);
}
return $html;
}
/**
* Fetch the content from a specific section of the given page.
*/
protected function fetchSectionOfPage(Page $page, string $sectionId): string
{
$topLevelTags = ['table', 'ul', 'ol', 'pre'];
$doc = new HtmlDocument($page->html);
// 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
$isTopLevel = in_array(strtolower($matchingElem->nodeName), $topLevelTags);
if ($isTopLevel) {
return $doc->getNodeOuterHtml($matchingElem);
}
return $doc->getNodeInnerHtml($matchingElem);
}
} }

View File

@ -14,7 +14,7 @@ class PageIncludeContent
*/ */
protected array $contents = []; protected array $contents = [];
protected bool $isTopLevel; protected bool $isTopLevel = false;
public function __construct( public function __construct(
string $html, string $html,

View File

@ -20,19 +20,19 @@ class PageIncludeParser
protected array $toCleanup = []; protected array $toCleanup = [];
public function __construct( public function __construct(
protected string $pageHtml, protected HtmlDocument $doc,
protected Closure $pageContentForId, protected Closure $pageContentForId,
) { ) {
} }
/** /**
* Parse out the include tags. * Parse out the include tags.
* Returns the count of new content DOM nodes added to the document.
*/ */
public function parse(): string public function parse(): int
{ {
$doc = new HtmlDocument($this->pageHtml); $nodesAdded = 0;
$tags = $this->locateAndIsolateIncludeTags();
$tags = $this->locateAndIsolateIncludeTags($doc);
foreach ($tags as $tag) { foreach ($tags as $tag) {
$htmlContent = $this->pageContentForId->call($this, $tag->getPageId()); $htmlContent = $this->pageContentForId->call($this, $tag->getPageId());
@ -48,12 +48,14 @@ class PageIncludeParser
} }
} }
$this->replaceNodeWithNodes($tag->domNode, $content->toDomNodes()); $replacementNodes = $content->toDomNodes();
$nodesAdded += count($replacementNodes);
$this->replaceNodeWithNodes($tag->domNode, $replacementNodes);
} }
$this->cleanup(); $this->cleanup();
return $doc->getBodyInnerHtml(); return $nodesAdded;
} }
/** /**
@ -61,9 +63,9 @@ class PageIncludeParser
* own nodes in the DOM for future targeted manipulation. * own nodes in the DOM for future targeted manipulation.
* @return PageIncludeTag[] * @return PageIncludeTag[]
*/ */
protected function locateAndIsolateIncludeTags(HtmlDocument $doc): array protected function locateAndIsolateIncludeTags(): array
{ {
$includeHosts = $doc->queryXPath("//body//*[text()[contains(., '{{@')]]"); $includeHosts = $this->doc->queryXPath("//*[text()[contains(., '{{@')]]");
$includeTags = []; $includeTags = [];
/** @var DOMNode $node */ /** @var DOMNode $node */
@ -125,7 +127,7 @@ class PageIncludeParser
foreach ($replacements as $replacement) { foreach ($replacements as $replacement) {
if ($replacement->ownerDocument !== $targetDoc) { if ($replacement->ownerDocument !== $targetDoc) {
$replacement = $targetDoc->adoptNode($replacement); $replacement = $targetDoc->importNode($replacement, true);
} }
$toReplace->parentNode->insertBefore($replacement, $toReplace); $toReplace->parentNode->insertBefore($replacement, $toReplace);
@ -190,7 +192,7 @@ class PageIncludeParser
return $parent; return $parent;
} }
$parent = $parent->parentElement; $parent = $parent->parentNode;
} while ($parent !== null); } while ($parent !== null);
return null; return null;

View File

@ -50,7 +50,7 @@ class CustomHtmlHeadContentProvider
$hash = md5($content); $hash = md5($content);
return $this->cache->remember('custom-head-export:' . $hash, 86400, function () use ($content) { return $this->cache->remember('custom-head-export:' . $hash, 86400, function () use ($content) {
return HtmlContentFilter::removeScripts($content); return HtmlContentFilter::removeScriptsFromHtmlString($content);
}); });
} }

