2019.12.05【ABAP随笔】 分组循环(LOOP AT Group) / REDUCE

ABAP 7.40新语法 LOOP AT Group 和 REDUCE

 *LOOP AT itab result [cond] GROUP BY key ( key1 = dobj1 key2 = dobj2 …
 *      [gs = GROUP SIZE] [gi = GROUP INDEX] )
 *      [ASCENDING|DESCENDING [AS TEXT]]
 *      [WITHOUT MEMBERS]
 *      [{INTO group}|{ASSIGNING <group>}]
 *      …
 *      [LOOP AT GROUP group|<group>
 *      …
 *      ENDLOOP.]
 *      …
 *ENDLOOP.
 *
 *… REDUCE type(
 *INIT result = start_value
 *           …
 *FOR for_exp1
 *FOR for_exp2
 *…
 *NEXT …
 *           result = iterated_value
 *… )

 "首先创建一个内表
 TYPES:BEGIN OF ty_data,
         id       TYPE i, "人员ID
         name     TYPE char10, "人员名称
         country  TYPE char10, "国家
         language TYPE char2, "语言
         age      TYPE i,
       END OF ty_data,
       ty_t_data TYPE STANDARD TABLE OF ty_data WITH EMPTY KEY.

 "数据
 DATA(gt_data) = VALUE ty_t_data(
 ( id =  name = 'Jerry' country = 'China' language = 'ZH' age =  )
 ( id =  name = 'Jack' country = 'China' language = 'ZH' age =  )
 ( id =  name = 'Nick' country = 'Korea' language = 'EN' age =  )
 ( id =  name = 'Rossi' country = 'Korea' language = 'EN' age =  )
 ( id =  name = 'Randy' country = 'Korea' language = 'EN' age =  )
 ( id =  name = 'Tab' country = 'China' language = 'ZH' age =  )
 ( id =  name = 'Lily' country = 'Korea' language = 'EN' age =  )
 ( id =  name = 'Lucy' country = 'China' language = 'EN' age =  )
 ( id =  name = 'Zera' country = 'China' language = 'EN' age =  )
 ( id =  name = 'Grace' country = 'China' language = 'EN' age =  )
 )
 .

 "REDUCE
 "1计算年龄最大
 DATA(lv_age_max_zh) = REDUCE i( INIT x =  FOR lw_data IN gt_data
                       WHERE ( language = 'ZH' ) NEXT x = nmax( val1 = x
                                                                val2 = lw_data-age )
                                                                ) .
 WRITE:/ |说中文的人中年龄最大的是：{ lv_age_max_zh  } |.

 "2.输出的reduce
 TYPES:outref TYPE REF TO if_demo_output.
 DATA(output) = REDUCE outref( INIT out = cl_demo_output=>new( )
                               text = 'Count up：'
                               FOR n =  UNTIL n >
                               NEXT out = out->write( text )
                               text = | { n } | ).
 output->display( ).

 "分组循环
 "1.ls_data这个工作区里面是没有内容的
 "2.<group>里面只有size index 和分组参数
 LOOP AT gt_data INTO DATA(ls_data) GROUP BY ( country = ls_data-country language = ls_data-language
   size = GROUP SIZE index = GROUP INDEX ) ASCENDING ASSIGNING FIELD-SYMBOL(<group>).

   WRITE:/ |Group:{ <group>-index } Country :{ <group>-country } language : { <group>-language }| &
   | Number lines :{ <group>-size } |.

   "3.按照<group>中的分组参数循环 GT_data中的数据
   LOOP AT GROUP <group> ASSIGNING FIELD-SYMBOL(<ls_member>).

     WRITE:/ | Name:{ <ls_member>-name } |.

   ENDLOOP.

   DATA(lv_age_max) = REDUCE i( INIT max =  FOR lw_member IN GROUP <group>
                              NEXT max = nmax( val1 = max val2 = lw_member-age ) ).

   DATA(lv_age_min) = REDUCE i( INIT min =  FOR lw_member IN GROUP <group>
                                NEXT min = nmin( val1 = min val2 = lw_member-age ) ).

   DATA(lv_age_sum) = REDUCE i( INIT sum =  FOR lw_member IN GROUP <group>
                                NEXT sum = sum + lw_member-age ).
   DATA(lv_age_avg) = lv_age_sum / <group>-size.

   WRITE:/ | 该组最大年龄，最小年龄和平均年龄分别为:{ lv_age_max } { lv_age_min } { lv_age_avg } |.
 ENDLOOP.

结果为:

个人感觉LOOP AT GROUP 和 REDUCE可以很好的代替 LOOP 中使用 AT END OF field 和 AT NEW field ( 这个是需要调整内表结构字段顺序的)。

平时用在分组求和，分组操作的时候还是很好用的。

-TAB 热爱技术 享受生活