Fwd: Ramping up for 2.8

Sent to the wrong address initially, so I'm forwarding it now --M

---------- Forwarded message ----------
From: martin odersky
Date: Thu, Nov 12, 2009 at 7:37 PM
Subject: Ramping up for 2.8
To: Paul Phillips , David Pollak
, scala-internalsl@epfl.ch, Antonio Cunei

Hi David, all:

We are getting ready to stabilize 2.8 to a degree where you can do
some serious release testing.
If the nightlies build OK, then the only remaining issues as far as I
am concerned is

 - Treatment of hash, but that's irrelevant for porting, as the
previous code using hashCode is not more
  broken by our fixes than it ever was.

 - Treatment of equality between primitive numbers and BigInt,
BigDecimal. If that's an issue, you should
  wait for Paul's patches.

Everybody else, if you are sitting on a critical ticket where you
suspect core tools will not build because of that, now is the time to
speak up!

So if the above are non-issues, you should be able to do release
testing of 2.8 as of tomorrow.

One other thing: To decrease porting risk, we will turn off
optimization for the nightlies at the moment, until all major tools
build and run fine against 2.8. Once that's the case we will turn
optimization on and ask you to please double check against optimized
code. Anything that breaks in the second round would most likely be
the fault of the optimizer, and we would try to fix those issues, of
course. The hope is that we can then ship the public beta of 2.8 with
optimized compiler and core libraries.

Cheers