LibRegex: Make codegen+optimisation for alternatives much faster

Just a little thinking outside the box, and we can now parse and optimise a million copies of "a|" chained together in just a second :^)
Author: https://github.com/alimpfard Commit: 97a333608e Pull-request: https://github.com/SerenityOS/serenity/pull/12653 Issue: https://github.com/SerenityOS/serenity/issues/12373 Issue: https://github.com/SerenityOS/serenity/issues/12615 Reviewed-by: https://github.com/trflynn89
2025-10-04 23:29:52 +00:00 · 2022-02-20 03:25:21 +03:30 · 2022-02-20 03:25:21 +03:30 · 97a333608e · 2024-07-17 18:29:29 +09:00
commit 97a333608e
parent 4be7239626
5 changed files with 150 additions and 65 deletions
--- a/Userland/Libraries/LibRegex/RegexBytecodeStreamOptimizer.h
+++ b/Userland/Libraries/LibRegex/RegexBytecodeStreamOptimizer.h
@ -14,6 +14,7 @@ namespace regex {
 class Optimizer {
 public:
    static void append_alternation(ByteCode& target, ByteCode&& left, ByteCode&& right);
+    static void append_alternation(ByteCode& target, Span<ByteCode> alternatives);
    static void append_character_class(ByteCode& target, Vector<CompareTypeAndValuePair>&& pairs);
 };