HTdescR: Hammett-Taft type sigma descriptor calculator

htbatch <- function (file, sigma.selection = "A", ...) {

  #reading the csv file as a dataframe

  qsardataframe <- read.csv(file, stringsAsFactors = FALSE,na.strings = "", encoding = "UTF-8")


  # colnames(qsardataframe)[colnames(qsardataframe)=="ï..no"] <- "no"

  # initializing the iterator

  i = 1
  n <- nrow (qsardataframe)

  for (i in 1:n) {

    if (is.na(qsardataframe$r1.meta1.smiles[i]) & is.na(qsardataframe$r1.ortho1.smiles[i]) & is.na(qsardataframe$r1.para1.smiles[i]) == TRUE) {

      t <- htdesc (smile = qsardataframe$r1.taft.smiles[i], HT.type = "taft", sigma.selection)
      qsardataframe$r1.taft.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r1.taft.mcs.index[i] <- t$tanimoto
      qsardataframe$r1.taft.value[i] <- t$value
      rm (t)



    } else {

      # if its not aromatic, than we will need to call htdesc

      #converting SMIset into SDFSet required for evaluation in getfragmentsigma

      # getting sigma value from htdescvalue function for r1taft

      # calling htdesc to fill substitute mcs values

      # if the structure is aromatic, than we will set these to default values

      # qsardataframe$r1.taft.smiles[i] <- "*C1=CC=CC=C1"
      # qsardataframe$r1.es.smiles[i] <- "*C1=CC=CC=C1"
      # qsardataframe$r1.ind.smiles[i] <- "*C1=CC=CC=C1"

      # calling htdesc to fill substitute mcs values

      t <- htdesc (smile = qsardataframe$r1.taft.smiles[i], HT.type = "taft", sigma.selection)
      qsardataframe$r1.taft.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r1.taft.mcs.index[i] <- t$tanimoto
      qsardataframe$r1.taft.value[i] <- t$value
      rm (t)

    }

    if (is.na(qsardataframe$r1.meta1.smiles[i]) == FALSE) {

      t <- htdesc (smile = qsardataframe$r1.meta1.smiles[i], HT.type = "meta", sigma.selection)
      qsardataframe$r1.meta1.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r1.meta1.mcs.index[i] <- t$tanimoto
      qsardataframe$r1.meta1.value[i] <- t$value
      rm (t)

    }

    if (is.na(qsardataframe$r1.meta2.smiles[i]) == FALSE) {

      t <- htdesc (smile = qsardataframe$r1.meta2.smiles[i], HT.type = "meta", sigma.selection)
      qsardataframe$r1.meta2.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r1.meta2.mcs.index[i] <- t$tanimoto
      qsardataframe$r1.meta2.value[i] <- t$value
      rm (t)

    }

    if (is.na(qsardataframe$r1.ortho1.smiles[i]) == FALSE) {

      t <- htdesc (smile = qsardataframe$r1.ortho1.smiles[i], HT.type = "ortho", sigma.selection)
      qsardataframe$r1.ortho1.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r1.ortho1.mcs.index[i] <- t$tanimoto
      qsardataframe$r1.ortho1.value[i] <- t$value
      rm (t)

    }

    if (is.na(qsardataframe$r1.ortho2.smiles[i]) == FALSE) {

      t <- htdesc (smile = qsardataframe$r1.ortho2.smiles[i], HT.type = "ortho", sigma.selection)
      qsardataframe$r1.ortho2.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r1.ortho2.mcs.index[i] <- t$tanimoto
      qsardataframe$r1.ortho2.value[i] <- t$value
      rm (t)

    }

    if (is.na(qsardataframe$r1.para1.smiles[i]) == FALSE) {

      t <- htdesc (smile = qsardataframe$r1.para1.smiles[i], HT.type = "para", sigma.selection)
      qsardataframe$r1.para1.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r1.para1.mcs.index[i] <- t$tanimoto
      qsardataframe$r1.para1.value[i] <- t$value
      rm (t)

    }

    if (is.na(qsardataframe$r2.meta1.smiles[i]) & is.na(qsardataframe$r2.ortho1.smiles[i]) & is.na(qsardataframe$r2.para1.smiles[i]) == TRUE) {

