hive用户自定义函数

--查看系统自带函数
hive> show functions;
OK
!
!=
%
&
*
+
-
/
<
<=
<>
=
==
>
>=
^
abs
acos
and
array
array_contains
ascii
asin
atan
avg
bigint
bin
boolean
case
ceil
ceiling
coalesce
collect_set
concat
concat_ws
context_ngrams
conv
corr
cos
count
covar_pop
covar_samp
create_union
date_add
date_sub
datediff
day
dayofmonth
degrees
div
double
e
elt
exp
explode
field
find_in_set
float
floor
from_unixtime
get_json_object
hash
hex
histogram_numeric
hour
if
in
index
instr
int
isnotnull
isnull
json_tuple
lcase
length
like
ln
locate
log
log10
log2
lower
lpad
ltrim
map
max
min
minute
month
negative
ngrams
not
or
parse_url
parse_url_tuple
percentile
percentile_approx
pi
pmod
positive
pow
power
radians
rand
reflect
regexp
regexp_extract
regexp_replace
repeat
reverse
rlike
round
rpad
rtrim
second
sentences
sign
sin
size
smallint
space
split
sqrt
std
stddev
stddev_pop
stddev_samp
str_to_map
string
struct
substr
substring
sum
tan
tinyint
to_date
trim
ucase
unhex
unix_timestamp
upper
var_pop
var_samp
variance
weekofyear
when
xpath
xpath_boolean
xpath_double
xpath_float
xpath_int
xpath_long
xpath_number
xpath_short
xpath_string
year
|
~
Time taken: 0.134 seconds

--查看函数描述信息
hive> describe function abs;
OK
abs(x) - returns the absolute value of x
Time taken: 0.214 seconds

hive> describe function substring;
OK
substring(str, pos[, len]) - returns the substring of str that starts at pos and is of length len
Time taken: 0.084 seconds

package com.lpxuan.hive.udf;

import org.apache.hadoop.hive.ql.exec.UDF;
import org.apache.hadoop.io.Text;

public class CompanyLength extends UDF{
    public int getCompanyLength(String company_name){
        if(company_name == null) {
            return -1;
        }
        else return company_name.length();
    }
}

lpxuan@hadoop1:~/mycode/com/lpxuan/hive/udf$ javac CompanyLength.java
lpxuan@hadoop1:~/mycode/com/lpxuan/hive/udf$ ls
CompanyLength.class  CompanyLength.java
lpxuan@hadoop1:~/mycode/com/lpxuan/hive/udf$ jar cvf CompanyLength.jar CompanyLength.class
标明清单(manifest)
增加：CompanyLength.class(读入= 392) (写出= 277)(压缩了 29%)
lpxuan@hadoop1:~/mycode/com/lpxuan/hive/udf$ ls
CompanyLength.class  CompanyLength.jar  CompanyLength.java

root@hadoop1:/home/lpxuan/mycode/com/lpxuan/hive/udf#  mv CompanyLength.jar /opt/hadoop/mytest/

hive> add jar /opt/hadoop/mytest/CompanyLength.jar
/hadoop/mytest/CompanyLength.jar to class path
Added resource: /opt/hadoop/mytest/CompanyLength.jar

create temporary function company_length as 'com.lpxuan.hive.udf.CompanyLength.getCompanyLength';