DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=4100>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=4100

System language settings and "BUILD FAILED": the <javac> task doesn't support 
"includes" attribute





------- Additional Comments From [EMAIL PROTECTED]  2001-10-20 02:25 -------
I guess someone should check through the Ant sources for suspicious uses of
toLowerCase() and toUpperCase()--I only checked the core package, not taskdefs
etc. The no-param versions should only be used on locale-aware text. Wherever
the input is assumed to be US-ASCII text (for Ant, probably this means all
method names, task names, etc.), some specific locale (e.g. Locale.US, really
anything other than Turkish) should be passed which ensures that the case
conversion is done in ASCII. Character.to*Case should also be safe.

Fortunately String.compareTo is not locale-sensitive, otherwise you could have
further differences in a variety of European locales.

I will attach a list of no-arg case conversions in current Ant dev sources. Most
have something to do with OS name; unless there is somebody testing for e.g.
"Irix" I doubt most of these would cause problems in Turkish. But many other
uses look suspicious.

Reply via email to