Files
sousa-gecko/parser/html/nsHtml5StringParser.h
Keith Cirkel 869e7cfd00 Bug 2062652 - Part 4 - Sanitize while parsing r=dom-core-reviewers,hsivonen
This change passes the Sanitizer to nsHtml5TreeBuilder & co, allowing for
consumption of the Sanitizer during parsing, rather than iterating over the
reified tree.

Differential Revision: https://phabricator.services.mozilla.com/D320597
2026-09-08 15:20:48 +00:00

128 lines
3.7 KiB
C++

/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
#ifndef nsHtml5StringParser_h
#define nsHtml5StringParser_h
#include "mozilla/UniquePtr.h"
#include "nsHtml5AtomTable.h"
#include "nsParserBase.h"
class nsHtml5OplessBuilder;
class nsHtml5TreeBuilder;
class nsHtml5Tokenizer;
class nsIContent;
namespace mozilla {
template <typename T>
class Maybe;
namespace dom {
class CustomElementRegistry;
class Document;
class Sanitizer;
} // namespace dom
} // namespace mozilla
class nsHtml5StringParser : public nsParserBase {
public:
NS_DECL_ISUPPORTS
/**
* Constructor for use ONLY by nsContentUtils. Others, please call the
* nsContentUtils statics that wrap this.
*/
nsHtml5StringParser();
/**
* Invoke the fragment parsing algorithm (innerHTML).
* https://html.spec.whatwg.org/#html-fragment-parsing-algorithm
* DO NOT CALL from outside nsContentUtils.cpp.
*
* @param aSourceBuffer the string being set as innerHTML
* @param aTargetNode the target container
* @param aContextLocalName local name of context node
* @param aContextNamespace namespace of context node
* @param aQuirks true to make <table> not close <p>
* @param aPreventScriptExecution true to prevent scripts from executing;
* don't set to false when parsing into a target node that has been bound
* to tree.
* @param aAllowDeclarativeShadowRoots allow the creation of declarative
* shadow roots.
* @param aSanitizer the Sanitizer API configuration to apply while parsing
* (the spec's "parser sanitizer configuration"), or nullptr not to sanitize
* while parsing.
* @param aSanitizerSafe safely sanitize, i.e.
* "remove javascript navigation URLs".
*/
nsresult ParseFragment(
const nsAString& aSourceBuffer, nsIContent* aTargetNode,
nsAtom* aContextLocalName, int32_t aContextNamespace, bool aQuirks,
bool aPreventScriptExecution, bool aAllowDeclarativeShadowRoots,
mozilla::Maybe<RefPtr<mozilla::dom::CustomElementRegistry>>
aCustomElementRegistry,
mozilla::dom::Sanitizer* aSanitizer = nullptr,
bool aSanitizerSafe = false);
/**
* Parse an entire HTML document from a source string.
* DO NOT CALL from outside nsContentUtils.cpp.
*
*/
nsresult ParseDocument(const nsAString& aSourceBuffer,
mozilla::dom::Document* aTargetDoc,
bool aScriptingEnabledForNoscriptParsing,
mozilla::dom::Sanitizer* aSanitizer,
bool aSanitizerSafe);
private:
virtual ~nsHtml5StringParser();
nsresult Tokenize(const nsAString& aSourceBuffer,
mozilla::dom::Document* aDocument,
bool aScriptingEnabledForNoscriptParsing,
bool aDeclarativeShadowRootsAllowed);
void TryCache();
void ClearCaches();
/**
* The tree operation executor
*/
RefPtr<nsHtml5OplessBuilder> mBuilder;
/**
* The HTML5 tree builder
*/
const mozilla::UniquePtr<nsHtml5TreeBuilder> mTreeBuilder;
/**
* The HTML5 tokenizer
*/
const mozilla::UniquePtr<nsHtml5Tokenizer> mTokenizer;
/**
* The scoped atom table
*/
nsHtml5AtomTable mAtomTable;
class CacheClearer : public mozilla::Runnable {
public:
explicit CacheClearer(nsHtml5StringParser* aParser)
: Runnable("CacheClearer"), mParser(aParser) {}
NS_IMETHOD Run() {
if (mParser) {
mParser->ClearCaches();
}
return NS_OK;
}
void Disconnect() { mParser = nullptr; }
private:
nsHtml5StringParser* mParser;
};
RefPtr<CacheClearer> mCacheClearer;
};
#endif // nsHtml5StringParser_h