Columnar operations like Min/Max of rows/columns in xdf file

MinMax.R

airline_demo_xdf_file <- file.path(rxGetOption("sampleDataDir"), "AirlineDemoSmall.xdf")

# Find min/max value of a particular column CRSDepTime in airlineXdfData
single_column_minmax <- rxDataStep(airline_demo_xdf_file, transformFunc=function(varlst) {
  rng <- range(varlst$CRSDepTime, na.rm = TRUE)
  .min <<- min(rng[1], .min)
  .max <<- max(rng[2], .max)
  NULL
},
transformVars="CRSDepTime",
transformObjects=list(.min=numeric(0), .max=numeric(0)),
returnTransformObjects=TRUE)

airline_single_column_minmax <- rxDataStep(airline_demo_xdf_file, transforms=list(
  min=rep(.min, .rxNumRows),
  max=rep(.max, .rxNumRows)), transformEnvir=list2env(single_column_minmax))

head(airline_single_column_minmax)

# Find min/max value of a set of columns (ArrDelay and CRSDepTime) in airlineXdfData
multiple_columns_minmax <- rxDataStep(airline_demo_xdf_file, transformFunc=function(varlst) {
  rng <- range(data.frame(varlst$ArrDelay,varlst$CRSDepTime), na.rm = TRUE)
  .min <<- min(rng[1], .min)
  .max <<- max(rng[2], .max)
  NULL
},
transformVars=c("CRSDepTime", "ArrDelay"),
transformObjects=list(.min=numeric(0), .max=numeric(0)),
returnTransformObjects=TRUE)

airline_multiple_columns_minmax <- rxDataStep(airline_demo_xdf_file, transforms=list(
  min=rep(.min, .rxNumRows),
  max=rep(.max, .rxNumRows)), transformEnvir=list2env(multiple_columns_minmax))

head(airline_multiple_columns_minmax)

# Find min/max value of a particular set of rows(say rows 2000 to 3000) in airlineXdfData 
rows_minmax <- rxDataStep(airline_demo_xdf_file, transformFunc=function(varlst) {
  rng <- range(data.frame(varlst$ArrDelay,varlst$CRSDepTime), na.rm = TRUE)
  .min <<- min(rng[1], .min)
  .max <<- max(rng[2], .max)
  NULL
},
transformObjects=list(.min=numeric(0), .max=numeric(0)),
returnTransformObjects=TRUE,
startRow = 2000,
numRows = 1000)

airline_rows_minmax <- rxDataStep(airline_demo_xdf_file, transforms=list(
  min=rep(.min, .rxNumRows),
  max=rep(.max, .rxNumRows)), transformEnvir=list2env(rows_minmax))

head(airline_rows_minmax)

# Find min/max value of a filtered set of rows in airlineXdfData (say DayOfWeek Sunday)
airline_filtered <- rxDataStep(airline_demo_xdf_file, rowSelection = DayOfWeek == "Sunday")

filtered_rows_minmax <- rxDataStep(airline_filtered, transformFunc=function(varlst) {
  rng <- range(data.frame(varlst$ArrDelay,varlst$CRSDepTime), na.rm = TRUE)
  .min <<- min(rng[1], .min)
  .max <<- max(rng[2], .max)
  NULL
},
transformVars=c("CRSDepTime", "ArrDelay"),
transformObjects=list(.min=numeric(0), .max=numeric(0)),
returnTransformObjects=TRUE)

airline_filtered_rows_minmax <- rxDataStep(airline_filtered, transforms=list(
  min=rep(.min, .rxNumRows),
  max=rep(.max, .rxNumRows)), transformEnvir=list2env(filtered_rows_minmax))

head(airline_filtered_rows_minmax)