inconsistency between spotfire data type and python data type

[wee lee]

Hi,

As far as I tested, data type is converted as below, some of them are inconsistent.

 

Format: Data type of input parameter from spotfire -> data type in Python -> data type of output parameter to Spotfire

Integer -> numpy.int64 -> Integer

Real -> numpy.float64 -> SingleReal(should be Real)

SingleReal -> str (should be numpy.float32) -> String (should be SingleReal)

Currency -> numpy.float64 -> SingleReal (should be Currency)

String -> str -> String

Date -> str(should be "date") -> String (should be Date)

Time -> str(should be "time") -> String (should be Time)

DateTime -> str(should be "datetime") -> String (should be DateTime)

Boolean -> numpy.bool_ -> Boolean

LongInteger -> numpy.int64 -> LongInteger/Integer depends on value (should be LongInteger always)

 

Pelase see attached example dxp.

Best regards,

Wei Li

[Kathleen Roan]

Please see "Spotfire and Python data type mapping"in https://docs.tibco.com/pub/sfire-analyst/latest/doc/pdf/TIB_sfire-analyst_python-packages.pdfid=0