R/putChar.R

In ecology-rocks/MakeMyForests: SORTIE-ND Preparation From Forest Inventory Data

#' Put Value in Expanded Dataframe According to Search Criteria
#' 
#' This function helps put a set of values into a large dataframe based on 
#' search criteria. This is useful for doing calculations that require 
#' species-specific values.
#' 
#' @param x This is your full dataframe that you search.
#' @param searchVals This is a dataframe that includes your search criteria. The
#'   final column is the responses that you want to insert.
#' @param responseName This is what you want your response Column (the one you get back) named.
#'   
#' @return This function returns a response dataframe with 1 column and rownumbers matching the initial rownumbers of x.
#' 
#' @note This may function unexpectedly if your searchVals criteria do not cover the entire length of x. E.g., if you miss 20 rows entirely, your response column will be a different length, and if you stick it back into TreesCA, it might just restart itself instead of putting "NA" or "NULL".
#' @note If you're having problems accessing something with your results, be sure that the str() is your desired type, and not the data.frame that this function passes.
#'   @examples 
#' myDf <- data.frame(type=c(rep("vegetable", 10), rep("fruit", 10)),
#' species=c(rep("broccoli",5), rep("beans", 5), rep("apples", 5),
#' rep("oranges", 5)), weights=as.numeric(runif(20, 0, 2))) 
#' uniques <- unique(myDf[,1:2])
#' uniques$val <- c("Gross", "Yum", "Ew", "Florida!")
#' putChar(myDf, uniques)
#' ## remove species and duplicates to demonstrate only one search column
#' uniques <- uniques[c(1,3), c(1,3)]
#' putChar(myDf, uniques, "test")
#' 
#' @export
putChar <- function(x, searchVals, responseName){
  ## recursively subset x through n-1 of cols of linkedVals
  searches <- searchVals[,1:(length(searchVals)-1)] ## remove last column
  searches <- as.data.frame(searches)
  if(ncol(searches) == 1){
    searchnames <- colnames(searchVals)[1]
  } else{ 
    searchnames <- colnames(searches)
  }
  response <- data.frame(rownames=NULL, response=NULL)
  
  for(i in 1:nrow(searches)){
    trimmedX <- x
    for(j in 1:length(searchnames)){
      trimmedX <- trimmedX[trimmedX[,searchnames[j]]==searches[i,j],]
    }
    names <- rownames(trimmedX)
    ## trimmedX is now subsetted properly. 
    ## put this value (searchVals[i,length(searchVals)]) in response at rownames
    response[names, responseName] <- searchVals[i,length(searchVals)]
  }
  return(response)
}