LibRegex: Ensure the GoBack operation decrements the code unit index

This was missed in commit 27d555b.
akimdi · Aug 18, 2021 · 325eabc · 325eabc
1 parent c4ee576
commit 325eabc
Show file tree

Hide file tree

Showing 3 changed files with 33 additions and 2 deletions.
diff --git a/Tests/LibRegex/Regex.cpp b/Tests/LibRegex/Regex.cpp
@@ -637,6 +637,8 @@ TEST_CASE(ECMA262_match)
  { "(a{4}){2}"sv, "aaaaaaaa"sv },
  { "(a{4}){2}"sv, "aaaaaabaa"sv, false },
  { "\\u{4}"sv, "uuuu" },
+ { "(?<=.{3})f"sv, "abcdef"sv, true, (ECMAScriptFlags)regex::AllFlags::Global },
+ { "(?<=.{3})f"sv, "abc😀ef"sv, true, (ECMAScriptFlags)regex::AllFlags::Global },
  // ECMA262, B.1.4. Regular Expression Pattern extensions for browsers
  { "{"sv, "{"sv, true, ECMAScriptFlags::BrowserExtended },
  { "\\5"sv, "\5"sv, true, ECMAScriptFlags::BrowserExtended },
@@ -694,6 +696,8 @@ TEST_CASE(ECMA262_unicode_match)
  { "\\u{1f600}"sv, "😀"sv, true, ECMAScriptFlags::Unicode },
  { "\\ud83d\\ud83d"sv, "\xed\xa0\xbd\xed\xa0\xbd"sv, true },
  { "\\ud83d\\ud83d"sv, "\xed\xa0\xbd\xed\xa0\xbd"sv, true, ECMAScriptFlags::Unicode },
+ { "(?<=.{3})f"sv, "abcdef"sv, true, ECMAScriptFlags::Unicode },
+ { "(?<=.{3})f"sv, "abc😀ef"sv, true, ECMAScriptFlags::Unicode },
  };
 
  for (auto& test : tests) {

diff --git a/Userland/Libraries/LibRegex/RegexByteCode.cpp b/Userland/Libraries/LibRegex/RegexByteCode.cpp
@@ -102,6 +102,17 @@ static void advance_string_position(MatchState& state, RegexStringView const& vi
  }
 }
 
+static void reverse_string_position(MatchState& state, RegexStringView const& view, size_t amount)
+{
+ VERIFY(state.string_position >= amount);
+ state.string_position -= amount;
+
+ if (view.unicode())
+ state.string_position_in_code_units = view.code_unit_offset_of(state.string_position);
+ else
+ state.string_position_in_code_units -= amount;
+}
+
 static void save_string_position(MatchInput const& input, MatchState const& state)
 {
  input.saved_positions.append(state.string_position);
@@ -226,12 +237,12 @@ ALWAYS_INLINE ExecutionResult OpCode_Restore::execute(MatchInput const& input, M
  return ExecutionResult::Continue;
 }
 
-ALWAYS_INLINE ExecutionResult OpCode_GoBack::execute(MatchInput const&, MatchState& state) const
+ALWAYS_INLINE ExecutionResult OpCode_GoBack::execute(MatchInput const& input, MatchState& state) const
 {
  if (count() > state.string_position)
  return ExecutionResult::Failed_ExecuteLowPrioForks;
 
- state.string_position -= count();
+ reverse_string_position(state, input.view, count());
  return ExecutionResult::Continue;
 }
 

diff --git a/Userland/Libraries/LibRegex/RegexMatch.h b/Userland/Libraries/LibRegex/RegexMatch.h
@@ -274,6 +274,22 @@ class RegexStringView {
  });
  }
 
+ size_t code_unit_offset_of(size_t code_point_index) const
+ {
+ return m_view.visit(
+ [&](StringView const& view) -> u32 {
+ Utf8View utf8_view { view };
+ return utf8_view.byte_offset_of(code_point_index);
+ },
+ [&](Utf32View const&) -> u32 { return code_point_index; },
+ [&](Utf16View const& view) -> u32 {
+ return view.code_unit_offset_of(code_point_index);
+ },
+ [&](Utf8View const& view) -> u32 {
+ return view.byte_offset_of(code_point_index);
+ });
+ }
+
  bool operator==(char const* cstring) const
  {
  return m_view.visit(