Re: [10] RFR: 8175233: Remove LambdaForm.debugName

2017-02-21 Thread John Rose
On Feb 20, 2017, at 6:24 AM, Claes Redestad wrote: > > the LambdaForm.debugName field is useful for debugging, but names > are generated and retained in this field also for production code, which > is then used to name generated methods and classes. Reviewed. This is a good change; thank you.

Re: [10] RFR: 8175233: Remove LambdaForm.debugName

2017-02-21 Thread Paul Sandoz
> On 21 Feb 2017, at 16:19, Claes Redestad wrote: >> >> LambdaForm >> — >> >> You might wanna consider colocating "lambdaName” with “generateDebugName”, >> then it’s easier to see that the latter is called from within a synchronized >> block of the former. >> > > yes, updated in-place. >

Re: [10] RFR: 8175233: Remove LambdaForm.debugName

2017-02-21 Thread Claes Redestad
Hi Paul, On 2017-02-21 23:46, Paul Sandoz wrote: On 20 Feb 2017, at 06:24, Claes Redestad wrote: Hi, the LambdaForm.debugName field is useful for debugging, but names are generated and retained in this field also for production code, which is then used to name generated methods and classes.

Re: [10] RFR: 8175233: Remove LambdaForm.debugName

2017-02-21 Thread Paul Sandoz
> On 20 Feb 2017, at 06:24, Claes Redestad wrote: > > Hi, > > the LambdaForm.debugName field is useful for debugging, but names > are generated and retained in this field also for production code, which > is then used to name generated methods and classes. > > This patch suggests to extract th

[10] RFR: 8168664 [JAXP] XALAN: xsl:element generates namespace prefixes if namespace is given as URI only

2017-02-21 Thread Langer, Christoph
Hi Joe, please finally review this fix for JDK10: http://cr.openjdk.java.net/~clanger/webrevs/8168664-10.1/ This version contains the Java property "jdk.xml.generatePrefix" which we should have no matter on which default we decide. In my customer base I have customers that need it either way -

Re: SubmissionPublisher - Subscriber#onComplete() not invoked when publisher is closed

2017-02-21 Thread Pavel Bucek
Thanks for all responses. I understand how it works and it makes sense, but I believe the javadoc is not exact; SubsmissionPublisher#close doesn't mention any condition for the Subscriber#onComplete() invocation, but there obviously is one. Thanks again, Pavel On 21/02/2017 12:49, Doug Lea

Re: SubmissionPublisher - Subscriber#onComplete() not invoked when publisher is closed

2017-02-21 Thread Michael McMahon
Maybe the spec could be tighter around this, but it's not unreasonable that there is a delay in receiving onComplete() notification because of the subscriber controlled flow control. Notifying onError() is not subject to flow control; so you might expect that it would be triggered immediately.

Re: SubmissionPublisher - Subscriber#onComplete() not invoked when publisher is closed

2017-02-21 Thread Doug Lea
On 02/21/2017 06:36 AM, Pavel Rappo wrote: Only if you want an answer from the concurrency uber geeks :-) There seems to be no need for a further answer anyway! Thanks for pointing out that Subscription.request must be called to receive any items, and given this, the example works as expected.

Re: SubmissionPublisher - Subscriber#onComplete() not invoked when publisher is closed

2017-02-21 Thread Michael McMahon
On 21/02/2017, 11:15, Pavel Rappo wrote: I believe, the most appropriate place for concurrency-related questions is http://altair.cs.oswego.edu/mailman/listinfo/concurrency-interest As for the question itself. I don't think this behaviour is a bug. SubmissionPublisher.close() seems to be

Re: SubmissionPublisher - Subscriber#onComplete() not invoked when publisher is closed

2017-02-21 Thread Pavel Rappo
Only if you want an answer from the concurrency uber geeks :-) > On 21 Feb 2017, at 11:32, Pavel Bucek wrote: > > Thanks for the link to the other mailing list - do I understand it correctly > that I should move this thread there?

Re: SubmissionPublisher - Subscriber#onComplete() not invoked when publisher is closed

2017-02-21 Thread Pavel Bucek
SubmissionPublisher#closeExceptionally does trigger Subscriber#onError, but based on javadoc, I cannot really be sure that it will be called, since it contains exactly the same wording as SubmissionPublisher#close /** * Unless already closed, issues {@link * Flow.Subscriber#onError(Throwable)

Re: SubmissionPublisher - Subscriber#onComplete() not invoked when publisher is closed

2017-02-21 Thread Pavel Rappo
I believe, the most appropriate place for concurrency-related questions is http://altair.cs.oswego.edu/mailman/listinfo/concurrency-interest As for the question itself. I don't think this behaviour is a bug. SubmissionPublisher.close() seems to be a graceful way of shutting down (in contrast

Re: SubmissionPublisher - Subscriber#onComplete() not invoked when publisher is closed

2017-02-21 Thread Michael McMahon
Sounds like a bug. It seems like the fact there isn't a call to Subscription.request() is what causes the problem. But by my reading of the spec, Subscriber.onComplete() should still be called, as it is known that " no additional Subscriber method invocations will occur". - Michael. On 21/02/

Re: SubmissionPublisher - Subscriber#onComplete() not invoked when publisher is closed

2017-02-21 Thread Pavel Bucek
there is a formatting issue in the code snippet, publisher.close() should be on the new line: { SubmissionPublisher publisher =new SubmissionPublisher<>(); publisher.subscribe(new Flow.Subscriber() { @Override public void onSubscribe(Flow.Subscription subscription) { }

SubmissionPublisher - Subscriber#onComplete() not invoked when publisher is closed

2017-02-21 Thread Pavel Bucek
Hi all, firstly - please let me know if this is is a wrong place to send this; I wasn't able to find list specific to concurrency. Consider following example: { SubmissionPublisher publisher =new SubmissionPublisher<>(); publisher.subscribe(new Flow.Subscriber() { @Override pu