View File

@ -9,16 +9,10 @@ use DOMNodeList;
class HtmlContentFilter class HtmlContentFilter
{ {
/** /**
* Remove all the script elements from the given HTML. * Remove all the script elements from the given HTML document.
*/ */
public static function removeScripts(string $html): string public static function removeScriptsFromDocument(HtmlDocument $doc)
{ {
if (empty($html)) {
return $html;
}
$doc = new HtmlDocument($html);
// Remove standard script tags // Remove standard script tags
$scriptElems = $doc->queryXPath('//script'); $scriptElems = $doc->queryXPath('//script');
static::removeNodes($scriptElems); static::removeNodes($scriptElems);
@ -53,6 +47,19 @@ class HtmlContentFilter
// Remove 'on*' attributes // Remove 'on*' attributes
$onAttributes = $doc->queryXPath('//@*[starts-with(name(), \'on\')]'); $onAttributes = $doc->queryXPath('//@*[starts-with(name(), \'on\')]');
static::removeAttributes($onAttributes); static::removeAttributes($onAttributes);
}
/**
* Remove scripts from the given HTML string.
*/
public static function removeScriptsFromHtmlString(string $html): string
{
if (empty($html)) {
return $html;
}
$doc = new HtmlDocument($html);
static::removeScriptsFromDocument($doc);
return $doc->getBodyInnerHtml(); return $doc->getBodyInnerHtml();
} }

View File

@ -8,7 +8,7 @@ use Tests\TestCase;
class PageContentTest extends TestCase class PageContentTest extends TestCase
{ {
protected $base64Jpeg = '/9j/2wBDAAMCAgICAgMCAgIDAwMDBAYEBAQEBAgGBgUGCQgKCgkICQkKDA8MCgsOCwkJDRENDg8QEBEQCgwSExIQEw8QEBD/yQALCAABAAEBAREA/8wABgAQEAX/2gAIAQEAAD8A0s8g/9k='; protected string $base64Jpeg = '/9j/2wBDAAMCAgICAgMCAgIDAwMDBAYEBAQEBAgGBgUGCQgKCgkICQkKDA8MCgsOCwkJDRENDg8QEBEQCgwSExIQEw8QEBD/yQALCAABAAEBAREA/8wABgAQEAX/2gAIAQEAAD8A0s8g/9k=';
public function test_page_includes() public function test_page_includes()
{ {
@ -57,38 +57,6 @@ class PageContentTest extends TestCase
$this->assertEquals('', $page->text); $this->assertEquals('', $page->text);
} }
public function test_page_includes_do_not_break_tables()
{
$page = $this->entities->page();
$secondPage = $this->entities->page();
$content = '<table id="table"><tbody><tr><td>test</td></tr></tbody></table>';
$secondPage->html = $content;
$secondPage->save();
$page->html = "{{@{$secondPage->id}#table}}";
$page->save();
$pageResp = $this->asEditor()->get($page->getUrl());
$pageResp->assertSee($content, false);
}
public function test_page_includes_do_not_break_code()
{
$page = $this->entities->page();
$secondPage = $this->entities->page();
$content = '<pre id="bkmrk-code"><code>var cat = null;</code></pre>';
$secondPage->html = $content;
$secondPage->save();
$page->html = "{{@{$secondPage->id}#bkmrk-code}}";
$page->save();
$pageResp = $this->asEditor()->get($page->getUrl());
$pageResp->assertSee($content, false);
}
public function test_page_includes_rendered_on_book_export() public function test_page_includes_rendered_on_book_export()
{ {
$page = $this->entities->page(); $page = $this->entities->page();

View File

@ -3,6 +3,7 @@
namespace Tests\Unit; namespace Tests\Unit;
use BookStack\Entities\Tools\PageIncludeParser; use BookStack\Entities\Tools\PageIncludeParser;
use BookStack\Util\HtmlDocument;
use Tests\TestCase; use Tests\TestCase;
class PageIncludeParserTest extends TestCase class PageIncludeParserTest extends TestCase
@ -214,13 +215,14 @@ class PageIncludeParserTest extends TestCase
); );
} }
protected function runParserTest(string $html, array $contentById, string $expected) protected function runParserTest(string $html, array $contentById, string $expected): void
{ {
$parser = new PageIncludeParser($html, function (int $id) use ($contentById) { $doc = new HtmlDocument($html);
$parser = new PageIncludeParser($doc, function (int $id) use ($contentById) {
return $contentById[strval($id)] ?? ''; return $contentById[strval($id)] ?? '';
}); });
$result = $parser->parse(); $parser->parse();
$this->assertEquals($expected, $result); $this->assertEquals($expected, $doc->getBodyInnerHtml());
} }
} }