September 5, 200025 yr Newbies Just learning this db and having a lot of fun with the relationships. I have created examples of one to one relationships and understand the portals to display the one to many. I still have yet to discover a good example of why I would want to use a self-join. Would it not be quicker to create 2 files and define a one to one or does it utilize less cpu to perform a self-join? I guess I have hit a wall because I don't understand how you define a relationship between different match fields. I am still thinking the fields must be equal. The help file did not make it clearer to me. Help anyone?
September 5, 200025 yr quote: Originally posted by appdemon: Just learning this db and having a lot of fun with the relationships. I have created examples of one to one relationships and understand the portals to display the one to many. I still have yet to discover a good example of why I would want to use a self-join. Would it not be quicker to create 2 files and define a one to one or does it utilize less cpu to perform a self-join? I guess I have hit a wall because I don't understand how you define a relationship between different match fields. I am still thinking the fields must be equal. The help file did not make it clearer to me. Help anyone? A self-join is not typically part of a relational database structure. A self-join is a convienent way to perform finds and summarizations. For instance, let's say that you are looking at customer records and want a count of all customers in the same zip code to be shown on the screen. Easist way is to do a self-join by Zip Code and setup a calculation Count (Self by ZipCode::CustomerID). Some procedures may want to work ONLY on the current record. Well you may want to insure that you are only working on ONE record, so you do a Go To Related Record [self by CustomerID, Show Only Related Records]. And I could go on and on. ------------------ =-=-=-=-=-=-=-=-=-=-=-=-= Kurt Knippel Senior Filemaker Developer http://www.database-resources.com mailto:[email protected] =-=-=-=-=-=-=-=-=-=-=-=-=
September 13, 200025 yr quote: Originally posted by appdemon: Just learning this db and having a lot of fun with the relationships. I have created examples of one to one relationships and understand the portals to display the one to many. I still have yet to discover a good example of why I would want to use a self-join. Would it not be quicker to create 2 files and define a one to one or does it utilize less cpu to perform a self-join? I guess I have hit a wall because I don't understand how you define a relationship between different match fields. I am still thinking the fields must be equal. The help file did not make it clearer to me. Help anyone? For a good discussion on the uses of self-joins and the many ways you can use them, check out Scriptology by Matt Petrowsky and John Mark Osborne. It's available at store.filemakerworld.com. I think it's the most valuable book on FileMaker available. Chuck
Create an account or sign in to comment