二、Context

    1.Supply/Demand vs. SELECT

    三、Internal Tables

    1. Using explicit work areas（显示的使用工作区）

    显示的指定工作区可以避免不必要的Move操作。见下列内表操作的语句：

    APPEND wa TO itab.
    INSERT wa INTO itab.
    COLLECT wa INTO itab.
    MODIFY itab FROM wa.
    READ TABLE itab INTO wa.
    LOOP AT itab INTO wa.

    ITAB = WA.

    APPEND ITAB.

    2.Linear search vs. binary search

    如果内表的数据超过20条，由于线性检索会遍历整个内表，这将会非常耗时。将内表排序并使用Binary Search，或者使用SORTED TABLE类型的内表。如果内表有n条记录，线性查找的次数为O( n )，Binary Search的查找次数为O( log2( n ) ).

    READ TABLE ITAB INTO WA WITH KEY K = 'X'

                    BINARY SEARCH.

    READ TABLE ITAB INTO WA

                    WITH KEY K = 'X'.

    3.Dynamic vs. static key access

    动态键值的读取比静态键值的读取要慢，

    READ TABLE ITAB INTO WA

                    WITH KEY K = 'X'.

    READ TABLE ITAB INTO WA

                    WITH KEY (NAME) = 'X'.

    4.Secondary indices

    If you need to access an internal table with different keys repeatedly, keep your own secondary indices. With a secondary index, you can replace a linear search with a binary search plus an index access.

    READ TABLE SEC_IDX INTO  SEC_IDX_WA

         WITH KEY DATE = SY-DATUM

              BINARY SEARCH.

    IF SY-SUBRC = 0.

      READ TABLE ITAB INTO WA

                      INDEX SEC_IDX_WA-INDX.

      " ...

    ENDIF.

    
    READ TABLE ITAB INTO WA

         WITH KEY DATE = SY-DATUM.

    IF SY-SUBRC = 0.

      " ...

    ENDIF.

    5.Key access to multiple lines

    LOOP ... WHERE比 LOOP/CHECK更快，因为LOOP ... WHERE只处理满足特定条件的数据。如果LOOP ... WHERE和FROM i1 and/or TO i2条件一起使用，性能会更好。

    LOOP AT ITAB INTO WA WHERE K = 'X'.

      " ...

    ENDLOOP.
   

    LOOP AT ITAB INTO WA.

      CHECK WA-K = 'X'.

      " ...

    ENDLOOP.