Hi Otávio,
Is there anyway you can experiment with publishing a webrev to dropbox or
github, preferably so that it is browsable, otherwise the zip?
Unfortunately the patches are hard to review, especially in response to reviews
where code is updated.
I realize until you have an openjdk
On Apr 27, 2014, at 2:22 PM, Claes Redestad claes.redes...@oracle.com wrote:
Possibly a few bytes in static class footprint, sure. Maybe this is something
javac should optimize (javap on some trivial examples suggests this doesn't
happen) rather than trying to root out all suboptimal cases,
On Apr 25, 2014, at 7:31 PM, Peter Levart peter.lev...@gmail.com wrote:
Hi Paul,
To make any sense of null return from the mergeFunction, which could mean
that an entry with that key is absent from resulting map, such Collector
would have to keep some more state - the collecting map (which
On 28/04/2014 1:05 PM, Otávio Gonçalves de Santana wrote:
In my opinion not, because Objects.requireNonNull is more readable than
just string.toString. This way is more understandable which field is
required and doesn't impact on performance.
An invocation of requireNonNull is potentially more
On 04/28/2014 10:20 AM, Paul Sandoz wrote:
On Apr 25, 2014, at 7:31 PM, Peter Levart peter.lev...@gmail.com wrote:
Hi Paul,
To make any sense of null return from the mergeFunction, which could mean that
an entry with that key is absent from resulting map, such Collector would have
to keep
On Apr 28, 2014, at 12:30 PM, Peter Levart peter.lev...@gmail.com wrote:
You could use the null return from Map.merge() as a signal to throw NPE, but
this is only 100% safe in to*Map() methods that don't take a mapSupplier.
Collections.toConcurrentMap(, SupplierM mapSupplier) could be
Yes you are right, but I mean it doesn't too much expensive enough to don't
use the requireNonNull.
IMHO.
On Mon, Apr 28, 2014 at 3:57 AM, David Holmes david.hol...@oracle.comwrote:
On 28/04/2014 1:05 PM, Otávio Gonçalves de Santana wrote:
In my opinion not, because Objects.requireNonNull is
Hi,
this is review request for: 8040754: Break the circular dependency
between SAAJ and JAXB
JBS: https://bugs.openjdk.java.net/browse/JDK-8040754
WEBREV: http://cr.openjdk.java.net/~mkos/8040754/jaxws.01/
It is basically moving classes in order to fix dependency between SAAJ,
JAX-B and
Done.
Link:
https://dl.dropboxusercontent.com/u/16109193/open_jdk/string_to_string_uncessary.zip
On Mon, Apr 28, 2014 at 4:34 AM, Paul Sandoz paul.san...@oracle.com wrote:
Hi Otávio,
Is there anyway you can experiment with publishing a webrev to dropbox or
github, preferably so that it is
link:
https://dl.dropboxusercontent.com/u/16109193/open_jdk/index_of_with_char.zip
On Mon, Apr 28, 2014 at 10:25 AM, Otávio Gonçalves de Santana
otavioj...@java.net wrote:
link:
https://dl.dropboxusercontent.com/u/16109193/open_jdk/index_of_with_char.zip
On Mon, Apr 28, 2014 at 4:42 AM,
I have to say that the expected NPE is quite hidden in the current code.
I am actually surprised someone was able to catch that. The use of
Objects.requreNonNull() is way more clear and I prefer that approach for
the sake of clarity.
On Mon, Apr 28, 2014 at 8:28 AM, Otávio Gonçalves de Santana
Miran,
This change looks good to me. Reviewed.
I applied the patch to the latest jdk9/dev and all tests pass as expected.
I see Mandy is sponsoring this for you.
-Chris.
On 28/04/14 13:01, Miroslav Kos wrote:
Hi,
this is review request for: 8040754: Break the circular dependency
between
On 24/04/2014 06:06, Zhong Yu wrote:
If a thread is created inside AccessController.doPrivileged(), it
seems to inherit AccessControlContext from the calling thread, is that
the expected/specified behavior?
It should inherit the acc from when it was created. If that context was
privileged by
On 04/28/2014 08:57 AM, David Holmes wrote:
On 28/04/2014 1:05 PM, Otávio Gonçalves de Santana wrote:
In my opinion not, because Objects.requireNonNull is more readable than
just string.toString. This way is more understandable which field is
required and doesn't impact on performance.
An
This follows from my posting of last week
http://mail.openjdk.java.net/pipermail/core-libs-dev/2014-April/026604.html
Firstly, I was informed that there are not supposed to be binary files in the
jdk9/jdk/test tree. My webrev listed in the above post includes one such file,
a GZIPped text file
Miran,
Looks good to me. I'll sponsor this patch for you.
Mandy
On 4/28/14 7:30 AM, Chris Hegarty wrote:
Miran,
This change looks good to me. Reviewed.
I applied the patch to the latest jdk9/dev and all tests pass as
expected.
I see Mandy is sponsoring this for you.
-Chris.
On
On 04/28/2014 07:53 PM, Brian Burkhalter wrote:
Meanwhile I suppose that I could convert my test file and use the
sun.misc.UU{En,De}coder classes to handle it in the test.
Uh-oh, we have java.util.Base64 these days.
But in this case, it's probably simpler to use this code snippet instead
On Apr 28, 2014, at 11:52 AM, Florian Weimer fwei...@redhat.com wrote:
On 04/28/2014 07:53 PM, Brian Burkhalter wrote:
Meanwhile I suppose that I could convert my test file and use the
sun.misc.UU{En,De}coder classes to handle it in the test.
Uh-oh, we have java.util.Base64 these days.
I am quite sure that you are right that these were deliberate so
I am adding Jim who was the author of all of the 2d ones to explain the
intent.
BTW Otávio, changes in 2D code like this should not slip by on the
core-libs list.
It ought to go to 2d-dev.
-phil.
On 4/28/2014 9:24 AM, Louis
For grins here’s an alternative implementation of your algorithm that is slower
but likely takes less memory:
private static BitSet primesBS(int n) {
int nbits = n + 1;
BitSet bs = new BitSet(nbits);
bs.set(0, 2, true);
for (int p = 2; p * p n;) {
We refer to these strings as constants in many places. If you have:
class Foo {
static final String bar = really long string;
}
class Blah {
// uses Foo.bar
}
Then Blah will incorporate that string by content, not by reference. I
don't know why this is done this way, but Strings are
Here is an updated patch
http://cr.openjdk.java.net/~bpb/8026236/webrev.02/
which has been revised to obviate the need for a file source of prime numbers.
Thanks,
Brian
On Apr 24, 2014, at 5:12 PM, Brian Burkhalter brian.burkhal...@oracle.com
wrote:
I have posted an updated patch here:
22 matches
Mail list logo
