RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-31 15:28:11 +00:00
Code Issues Activity Actions

LibWeb: Make HTML parser flush all pending tokens in "in table text"

Browse source 
There were multiple bugs in the parsing algorithm for handling text
occurring inside a `table` element:

- When there was pending non-whitespace text inside a table, we only
  flushed one token instead of all pending tokens.

- Also, we didn't even flush one of the right tokens, but instead the
  token that caused the flush to happen.

- Once we started flushing the right tokens, it turned out we had not
  yet implemented character insertion points expressed as "before X".

- Finally, we were not exiting the "in table text" mode after flushing
  pending tokens, effectively getting us stuck in that mode until EOF.
This commit is contained in:
Andreas Kling 2023-07-03 10:00:20 +02:00
parent 8c3e5137f7
commit 5cdb394400
3 changed files with 25 additions and 12 deletions
Download patch file Download diff file Expand all files Collapse all files

1
Tests/LibWeb/Text/expected/html-parser-text-in-table-hoisting.txt Normal file
View file

@ -0,0 +1 @@
PASS

10
Tests/LibWeb/Text/input/html-parser-text-in-table-hoisting.html Normal file
View file

@ -0,0 +1,10 @@
<script src="include.js"></script>
<body><table><tr>PASS</tr></table></body>
<script>
test(() => {
// Remove the table. "PASS" should still be visible,
// as the HTML parser inserts it *before* the table
// under these circumstances.
document.querySelector("table").remove()
});
</script>

26
Userland/Libraries/LibWeb/HTML/Parser/HTMLParser.cpp
View file

@ -1001,7 +1001,11 @@ DOM::Text* HTMLParser::find_character_insertion_node()
{ {
auto adjusted_insertion_location = find_appropriate_place_for_inserting_node(); auto adjusted_insertion_location = find_appropriate_place_for_inserting_node();
if (adjusted_insertion_location.insert_before_sibling) { if (adjusted_insertion_location.insert_before_sibling) {
TODO(); if (adjusted_insertion_location.insert_before_sibling->previous_sibling() && adjusted_insertion_location.insert_before_sibling->previous_sibling()->is_text())
return static_cast<DOM::Text*>(adjusted_insertion_location.insert_before_sibling->previous_sibling());
auto new_text_node = realm().heap().allocate<DOM::Text>(realm(), document(), "").release_allocated_value_but_fixme_should_propagate_errors();
adjusted_insertion_location.parent->insert_before(*new_text_node, *adjusted_insertion_location.insert_before_sibling);
return new_text_node;
} }
if (adjusted_insertion_location.parent->is_document()) if (adjusted_insertion_location.parent->is_document())
return nullptr; return nullptr;
@ -2661,20 +2665,18 @@ void HTMLParser::handle_in_table_text(HTMLToken& token)
// are character tokens that are not ASCII whitespace, then this is a parse error: // are character tokens that are not ASCII whitespace, then this is a parse error:
// reprocess the character tokens in the pending table character tokens list using // reprocess the character tokens in the pending table character tokens list using
// the rules given in the "anything else" entry in the "in table" insertion mode. // the rules given in the "anything else" entry in the "in table" insertion mode.
for (auto& pending_token : m_pending_table_character_tokens) { if (any_of(m_pending_table_character_tokens, [](auto const& token) { return !token.is_parser_whitespace(); })) {
VERIFY(pending_token.is_character()); log_parse_error();
if (!pending_token.is_parser_whitespace()) { for (auto& pending_token : m_pending_table_character_tokens) {
log_parse_error();
m_foster_parenting = true; m_foster_parenting = true;
process_using_the_rules_for(InsertionMode::InBody, token); process_using_the_rules_for(InsertionMode::InBody, pending_token);
m_foster_parenting = false; m_foster_parenting = false;
return;
} }
} } else {
// Otherwise, insert the characters given by the pending table character tokens list.
// Otherwise, insert the characters given by the pending table character tokens list. for (auto& pending_token : m_pending_table_character_tokens) {
for (auto& pending_token : m_pending_table_character_tokens) { insert_character(pending_token.code_point());
insert_character(pending_token.code_point()); }
} }
// Switch the insertion mode to the original insertion mode and reprocess the token. // Switch the insertion mode to the original insertion mode and reprocess the token.