library(foreign)
library(dplyr)

# Interviewer-Questionnaire with Inwer ID (intnum):
dataset <- read.csv2("vsi.csv", dec=".", stringsAsFactors=F)

extract_ap <- function (source) {
  return (c(substr(source,0,7),
           substr(source,16,23),
           substr(source,32,39),
           substr(source,48,55),
           substr(source,64,71)));  
}

match_within <- function (t1,t2,t3) {
  
  matched_within = 0
  
  for (a in seq(1,5)) {
    
    if (t1[a] %in% t2 && t1[a] %in% t3) {
      matched_within=matched_within + 1
    }
  }
  if (matched_within >= 3) {
    return (TRUE)
  }
  else {
    return (FALSE)
    
  }
  
}

match_outside <- function (a_t1, a_t2, a_t3, b_t1, b_t2, b_t3) {

  matched = FALSE
  
  # only take cases that match !WITHIN!
  if (match_within(a_t1, a_t2, a_t3) == FALSE || match_within(b_t1, b_t2, b_t3) == FALSE) {
    return (FALSE) # no match, as we can't even do it within
  }
    
  
  matches=0
  
  # we iterate through all 5 APs. If at least three are repeated accross
  # all time points in both surveys (a_t1... b_t3) we have a match
  
  for (a in seq(1,5)) {
    
    if (a_t1[a] %in% b_t1 && a_t1[a] %in% b_t2 && a_t1[a] %in% b_t3 &&
        a_t2[a] %in% b_t1 && a_t2[a] %in% b_t2 && a_t2[a] %in% b_t3 &&
        a_t3[a] %in% b_t1 && a_t3[a] %in% b_t2 && a_t3[a] %in% b_t3) 
    {
      matches=matches+1
    }
  }
  
  if (matches >= 3) {
    return (TRUE)
  }
  else {
    return (FALSE)
  }
}

# non-optimal, case by case

for(current_row in 1:nrow(dataset)) {
  vsi_t1_exploded <- extract_ap(dataset[current_row,]$VSI1);
  vsi_t2_exploded <- extract_ap(dataset[current_row,]$VSI2);
  vsi_t3_exploded <- extract_ap(dataset[current_row,]$VSI3);
  
  # checking within
  if (!match_within(vsi_t1_exploded,vsi_t2_exploded,vsi_t3_exploded)) {
    print (paste("CASE ", current_row, " SEEM NOT TO BE TAKEN AT A SINGLE LOCATION!"))
  }
  
  # compare against all other cases
  for (compare_against in current_row+1:nrow(dataset)) {

    t1_oth <- extract_ap(dataset[compare_against,]$VSI1);
    t2_oth <- extract_ap(dataset[compare_against,]$VSI2);
    t3_oth <- extract_ap(dataset[compare_against,]$VSI3);    

    if (match_outside (vsi_t1_exploded, vsi_t2_exploded, vsi_t3_exploded, t1_oth, t2_oth, t3_oth)) {
      print (paste ("WARNING, CASE ", current_row, " SEEMS TO BE TAKEN AT SAME LOCATION AS ", compare_against));
    }
  }
}