explosion of tiny tables representing multiple fields--Is this necessary? - Mailing list pgsql-general

From Benjamin Weaver
Subject explosion of tiny tables representing multiple fields--Is this necessary?
Date
Msg-id 20061102163649.8BB195A078@webmail218.herald.ox.ac.uk
Whole thread Raw
Responses Re: explosion of tiny tables representing multiple fields--Is this necessary?  (Martijn van Oosterhout <kleptog@svana.org>)
List pgsql-general
Dear PostGreSQL experts,


I am working with text objects. A text object will have lots of fields that
are potentially multiple. There may be more than one author, more than one
modern editor, more than one edition number, etc.

These potentially multiple fields are, in my schema, nothing more than
strings.

For my purposes, multiple strings of this kind would consist of
composed/aggregated objects stored as a member in an instance of a
MyTextObject. The member would be some kind of collection. Like this:



public Class MyTExtObject

Vector authors;  //simple String for author name
Vector editionNumbers;  //simple string for editionNumber
...


}//MyTextObject



My question: is there any legitimate way around creating lots of tiny tables,
one for each multiple field, when the multiple fields are nothing more than
strings?

We will be doing lots of involved queries, and although I have a fair amount
of experience with SQL, this profusion of tables would be a pain to implement.
But I don't see any way around creating them when the relation between the
MyTextObject and the various fields is one-to-many.




--
Benjamin Weaver
Faculty Research Associate, Imaging Papyri Projects, Herculaneum Society,
Oxford
email:  benjamin.weaver@classics.ox.ac.uk
phone:  (0)1865 288260


pgsql-general by date:

Previous
From: Shane Ambler
Date:
Subject: Re: new and old not available in rule subselects?
Next
From: "Bobus"
Date:
Subject: fetching unique pins in a high-transaction environment...