[Mono-bugs] [Bug 480152] string.Normalize() frequently produces incorrect output

[bugzilla_noreply at novell.com]

http://bugzilla.novell.com/show_bug.cgi?id=480152

http://bugzilla.novell.com/show_bug.cgi?id=480152#c32


--- Comment #32 from Damien Diederen <dd at crosstwine.com> 2010-04-20 20:26:09 UTC ---
Hello everybody,

I have been working some more on this bug, and have discovered a
couple other edge cases in the current composition algorithm.  After a
few retries, I realized that it just couldn't be “patched” to do the
right thing, so I bit the bullet and rewrote it by following the spec.

This, a couple other fixes, and the addition of the Hangul composition
algorithm (complementing Gabriel's work on decomposition) causes the
number of failures in the BMP subset of the Unicode canonicalization
test suite to drop from 4k+ to 888.

(Another, still experimental, and more controversial series of mine
brings that number down to zero, but it still needs a bit of work
before I post it as an RFC.)

As far as I can tell, this series is ready for inclusion in the Mono
repository.  Could somebody have a look at it, and comment/question/
inform me about the next steps?  Thanks!

(The patches will follow as attachments, they should apply cleanly to
trunk/mcs at 155829.)

Cheers,
Damien.

-- 
Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.