ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefan Bodewig <>
Subject Re: log4j failing on Gump, maybe something in recent commits
Date Wed, 17 Feb 2010 16:16:58 GMT
On 2010-02-17, Antoine Levy Lambert <> wrote:

> I did not even know that he have a utility class called VectorSet.

That's the one introduced to speed up DirectoryScanner without breaking
backwards compatibility.  A Vector-subclass (so we could keep the
protected Vector fields) with O(1) contains().

> I did not know that Locale.ENGLISH is supposed to be the default
> locale.

At least it is available everywhere and it does what you'd expect it to

> In which cases does String#toLowerCase() or toUpperCase() behave
> differently depending upon the locale ?


The canonical example is the letter i and the Turkish locale.  In Turkey
there are two "i"s - at least if you look at it with "western" eyes.
One with a dot on top of it and one without, and both come in upper- and
lowercase versions.  So an uppercase "i" is "I" with a dot on top of it
and a lowercase "I" an "i" without a dot.

This means in a Turkish locale "FILESET".equalsIgnoreCase("fileset")
returns false - as does "FILESET".toLowerCase().equals("fileset").


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message