Talk:MD5

The UTF8_MD5 seem to work OK, only as long as the string contains only letters.

For example UTF8_MD5("Anylyn Hax") returns "4af17d386f8a3ea9cb861afdf8dbae00", and thats correct.

But if I feed it a String containing numbers like UTF8_MD5("Hf0jUQFz32pIVRa") this should return "d06d41d2f783f39934bfc3c1969a30b2", but it returns "d998312450ba64589563cc563a6cc0cf", same with UTF8_MD5("666") it returns "693e9af84d3dfcc71e640e005bdc5e2e" instead of "fae0b27c451c728867a567e8c1bb4e53"

Can you explain me this ?

--Anylyn Hax 17:19, 27 December 2011 (PST)

It must be something to do with the order of operations as I get the correct results in LSLEditor. -- Strife ^{(talk|contribs)} 16:54, 28 December 2011 (PST)

I haven't spotted any instances of optimization that could be causing this. I have spotted one operation where parentheses weren't used to remove all ambiguity but it's excedingly unlikely to be the cause. -- Strife ^{(talk|contribs)} 17:27, 28 December 2011 (PST)

At this point I'm inclined to believe it to be an LSL issue. Likely there is a known bug/caveat that I've overlooked that LSLEditor doesn't abide by. Until I can get in world to confirm your findings and debug I am at a loose as to where to start. -- Strife ^{(talk|contribs)} 17:34, 28 December 2011 (PST)