Sven K�hler wrote: > > > 1. In the code there is a very nervous part: in case a number > > has more than 16 non-zero digits and should be handled > > with one of a bunch of functions, there is panic and the result > > is that '?'. > > This will be changed in one of the next releases > > so using double (aka float(38)) is a very bad idea for use > with sin(), > right? > > > 2. A workaround for this is > > select sin( float(sin(52)*sin(2), 15) ) > > what does sin(52)*sin(2) produce? does it produce a float(38) or > something like that?
Yes, it does > > i might ask the question, why sin() (or the *-operator) > returns values, > that sin() cannot handle. * produces a result as precise as possible (if no SIN was around, most of you would like that behaviour). It is the fault of sin to be too restrictive when checking its input. Elke SAP Labs Berlin _______________________________________________ sapdb.general mailing list [EMAIL PROTECTED] http://listserv.sap.com/mailman/listinfo/sapdb.general
