[
https://issues.apache.org/jira/browse/FLINK-20174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17233755#comment-17233755
]
Steven Zhen Wu edited comment on FLINK-20174 at 11/17/20, 5:23 PM:
-------------------------------------------------------------------
I tried the change of setting up InputFiles map for DeleteFilter per
FileScanTask. I was wrong earlier. `TestFlinkInputFormatReaderDeletes` works
fine after [the change|https://github.com/stevenzwu/iceberg/pull/4].
So I guess the only reason left is the throughput benefit from combining small
files into a decent sized CombinedScanTask, which is still an important reason.
was (Author: stevenz3wu):
I tried the change of setting up InputFiles map for DeleteFilter per
FileScanTask. I was wrong earlier. It actually works fine. You can see the
change here.
https://github.com/stevenzwu/iceberg/pull/4
So I guess the only reason left is the throughput benefit from combining small
files into a decent sized CombinedScanTask, which is still an important reason.
> Make BulkFormat more extensible
> -------------------------------
>
> Key: FLINK-20174
> URL: https://issues.apache.org/jira/browse/FLINK-20174
> Project: Flink
> Issue Type: Improvement
> Components: Connectors / FileSystem
> Affects Versions: 1.12.0
> Reporter: Steven Zhen Wu
> Priority: Major
>
> Right now, BulkFormat has the generic `SpitT` type extending from
> `FileSourceSplit`. We can make BulkFormat taking the generic `SplitT` type
> extending from `SourceSplit`. This way, IcebergSourceSplit doesn't have to
> extend from `FileSourceSplit` and Iceberg source can reuse this BulkFormat
> interface as [~lzljs3620320] suggested. This allows Iceberg source to take
> advantages high-performant `ParquetVectorizedInputFormat` provided by Flink.
> [~sewen] [~lzljs3620320] if you are onboard with the change, I would be happy
> to submit a PR. Since it is a breaking change, maybe we can only add it to
> master branch after 1.12 release branch is cut?
> The other related question is the two `createReader` and `restoreReader`
> APIs. I understand the motivation. I am just wondering if the separation is
> necessary. if the SplitT has the CheckpointedLocation, the seek operation can
> be handled internal to `createReader`. We can also define an abstract
> `FileSourceSplitBase` that adds a `getCheckpointedPosition` API to the
> `SourceSplit`.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
