I'm wondering if Alpha Five V11 is more restrictive than V1 in the following situation.
I have an invoice application with the invoice header record as the parent, 1 to 1 with the customer record, and 1 to many with the invoice line items.
The customer number has a validation field rule which forces it to be unique. This is a lookup field in the set and the customer fields fill the invoice header. This should be irrelevant but the invoice number is an autoincrement field.
The invoice form seems to work fine in most regards, but if I change any existing invoice header field and then either advance to the next record or explicitly save the record I get an error "Customer ID Must be Unique". That message can only come from the Customer table's validation field rule on Customer No. The Invoice set's field rule for that field shows the same validation rule. I honestly do not know whether that field rule exists separately in both the Customer table and the Invoice set or whether Alpha Five simply shows it when you look at the set's field rules.
This problem does not exist in A5 V1.
I don't know why A5 would think that a duplicate customer record is being created when the fields of the set are saved. The invoice form uses a different customer no. field from the one in the customer file, though of course they have the same value.
From what little I've just told you, can you suggest any possible culprits, especially something in the definitions that V1 would forgive and V11 not so forgiving.
I wanted to get this question posted first but I'm now going to explore whether the Customer table's field rule truly exists twice, once in the Customer table and again in the Invoice set. If it doesn't need to be in the set, I'll remove it. It would seem to me that it's sufficient to have the field rule only in the Customer table to protect it there, because if the user tried to add a duplicate record it would happen outside the set as a direct update to the lookup table.
I hope this is sufficient information for you to give me a couple problem areas to investigate.
Thanks so much ... Sam
I have an invoice application with the invoice header record as the parent, 1 to 1 with the customer record, and 1 to many with the invoice line items.
The customer number has a validation field rule which forces it to be unique. This is a lookup field in the set and the customer fields fill the invoice header. This should be irrelevant but the invoice number is an autoincrement field.
The invoice form seems to work fine in most regards, but if I change any existing invoice header field and then either advance to the next record or explicitly save the record I get an error "Customer ID Must be Unique". That message can only come from the Customer table's validation field rule on Customer No. The Invoice set's field rule for that field shows the same validation rule. I honestly do not know whether that field rule exists separately in both the Customer table and the Invoice set or whether Alpha Five simply shows it when you look at the set's field rules.
This problem does not exist in A5 V1.
I don't know why A5 would think that a duplicate customer record is being created when the fields of the set are saved. The invoice form uses a different customer no. field from the one in the customer file, though of course they have the same value.
From what little I've just told you, can you suggest any possible culprits, especially something in the definitions that V1 would forgive and V11 not so forgiving.
I wanted to get this question posted first but I'm now going to explore whether the Customer table's field rule truly exists twice, once in the Customer table and again in the Invoice set. If it doesn't need to be in the set, I'll remove it. It would seem to me that it's sufficient to have the field rule only in the Customer table to protect it there, because if the user tried to add a duplicate record it would happen outside the set as a direct update to the lookup table.
I hope this is sufficient information for you to give me a couple problem areas to investigate.
Thanks so much ... Sam
Comment