Using one-to-one relationship

Anonymous
2011-02-07T01:32:55+00:00

For various reasons, I need to break up a large table into smaller tables.  I've set up the smaller tables with one-to-one relationships.

If I assign a form's record source using just one of the smaller tables (and its fields), the form opens correctly showing the first record.  And, I can move through all the records using the Access record controls.

If I add one or more of the smaller tables (which participate in the one-to-one relationship with the original table), the form opens but no records are available - i.e., it opens as if I asked it to display a new record, and it doesn't recognize that there are other records available.

The one-to-one relationships are correctly set, with each table's primary key being a single key.

I have form controls that need to be bound to several of the one-to-one tables, so several tables need to be the form's record source.

As an example:  Table A has an autonumber primary key field, AID, and another field, Aname.  Table B has a primary key field BID (numeric long integer type) joined to AID in a one-to-one relationship (with referential integrity), and another field, Baddress.

If Table A is a recordsource for a form, the value of field Aname appears on the form and I can move through all Table A records.  As soon as I add Table B (along with Table A) as a recordsource, the form opens with no records available (i.e., a blank form as if I told it to go to a new reocrd).

I realize I can resolve this through a single large table, but that's what I'm trying to avoid. 

Has anyone worked with one-to-one relationships enough to have run into this problem?  And come up with a solution?

Thanks for any help.

Mike

Microsoft 365 and Office | Access | For home | Windows

Locked Question. This question was migrated from the Microsoft Support Community. You can vote on whether it's helpful, but you can't add comments or replies or follow the question.

0 comments No comments
Answer accepted by question author
ScottGem 68,840 Reputation points Volunteer Moderator
2011-02-10T19:13:03+00:00

Mike,

I'm going to agree with Roger yet again. Helping you has been a pleasure and we will be here to support you as your project continues.

The main difference between bound and unbound forms is that, with a bound form, Access handles all the I/O between the form and the table. With an unbound form, the user has to write code (VBA), to handle the I/O. The complexity of that code can vary widely depending on the usage and purpose of a form. For example, a form solely for entering new records is fairly easy to code. But a form for retrieving data and populating the unbound controls, can be very complex. The coding involved depends on the table structure and form layout. Personally, I'm lazy and tend to use bound forms wherever I can.

There is only ONE circumstance in Access where a child record can be created with an automatically populated FK. And that's when using a subform. When you correctly link a subform to the main (parent) form, then any records created in the subform have the FK automatically populated. This is one of the reasons I said that subforms are a very powerful feature. But this only works with bound forms. So, in every other circumstance, when you add a record to a child table, you have to make sure the FK is populated correctly.


Hope this helps, Scott<> P.S. Please post a response to let us know whether our answer helped or not. Microsoft Access MVP 2010 Blog: http://scottgem.wordpress.com Author: Microsoft Office Access 2007 VBA Technical Editor for: Special Edition Using Microsoft Access 2007 and Access 2007 Forms, Reports and Queries

Was this answer helpful?

1 person found this answer helpful.
0 comments No comments
Answer accepted by question author
Anonymous
2011-02-10T18:15:24+00:00

Mike,

It's posts like this that makes the whole process of answering questions here worth while.  I think we've seen a lightbulb go off above your head <grin>

As you no doubt realize, there are two distinct knowledge sets necessary for database development: design and implementation.  The Design set is universal and can be applied to any DBMS implementation.  The Implementation set is specific to the DBMS.

For Design knowledge, I recommend Michael Hernandez' "Database Design for Mere Mortals".  I taught 12 years of college courses in db design from that book.  It was a revelation to me.

For Access Implementation knowledge, I recommend "Access Cookbook" by Ken Getz, Paul Litwin, and Andy Baron.  I learn best from examples and this book is a set of common problems and solutions in Access development.  It's not really a book that you read through.  It's best used to find how to solve a specific problem.

I created the samples on my website (www.rogersaccesslibrary.com) with the same idea in mind. 

Good luck!


-- Roger Carlson

MS Access MVP 2006-2010

www.rogersaccesslibrary.com

If you want a detailed answer, ask a detailed question!

