Re: Patch: Update of scripts/unicode-muncher.pl

[Tom Tromey <tromey at redhat dot com>]

>>>>> "Michael" == Michael Koch <konqueror@gmx.de> writes:

Michael> I merged the attached patch from GNU classpath to get 
Michael> scripts/unicode-muncher.pl version 2.1 (updating from 2.0) and to 
Michael> merge gnu.java.lang.CharData.
Michael> Okay for mainline ?

I have a couple comments on this.

First, I don't think we can redistribute files like
SpecialCasing-2.txt.  I thought the Unicode.org license prohibited
this.  That is why we don't have the unicode data checked in to cvs.

Second, libgcj doesn't directly use CharData.  I see that it is
listed in ordinary_java_source_files -- but this is wrong.  It just
makes the library bigger.  Could you fix this?

Instead, we have a special program that uses CharData.java to build
our own C++ header describing all the character data.  This is used
by our Character implementation.  See the comments in Character and
also scripts/MakeCharTables.java.

So, this isn't ok as-is, but something similar to it will be...
thanks.

Tom