More precision than a double - for a 17 digit integer

[jason_scarlett]

I've loaded two tables from Oracle that have an ID field whichs is number(20) in Oracle. When it loads into Alteryx (vai ODBC) the incoming Alteryx format is "FixedDecimal 22.0" ... ther are no errors/warnings until I use a Filter with the expression "[ENCOUNTER_ID]  = [ENCOUNTER_ID]" ... when I run it, Alteryx gives me the warning: "Filter (54) ENCOUNTER_ID: 85572100030464414 had more precision than a double. Some precision was lost."

I tried using a Select block to change them to 'int64', 'float' and 'double', but but I still get the same warning. Why is Alteryx complaining about a 'double'? Do I need to convert it to a string?

[AdamR_AYX · Accepted solution]

Unfortunately the Alteryx formula library does not have support for fixed decimals (or int64s) and so attempts to convert them to doubles before doing anything with them.  Which is the point that you get the field conversion error.

If you just need to do an equality comparison then yes using a select tool to convert them to strings would be the solution.

[jason_scarlett]

Thanks. Hopefully this is something that is added in future releases.

Is there a difference in perfromance when using formulas with strings versus numbers?

e.g. Would a formula with a single equality check be faster with a 5-digit number column or with a 5-character string column?

Jason