1.Lookup join
概念：Lookup join是针对于由作业流表触发，关联右侧维表来补全数据的场景 。
默认情况下，在流表有数据变更，都会触发维表查询（可以通过设置维表是否缓存，来减轻查询压力），由于不保存状态，因此对内存占用较小。（以上来自网络）
具体配置如下：
SET execution.checkpointing.interval=5000;
SET state.checkpoints.dir=hdfs://hadoop01:9000/flink/checkpoints/2023090510549999;
SET execution.runtime-mode=streaming;
// 定义维表，维表 可以有主键,不能有水位线字段
CREATE TEMPORARY TABLE source_dim_dept (
`id` BIGINT,
`dept` STRING,
PRIMARY KEY(id) NOT ENFORCED
) WITH (
'connector'='jdbc',
'url'='jdbc:mysql://1.1.1.1:3306/data_storage?autoReconnect=true&useUnicode=true&characterEncoding=utf-8&allowMultiQueries=true&serverTimezone=Asia/Shanghai',
'username'='data_manage',
'password'='xxxxxx',
'table-name'='source_dim_dept ',
'lookup.cache'='PARTIAL',
'lookup.partial-cache.expire-after-write'='30s',
'lookup.cache.ttl' = '30s',
'lookup.cache.max-rows'='5',
'lookup.partial-cache.max-rows'='5'
);
// 流表 -流表必须要有 PROCTIME(),和 水位线字段
CREATE TABLE source_kafka_stream_data(
`id` BIGINT,
`ts` BIGINT,
`price` FLOAT,
proc_time AS PROCTIME(),
`auto_row_time` AS cast(CURRENT_TIMESTAMP as timestamp(3)),
WATERMARK FOR `auto_row_time` AS `auto_row_time` - INTERVAL '0' SECONDS
) WITH (
'connector'='kafka',
'properties.group.id'='data_processing_producer',
'scan.startup.mode'='group-offsets',
'properties.auto.offset.reset'='latest',
'topic'='data-paimon-test',
'properties.bootstrap.servers'='1.1.1.1:9192',
'format'='json'
);
CREATE VIEW transform_tableJoin_XeH7VK1o2U AS
select `status`,`id`,`ts`,`price`, PROCTIME() as auto_row_time from (
select c.`dept` as `status`,c.`id` as `id`,O.`ts` as `ts`,O.`price` as `price`
from source_kafka_stream_data AS O
JOIN source_dim_dept FOR SYSTEM_TIME AS OF O.proc_time AS c
*注意坑在这里：上述的 连接中流表的时间字段一定要用 PROCTIME() 类型的 AS OF O.proc_time，如果 用水位线字段则Flink 会转为TemperalJoin
而不是Lookup join
// 以下是输出，无特殊配置
CREATE CATALOG paimon WITH (
'type' = 'paimon',
'warehouse' = 'hdfs://hadoop01:9000/painmon/data-processing/paimon_ods'
);
USE CATALOG paimon;
create database if not exists paimon.paimon_ods_db;
drop table if exists paimon_ods_db.paimon_test_stream_join;
CREATE TABLE if not exists paimon_ods_db.paimon_test_stream_join(
`uuid` STRING,
`status` STRING,
`id` BIGINT,
`ts` BIGINT,
`price` DOUBLE
) WITH (
'sink.parallelism'='8',
'bucket'='8',
'bucket-key'='uuid',
'write-mode'='append-only',
'sink.use-managed-memory-allocator'='true',
'sink.managed.writer-buffer-memory'='512MB',
'num-sorted-run.compaction-trigger'='20',
'write-buffer-size'='1024MB',
'write-buffer-spillable'='true'
);
INSERT INTO paimon_ods_db.paimon_test_stream_join select uuid(),`status`,`id`,`ts`,`price` from default_catalog.default_database.transform_tableJoin_XeH7VK1o2U;
2.Temporal join 时态表连接
时态表是一个随时间演变的表，在Flink中也称为动态表。
时态表中的行与一个或多个时态周期相关联，并且所有Flink表都是时态的（动态的）。时态表包含一个或多个版本化的表快照，它可以是跟踪更改的更改历史表（例如数据库更改日志，包含所有快照），也可以是具体化更改的维表（例如包含最新快照的数据库表）。
