The canonical data set is in a maven repo:
<dependency>
<groupId>pentaho</groupId>
<artifactId>mondrian-data-foodmart-hsqldb</artifactId>
<version>0.2</version>
<scope>test</scope>
</dependency>
It is an hsqldb database inside a jar file.
It was what Calcite uses when it runs its tests from hsqldb; it is also the
source used by the MondrianFoodMartLoader program to load into other database.
Loader scripts and dumps for other databases (including the MySQL script I just
created) are derived, and I don’t want to get involved in managing them,
whether on github, s3 or anywhere.
I thought you mind just check in the MySQL dump as part of the script that
creates the VM. If you feel like going the extra 10 miles, you could copy
Mondrian’s loader (although it is embedded in Mondrian right now, it basically
consists of just MondrianFoodMartLoader.java and loadFoodMart.sh).
Julian
On Jan 9, 2015, at 6:39 AM, Ted Dunning <[email protected]> wrote:
> On Fri, Jan 9, 2015 at 1:18 AM, Vladimir Sitnikov <
> [email protected]> wrote:
>
>> Any ideas, where permanent data set can be stored?
>>
>
> S3 comes to mind.
