ODI - Reverse Engineering Hive Tables
By David Allan-Oracle on Dec 28, 2012
ODI can reverse engineer Hive tables via the standard reverse engineer and also an RKM to reverse engineer tables defined in Hive, this makes it very easy to capture table designs in ODI from Hive for integrating. To illustrate I will use the movie lens data set which is a common data set used in Hadoop training.
I have defined 2 tables in Hive for movies and their ratings as below, one file has fields delimited with '|' the other is tab delimited.
- create table movies (movie_id int, movie_name string, release_date string, vid_release_date string,imdb_url string) row format delimited fields terminated by '|';
- create table movie_ratings (user_id string, movie_id string, rating float, tmstmp string) row format delimited fields terminated by '\t';
For this example I have loaded the Hive tables manually from my local filesystem (into Hive/HDFS) using the following LOAD DATA Hive commands and the movie lens data set mentioned earlier;
- load data local inpath '/home/oracle/data/u.item' OVERWRITE INTO TABLE movies;
- load data local inpath '/home/oracle/data/u.data' OVERWRITE INTO TABLE movie_ratings;
The data set in the file u.item data file looks like the following with '|' delimiter;
- 1|Toy Story (1995)|01-Jan-1995||http://us.imdb.com/M/title-exact?Toy%20Story%20(1995)|0|0|0|1|1|1|0|0|0|0|0|0|0|0|0|0|0|0|0
- 2|GoldenEye (1995)|01-Jan-1995||http://us.imdb.com/M/title-exact?GoldenEye%20(1995)|0|1|1|0|0|0|0|0|0|0|0|0|0|0|0|0|1|0|0
- 3|Four Rooms (1995)|01-Jan-1995||http://us.imdb.com/M/title-exact?Four%20Rooms%20(1995)|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|1|0|0
In ODI I can define my Hive data server and logical schema, here is the JDBC connection for my Hive database (I just used the default);
I can then define my model and perform a selective reverse using standard ODI functionality, below I am reversing just the movies table and the movie ratings table;
After the reverse is complete, the tables will appear in the model in the tree, the data can be inspected just like regular datastores;
From here we see the data in the regular data view;
The ODI RKM for Hive performs logging that is useful in debugging if you hit issues with the reverse engineer. This is a very basic example of how some of the capabilities hang together, ODI can also be used to design the load of the file into Hive, transformations within it and subsequent loads using Oracle Loader for Hadoop into Oracle and on and on.