>Jesse Glick wrote:
>> [EMAIL PROTECTED] wrote:
>>>> It seems safe and useful for 1.7 .
>>>
>>> +1
useful: +1, many people are writing on Java5, Java6 is ante portas.
>> Any other +/-? It's not really clear to me what, if any, criteria
>> there are for RFEs this late in the cycle. I *think* the patch is
safe
>> but you never know.
Yes - very late. It's not a checkstyle or doco-patch ...
You have done a complete test I think?
>the only thing I see this breaking is any class which has a
>java5 enum of a specific name, and a legacy enum with the same
>name. Is this likely? Not yet.
You mean that?
public class T extends org.apache.tools.ant.Task {
public enum ThingEn { first, second, third }
private ThingEn enThing;
public void setThing(ThingEn e) {
thing = e;
}
public ThingEA extends EnumeratedAttribute {
public String[] getValues() { return new String[]{"first", "second",
"third"}; }
}
private ThingEA eaThing;
public void setThing(ThingEA e) {
eaThing = e;
}
}
--> two "setThing()" methods
>Oh, and you end up writing tasks that are java5+ only, of course.
Nothing for Ant itself - but if the writer wants to use Java5, he will
do that.
But supporting enum it would be easier to forget, that enum is Java5+
... that's right.
Maybe we could spent a refactoring "convert enum to EnumeratedAttribute"
;-)
>If the change does go in, I'd have to update that bit of Ant
>in Action (that's its new name, BTW), which is not too late.
hihi .... "Author in Action" ;-)
Jan
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
