原帖：http://www.cnblogs.com/nayitian/p/3231734.html

wmsys.wm_concat

Definition: The Oracle PL/SQL WM_CONCAT function is used to aggregate data from a number of rows into a single row, giving a list of data associated with a specific COMMENT_BODY. In effect, it cross-tabulates a comma delimited list.

Note that WM_CONCAT is undocumented and unsupported by Oracle, meaning it should not be used in production systems. The LISTAGG function, which can produce the same output asWM_CONCAT is both documented and supported by Oracle.

1. 现有数据结构

BBSDETAIL表(主表)

DETAIL_ID NOT NULL NUMBER　　主键

TITLE NOT NULL VARCHAR2(100)

BBSCOMMENT表(从表)

DETAIL_ID NOT NULL NUMBER　　外键

COMMENT_BODY NOT NULL VARCHAR2(500)

COMMENT_TIME NOT NULL DATE

2. 实现功能一(从表多行记录合并为一行，不要求排序)

--多行合并为一行，不要求排序

select DETAIL_ID,wmsys.wm_concat('{'||COMMENT_BODY||'}')

from BBSCOMMENT

group by DETAIL_ID;

输出：

13859 {东西好，送货快 },{好 },{物流有些慢 }

14938 {卖家还是挺热心的，以后再来 },{东西不错 }

3. 实现功能二(从表多行记录合并为一行后，与主表做一连接)

--将上述SQL语句与主表做一个连接查询

select bd.DETAIL_ID,TITLE,bcm.COMMENT_INFO

from BBSDETAIL bd,(select DETAIL_ID,wmsys.wm_concat('{'||COMMENT_BODY||'}') as COMMENT_INFO from BBSCOMMENT group by DETAIL_ID) bcm

where bd.DETAIL_ID=bcm.DETAIL_ID(+);

输出：

13859 苏泊尔电压力锅配件 {东西好，送货快 },{好 },{物流有些慢 }

14938 Nike/耐克男性跑步鞋跑步 {卖家还是挺热心的，以后再来 },{东西不错 }

4. 实现功能三(从表多行记录合并为一行，并按评价时间排序)

--多行合并为一行，要求排序(最新的评论在前面)

select DETAIL_ID, max(r)

from (select DETAIL_ID, wmsys.wm_concat(COMMENT_BODY||'('||to_char(COMMENT_TIME,'yyyy-mm-dd hh:mi:ss')||')')

OVER(PARTITION BY DETAIL_ID ORDER BY COMMENT_TIME desc) r from BBSCOMMENT)

group by DETAIL_ID;

输出：

13859 东西好，送货快(2013-02-19 06:27:37),好(2012-01-14 02:23:46),物流有些慢(2012-01-01 12:00:25)

14938 卖家还是挺热心的，以后再来(2011-11-27 05:28:27),东西不错(2011-10-11 05:09:06)

5. 实现功能四(行变列：分两列显示从表两种汇总结果，排序，并保证两列中数据的对应关系)

--分两列显示两种汇总结果，并排序，保证对应关系

select DETAIL_ID,COMMENT_TIME,COMMENT_BODY from (

select

DETAIL_ID,

WMSYS.WM_CONCAT(to_char(COMMENT_TIME,'yyyy-mm-dd hh:mi:ss'))

OVER(PARTITION BY DETAIL_ID ORDER BY COMMENT_TIME) COMMENT_TIME,

WMSYS.WM_CONCAT('{'||COMMENT_BODY||'}')

OVER(PARTITION BY DETAIL_ID ORDER BY COMMENT_TIME) COMMENT_BODY,

row_number() OVER(PARTITION BY DETAIL_ID ORDER BY COMMENT_TIME desc) rs

from BBSCOMMENT) where rs=1;

输出：

13859 2013-02-19 06:27:37,2012-01-14 02:23:46,2012-01-01 12:00:25 {东西好，送货快},{好},{物流有些慢}

14938 2011-11-27 05:28:27,2011-10-11 05:09:06 {卖家还是挺热心的，以后再来},{东西不错}