beenull/ladybird
1
0
Fork 0
mirror of https://github.com/LadybirdBrowser/ladybird.git synced 2024-11-22 15:40:19 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 6

LibPDF: Support parsing page tree nodes that are in object streams

Browse source 
conditionally_parse_page_tree_node used to assume that the xref table
contained a byte offset, even for compressed objects. It now uses the
common facilities for parsing objects, at the expense of some
performance.
This commit is contained in:
Julian Offenhäuser 2022-08-20 15:17:15 +02:00 committed by Sam Atkins
parent 6225c03256
commit 77f5f7a6f4
Notes: sideshowbarker 2024-07-17 06:58:19 +09:00 
Author: https://github.com/janso3
Commit: https://github.com/SerenityOS/serenity/commit/77f5f7a6f4
Pull-request: https://github.com/SerenityOS/serenity/pull/14873
Reviewed-by: https://github.com/mattco98 ✅
2 changed files with 20 additions and 42 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

59
Userland/Libraries/LibPDF/DocumentParser.cpp
View file

@ -644,51 +644,26 @@ bool DocumentParser::navigate_to_after_startxref()
PDFErrorOr<RefPtr<DictObject>> DocumentParser::conditionally_parse_page_tree_node(u32 object_index)
{
VERIFY(m_xref_table->has_object(object_index));
auto byte_offset = m_xref_table->byte_offset_for_object(object_index);
auto dict_value = TRY(parse_object_with_index(object_index));
auto dict_object = dict_value.get<NonnullRefPtr<Object>>();
if (!dict_object->is<DictObject>())
return error(String::formatted("Invalid page tree with xref index {}", object_index));
m_reader.move_to(byte_offset);
TRY(parse_number());
TRY(parse_number());
if (!m_reader.matches("obj"))
return error(String::formatted("Invalid page tree offset {}", object_index));
auto dict = dict_object->cast<DictObject>();
if (!dict->contains_any_of(CommonNames::Type, CommonNames::Parent, CommonNames::Kids, CommonNames::Count))
// This is a page, not a page tree node
return RefPtr<DictObject> {};
m_reader.move_by(3);
m_reader.consume_whitespace();
if (!dict->contains(CommonNames::Type))
return RefPtr<DictObject> {};
auto type_object = TRY(dict->get_object(m_document, CommonNames::Type));
if (!type_object->is<NameObject>())
return RefPtr<DictObject> {};
auto type_name = type_object->cast<NameObject>();
if (type_name->name() != CommonNames::Pages)
return RefPtr<DictObject> {};
VERIFY(m_reader.consume('<') && m_reader.consume('<'));
m_reader.consume_whitespace();
HashMap<FlyString, Value> map;
while (true) {
if (m_reader.matches(">>"))
break;
auto name = TRY(parse_name());
auto name_string = name->name();
if (!name_string.is_one_of(CommonNames::Type, CommonNames::Parent, CommonNames::Kids, CommonNames::Count)) {
// This is a page, not a page tree node
return RefPtr<DictObject> {};
}
auto value = TRY(parse_value());
if (name_string == CommonNames::Type) {
if (!value.has<NonnullRefPtr<Object>>())
return RefPtr<DictObject> {};
auto type_object = value.get<NonnullRefPtr<Object>>();
if (!type_object->is<NameObject>())
return RefPtr<DictObject> {};
auto type_name = type_object->cast<NameObject>();
if (type_name->name() != CommonNames::Pages)
return RefPtr<DictObject> {};
}
map.set(name->name(), value);
}
VERIFY(m_reader.consume('>') && m_reader.consume('>'));
m_reader.consume_whitespace();
return make_object<DictObject>(map);
return dict;
}
}

3
Userland/Libraries/LibPDF/ObjectDerivatives.h
View file

@ -113,6 +113,9 @@ public:
template<typename... Args>
bool contains(Args&&... keys) const { return (m_map.contains(keys) && ...); }
template<typename... Args>
bool contains_any_of(Args&&... keys) const { return (m_map.contains(keys) || ...); }
ALWAYS_INLINE Optional<Value> get(FlyString const& key) const { return m_map.get(key); }
Value get_value(FlyString const& key) const