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 4fca3bef [hotfix][doc] Document partial-update.ignore-delete
4fca3bef is described below
commit 4fca3bef71ef9068acd8f4afe00e795deebd0481
Author: JingsongLi <[email protected]>
AuthorDate: Mon Jan 9 14:26:54 2023 +0800
[hotfix][doc] Document partial-update.ignore-delete
---
docs/content/docs/features/table-types.md | 19 +++++++++++++++++--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/docs/content/docs/features/table-types.md
b/docs/content/docs/features/table-types.md
index 2826ab44..c2911e66 100644
--- a/docs/content/docs/features/table-types.md
+++ b/docs/content/docs/features/table-types.md
@@ -50,9 +50,20 @@ Specifically, if the latest record is a `DELETE` record, all
records with the sa
By specifying `'merge-engine' = 'partial-update'`, users can set columns of a
record across multiple updates and finally get a complete record. Specifically,
value fields are updated to the latest data one by one under the same primary
key, but null values are not overwritten.
-For example, let's say Table Store receives three records `<1, 23.0, 10,
NULL>`, `<1, NULL, NULL, 'This is a book'>` and `<1, 25.2, NULL, NULL>`, where
the first column is the primary key. The final result will be `<1, 25.2, 10,
'This is a book'>`.
+For example, let's say Table Store receives three records:
+- `<1, 23.0, 10, NULL>`-
+- `<1, NULL, NULL, 'This is a book'>`
+- `<1, 25.2, NULL, NULL>`
-NOTE: For streaming queries, `partial-update` merge engine must be used
together with `full-compaction` [changelog producer]({{< ref
"docs/features/table-types#changelog-producers" >}}).
+If the first column is the primary key. The final result will be `<1, 25.2,
10, 'This is a book'>`.
+
+{{< hint info >}}
+For streaming queries, `partial-update` merge engine must be used together
with `full-compaction` [changelog producer]({{< ref
"docs/features/table-types#changelog-producers" >}}).
+{{< /hint >}}
+
+{{< hint info >}}
+Partial cannot receive `DELETE` messages because the behavior cannot be
defined. You can configure `partial-update.ignore-delete` to ignore `DELETE`
messages.
+{{< /hint >}}
#### Aggregation
@@ -91,6 +102,10 @@ Current supported aggregate functions and data types are:
* `listagg`: supports STRING data type.
* `bool_and` / `bool_or`: support BOOLEAN data type.
+{{< hint info >}}
+For streaming queries, `aggregation` merge engine must be used together with
`full-compaction` [changelog producer]({{< ref
"docs/features/table-types#changelog-producers" >}}).
+{{< /hint >}}
+
### Changelog Producers
Streaming queries will continuously produce latest changes. These changes can
come from the underlying table files or from an [external log system]({{< ref
"docs/features/external-log-systems" >}}) like Kafka. Compared to the external
log system, changes from table files have lower cost but higher latency
(depending on how often snapshots are created).
