[
https://issues.apache.org/jira/browse/FLINK-28808?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Chesnay Schepler updated FLINK-28808:
-------------------------------------
Description:
Since we shouldn't assume that jackson mappers are serializable (due to
non-serializable extension) the RowDataToCsvFormatConverterContext also
shouldn't be serializable.
With that in mind the CsvFileFormatFactory should create the context when the
writer is created, not beforehand.
was:
Since we shouldn't assume that jackson mappers are serializable (due to
non-serializable extension) the RowDataToCsvFormatConverterContext also
shouldn't be serializable.
With that in mind the CsvFileFormatFactory should create the context when the
writer is created, not beforehand.
> CsvFileFormatFactory#createEncodingFormat should create ConverterContext on
> server-side
> ---------------------------------------------------------------------------------------
>
> Key: FLINK-28808
> URL: https://issues.apache.org/jira/browse/FLINK-28808
> Project: Flink
> Issue Type: Technical Debt
> Components: Formats (JSON, Avro, Parquet, ORC, SequenceFile)
> Reporter: Chesnay Schepler
> Assignee: Chesnay Schepler
> Priority: Major
> Fix For: 1.16.0
>
>
> Since we shouldn't assume that jackson mappers are serializable (due to
> non-serializable extension) the RowDataToCsvFormatConverterContext also
> shouldn't be serializable.
> With that in mind the CsvFileFormatFactory should create the context when the
> writer is created, not beforehand.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
