Optimize String ==, ends_with and begins_with by using memcmp.

[Ivorforce]

This should behave identically as before.

Benchmark

The new implementation can be about 4x as fast as before:

	String s1 = "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.";
	String s2 = s1;

	{
		auto t0 = std::chrono::high_resolution_clock::now();
		for (int i = 0; i < 10000000; i ++) {
			// Test
			bool b = s1 == s2;
		}
		auto t1 = std::chrono::high_resolution_clock::now();
		std::cout << std::chrono::duration_cast<std::chrono::milliseconds>(t1 - t0).count() << "ms\n";
	}
	{
		auto t0 = std::chrono::high_resolution_clock::now();
		for (int i = 0; i < 10000000; i ++) {
			// Test
			bool b = s1.begins_with(s2);
		}
		auto t1 = std::chrono::high_resolution_clock::now();
		std::cout << std::chrono::duration_cast<std::chrono::milliseconds>(t1 - t0).count() << "ms\n";
	}
	{
		auto t0 = std::chrono::high_resolution_clock::now();
		for (int i = 0; i < 10000000; i ++) {
			// Test
			bool b = s1.ends_with(s2);
		}
		auto t1 = std::chrono::high_resolution_clock::now();
		std::cout << std::chrono::duration_cast<std::chrono::milliseconds>(t1 - t0).count() << "ms\n";
	}

• ~1950ms each on master (5b52b4b)
• ~510ms each on this PR (2e60078)

Notes

This PR only optimizes string to string comparisons. It does not optimize comparison to string literals (which are char[]).

[kiroxas]

Nice, and the line count is very good !

[JekSun97]

I haven't looked at the GDScript implementation, but does this performance improvement also apply to GDScript syntax rule parsing?

[Ivorforce]

I haven't looked at the GDScript implementation, but does this performance improvement also apply to GDScript syntax rule parsing?

I don't know for sure, but I think it's unlikely that it will observably affect it. The reason is that most C++ code uses string literals, which are char[] instead of char32_t[] and thus they can't benefit from this change.

[hpvb]

I originally had some questions about the change from get_data() but because of the checks above length() != len and is_empty() this should be safe.

[akien-mga]

Thanks!