On 3/20/14 3:58 PM, Petr Pchelko wrote:
Hello, Sergey.
Class files:
KeyEvent: 21018 bytes -> 21027 bytes
InputEvent: 5236 bytes -> 5240 bytes
MouseEvent: 6882 bytes -> 6921 bytes
TimerQueue: 5176 bytes -> 5968 bytes
The changes are insignificant. Only TimerQueue have grown a bit.
Performance:
KeyEvent.toString ~197 ops/ms -> ~195 ops/ms
Results for other event classes are the same.
TimerQueue: I’ve made a piece of similar code, and the Streams code works
around 5% slower then old-style code.
But the new-style looks so much cleaner and better)
But probably it could be faster as well? I think it will be better to
improve it a little bit more at least 15-20%.
With best regards. Petr.
20 марта 2014 г., в 3:17 после полудня, Sergey Bylokhov
<sergey.bylok...@oracle.com> написал(а):
Hi, Petr.
It will be good to know how performance is affected(is new code faster or
slower). Same for generated size of byte code.
On 3/20/14 11:57 AM, Petr Pchelko wrote:
Hello, AWT team.
Please review a simple cleanup fix for the issue:
https://bugs.openjdk.java.net/browse/JDK-8037912
The fix is available at: http://cr.openjdk.java.net/~pchelko/9/8037912/webrev/
Recently I’ve found out that a new StringJoiner class was added to JDK 8 and
wanted to try that out.
With best regards. Petr.
--
Best regards, Sergey.
--
Best regards, Sergey.