Was this answer helpful?

1 person found this answer helpful.
0 comments No comments

44 additional answers

Sort by: Most helpful
  1. ScottGem 68,840 Reputation points Volunteer Moderator
    2011-02-11T20:13:37+00:00

    There is only ONE circumstance in Access where a child record can be created with an automatically populated FK.

    Actually Scott, that's not really the case.  The following query using a referenced Statuses table and a referencing Transactions table is updatable and can be used to insert a new Statuses row and a matching Transactions row without having to insert a value into the foreign key Transactions.Status column, or to insert a new matching row into Transactions where the referenced row in Statuses already exists:

    Ken Sheridan, Stafford, England

    Thanks Ken,

    I'll have to try that. I've never tried creating a record in a multi-table query to see if both key fields would populate. Are you saying this would work on a form bound to such a query, or when doing an Append query or both? I'm curious though whether that's a SQL thing or an Access SQL thing.


    Hope this helps, Scott<> P.S. Please post a response to let us know whether our answer helped or not. Microsoft Access MVP 2010 Blog: http://scottgem.wordpress.com Author: Microsoft Office Access 2007 VBA Technical Editor for: Special Edition Using Microsoft Access 2007 and Access 2007 Forms, Reports and Queries

    Was this answer helpful?

    0 comments No comments
  2. Anonymous
    2011-02-11T19:21:45+00:00

    There is only ONE circumstance in Access where a child record can be created with an automatically populated FK.

    Actually Scott, that's not really the case.  The following query using a referenced Statuses table and a referencing Transactions table is updatable and can be used to insert a new Statuses row and a matching Transactions row without having to insert a value into the foreign key Transactions.Status column, or to insert a new matching row into Transactions where the referenced row in Statuses already exists:

    SELECT Statuses.Status, Transactions.Status,

    Transactions.TransactionDate, Transactions.TransactionAmount

    FROM Statuses INNER JOIN Transactions

    ON Statuses.Status = Transactions.Status;

    The steps differ slightly in the two scenarios:

    1.  When inserting a new Statuses row and a matching Transactions row the value is inserted into the Statuses.Status column and into the non-key columns from Transactions.  The Transactions.Status column receives the correct value automatically.

    2.  When inserting a new row into Transactions referencing an existing row in Statuses the reverse is the case as regards the keys, the value being inserted into Transactions.Status.  The result table will show the same value in the Statuses.Status column.

    In a one-to-one relationship type only scenario 1 applies of course, so it can be done with a single query, and there is no need to show the foreign key from the referencing table at all.  However, as Roger has pointed out, this relationship type is not often used correctly.  Its correct use is to model a type/sub-type, but this implies the existence of at least two sub-types, each sharing the type's attributes, but not each others.  When it is used where there is only one 'sub-type' it's not really a sub-type at all but simply a subset of the attributes of a single entity type, and, as has been pointed out here this more often than not is simply because of a large number of columns resulting from a non-normalized table which needs decomposition.

    In  a one-to-many relationship type (scenario 2), while possible, it's a cumbersome and unsatisfactory way of inserting data. 

    So, what I'm really saying is that while it can be done, it really should not be, and you are absolutely right to stress the use of subforms in this context.


    Ken Sheridan, Stafford, England

    Was this answer helpful?

    0 comments No comments
  3. ScottGem 68,840 Reputation points Volunteer Moderator
    2011-02-11T18:17:50+00:00

    Okay.  I copied it directly from your prior post.  I'll play with it, but as stated, it's not a valid string assignment statement with the , (comma) outside of the quotation marks.

    Mike

    Whoops, I see the mistake now. Roger's correction is correct.


    Hope this helps, Scott<> P.S. Please post a response to let us know whether our answer helped or not. Microsoft Access MVP 2010 Blog: http://scottgem.wordpress.com Author: Microsoft Office Access 2007 VBA Technical Editor for: Special Edition Using Microsoft Access 2007 and Access 2007 Forms, Reports and Queries

    Was this answer helpful?

    0 comments No comments