----- Original Message ----- From: vijaya raj To: Rhino Sent: Monday, August 02, 2004 9:31 AM Subject: Re: help-in-normalizing
dear rhino and whil thanks for that valuable tip. but, dont i have to wory about the repeating id's (are they not redundant...) in this table, excuse me if this query looks silly... PLANT_MOLECULE plant_id molecule_id quantity --------- -------------- -------- 1 A 3.0E12 1 D 4.7E09 2 D 2.1E05 2 B 9.8E22 2 C 0.8E08 3 E 0.1E31 3 D 6.2E03 First of all, this is a table, not a query. There is no duplication in the table either. Although various plant_ids and molecule_ids appear several times in the table, each row represents one COMBINATION of plant_id and molecule_id. In other words, plant_id 3, contains two molecules, E and D. Although the other plants also contain molecule D, as well as other molecules, this does not affect plant_id 3, as long as the primary key of plant_molecule is the COMBINATION of plant_id and molecule_id. Remember, if the primary key contains both the plant_id and the molecule_id, the table will allow only one occurrence of a given combination of plant_id and molecule_id and you will never be able to get two or more rows that have the plant_id 3 / molecule_id D combination. Don't let the quantity column bother you too much; I only mentioned it because sometimes you need to have additional information about the combination of plant_id and molecule_id, such as the number of molecules per gram of a specific plant. (Actually, you probably don't care how many grams of molecules occur in a given plant: I was just trying to come up with an example to illustrate how such a quantity should be added to your design.) You may never need any additional column in the plant_molecule table beyond the plant_id and molecule_id columns so just ignore the quantity column altogether unless you need something like that. Rhino
