## File : 03_compute-sum-of-classes_2009.r
## Description : Calculer les synchrones pour chaque groupe obtenu par le
##               clustering. 

rm(list = ls())

MOJARRITA <- Sys.info()[4] ==  "mojarrita"

if(MOJARRITA){ 
  setwd("~/Documents/projects/2014_EDF-Orsay-Lyon2/codes/")
} else {
  setwd("~/2014_EDF-Orsay-Lyon2/codes/")
}


## 1. Read auxiliar data files ####

identifiants <- read.table("identifs.txt")[ ,1]
dates0       <- read.table("datesall.txt")[, 1]
dates        <- dates0[grep("2009", dates0)]
rm(dates0)

n <- length(identifiants)
p <- length(dates)

if(MOJARRITA) { 
  blocks <- c(rep(6500, 3), 5511) 
  } else {
  blocks <- 25011  
  } 

# Fit of the clustering : clfit 
load('../res/clfitdf200.Rdata') # Loads res that containts  
                                  # clusterings memberships
res <- as.data.frame(res)

resRDN <- as.data.frame(lapply(res, sample))
#save(file = "../res/clfitdf200RDN.Rdata", resRDN)
load("../res/clfitdf200RDN.Rdata")

lres <- length(res)
K       <- 200 #nrow(clfit$clusinfo)
rm(res)

## 2. Process the large file ####

close(con)
con <- file("~/tmp/2009_full.txt")  # Establish a connection to the file
open(con, "r")                      # Open the connection

for(b in seq_along(blocks)){        # Reading loop
  nb <- blocks[b]
  actual <- readLines(con = con, n = nb )
  actual_split <- strsplit(actual, " ")
  rm(actual)
  
  #auxmat <- matrix(unlist(strsplit(actual, " ")), ncol = p + 1, byrow = TRUE)
  #datamat <- t(apply(auxmat[, -1], 1, as.numeric)) # the NA introduced by 
                                                    # as.numeric
  auxlist <- lapply(actual_split, function(x) as.numeric(x[-1]))
  datamat <- matrix(unlist(auxlist), ncol = p, byrow = TRUE)
  rm(auxlist)

                                                   # are NA strings
  #rownames(datamat) <- substr(auxmat[, 1], 2, 7)
  auxnames          <- unlist(lapply(strsplit(actual, " "), "[", 1))
  rownames(datamat) <- substr(auxnames, 2, 7)

  rm(auxmat, actual_split)

  synchros <- lapply(resRDN, 
                     function(ll) { 
                        aux <- matrix(0, ncol = p, nrow = K)
                        for(k in 1:K) {
                          clustk <- which(ll == k)
                          if(length(clustk) > 1) {
                            aux[k, ] <- colSums(datamat[clustk, ])
                          } else {
                            aux[k, ] <- datamat[clustk, ]
                          }
                        }
                        aux
                      })
}

close(con)                # close connection to the file

# save(synchros, file = "~/tmp/2009synchrosdf200RND")