Microsoft KB Archive/103990

= ACC: Update Query on More Than One Table Gives Unexpected Data =

Article ID: 103990

Article Last Modified on 1/18/2007

-

APPLIES TO


 * Microsoft Access 1.0 Standard Edition
 * Microsoft Access 1.1 Standard Edition
 * Microsoft Access 2.0 Standard Edition
 * Microsoft Access 95 Standard Edition
 * Microsoft Access 97 Standard Edition

-



This article was previously published under Q103990



Novice: Requires knowledge of the user interface on single-user computers.



SYMPTOMS
An update query that tries to update more than one table may yield unexpected results.



RESOLUTION
You should not update more than one table in an update query.



STATUS
This behavior is by design.



MORE INFORMATION
If an update query in Microsoft Access is used to update two (or more) tables on the "many" side of a relationship, the resulting updated fields will be multiples of the updated value. Each value will be updated one time for each updated record in the corresponding table.

Steps to Reproduce Behavior
  Create the following three tables:

     Table: Vendors -     Field Name: Name Data Type: Text

Table: New Parts Field Name: Item Data Type: Text Field Name: Vendor Data Type: Text Field Name: Cost Data Type: Number

Table: Rebuilt Parts Field Name: Item Data Type: Text Field Name: Vendor Data Type: Text Field Name: Cost Data Type: Number   Add the following two records to the Vendors table:

     Name Bob Jill   Enter the following three records to the New Parts table:

    Item       Vendor       Cost -    fender      Bob        100.00 hood       Jill        50.00 trunk      Bob        100.00   Enter the following four records in the Rebuilt Parts table:

<pre class="fixed_text">    Item       Vendor      Cost Pump       Bob        10.00 Fan        Jill       45.00 Pulley     Bob        20.00 Piston     Bob        30.00 </li>  Create the following query based on the three tables that you created above:

<pre class="fixed_text">     Query:  Example Query ---     Type: Update Query Join: [New Parts].[Vendor] <-> [Vendors].[Name] Join: [Rebuilt Parts].[Vendor] <-> [Vendors].[Name] Field: Name Table: Vendors Criteria: "Bob" Field: Cost Table: New Parts Update To: [New Parts].[Cost] * 2.0 Field: Cost Table: Rebuilt Parts Update To: [Rebuilt Parts].[Cost] * 2.0 </li> Run the query, and then save and close it.</li> Open the Rebuilt Parts table.

Note that the costs for items listing Bob as the vendor have been updated to four times their original values, rather than to two times as specified in the update query. Each updated record in the Rebuilt Parts table has been updated once for each updated record in the New Parts table.</li> Open the New Parts table.

Note that the costs for items listing Bob as the vendor have been updated to eight times their original values, rather than to two times as specified in the update query. Each updated record in the New Parts table has been updated once for each updated record in the Rebuilt Parts table.</li></ol>

This effect is called a cross product (or Cartesian product), in which the values in Table B are updated one time for each item meeting the same criteria in Table A.

<div class="references_section">