Approved for 8u-dev and 7u-dev.
regards,
Sean.
On 26/09/2014 11:28, Anton Nashatyrev wrote:
Hello,
please approve the backport to 8u-dev and 7u-dev (the patch is
absolutely the same for 8u-dev and 7u-dev)
JBS: https://bugs.openjdk.java.net/browse/JDK-8058473
JDK9 changeset:
http://hg.openjdk.java.net/jdk9/client/jdk/rev/c87d6fdf8376
Backport to 8u:
http://cr.openjdk.java.net/%7Eanashaty/8058473/8/webrev.00/
Review thread:
http://mail.openjdk.java.net/pipermail/awt-dev/2014-September/008516.html
Thanks!
Anton.
Hello Anton,
the backport looks fine.
Thanks,
Andrew
On 9/25/2014 4:16 PM, anton nashatyrev wrote:
Hi Sergey, Andrew,
could you please additionally review the fix backport to 8u-dev
(the fixed util method as been moved to another class) ?
JDK9 fix: http://cr.openjdk.java.net/%7Eanashaty/8058473/9/webrev.00/
Backport to 8u:
http://cr.openjdk.java.net/%7Eanashaty/8058473/8/webrev.00/
Thanks!
Anton.
On 24.09.2014 18:15, Andrew Brygin wrote:
Hello Anton,
the fix looks fine to me.
Thanks,
Andrew
On 9/24/2014 1:51 PM, anton nashatyrev wrote:
Sergey, thanks for review!
Any other volunteer?
On 19.09.2014 17:27, Sergey Bylokhov wrote:
Hi, Anton.
Then the fix looks good, thanks for clarification.
On 19.09.2014 17:25, anton nashatyrev wrote:
We don't treat the 'text/plain; class=java.lang.Object' as
TextType (DataFlavor.isFlavorTextType()) since it has unknown
representation class. Such flavors fall into 'other' category.
On 19.09.2014 17:16, Sergey Bylokhov wrote:
Hi, Anton.
Why "unknown/flavor" is before "text/plain" in the fixed version?
text/plain; class=java.io.Serializable; charset=unicode
text/html; class=java.io.Serializable; charset=unicode
unknown/flavor; class=java.io.InputStream
unknown/flavor; class=java.lang.Object
text/unknown; class=java.lang.Object; charset=unicode
text/unknown; class=java.lang.Object
text/plain; class=java.lang.Object; charset=unicode
text/plain; class=java.lang.Object
On 19.09.2014 15:58, anton nashatyrev wrote:
Hi Sergey,
On 18.09.2014 17:30, Sergey Bylokhov wrote:
- Can you try to change sort TimSort algorithm to mergerSort.
What the order will be in this case ?(see JDK-8048887).
The mergeSort seems to be more relaxed regarding the Comparator
correctness (BTW TimSort also doesn't fail each time - this
depends on initial elements positions). The order of course is
not correct with the incorrect comparator:
MergeSort with current comparator Merge/TimSort with fixed
Comparator
---------------------------------
-----------------------------------
application/unknown; class=java.io.InputStream
application/unknown; class=java.io.InputStream
application/unknown; class=java.io.InputStream
application/unknown; class=java.io.InputStream
application/unknown; class=java.lang.Object
application/unknown; class=java.lang.Object
application/x-java-jvm-local-objectref;
class=java.io.InputStream
application/x-java-jvm-local-objectref; class=java.io.InputStream
application/x-java-jvm-local-objectref;
class=java.io.InputStream
application/x-java-jvm-local-objectref; class=java.io.InputStream
application/x-java-jvm-local-objectref; class=java.lang.Object
application/x-java-jvm-local-objectref; class=java.lang.Object
text/unknown; class=java.io.Serializable;
charset=unicode text/html; class=java.io.Reader
text/plain; class=java.io.Serializable;
charset=unicode text/html; class=java.lang.String
text/html; class=java.io.Serializable;
charset=unicode text/html;
class=java.io.InputStream; charset=unicode
unknown/flavor; class=java.io.InputStream text/html;
class=java.io.InputStream; charset=unicode
unknown/flavor; class=java.io.InputStream text/html;
class=java.io.InputStream
text/html; class=java.io.Reader text/html;
class=java.io.InputStream; charset=cp1251
text/html; class=java.lang.String text/plain;
class=java.io.Reader
text/html; class=java.io.InputStream;
charset=unicode text/plain; class=java.lang.String
text/html; class=java.io.InputStream;
charset=unicode text/plain;
class=java.io.InputStream; charset=unicode
text/html; class=java.io.InputStream text/plain;
class=java.io.InputStream; charset=unicode
text/html; class=java.io.InputStream;
charset=cp1251 text/plain;
class=java.io.InputStream
text/plain; class=java.io.InputStream;
charset=unicode text/plain;
class=java.io.InputStream; charset=cp1251
text/plain; class=java.io.InputStream;
charset=unicode text/unknown; class=java.io.Reader
text/plain; class=java.io.InputStream text/unknown;
class=java.lang.String
text/plain; class=java.io.InputStream;
charset=cp1251 text/unknown;
class=java.io.InputStream; charset=unicode
text/unknown; class=java.io.InputStream;
charset=unicode text/unknown;
class=java.io.InputStream; charset=unicode
text/unknown; class=java.io.InputStream;
charset=cp1251 text/unknown;
class=java.io.InputStream
unknown/flavor; class=java.lang.Object text/unknown;
class=java.io.InputStream; charset=cp1251
text/unknown; class=java.lang.Object;
charset=unicode text/unknown;
class=java.io.Serializable; charset=unicode
text/unknown; class=java.io.Reader text/plain;
class=java.io.Serializable; charset=unicode
text/unknown; class=java.lang.String text/html;
class=java.io.Serializable; charset=unicode
text/unknown; class=java.io.InputStream;
charset=unicode unknown/flavor;
class=java.io.InputStream
text/unknown; class=java.io.InputStream unknown/flavor;
class=java.io.InputStream
text/unknown; class=java.lang.Object unknown/flavor;
class=java.lang.Object
text/plain; class=java.lang.String text/unknown;
class=java.lang.Object; charset=unicode
text/plain; class=java.lang.Object;
charset=unicode text/unknown;
class=java.lang.Object
text/plain; class=java.lang.Object text/plain;
class=java.lang.Object; charset=unicode
text/plain; class=java.io.Reader text/plain;
class=java.lang.Object
text/html; class=java.lang.Object;
charset=unicode text/html;
class=java.lang.Object; charset=unicode
text/html; class=java.lang.Object text/html;
class=java.lang.Object
- Can you run all related regression and jck tests.
Executed all 'datatransfer' regression and JCK tests - no
regressions detected.
Thanks.
On 18.09.2014 17:04, anton nashatyrev wrote:
Hello,
could you please review the following fix:
fix: http://cr.openjdk.java.net/~anashaty/8058473/9/webrev.00/
<http://cr.openjdk.java.net/%7Eanashaty/8058473/9/webrev.00/>
bug: https://bugs.openjdk.java.net/browse/JDK-8058473
Problem: DataFlavorComparator violates Comparator contract.
Fix: make the DataFlavorComparator transitive conformant
(please see the bug comment for details)
Thanks!
Anton.
