| ID | FIELD1 | FIELD2 | FIELD3 | FIELD4 | Key |
| 1 | *** | *** | *** | *** | meat1 |
| 2 | *** | *** | *** | *** | meat1 |
| 3 | *** | *** | *** | *** | meat1 |
| 4 | *** | *** | *** | *** | meat1 |
| 5 | *** | *** | *** | *** | fruit2 |
| 6 | *** | *** | *** | *** | fruit2 |
| 7 | *** | *** | *** | *** | fruit2 |
| 8 | *** | *** | *** | *** | fruit2 |
| 9 | *** | *** | *** | *** | fruit2 |
| 10 | *** | *** | *** | *** | food3 |
| 11 | *** | *** | *** | *** | food3 |
现在有如上所示的结果
要求：要对这个已经按key进行排序了的数组进行操作，相同key的项进行处理。
提示：这个是很典型的母子表的结构，也就是说其实它是两张表的合并，可以这样处理成两个数组，方便数组里面对块的操作
array1:ID|Key
| ID | Key |
| 1 | meat1 |
| 2 | meat1 |
| 3 | meat1 |
| 4 | meat1 |
| 5 | fruit2 |
| 6 | fruit2 |
| 7 | fruit2 |
| 8 | fruit2 |
| 9 | fruit2 |
| 10 | food3 |
| 11 | food3 |
array2:key => array(ID,FIELD1,FIELD2,FIELD3,FIELD4,FIELD5,Key)
| ID | FIELD1 | FIELD2 | FIELD3 | FIELD4 | Key | |
| meat1=> | 1 | *** | *** | *** | *** | meat1 |
| 2 | *** | *** | *** | *** | meat1 | |
| 3 | *** | *** | *** | *** | meat1 | |
| 4 | *** | *** | *** | *** |
