Currently there are DDL scripts for Oracle, Postgresql and MySql. However I was wondering whether these scripts could be used to create a database schema in Hive?
It’s Open Source. If something is missing and you need it done - it becomes your job.
But seriously: It would be great if whoever is working with it put it in here.
A bit off-topic, but still might be of interest for you, @Vimala_Jacob :
Achilles is one of the OHDSI tool. To feed it with data, you have to summarize the content of your OMOP CDM-compliant database first. This sumarization is performed from R by a function in the R package “Achilles”. That function works with data stored in Oracle, Postgre, MySql and a few more DBMS’s, but not in Hive.
I was analyzing what would it take to extend the function in Achilles package for data stored in Hive. We’re using Kerberos for authentification, and I worked out the connection using JDBC connector (via R package RJDBC). The next step would be to extend the functions DatabaseConnector.R and Achilles.R, which calls (four levels below) a translation table that converts queries between various SQL dialects. We’d have to extend the table by a translation of HIVE commands to SQL Server syntax.
It could be that members who’re more knowledgeable about Achilles architecture are aware of what I described; I am sharing it for the others who were not aware (as I was).