...
If the temporal attribute(s) on which an expression in a select operator is evaluated is a GenericTemporalType, i.e., a map from points in times to values, the trimTemporal function is useful. The select operator only changes the PredictionTimes metadata. For example, it limits the metdata from containins 100 points in time to 1 point in time. In that case, a GenericTemporalType temporal attribute contains 100 values but only needs to hold 1 value. The trimTemporal function removes all values instead of the ones which are within the metadata. This can reduce the memory consumption and makes the temporal attributes simpler to read in the output. On an algebraic level, this function changes nothing, as future functions will only use those values which are valid in the predictionTimes interval, anyway.
| Code Block |
|---|
testSelect = SELECT({
predicate = 'tdistance > 15000'
},calcDistance)
testTrim = MAP({
expressions = ['TrimTemporal(tdistance, PredictionTimes)']
},testSelect) |
...