      # calling htdesc to fill substitute mcs values

      t <- htdesc (smile = qsardataframe$r2.taft.smiles[i], HT.type = "taft", sigma.selection)
      qsardataframe$r2.taft.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r2.taft.mcs.index[i] <- t$tanimoto
      qsardataframe$r2.taft.value[i] <- t$value
      rm (t)

    } else {

      # if its not aromatic, than we will need to call htdesc
      # coercing smiles (factors in dataframe object) into a list
      # x <- list (qsardataframe$r1.taft.smiles[i])
      # coercing the list containing smiles into ChemmineR's SMIset
      # y <- as(x, "SMIset")
      #converting SMIset into SDFSet required for evaluation in getfragmentsigma

      # getting sigma value from htdescvalue function for r1taft

      # calling htdesc to fill substitute mcs values

      # if the structure is aromatic, than we will set these to default values

      # qsardataframe$r2.taft.smiles[i] <- "*C1=CC=CC=C1"
      # qsardataframe$r2.es.smiles[i] <- "*C1=CC=CC=C1"
      # qsardataframe$r2.ind.smiles[i] <- "*C1=CC=CC=C1"

      t <- htdesc (smile = qsardataframe$r2.taft.smiles[i], HT.type = "taft", sigma.selection)
      qsardataframe$r2.taft.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r2.taft.mcs.index[i] <- t$tanimoto
      qsardataframe$r2.taft.value[i] <- t$value
      rm (t)
    }

    if (is.na(qsardataframe$r2.meta1.smiles[i]) == FALSE) {

      t <- htdesc (smile = qsardataframe$r2.meta1.smiles[i], HT.type = "meta", sigma.selection)
      qsardataframe$r2.meta1.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r2.meta1.mcs.index[i] <- t$tanimoto
      qsardataframe$r2.meta1.value[i] <- t$value
      rm (t)

    }

    if (is.na(qsardataframe$r2.meta2.smiles[i]) == FALSE) {

      t <- htdesc (smile = qsardataframe$r2.meta2.smiles[i], HT.type = "meta", sigma.selection)
      qsardataframe$r2.meta2.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r2.meta2.mcs.index[i] <- t$tanimoto
      qsardataframe$r2.meta2.value[i] <- t$value
      rm (t)

    }

    if (is.na(qsardataframe$r2.ortho1.smiles[i]) == FALSE) {

      t <- htdesc (smile = qsardataframe$r2.ortho1.smiles[i], HT.type = "ortho", sigma.selection)
      qsardataframe$r2.ortho1.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r2.ortho1.mcs.index[i] <- t$tanimoto
      qsardataframe$r2.ortho1.value[i] <- t$value
      rm (t)

    }

    if (is.na(qsardataframe$r2.ortho2.smiles[i]) == FALSE) {

      t <- htdesc (smile = qsardataframe$r2.ortho2.smiles[i], HT.type = "ortho", sigma.selection)
      qsardataframe$r2.ortho2.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r2.ortho2.mcs.index[i] <- t$tanimoto
      qsardataframe$r2.ortho2.value[i] <- t$value
      rm (t)

    }

    if (is.na(qsardataframe$r2.para1.smiles[i]) == FALSE) {

      t <- htdesc (smile = qsardataframe$r2.para1.smiles[i], HT.type = "para", sigma.selection)
      qsardataframe$r2.para1.sub.smiles[i] <- as.character (t$sub)
      qsardataframe$r2.para1.mcs.index[i] <- t$tanimoto
      qsardataframe$r2.para1.value[i] <- t$value
      rm (t)

    }
  }

  closeAllConnections()
# I am not sure why this is here, but dont remove it!!
  return (qsardataframe)
  #work still left
  # Low priority
    # insert if statement for es and ind
    # insert output file format as a function attribute

  # medium priority

    # use appropriate HT.type after creating the data files
    # use appropriate sigma.selection method

  # high priority

    # replace NA with *H from .smiles columns from those chemicals where its proven that r1 or r2 are aromatic
    # replace NA with 0 for .values cells
    # you can leave NA for .mcs.index and .sub.smiles for ones where we didnt trigger getsmiles function
    # insert a summation method which adds up values for hammetts

}