This is a quick&dirty implementation of a MongoDB storage handler for Apache HIVE.
currently only support Hive primitive types: string, int, smallint....
Whitespace should not be used in between entries in the "mongo.column.mapping" string, since these will be interperted as part of the column name, which is not what you want.
if you want "insert overwrite" feature, you must have a field named be mapped to "_id" field (Object Id in MongoDB collections).
Some code are borrowed/referenced from Balshor's Google Spreadsheet Handler(https://github.com/balshor/gdata-storagehandler) and HyperTable Hive extension(http://code.google.com/p/hypertable/wiki/HiveExtension), thanks for the help.
Here's a simple guide on how to build, hope it helps(thanks WalterDalton for providing the information):
- 1. make sure you have java sdk installed (otherwise download and install from http://www.oracle.com/technetwork/java/index.html) , $JAVA_HOME env variable is point to the installed directory and $JAVA_HOME/bin/ is included in $PATH env variable;
- 2. download maven from http://maven.apache.org and install to a directory (let's say $MAVEN_HOME), add $MAVEN_HOME/bin to $PATH
- 3. git clone Hive-Mongo to a directory; launch a cmd shell, cd that directory and execute "mvn package"; if everything is OK, you can find "hive-mongo-0.0.1-SNAPSHOT.jar" in the "target" directory. There also have a jar named "hive-mongo-0.0.1-SNAPSHOT-jar-with-dependencies.jar" which is a combo; with this one you do not need to include mongo-java-driver-2.6.3.jar and guava-r06.jar.
> $HIVE_HOME/bin/hive --auxpath /home/yc.huang/mongo-java-driver-2.6.3.jar,/home/yc.huang/guava-r06.jar, /home/yc.huang/hive-mongo-0.0.3-SNAPSHOT.jar hive> create external table mongo_users(id int, name string, age int) stored by "org.yong3.hive.mongo.MongoStorageHandler" with serdeproperties ( "mongo.column.mapping" = "_id,name,age" ) tblproperties ( "mongo.host" = "192.168.0.5", "mongo.port" = "11211", "mongo.db" = "test", "mongo.user" = "testUser", "mongo.passwd" = "testPasswd", "mongo.collection" = "users" ); OK Time taken: 4.093 seconds hive> insert overwrite table mongo_users select id, name,age from hive_test; Total MapReduce jobs = 1 Launching Job 1 out of 1 Number of reduce tasks is set to 0 since there's no reduce operator Starting Job = job_201111021553_13715, Tracking URL = http://JobTracker:50030/jobdetails.jsp?jobid=job_201111021553_13715 Kill Command = /root/dev/hadoop-0.20.2/bin/../bin/hadoop job -Dmapred.job.tracker=JobTracker:9001 -kill job_201111021553_13715 2011-11-17 18:01:25,849 Stage-0 map = 0%, reduce = 0% 2011-11-17 18:01:28,876 Stage-0 map = 100%, reduce = 0% 2011-11-17 18:01:31,893 Stage-0 map = 100%, reduce = 100% Ended Job = job_201111021553_13715 4 Rows loaded to mongo_users OK Time taken: 14.37 seconds hive> select * from mongo_users; OK 1 Tom 28 2 Alice 18 3 Bob 29 101 Scott 10 Time taken: 0.171 seconds