This is an automated email from the ASF dual-hosted git repository.
lzljs3620320 pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/flink-table-store.git
The following commit(s) were added to refs/heads/master by this push:
new cd282aa5 [hotfix] Document sequence.field need to be updated
cd282aa5 is described below
commit cd282aa59f7f31bd4f361acd9e2525043ad8811c
Author: JingsongLi <[email protected]>
AuthorDate: Fri Mar 3 13:51:12 2023 +0800
[hotfix] Document sequence.field need to be updated
---
docs/content/docs/concepts/primary-key-table.md | 15 +++++++++++----
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/docs/content/docs/concepts/primary-key-table.md
b/docs/content/docs/concepts/primary-key-table.md
index 8005f303..bb745258 100644
--- a/docs/content/docs/concepts/primary-key-table.md
+++ b/docs/content/docs/concepts/primary-key-table.md
@@ -36,6 +36,12 @@ By [defining primary keys]({{< ref
"docs/how-to/creating-tables#tables-with-prim
When Table Store sink receives two or more records with the same primary keys,
it will merge them into one record to keep primary keys unique. By specifying
the `merge-engine` table property, users can choose how records are merged
together.
+{{< hint info >}}
+Set `table.exec.sink.upsert-materialize` to `NONE` always in Flink SQL
TableConfig, sink upsert-materialize may
+result in strange behavior. When the input is out of order, we recommend that
you use
+[Sequence Field]({{< ref "docs/concepts/primary-key-table#sequence-field" >}})
to correct disorder.
+{{< /hint >}}
+
### Deduplicate
`deduplicate` merge engine is the default merge engine. Table Store will only
keep the latest record and throw away other records with the same primary keys.
@@ -159,6 +165,11 @@ Full compaction changelog producer can produce complete
changelog for any type o
By default, the primary key table determines the merge order according to the
input order (the last input record will be the last to merge). However, in
distributed computing,
there will be some cases that lead to data disorder. At this time, you can use
a time field as `sequence.field`, for example:
+{{< hint info >}}
+When the record is updated or deleted, the `sequence.field` must become larger
and cannot remain unchanged. For example,
+you can use [Mysql Binlog operation
time](https://ververica.github.io/flink-cdc-connectors/master/content/connectors/mysql-cdc.html#available-metadata)
as `sequence.field`.
+{{< /hint >}}
+
{{< tabs "sequence.field" >}}
{{< tab "Flink" >}}
@@ -179,7 +190,3 @@ CREATE TABLE MyTable (
{{< /tabs >}}
The record with the largest `sequence.field` value will be the last to merge,
regardless of the input order.
-
-{{< hint info >}}
-We recommend you set `sequence.field` to table to correct disorder. Set
`table.exec.sink.upsert-materialize` to `NONE` always to avoid any materialize
operator being added by Flink SQL, which may result in strange behavior.
-{{< /hint >}}
