Do we think this will be useful for users or do we first want to
introduce this for internal use cases, such as the Table API/SQL runner?
Aljoscha
On 15.10.20 10:35, Sean Z wrote:
Hi Jark,
Thanks for the reply and sharing thoughts.
Yes, negative long will make things complicated. We had the e
Hi Jark,
Thanks for the reply and sharing thoughts.
Yes, negative long will make things complicated. We had the exact same
issue when implementing our own sortedMapState.
This could be solved by some special pre-defined serializers, and it looks
like that's what blink did [1] as you mentioned.
Bl
Hi,
Thanks for bringing this discussion.
I think limiting the key type to Long can't resolve the comparison problem,
because the bytes order and value order of negative numbers is different.
Unless, we limit the key type to positive Long. But how to check this
before submitting a job?
In Blink c
Thanks for the reply! Look forward to learning more about this prototype.
Is there any way that we could track this TemporalState like Jira issue? or
should we start to create one in Jira? so anyone has interest like me,
could be part of the loop. Besides, is there any written docs/code about
the p
I'm very interested in this topic, and have even done some prototyping of
solution 1 -- limiting the key type to Long -- which Nico Kruber and I
called TemporalState in our prototype.
I look forward to sharing what we learned, and to discussing this further,
but I am completely overwhelmed with Fl
Hi devs,
Current DataStream API doesn't have SortedMapState supported. There are
lots of use cases based on sorted time-series data like range-query or
higher/lower key fetch, and ordered data seems like a nature of time-series
stream processing. Therefore, we propose to support the KeyedSortedMap
