[
https://issues.apache.org/jira/browse/FLINK-9712?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16630330#comment-16630330
]
ASF GitHub Bot commented on FLINK-9712:
---------------------------------------
twalthr commented on a change in pull request #6741: [FLINK-9712][table,docs]
Document processing time Temporal Table Joins
URL: https://github.com/apache/flink/pull/6741#discussion_r220891980
##########
File path: docs/dev/table/streaming/joins.md
##########
@@ -0,0 +1,93 @@
+---
+title: "Joins"
+nav-parent_id: streaming_tableapi
+nav-pos: 3
+---
+<!--
+Licensed to the Apache Software Foundation (ASF) under one
+or more contributor license agreements. See the NOTICE file
+distributed with this work for additional information
+regarding copyright ownership. The ASF licenses this file
+to you under the Apache License, Version 2.0 (the
+"License"); you may not use this file except in compliance
+with the License. You may obtain a copy of the License at
+
+ http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing,
+software distributed under the License is distributed on an
+"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+KIND, either express or implied. See the License for the
+specific language governing permissions and limitations
+under the License.
+-->
+
+In batch processing joins are relatively easy, since we are working on a
bounded completed data sets.
+In stream processing things are a little bit more complicated, especially when
it comes to the issue how to handle that data can change over time.
+Because of that, there are couple of ways to actually perform the join using
either Table API or SQL.
+
+For more information regarding syntax please check Joins sections in [Table
API](../tableApi.html#joins) and [SQL](../sql.html#joins).
+
+* This will be replaced by the TOC
+{:toc}
+
+Regular Joins
+-------------
+
+This is the most basic case in which any new records or changes to either side
of the join input are visible and are affecting whole join result.
+If there is a new record on the left side, it will be joined with all of the
previous and future records on the other side.
+
+Such semantic has an important limitation:
+it requires to keep both sides of the join input on the state indefinitely and
resource usage will grow indefinitely as well.
+
+Example:
+{% highlight sql %}
+SELECT * FROM Orders
+INNER JOIN Product
+ON Orders.productId = Product.id
+{% endhighlight %}
+
+Time-windowed Joins
+-------------------
+
+In this case we are restricting scope of the join to some time window.
+This allows Flink to remove old values from the state (using
[watermarks](time_attributes.html) without affecting the correctness of the
result.
+
+Example:
+{% highlight sql %}
+SELECT *
+FROM
+ Orders o,
+ Shipments s
+WHERE o.id = s.orderId AND
+ o.ordertime BETWEEN s.shiptime - INTERVAL '4' HOUR AND s.shiptime
+{% endhighlight %}
+
+Temporal Table Joins
+--------------------
+
+Temporal Table Joins allow to join a stream (left/probe side) with a table
(right/build side) that changes over time.
+For each record from the probe side, it will be joined only with the latest
version of the build side.
+That means (in contrast to [Regular Joins](#regular-joins)) if there is a new
record on the build side,
+it will not affect the previous past results of the join.
+This again allow Flink to limit the number of elements that must be kept on
the state.
+In order to support updates (overwrites) of previous values on the build side
table, this table must define a primary key.
+
+Compared to [Time-windowed Joins](#time-windowed-joins),
+Temporal Table Joins are not defining a time window within which bounds the
records will be joined.
+Records from the probe side are joined with the most recent versions of the
build side and records on the build side might be arbitrary old.
+As time passes the previous, no longer needed, versions of the record (for the
given primary key) will be removed from the state.
+
+Example:
+{% highlight sql %}
+SELECT
+ o.amount * r.rate AS amount
+FROM
+ Orders AS o,
+ LATERAL TABLE (Rates(o.rowtime)) AS r
+WHERE r.currency = o.currency
+{% endhighlight %}
+
+For more information about this concept please check [Temporal
Tables](temporal_tables.html) page.
Review comment:
I miss the buzzword `stream enrichment` in this section. We should mention
that somewhere.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> Support enrichment joins in Flink SQL/Table API
> -----------------------------------------------
>
> Key: FLINK-9712
> URL: https://issues.apache.org/jira/browse/FLINK-9712
> Project: Flink
> Issue Type: New Feature
> Components: Table API & SQL
> Affects Versions: 1.5.0
> Reporter: Piotr Nowojski
> Assignee: Piotr Nowojski
> Priority: Major
> Labels: pull-request-available
>
> As described here:
> https://docs.google.com/document/d/1KaAkPZjWFeu-ffrC9FhYuxE6CIKsatHTTxyrxSBR8Sk/edit?usp=sharing
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
