> 'CAKE','SUGAR',10., > 'CAKE','FLOUR',50., > 'CAKE','EGG',25., > > When I change row 'EGG' from 25 to 20, I get the rule's message .... > No problem on SUGAR and FLOUR ... > > Any idea why ?
A quick guess, the problem is with the definition for CleCompo. If R:Base can't match the value in the in-memory buffer with the value on disk it will NOT exclude the existing record in the table. Then, when you change the Egg to 20%, your total will be 105% (85% in the table and 20% in the in-memory buffer). Why might R:Base have trouble matching the in-memory value with table value, and only for the last row? Again, I can only guess, but my guess is that the in-memory value is padded with an extra blank character. R:Base never uses fewer than 4 bytes to represent a text field, and EGG is one short of four, so perhaps the in-memory version is padded. Of course, that is just a guess. When I tried the rule, I used an INTEGER autonumber column for my key. I strongly recommend this because you will avoid any comparison problems, it will take much less room in your database (four bytes per row instead of sixteen for the CleCompo column), the key will take less room in your index file, and key-building and key based retrieval will be faster. Also, I am a big proponent of using "non-intelligent" keys -- keys that do not have any meaning to the database. -- Larry _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com ================================================ TO SEE MESSAGE POSTING GUIDELINES: Send a plain text email to [EMAIL PROTECTED] In the message body, put just two words: INTRO rbase-l ================================================ TO UNSUBSCRIBE: send a plain text email to [EMAIL PROTECTED] In the message body, put just two words: UNSUBSCRIBE rbase-l ================================================ TO SEARCH ARCHIVES: http://www.mail-archive.com/rbase-l%40sonetmail.com/
