[ome-users] DBLocation in STD definitions

Graham Klyne Graham.Klyne at zoo.ox.ac.uk
Mon Jul 25 17:16:28 BST 2005


At 10:40 25/07/05 -0400, Ilya Goldberg wrote:


>On Jul 22, 2005, at 10:36 AM, Graham Klyne wrote:
>
>>I see that the definition of a STD requires a DBLoication attribute, 
>>which I assume is a reference to where the corresponding datum is stored 
>>in the relational DB schema.
>
>This is a declaration of where in the DB you would like to store your data 
>- not a declaration of where the data is already stored.  In other words, 
>the DBLocation should be a new table/column for your data.

Ah yes, of course.

>... The new tables and columns will be created for you when you import an 
>STD document.
>
>We are looking to making this specification optional as we move to a pure 
>class-per-table structure.  Currently, a table can contain more than one 
>class, which is a needless complication.  In principle, there can be 
>table-spanning classes as well, though we've never actually tried this.

I like the idea of class-per-table.  (Where class==STD?)

>>Where can I find information about the DB schema, and in particular 
>>pointers to where I might specify that new values be stored?
>
>You can generate the schema yourself by executing
>doc/postgresql_autodoc.pl -d ome
>this will generate docbook, SVG, html, etc for the DB schema.

Hmmm, this segfaults for me.

>Or you can probe it using any of the standard DB Schema mapping tools 
>(Visio for example).

OK, I guess I can find what I need using psql.  On the basis of your above 
comments, I guess I can use a new table for our annotations, and

Thanks.

#g


---
Graham Klyne
Image Bioinformatics Research Group (http://www.bioimage.org/)
Department of Zoology, University of Oxford
South Parks Road, Oxford OX1 3PS, UK
E-mail: <Graham.Klyne at zoo.ox.ac.uk>
Direct phone: +44-(0)1865-281991
Departmental fax: +44-(0)1865-310447




More information about the ome-users mailing list