OK, This is a similar solution to a problem posted last month (he was trying to match debits to credits). Here's how it works:
Create a temporary table to match your fruits_animals table except you want to put UNIQUE INDEXES on both columns individually. Then you run an INSERT IGNORE to copy the rows from fruits_animals into your temp table. What you will have when the INSERT IGNORE completes is a list that contains all of your animals listed only once and all of the fruits listed only once but only if that animal/fruit combination already existed. CREATE TEMPORARY TABLE tmpDedupe( animal_id int not null, fruit_id int not null, UNIQUE INDEX (animal_id), UNIQUE INDEX (fruit_id) ) INSERT IGNORE tmpDedupe (animal_id, fruit_id) SELECT (id_fruits, id_animals) FROM fruits_animals SELECT * FROM tmpDedupe ORDER BY RAND() LIMIT 100 DROP TABLE tmpDedupe Make sense? Yours, Shawn Green Database Administrator Unimin Corporation - Spruce Pine "Gerske, Sebastian" <[EMAIL PROTECTED]> wrote on 07/20/2004 09:22:30 AM: > well neither a,b or c :P > > i have lets say 1000 different fruits > and 1000 different animals > with many to man relations > now i want to extract 100 differnt fruits held by 100 different animals > without dupes of fruit or animal > > > <[EMAIL PROTECTED]> schrieb im Newsbeitrag > news:[EMAIL PROTECTED] > > Sebastian, > > > > I don't think we completely understand your question....or someone would > > have responded long before now. Let me see if I can rephrase the > > situation and maybe we can get a response. > > > > You have two tables that contain objects (your example: fruits and > > animals) and a table that relates them (example: fruits_animals) in a > > many-to-many relationship. > > > > Is your situation: > > A) You want to return 100 un-duplicated random combinations of the > > objects, regardless of if they have an entry in the relationship table? > > - OR - > > B) You want a list of 100 un-duplicated random rows from the relationship > > table? > > - OR - > > C) Each time you run your query containing "ORDER BY RAND()" , you get the > > exact same set of records back? > > > > Yours, > > Shawn Green > > Database Administrator > > Unimin Corporation - Spruce Pine > > > > > > "Gerske, Sebastian" <[EMAIL PROTECTED]> wrote on 07/20/2004 06:18:11 AM: > > > > > help me please :/ > > > > > > > > > ""Gerske, Sebastian"" <[EMAIL PROTECTED]> schrieb im Newsbeitrag > > > news:[EMAIL PROTECTED] > > > > Hello list, > > > > > > > > i still need a solution for my 3 table join rand problem, im using > > version > > > > 4.1 now so subqueries are possible: > > > > > > > > explanation: > > > > > > > > i have 3 tables lets say: > > > > > > > > fruits: > > > > > > > > id, name > > > > 1 banana > > > > 2 apple > > > > 3 strawberry > > > > > > > > fruits_animals > > > > > > > > id, id_fruits, id_animals > > > > 1 1 2 > > > > 2 1 3 > > > > 3 3 1 > > > > 4 2 4 > > > > > > > > animals > > > > > > > > id, name > > > > 1 cat > > > > 2 dog > > > > 3 mouse > > > > 4 elephant > > > > > > > > > > > > My problem is that if i join the tables and order them by rand i > > always > > > get > > > > one result something like: > > > > strawberry, cat (fruit id = 3, animal id = 1 ) > > > > or > > > > banana, dog (fruit id = 1, animal id = 2) > > > > but never > > > > banana, mouse( fruit id = 1, animal id = 3 ) > > > > > > > > and need to select 100 different relations without redundancies hows > > that > > > > possible ? > > > > > > > > regards sebastian gerske > > > > > > > > > > > > > > > > > > > > -- > > > MySQL General Mailing List > > > For list archives: http://lists.mysql.com/mysql > > > To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED] > > > > > > > > > -- > MySQL General Mailing List > For list archives: http://lists.mysql.com/mysql > To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED] >
