[Mono-devel-list] ICU and Formats

[David Sheldon]

Hi, 

I have been trying to solve 
  http://bugs.ximian.com/show_bug.cgi?id=53022

With the aim of reducing the number of test case failures in the corlib.

I have made one change that improves the parsing of the
NumberFormatInfo. However it appears that the underlying data in the ICU
locales does not match up with that in the MS runtime. 

I wrote a basic program to produce locale data from the MS runtime, and
with that installed in ICU, the formatting of numbers is far better.

I would appreciate feedback on the solution here, as it looks like we
will have to have our own ICU lookup tables, and will have to augment my 
locale generating code to produce all of the information required by
each MS.NET locale. 

Would it be better for us to put the data from the other runtime into
our own format in between, rather than using ICU at all. Would there be
issues with using the MS runtime in this way? 

David
-- 
About the use of language: it is impossible to sharpen a pencil with a blunt
ax.  It is equally vain to try to do it with ten blunt axes instead.
		-- Edsger Dijkstra