develooper Front page | perl.perl5.porters | Postings from November 2008

Re: char16 datatype

Thread Previous | Thread Next
From:
Dr.Ruud
Date:
November 14, 2008 01:07
Subject:
Re: char16 datatype
Message ID:
20081114090701.18291.qmail@lists.develooper.com
demerphq schreef:

> Ultimately UTF-8 was a kludge,
> developed practically overnight to ensure that there would be a
> unicode representation that was unix legacy compatible, with the long
> term intention of replacing it with something better.

I would never put phrases like "kludge" and "intention to replace" on
UTF-8. Even if those were at the start (and I actually doubt they ever
were), that is all irrelevant now, because the kid has a life of its
own.

UTF-8 is good wherever you have a stream of integers and most of them
are small.

Now only if RAM was that smart, that we could process the start bytes of
a UTF-8 encoded buffer as a continous stream. We just need smarter
hardware! :) (each byte would be subaddressable, or be a stack)

Temporarily storing the continuation bytes separately is obviously
already possible.

-- 
Affijn, Ruud

"Gewoon is een tijger."


Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About