输入数据为
+----------------------+--------------------------------+
| movie_name | Genres |
+----------------------+--------------------------------+
| digimon | Adventure|Animation|Children's |
| Slumber_Party_Massac | Horror |
+----------------------+--------------------------------+
我需要像
这样的输出+----------------------+--------------------------------+-----------------+
| movie_name | Genres | count_of_genres |
+----------------------+--------------------------------+-----------------+
| digimon | Adventure|Animation|Children's | 3 |
| Slumber_Party_Massac | Horror | 1 |
+----------------------+--------------------------------+-----------------+
最佳答案
select *
,size(split(coalesce(Genres,''),'[^|\\s]+'))-1 as count_of_genres
from mytable
该解决方案涵盖不同的用例，包括 -
- NULL 值
- 空字符串
- 空标记(例如
Adventure||Animation或Adventure| |Animation)
关于hadoop - 如何使用配置单元计算由 "|"分隔符分隔的每列中的单词数？，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/43573144/