Thanks for the comments, Petr. I'm OK with the fix.
--
best regards,
Anthony
On 12/06/2013 04:46 PM, Petr Pchelko wrote:
Hello, Anthony.
1618 // common practice (Wine, SWT) seems to be to
Is that Wine - the www.winehq.org ? Or is it a typo?
This comment was there for ages, so I can't answer this question.
Does it make sense to compile two lists for the if/else statements: one before
JDK-7075105, and the other one - the current one, and compare them to make sure
this is the last regression of this kind?
This make sense and I've tried to do this. The list of the clauses is the same,
but they are separated to translateBytes/translateStream. The problem is that
some of the if-else clauses are valid only for bytes
and some only for stream and this seem to be correct. I can't predict what
will happen if I accidentally add an extra if-else clause to the place where it
does not belong. We are going to native it this clauses,
so I can easily introduce a crash. I think it's too risky to try fixing it now.
Early in JDK 9 I'm going to rewrite this code and we could backport it to 8u20.
With best regards. Petr.
On 06.12.2013, at 16:25, Anthony Petrov <[email protected]> wrote:
Hi Petr,
The fix looks OK. A couple of remarks:
1.
1618 // common practice (Wine, SWT) seems to be to
Is that Wine - the www.winehq.org ? Or is it a typo?
2. Does it make sense to compile two lists for the if/else statements: one
before JDK-7075105, and the other one - the current one, and compare them to
make sure this is the last regression of this kind?
--
best regards,
Anthony
On 12/06/2013 04:17 PM, Petr Pchelko wrote:
Hello, AWT Team.
Please review the fix for the issue:
https://bugs.openjdk.java.net/browse/JDK-8029565
The fix is available at:
http://cr.openjdk.java.net/~pchelko/8029565/webrev/
This is a yet another regression of the JDK-7075105 and a yet another
copy/paste fix.
I'm going to file a bug to perform a major refactoring of the DataTransferer
code early in JDK 9 as now it's a mess.
With best regards. Petr.
