How we made Haskell search strings as fast as Rust

Re:How we made Haskell search strings as fast as Rust by gowld on Wednesday March 13, 2019 @06:07PM (#19384411)

32 bit-representations (compared to UCS-2), not 32 bits worth of values, so it can represent all of Unicode.

UCS-2 is limited to only 16-bit representation, making it _impossible_ to express Unicode values outside the basic multilingual plane (BMP).

[ Parent | on HN ]

Re:How we made Haskell search strings as fast as Rust by stouset on Wednesday March 13, 2019 @07:21PM (#19384982)

Correct, this is what I meant. UTF-16 can use up to 32 bits to represent a larger number of characters, but cannot uniquely represent 2^32 possible characters.

[ Parent | on HN ]