Thanks Timo 🙏
On Fri, Jun 4, 2021, 17:13 Timo Walther wrote:
> Hi Yuval,
>
> I would recommend option 2. Because esp. when it comes to state you
> should be in control what is persisted. There is no guarantee that the
> ExternalSerializer will not change in the future. It is only meant for
> shi
Hi Yuval,
I would recommend option 2. Because esp. when it comes to state you
should be in control what is persisted. There is no guarantee that the
ExternalSerializer will not change in the future. It is only meant for
shipping data as the input of the next operator.
I would recommend to wr
Hi Timo,
Thank you for the response.
The tables being created in reality are based on arbitrary SQL code such
that I don't know what the schema actually is to create the TypeInformation
"by hand" and pass it on to the DataStream API.
This leaves me with option 1, which leads to another question:
Hi Yuval,
TypeConversions.fromDataTypeToLegacyInfo was only a utility to bridge
between TypeInformation and DataType until TypeInformation is not
exposed through the Table API anymore.
Beginning from Flink 1.13 the Table API is able to serialize the records
to the first DataStream operator
When upgrading to Flink 1.13, I ran into deprecation warnings on
TypeConversions
[image: image.png]
The deprecation message states that this API will be deprecated soon, but
does not mention the alternatives that can be used for these
transformations.
My use case is that I have a table that need
