On 2010-02-17, Jesse Glick <Jesse.Glick@(protected):

> bodewig@(protected):
>> if user input is compared to fixed strings in a case insensitive manner we should use the default locale. Use Locale.ENGLISH consistently

> Perhaps I'm not understanding the point of this.

>> URL: http://svn.apache.org/viewvc/ant/core/trunk/src/main/org/apache/tools/ant/Project.java?rev=910955&r1=910954&r2=910955&view=diff
>>    public static boolean toBoolean(String s) {
>> -     return ("on".equalsIgnoreCase(s)
>> -           || "true".equalsIgnoreCase(s)
>> -           || "yes".equalsIgnoreCase(s));
>>>     String lc = s == null ? null : s.toLowerCase(Locale.ENGLISH);
>>>     return ("on".equals(lc)
>>>           || "true".equals(lc)
>>>           || "yes".equals(lc));

> This should result in exactly the same behavior, except slower. So... why?

Because I misread the Javadocs, it seems. And I didn't perform any
tests to back what I thought I read. The point I didn't see was that
Character.toUpperCase is not Locale sensitive (which isn't spelled out
explicitly).

I'll revert most of the changes and transform the comparisions to
equalsIgnoreCase where appropriate.

Stefan

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@(protected)
For additional commands, e-mail: dev-help@(protected)