Hi Aljoscha

thanks for the comment.
is wrapping by a PurgingTrigger.of() the same as doing "return
TriggerResult.FIRE_AND_PURGE;"
inside of a custom trigger?

gave it a test and the result seems the opposite of what I meant...
instead of throwing away previous windows' contents, I wanna keep them
all the way till the end.
that way I can get the cumulative counts of all input.

wonder how to achieve it.
anyone?

jad


On Fri, Jul 14, 2017 at 12:16 AM, Aljoscha Krettek <aljos...@apache.org>
wrote:

> Window contents are only purged from state if the Trigger says so or if
> the watermark passes the garbage collection horizon for a given window.
> With GlobalWindows, the GC horizon is never reached, that leaves Triggers.
>
> You can create a Trigger that purges every time it fires by wrapping it in
> a PurgingTrigger, i.e.
>
> .window(PurgingTrigger.of(<my trigger>))
>
> Best,
> Aljoscha
>
> On 13. Jul 2017, at 14:00, jad mad <jadmad0...@gmail.com> wrote:
>
> Hi Prashant,
>
> env.setStreamTimeCharacteristic(TimeCharacteristic.EventTime);
>
> actually I could make my custom trigger to fire periodically.
> The problem is the element set stored in the iterable variable
> is always uniform which is not what I'm expecting...
>
> private static class MyWindowFunction_Window...
>          ...
>        @Override
>         public void apply(Tuple tuple, W window, Iterable<MyClass>
> iterable,
>              ...
>              for(MyClass element : iterable)
>
> does anyone have any idea on this?
> thanks a lot in advance,
> jad
>
>
> On Thu, Jul 13, 2017 at 10:55 AM, prashantnayak <
> prash...@intellifylearning.com> wrote:
>
>> Hi
>>
>> We've have custom operators using global windows and are using event time.
>>
>> How are you specifying event time as the time characteristic?
>>
>> Prashant
>>
>>
>>
>> --
>> View this message in context: http://apache-flink-user-maili
>> ng-list-archive.2336050.n4.nabble.com/global-window-trigg
>> er-tp14206p14239.html
>> Sent from the Apache Flink User Mailing List archive. mailing list
>> archive at Nabble.com.
>>
>
>
>

Reply via email to