ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Steve Loughran" <stev...@iseran.com>
Subject Re: Why stop at JDK1.2? Only semi-serious question...
Date Sun, 14 Jul 2002 04:21:26 GMT

We had to use 1.2 last year for production as it was what ops used, and yes,
we used ant server side to run the deployment stuff.

I dont think there is enough compelling to move to 1.3 over 1.2. dynamic
proxies are slick for things like XML to soap bindings where you want a fast
binding, but except for those tasks called many times in a build (echo,
copy), it probably isnt worth the effort.

Java1.4 does add some interesting stuff for a change; I'd like to add
asserts to my code again, NIO would be slick for something high performance
too. But we have to hold off using that till it is widespread, which will
take ages. And once you add assert statements, you are definately discarding
backwards compat.

-steve

----- Original Message -----
From: "Jon Skeet" <jon.skeet@peramon.com>
To: "Ant developers (E-mail)" <ant-dev@jakarta.apache.org>
Sent: Saturday, July 13, 2002 14:41
Subject: Why stop at JDK1.2? Only semi-serious question...


All this discussion about whether or not to drop JDK 1.1 support has
just got me thinking: why stop there? Just as almost everyone has a
Java2 version of the JDK, doesn't almost everyone have 1.3+ ?

Don't jump on me at once: I know not everyone has. However, I don't know
how large that divide is, nor do I know how much benefit requiring 1.3+
would give us. I just think it's worth asking the question *now* rather
than waiting until there's a situation which would be easier to handle
in 1.3.

I suspect the answer is that there's not a lot that 1.3 gives us that
1.2 doesn't, in terms of API. Here's what the 1.3 "Summary of New
Features" list has:

o JNDI (probably irrelevant to the core)
o RMI (ditto)
o Various client-only things such as Swing, AWT, Java Sound etc
enhancements
o Security (probably not an issue?)
o Networking (probably irrelevant to the core)
o Reflection enhancements (dynamic proxy classes - these have always
struck me as cool, but I've never actually used them. I don't know how
to predict whether or not they'd be useful to Ant.)
o Serialization (I try to avoid this in general, and I suspect Ant
doesn't need to use it)
o Collections framework enhancements - nothing huge, as far as I can see
o Enhancements in java.{util,math,lang} - as far as I can see, only
java.util.Timer is likely to be particularly useful to Ant, and can be
reasonably easily implemented by us anyway, I'm sure.
o JPDA (Java Platform Debugger Architecture) - More likely to be useful
for extra tasks, I'd have thought (I hadn't considered using Ant in the
context of a debugger before - might think about that some more...)
o Tool and performance changes are irrelevant to the core of Ant, I
suspect
--
To unsubscribe, e-mail:   <mailto:ant-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:ant-dev-help@jakarta.apache.org>



--
To unsubscribe, e-mail:   <mailto:ant-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:ant-dev-help@jakarta.apache.org>


Mime
View raw message