时态表可以分为版本表和普通表。
版本表：如果时态表中的记录可以追踪和并访问它的历史版本，这种表我们称之为版本表，来自数据库的 changelog （如mysql binlog）可以定义成版本表，版本表内的数据始终不会自动清理，只能通过upsert触发。
普通表：如果时态表中的记录仅仅可以追踪并和它的最新版本，这种表我们称之为普通表，来自数据库 或 HBase 、redis的表可以定义成普通表。
（以上来自网络）
SET execution.checkpointing.interval=5000;
SET state.checkpoints.dir=hdfs://hadoop01:9000/flink/checkpoints/2023090510549999;
SET execution.runtime-mode=streaming;
// 定义维表，维表一定要有主键和水位线字段
CREATE TEMPORARY TABLE source_dim_dept (
`id` BIGINT,
`dept` STRING,
`auto_row_time` AS cast(CURRENT_TIMESTAMP as timestamp(3)),
WATERMARK FOR `auto_row_time` AS `auto_row_time` - INTERVAL '0' SECONDS
PRIMARY KEY(id) NOT ENFORCED
) WITH (
'connector'='jdbc',
'url'='jdbc:mysql://1.1.1.1:3306/data_storage?autoReconnect=true&useUnicode=true&characterEncoding=utf-8&allowMultiQueries=true&serverTimezone=Asia/Shanghai',
'username'='data_manage',
'password'='xxxxx',
'table-name'='source_dim_dept ',
'lookup.cache'='PARTIAL',
'lookup.partial-cache.expire-after-write'='30s',
'lookup.cache.ttl' = '30s',
'lookup.cache.max-rows'='5',
'lookup.partial-cache.max-rows'='5'
);
// 流表 -流表要有 水位线字段
CREATE TABLE source_kafka_stream_data(
`id` BIGINT,
`ts` BIGINT,
`price` FLOAT,
proc_time AS PROCTIME(),
`auto_row_time` AS cast(CURRENT_TIMESTAMP as timestamp(3)),
WATERMARK FOR `auto_row_time` AS `auto_row_time` - INTERVAL '0' SECONDS
) WITH (
'connector'='kafka',
'properties.group.id'='data_processing_producer',
'scan.startup.mode'='group-offsets',
'properties.auto.offset.reset'='latest',
'topic'='data-paimon-test',
'properties.bootstrap.servers'='1.1.1.1:9192',
'format'='json'
);
CREATE VIEW transform_tableJoin_XeH7VK1o2U AS
select `status`,`id`,`ts`,`price`, PROCTIME() as auto_row_time from (
select c.`dept` as `status`,c.`id` as `id`,O.`ts` as `ts`,O.`price` as `price`
from source_kafka_stream_data AS O
JOIN source_dim_dept FOR SYSTEM_TIME AS OF O.auto_row_timeAS c
*注意坑在这里：上面的Lookup Join 区别也在这里，连接中流表的时间字段一定要用 水位线字段 类型的 AS OF O.auto_row_time
// 以下是输出，无特殊配置
CREATE CATALOG paimon WITH (
'type' = 'paimon',
'warehouse' = 'hdfs://hadoop01:9000/painmon/data-processing/paimon_ods'
);
USE CATALOG paimon;
create database if not exists paimon.paimon_ods_db;
drop table if exists paimon_ods_db.paimon_test_stream_join;
CREATE TABLE if not exists paimon_ods_db.paimon_test_stream_join(
`uuid` STRING,
`status` STRING,
`id` BIGINT,
`ts` BIGINT,
`price` DOUBLE
) WITH (
'sink.parallelism'='8',
'bucket'='8',
'bucket-key'='uuid',
'write-mode'='append-only',
'sink.use-managed-memory-allocator'='true',
'sink.managed.writer-buffer-memory'='512MB',
'num-sorted-run.compaction-trigger'='20',
'write-buffer-size'='1024MB',
'write-buffer-spillable'='true'
);
INSERT INTO paimon_ods_db.paimon_test_stream_join select uuid(),`status`,`id`,`ts`,`price` from default_catalog.default_database.transform_tableJoin_XeH7VK1o2U;