expandrows_issue

### Checking 'expandrows' function:

## Subsetted example dataset:

mydf <- structure(list(Timestamp = c("2/25/2015 15:14:09", "2/25/2015 15:16:49", 
                             "2/25/2015 15:17:06", "2/25/2015 15:18:10", "2/25/2015 15:20:16", 
                             "2/25/2015 15:21:34", "3/1/2015 14:47:32", "3/1/2015 14:49:00", 
                             "3/1/2015 14:50:50", "3/1/2015 14:56:46", "3/1/2015 14:58:08"
), Actor = c("23", "15, 28", "6", "27", "23", "29", "30", "13", 
             "26", "13", "20"), Behavior = c("Fighting", "Fighting", "Chasing", 
                                             "Chasing", "Fighting, Chasing", "Chasing", "Chasing, Subordinate", 
                                             "Chasing", "Chasing", "Fighting", "Chasing"), Recipient = c("4", 
                                                                                                         "15, 28", "17", "22", "2", "26", "8", "16", "29", "6, 16", "3"
                                             )), .Names = c("Timestamp", "Actor", "Behavior", "Recipient"), row.names = c("7", 
                                                                                                                          "8", "9", "10", "11", "12", "271", "272", "273", "274", "275"
                                             ), class = "data.frame")




mydf




#### Option 1.  Using function loaded directly into global environment:


#### This is Ananda Mahto's function based on answer here:
#### http://stackoverflow.com/questions/25901851/duplicating-modifying-rows-of-a-dataframe-dependent-on-observations-r

expandrows1 <- function(df){
  
  library(splitstackshape)  
  library(data.table)  
  
  
  temp <- cSplit(cSplit(cbind(id = 1:nrow(df), df),
                        "Actor", ",", "long"), 
                 "Recipient", ",", "long")
  

  ## Convert "Actor" and "Recipient" to numeric
  SD <- c("Actor", "Recipient")
  temp[, (SD) := lapply(.SD, as.numeric), .SDcols = SD]
  

  ## Sort Actors and Recipients, and check for duplicates and any points where Actors equal Recipients  
  temp[, toDrop := duplicated(
    paste(pmin(Actor, Recipient), pmax(Actor, Recipient))) |
      Actor == Recipient, by = id]
  
  ## Create your "score" column
  temp[, score := ifelse(any(toDrop), 0.5, 1), by = id]
  
  ## Subset and drop the irrelevant columns
  out <- temp[!temp[, toDrop, with = TRUE]][, toDrop := NULL]
  
  return(out)
}

expandrows1(mydf)  # works !

#    id          Timestamp Actor             Behavior Recipient score
#1:   1 2/25/2015 15:14:09    23             Fighting         4   1.0
#2:   2 2/25/2015 15:16:49    15             Fighting        28   0.5
#3:   3 2/25/2015 15:17:06     6              Chasing        17   1.0
#4:   4 2/25/2015 15:18:10    27              Chasing        22   1.0
#5:   5 2/25/2015 15:20:16    23    Fighting, Chasing         2   1.0
#6:   6 2/25/2015 15:21:34    29              Chasing        26   1.0
#7:   7  3/1/2015 14:47:32    30 Chasing, Subordinate         8   1.0
#8:   8  3/1/2015 14:49:00    13              Chasing        16   1.0
#9:   9  3/1/2015 14:50:50    26              Chasing        29   1.0
#10: 10  3/1/2015 14:56:46    13             Fighting         6   1.0
#11: 10  3/1/2015 14:56:46    13             Fighting        16   1.0
#12: 11  3/1/2015 14:58:08    20              Chasing         3   1.0

#although above does receive this warning:
# *** NB: by=.EACHI is now explicit. See README to restore previous behaviour.



#### Option 2.  Using 'curleylab' R package:

# this only contains one function - 'expandrows'

### install:
devtools::install_github('jalapic/curleylab')

library(curleylab) #load

expandrows(mydf)  ### does not work...

### ERROR MESSAGE: 

#  Error in `:=`((SD), lapply(.SD, as.numeric)) : 
#  Check that is.data.table(DT) == TRUE. Otherwise, := and `:=`(...) are defined for use in j, once only and in particular ways. See help(":=").