I see, alright.
Thank you for all the help!
On Mon, Dec 28, 2020 at 3:57 PM Timo Walther wrote:
> Hi Yuval,
>
> FLIP-136 will resolve the remaining legacy type issues for DataStream
> API. After that we can drop legacy types because everything can be
> expressed with DataType. DataType is a sup
Hi Yuval,
FLIP-136 will resolve the remaining legacy type issues for DataStream
API. After that we can drop legacy types because everything can be
expressed with DataType. DataType is a super type of TypeInformation.
Registering types (e.g. giving a RAW type backed by io.circe.Json a name
su
Hi Timo,
I will look at the Catalog interface and see what it requires.
For the meanwhile I'll go back to the old API. Do you expect FLIP-136 to
resolve the issue around legacy types? Will it's implementation allow to
register LEGACY types? or a new variation of them?
On Mon, Dec 28, 2020 at 12:4
I would recommend to use the old UDF stack for now. You can simply call
`StreamTableEnvironment.registerFunction` instead of
`createTemporarySystemFunction`. Then the UDF returns a legacy type that
the DataStream API understands.
Have you thought about implementing your own catalog instead of
Timo, an additional question.
I am currently using TypeConversions.fromLegacyInfoToDataType. However,
this still does not allow me to register this table with the catalog since
any representation of LEGACY isn't supported (I don't see it generating any
other DataType other than RAW(.., LEGACY). Is
Hi Timo,
Thanks for the explanation. Passing around a byte array is not possible
since I need to know the concrete type later for serialization in my sink,
so I need to keep the type.
What I'm trying to achieve is the following:
I have a scalar UDF function:
[image: image.png]
This function is
Hi Yuval,
the legacy type has no string representation that can be used in a SQL
DDL statement. The current string representation LEGACY(...) is only a
temporary work around to persist the old types in catalogs.
Until FLIP-136 is fully implemented, toAppendStream/toRetractStream
support only
Hi Danny,
Yes, I tried implementing the DataTypeFactory for the UDF using
TypeInformationRawType (which is deprecated BTW, and there's no support for
RawType in the conversion), didn't help.
I did manage to get the conversion working using
TableEnvironment.toAppendStream (I was previously directl
> SQL parse failed. Encount
What syntax did you use ?
> TypeConversions.fromDataTypeToLegacyInfo cannot convert a plain RAW type
back to TypeInformation.
Did you try to construct type information by a new
fresh TypeInformationRawType ?
Yuval Itzchakov 于2020年12月24日周四 下午7:24写道:
> An expansion to
An expansion to my question:
What I really want is for the UDF to return `RAW(io.circe.Json, ?)` type,
but I have to do a conversion between Table and DataStream, and
TypeConversions.fromDataTypeToLegacyInfo cannot convert a plain RAW type
back to TypeInformation.
On Thu, Dec 24, 2020 at 12:59 PM
Hi,
I have a UDF which returns a type of MAP')>. When I try to register this type with Flink via the
CREATE TABLE DDL, I encounter an exception:
- SQL parse failed. Encountered "(" at line 2, column 256.
Was expecting one of:
"NOT" ...
"NULL" ...
">" ...
"MULTISET" ...
"ARRAY"
11 matches
Mail list logo