Skip to content

Commit e72800d

Browse files
author
liuwenli
committed
update doc
1 parent 1003c2e commit e72800d

File tree

1 file changed

+9
-3
lines changed

1 file changed

+9
-3
lines changed

docs/user-guide.md

+9-3
Original file line numberDiff line numberDiff line change
@@ -294,7 +294,9 @@ Lookup SQL具体可参考Spark Flow Transformation的Lookup SQL章节
294294

295295
Flink SQL 用于处理 Source Namespace 数据,from 后面直接接表名即可。Flink支持window,UDF和UDAF操作
296296

297-
####### Window
297+
298+
299+
###### Window
298300

299301
process time处理方式中window中相应的字段名称为processing_time。例:SELECT name, SUM(key) as keysum from ums GROUP BY TUMBLE(processing_time, INTERVAL '1' HOUR), name;
300302

@@ -310,7 +312,9 @@ event time处理方式中window中相应的字段名称为ums_ts_。例:SELECT
310312

311313
在Transformation Config中可对这三个参数进行配置,配置格式为json。例如:{"min_idle_state_retention_time":"10","max_idle_state_retention_time":"20","preserve_message_flag":"true"}
312314

313-
###### #UDF
315+
316+
317+
###### UDF
314318

315319
Wormhole Flink UDF支持普通的java程序,而不需要按照Flink官方文档的格式实现UDF。UDF名称大小写敏感。UDF相应的字段需要使用as指定新字段的名称。例如:
316320

@@ -324,7 +328,9 @@ Java程序:
324328
使用UDF的Flink SQL:
325329

326330
select intvalue, fInt(intvalue) as fint from mytable;
327-
###### #UDAF
331+
332+
333+
###### UDAF
328334

329335
(1)使用UDAF需要进行以下操作
330336

0 commit comments

Comments
 (0)