This J application benefits the symbolic readability of J analysing
power&temperature correlation. I think this J makes it very nice code:
...
PEmax=: ((~.CLD)i.CLD) { CLD mean/. P - PGR - PTCmax
PEmaxf=: PGRf + ((~.CLD)i.CLDf) { CLD mean/. P - PGR - PTCmax
PEmin=: ((~.CLD)i.CLD) { CLD mean/. P - PGR - PTCmin
PEminf=: PGRf + ((~.CLD)i.CLDf) { CLD mean/. P - PGR - PTCmin
)
For me this describes preciesly the process and P, PGR, PTCmax etc.
are meanful names of variables or arrays of variables.
The problem is that if I accidentially forget to define ie PTCmax, J
does not inform this by "value error". Instead J builds an function
construction PEmax and goes to the next line. So it is sometimes very
difficult to find the error from a long skript. I wonder if somebody
else has met this difficulty and found ideas to avoid such errors.
Anssi
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
