说在前面的话
hive的正则表达式，是非常重要！作为大数据开发人员，用好hive，正则表达式，是必须品！
Hive中的正则表达式还是很强大的。数据工作者平时也离不开正则表达式。对此，特意做了个hive正则表达式的小结。所有代码都经过亲测，正常运行。
1.regexp
语法: A REGEXP B
操作类型: strings
描述: 功能与RLIKE相同
select count(*) from olap_b_dw_hotelorder_f where create_date_wid not regexp '\\d{8}'
与下面查询的效果是等效的：
select count(*) from olap_b_dw_hotelorder_f where create_date_wid not rlike '\\d{8}';
2.regexp_extract
语法: regexp_extract(string subject, string pattern, int index)
返回值: string
说明：将字符串subject按照pattern正则表达式的规则拆分，返回index指定的字符。
hive>selectregexp_extract('IloveYou','I(.*?)(You)',1) from test1 limit1;
Total jobs =1
...
Total MapReduce CPU Time Spent:7 seconds340 msec
OK
love
Time taken:28.067 seconds, Fetched:1 row(s)
hive>selectregexp_extract('IloveYou','I(.*?)(You)',2) from test1 limit1;
Total jobs =1
...
OK
You
Time taken:26.067 seconds, Fetched:1 row(s)
hive>select regexp_extract('IloveYou','(I)(.*?)(You)',1) from test1 limit1;
Total jobs =1
...
OK
I
Time taken:26.057 seconds, Fetched:1 row(s)
hive>select regexp_extract('IloveYou','(I)(.*?)(You)',0) from test1 limit1;
Total jobs =1
...
OK
IloveYou
Time taken:28.06 seconds, Fetched:1 row(s)
hive>select regexp_replace("IloveYou","You","") from test1 limit1;
Total jobs =1
...
OK
Ilove
Time taken:26.063 seconds, Fetched:1 row(s)
3.regexp_replace
语法: regexp_replace(string A, string B, string C)
返回值: string
说明：将字符串A中的符合Java正则表达式B的部分替换为C。注意，在有些情况下要使用转义字符,类似Oracle中的regexp_replace函数。
hive> selectregexp_replace("IloveYou","You","") from test1 limit1;
Total jobs =1
...
OK
Ilove
Time taken:26.063 seconds, Fetched:1 row(s)
hive>selectregexp_replace("IloveYou","You","lili") from test1 limit1;
Total jobs =1
...
OK
Ilovelili
Hive里的正则表达式
如，/confluence/display/Hive/GettingStarted
输入regex可查到
CREATE TABLE apachelog (
host STRING,
identity STRING,
user STRING,
time STRING,
request STRING,
status STRING,
size STRING,
referer STRING,
agent STRING)
ROW FORMAT SERDE 'org.apache.hadoop.hive.serde2.RegexSerDe'
WITH SERDEPROPERTIES (
"input.regex" = "([^ ]*) ([^ ]*) ([^]*) \[()\] ([^ \"]*|\"[^\"]*\") (-|[0-9]*) (-|[0-9]*)(?: ([^ \"]*|\".*\") ([^ \"]*|\".*\"))?"
)
STORED AS TEXTFILE;
下面就是hive里的正则表达式，9个字段，对应定义那边也要9个
"input.regex" = "([^ ]*) ([^ ]*) ([^.]*) \[(.*)\] "(.*)" (-|[0-9]*) (-|[(0-9]*) "(.*)" "(.*)""
([^ ]*) ([^ ]*) ([^.]*) \[(.*)\] "(.*)" (-|[0-9]*) (-|[(0-9]*) "(.*)" "(.*)"
([^ ]*) ([^ ]*) ([^.]*) \\[(.*)\\] "(.*)" (-|[0-9]*) (-|[(0-9]*) \"(.*)\" \"(.*)\"
数据来源，
yarn-root-nodemanager-master.log
或
yarn-spark-nodemanager-master.log
yarn-hadoop-nodemanager-master.log
这里，有个正则表达式的好工具！
RegexBuddy.exe
很好用的这款软件！双击它即可。
如上图所示颜色，代表我们测试的正则表达式，是正确的！
本文转自大数据躺过的坑博客园博客，原文链接：/zlslch/p/6102789.html，如需转载请自行联系原作者
