I am trying to use 2 columns returned by a magic property (identifiers and
values, respectively) and to sum the values of like identifiers.
I wrote a recursive SPIN function that does the job if the arguments are
rdf:Lists (since I can recurse on rdf:List). But the magic property that
provides the columns is not doing so in the form of rdf:List. So I am
looking for alternative approaches to solving the problem. Has anyone done
this kind of thing before? I am provided detailed information below:
The original magic property (below) grouped and summed the results nicely,
but could not be run as a web service, so we turned to using the
efunc:getQuantitySet results as input to another magic property that would
group and sum.
*SELECT* ?uID (*SUM* (?demand) *AS* ?total)
*WHERE* {
"test1" *efunc:relevantBuildings* ?resourceID .
*BIND* (efunc:getLargestModNum(?arg1, "BID_LOAD", ?resourceID) *AS*
?modNum) .
( ?eventID ?modNum ?resourceID "BID_LOAD" ) *efunc:getQuantitySet* (
?uID ?demand ) .
}
*GROUP* *BY* ?uID
*ORDER* *BY* (xsd:integer (?uID))
So I defined a new magic property to get just the results of the
efunc:getQuantitySet magic property but without the GROUP and SUM
(efunc:getLatestForecasts):
*SELECT* *DISTINCT* ?resourceID ?uID ?demand
*WHERE* {
*?arg1* *efunc:relevantBuildings* ?resourceID .
*BIND* (efunc:getLargestModNum(*?arg1*, "BID_LOAD", ?resourceID) *AS*
?modNum) .
( ?eventID ?modNum ?resourceID "BID_LOAD" ) *efunc:getQuantitySet* (
?uID ?demand ) .
}
Which, when executed as follows:
*SELECT* ?resourceID ?uID ?demand
*WHERE* {
*BIND* ("178805" *AS* ?arg1) .
?arg1 *efunc:getLatestForecasts* (?resourceID ?uID ?demand) .
}
produces content like this:
[resourceID] uID demand
(s) AAS (s) 15 (F) 0.055
(s) AAS (s) 7 (F) 0.093
(s) CIEE (s) 0 (F) 0.056
(s) CIEE (s) 8 (F) 0.073
Test the sum SPIN function (efunc:sumForecasts) with expected arguments:
*SELECT* ?result
*WHERE* {
*?arg2* rdf:first ?uifirst .
*?arg2* rdf:rest ?uirest .
*?arg3* rdf:first ?fcfirst .
*?arg3* rdf:rest ?fcrest .
*BIND* (*IF*((*?arg1* = ?uifirst), *IF*((?uirest = rdf:nil), ?fcfirst, (
?fcfirst + efunc:sumForecasts(*?arg1*, ?uirest, ?fcrest, *?arg4*))), *IF*((
*?arg4* != 0), ?fcfirst, *IF*((?uirest != rdf:nil), efunc:sumForecasts(
*?arg1*, ?uirest, ?fcrest, *?arg4*), *?arg4*))) *AS* ?result) .
}
Which, when executed as follows:
*SELECT* *DISTINCT* ?item ?sum
*WHERE* {
*BIND* (nmtest:intervalList *AS* ?arg2) .
*BIND* (nmtest:forecastList *AS* ?arg3) .
?arg2 (rdf:rest)*/rdf:first ?item .
*BIND* (efunc:sumForecasts(?item, ?arg2, ?arg3, 0) *AS* ?sum) .
} *ORDER* *BY* ?item
on the following rdf:Lists:
nmtest:intervalList: ["0", "0", "1", "2", "2", "3"]
nmtest:forecastList: [0.056, 0.056, 0.054, 0.052, 0.052, 0.4]
produces content like this:
item sum
(s) 0 0.112
(s) 1 0.054
(s) 2 0.104
(s) 3 0.4
It does not seem reasonable to convert the results of efunc:getQuantitySet
to rdf:List, so I am looking into a Javascript version to take the original
results of efunc:getQuantitySet and sum them. Any advice would be most
appreciated.
Jack Hodges, Siemens
--
You received this message because you are subscribed to the Google Groups
"TopBraid Suite Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to topbraid-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
