Numeric colums with values >= 1e7 make solution FALSE

[edwindj]

rules <- validator( profit == turnover - cost
                  , cost >= 0.6 * turnover
                  , turnover >= 0
                  )

data <- data.frame(profit = 1e10, cost = 200, turnover = 300)
el <- locate_errors(data, rules)
el$errors
##     profit cost turnover
##[1,]   TRUE TRUE     TRUE

el$status
# 2 

This large a value makes the MIP problem unsolveable.
Thanks to Garðar Páll Gíslason for reporting!

[edwindj]

This issue is fixed in github. Issue will be closed when on CRAN.

[edwindj]

New behavior is:

rules <- validator( profit == turnover - cost
                  , cost >= 0.6 * turnover
                  , turnover >= 0
                  )

data <- data.frame(profit = 1e10, cost = 200, turnover = 300)
el <- locate_errors(data, rules)

el$errors
##     profit cost turnover
##[1,]   TRUE FALSE     FALSE

el$status
# 0

[edwindj]

Is fixed

[edwindj]

(it generates a warning though)