presto-third-functions

Introduction

some custom presto functions

Build

requires

• Java 8 Update 60 or higher (8u60+)
• Maven 3.3.9+ (for building)

cd ${project_home}
mvn clean package

If you want skip unit tests, please run:

mvn clean package -DskipTests

It will generate presto-third-functions-{version}-shaded.jar in target directory.

You can also directly download file from release page.

version description

functions

1. string functions

2. array functions

I had already proposed a pull request about array_union. Currently, it be merged to presto master branch. So, If your presto version > 0.151, it already include array_union function.

To support presto-0.150+, from 0.3.0, it had rename to arr_union. (from 0.5.0, I had delete arr_union function, please use array_union.)

3. date functions

4. JSON functions

5. MAP functions

6. geographic functions

关于互联网地图坐标系的说明见: 当前互联网地图的坐标系现状

7. china id card functions

8. other functions

Use

put presto-third-functions-{version}-shaded.jar into ${presto_home}/plugin/hive-hadoop2 and restart presto server. for example:

1. restart presto

mv presto-third-functions-{version}-shaded.jar /home/presto/presto-server-0.147/plugin/hive-hadoop2/
cd /home/presto/presto-server-0.147
bin/launcher restart

2. set presto command

cd /home/presto/
ln -s presto-client-0.147/presto-cli-0.147-executable.jar presto-cli
export JAVA_HOME=/home/java8/jdk1.8.0_91/;
export PATH=/home/java8/jdk1.8.0_91/bin/:$PATH;
alias presto="/home/presto/presto-cli --server localhost:8080 --catalog hive --schema default"

3. example

3.1 string functions

presto:default> select pinyin(country) from (values '中国') as t(country);
  _col0
----------
 zhongguo
(1 row)

Query 20160707_073649_00006_iya2r, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]

presto:default> select md5(col1), sha256(col1) from (values 'aaronshan') as t(col1)\G;
-[ RECORD 1 ]-----------------------------------------------------------
_col0 | 95686bc0483262afe170b550dd4544d1
_col1 | d16bb375433ad383169f911afdf45e209eabfcf047ba1faebdd8f6a0b39e0a32

Query 20160712_071936_00006_hkbes, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]

3.2 array functions

presto:default> select array_union(arr1, arr2) from (values (ARRAY [1,3,5,null], ARRAY [2,3,4,null])) as t(arr1, arr2);
         _col0
-----------------------
 [1, 3, 5, null, 2, 4]
(1 row)

Query 20160713_061707_00004_82kmt, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]

presto:default> select value_count(arr1, 'a') from (values (ARRAY['a', 'b', 'a'])) t(arr1);
 _col0
-------
     2
(1 row)

Query 20160721_111719_00008_xgf26, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]

3.3 date functions

presto
presto:default> select dayofweek(my_day) from (values '2016-07-07') as t(my_day);
 _col0
-------
     4
(1 row)

Query 20160707_073523_00005_iya2r, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]

3.4 JSON functions

presto:default> select json_array_extract(arr1, '$.book.id') from (values ('[{"book":{"id":"12"}}, {"book":{"id":"14"}}]')) t(arr1);
    _col0
--------------
 ["12", "14"]
(1 row)

Query 20160721_105423_00006_xgf26, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]

presto:default> select json_array_extract_scalar(arr1, '$.book.id') from (values ('[{"book":{"id":"12"}}, {"book":{"id":"14"}}]')) t(arr1);
  _col0
----------
 [12, 14]
(1 row)

Query 20160721_105426_00007_xgf26, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]

3.5 MAP functions

presto:default> select map1, value_count(map1, 'a') from (values (map(ARRAY[1,2,3], ARRAY['a', 'b', 'a']))) t(map1);
      map1       | _col1
-----------------+-------
 {1=a, 2=b, 3=a} |     2
(1 row)

Query 20160721_111906_00011_xgf26, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]

3.6 china id card functions

presto:default> select id_card_info(card) from (values '110101198901084517') as t(card);
                                      _col0
----------------------------------------------------------------------------------
 {"area":"东城区","valid":true,"province":"北京市","gender":"男","city":"北京市"}
(1 row)

Query 20160712_071700_00004_hkbes, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]

3.7 geographic functions

presto:default> select gcj_to_bd(lat,lng), bd_to_gcj(lat,lng), wgs_to_gcj(lat,lng), gcj_to_wgs(lat,lng), gcj_extract_wgs(lat,lng) from (values (39.915, 116.404)) as t(lat, lng)\G;
-[ RECORD 1 ]----------------------------------------------
_col0 | {"lng":116.41036949371029,"lat":39.92133699351022}
_col1 | {"lng":116.39762729119315,"lat":39.90865673957631}
_col2 | {"lng":116.41024449916938,"lat":39.91640428150164}
_col3 | {"lng":116.39775550083061,"lat":39.91359571849836}
_col4 | {"lng":116.39775549316407,"lat":39.913596801757805}

Query 20160712_024714_00003_9rund, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]

3.8 other functions

presto:default> select is_null(col0),is_null(col1),is_null(col2),is_null(col3) from (values ('test', 1, 0.5, ARRAY [1]),(null, null, null, null)) as t(col0, col1, col2,col3);
 _col0 | _col1 | _col2 | _col3
-------+-------+-------+-------
 false | false | false | false
 true  | true  | true  | true
(2 rows)

Query 20160713_061435_00003_82kmt, FINISHED, 1 node
Splits: 1 total, 0 done (0.00%)
0:00 [0 rows, 0B] [0 rows/s, 0B/s]