During the load of a df I get the error message "Error during ADD FIELD tx2_max, Invalid Initial value".
ADD FIELD "tx2_max" OF "tx2_mstr" AS decimal DESCRIPTION "The maximum amount for which the tax can be applied." FORMAT ">>>,>>>,>>>,>>9.99<" INITIAL "999,999,999,999.99" LABEL "Maximum Taxable Amount" POSITION 9 MAX-WIDTH 38 COLUMN-LABEL "Max Amount" DECIMALS 10 ORDER 140 MANDATORY
After investigation I found out that the numeric format of the session was "european".
So it seems that dumping a DF does not look at the numeric format, but loading a df is, correct?
An initial value with thousands separators?
hmmm, well spotted.
So I did some testing with decimal separators in initial value:
I created a table 'test' with 2 fields. Field 'deTest1' with initial value '9,99' and field 'deTest2' with initial value '9.99'.
This table was created with num-format 'European'.
I then did 'create test' and 'find first test. Display test'.
deTest1 = 9,99
deTest2 = 999.00
Then I change the num-format to 'American' and display the record again.
Result is the same, only the separator is different:
deTest1 = 9.99
Then I deleted this record and created a new record again with num-format 'American'.
deTest2 = 999.00 (I would expect deTest1 = 999.00 and deTest2 = 9.99)
Then I change the num-format to 'European' and display the record again.
deTest2 = 999,00
After these test I change the num-format to 'American' and deleted the table and add the table again with the same fields and the same initial value, so field 'deTest1' with initial value '9,99' and field 'deTest2' with initial value '9.99'.
Then I created a new record.
deTest1 = 999.00
deTest2 = 9.99
The result is the other way around when I created the table with num-format is 'european'????
Changing the num-format does not change the value, it only change the decimal separator from '.' to ','.
But it seems that num-format is very important when you create a decimal field with a decimal initial value.