Re: [MonetDB-users] rdf.sql missing from Mar2011 branch

That explains it, but again why is this information accessible when querying the database schema, using for example command \D. I mean command \D does not return information about tables of sys such as sys.schemas, sys.types, sys.functions, sys.triggers, etc. I would expect the same behavior for the case of the rdf.table. Thanks, Babis On Fri, Feb 18, 2011 at 3:51 PM, Lefteris wrote:

On Fri, Feb 18, 2011 at 2:27 PM, Babis Nikolaou wrote:

...

On Fri, Feb 18, 2011 at 3:05 PM, Lefteris wrote:

...

On Fri, Feb 18, 2011 at 1:59 PM, Fabian Groffen wrote:

...

On 18-02-2011 14:58:12 +0200, Babis Nikolaou wrote:

...

...

> When you create a new database after this file was > installed, it should get executed automatically.  In other words, newly > created database should have the schema.

What is the design decision for this? If i would like to have other databases together with RDF ones, why should i have a predefined RDF schema and then drop it?

This is not a predefined RDF schema. In order for the internal relational representation of RDF to work, there is a need of some information to be added to the SQL cataloque. That is the purpose of the file 30_rdf.sql It does not affect the functionality of the database nor the freedom to define whatever databases and schemas you want.

But what is exactly that you want to do? I don't think it has to do anything with the specific file (plus you can always drop it and rerun it when you want indeed).

...

why would you drop it?

Because, it may be the case that i don't want to store RDF data in this new database. Let me explain further:

suppose I create a new database. Then, running mclient and connecting to this database I can run the following query

select * from rdf.graph;

And also in all databases if you run:

select * from sys.tables;

you will get an answer no matter what.

This is part of the cataloque that internaly the database is using to know where to find stuff, such as the names of the tables, or the names of the graphs. It has nothing to do with user defined schemas or structured and unstructured data. Think it of simply having the schema rdf reserved by the database in the same was as the schema sys is reserved and many others too.

...

and get as answer the following relation:

+-------+--------+ | gname | gid | +====+===+ +——+------+

This means, that each database a create, regardless the type of data (RDF, structured) that will be stored, it is configured to a predefined database schema, the one that 30_rdf.sql creates: containing a two-columned table with name rdf.graph.

It does not seem logical to me to drop this schema and then create the preferred one to store relational data.

You do not drop anything to create your database. It was just like this, in the old times you had to manually run a script to update the internal catalog information, now this is done automatically by a script, in the future it will be done during RDF shredding throw the internals of monetdb (aka c functions) the same as everything else is done. RDF is just still experimental and some things are done by hand to start with.

...

Hope that i made my point clear now.

Thanks, Babis

...

...

------------------------------------------------------------------------------ The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: Pinpoint memory and threading errors before they happen. Find and fix more than 250 security defects in the development cycle. Locate bottlenecks in serial and parallel code that limit performance. http://p.sf.net/sfu/intel-dev2devfeb _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users

------------------------------------------------------------------------------ The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: Pinpoint memory and threading errors before they happen. Find and fix more than 250 security defects in the development cycle. Locate bottlenecks in serial and parallel code that limit performance. http://p.sf.net/sfu/intel-dev2devfeb _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users