Re: Gump and Unicode

2003-06-12 Thread Jordi Salvat i Alabart
I once had the same problem with the JMeter sources, tried to add that encoding attribute to the java task, and it didn't help. It was helping on my platform, but not on Gump. I never learned why. If you attempt it here, I'll be interested to know if it works. Salut, Jordi. En/na Conor

Re: Gump and Unicode

2003-06-12 Thread Tetsuya Kitahata
Tim, I looked at the code in codec. It is obvious that GUMP error would occur the same as Jakarta-Jetspeed and other projects experienced once. (Also, in my Japanese environment, it is garbled character) This means that we Japanese can not build codec. (Default codepage is different) ...

Re: Gump and Unicode

2003-06-12 Thread Santiago Gala
Sam Ruby escribió: Santiago Gala wrote: Conor MacNeill escribió: (...) Just to be clear this is not a Gump issue - I think the problem would appear whenever you try to compile on any platform with a different default encoding. Yes. For this reason, I'm encouraging people to start using

RE: Gump and Unicode

2003-06-12 Thread Tim Vernum
export LC_ALL=en_US.UTF-8 in /etc/.profile (or .bash_profile... depending on shell) of the user under which the processes run should map I'm not Solaris Expert, so I can't comment on this. My installs of Solaris 2.6 and 8 support the en_US.UTF-8 locale, so I suspect the process is the

Re: Gump and Unicode

2003-06-11 Thread Santiago Gala
Tim O'Brien escribió: commons-codec fails to compile in Gump because it contains an Ntilde among other characters used in languages other than English. Any ideas? I seem to recall that java source code is supposed to be written in unicode, but I could be wrong. The '\u' convention is

Re: Gump and Unicode

2003-06-11 Thread Santiago Gala
Santiago Gala escribió: Tim O'Brien escribió: commons-codec fails to compile in Gump because it contains an Ntilde among other characters used in languages other than English. Any ideas? I seem to recall that java source code is supposed to be written in unicode, but I could be wrong. The

Re: Gump and Unicode

2003-06-11 Thread Brian Ewins
Use the unicode escapes rather than the character literals in the code? You won't get DoubleMetaphone.java to compile unless you pass the encoding flag to javac. The two letters appear to be \u00C7, \u00D1 - capital C with a cedilla and capital N with a tilde? Putting case '\u00C7': case

Re: Gump and Unicode

2003-06-11 Thread Conor MacNeill
On Thu, 12 Jun 2003 01:05 am, Brian Ewins wrote: Use the unicode escapes rather than the character literals in the code? You won't get DoubleMetaphone.java to compile unless you pass the encoding flag to javac. The two letters appear to be \u00C7, \u00D1 - capital C with a cedilla and

Re: Gump and Unicode

2003-06-11 Thread Santiago Gala
Conor MacNeill escribió: (...) Just to be clear this is not a Gump issue - I think the problem would appear whenever you try to compile on any platform with a different default encoding. Yes. For this reason, I'm encouraging people to start using utf-8 as default encoding in any server