{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "if(!require(readr))   {\n",
    "   install.packages(\"readr\")\n",
    "   library(readr)\n",
    "} else {\n",
    "  library(readr)\n",
    "}\n",
    "\n",
    "## The three lines below are user supplied. dataDir is where the data currently is located (either mounted FTP directory or \n",
    "## some temporary directory where you've copied the files to locally), facilityMD5FileName is the MD5 file supplied \n",
    "## by the sequencing facility, considered \"correct\" for the purposes of this script, and owlDir is the directory where files \n",
    "## will be copied to in Owl\n",
    "dataDir <- \"~/Documents/OwlUploader/\"\n",
    "facilityMD5FileName <- \"md5sums.txt\"\n",
    "owlDir <- \"~/Documents/OwlUploader/testDir\"\n",
    "\n",
    "if(length(grep(\"apple\", R.Version()$platform) == 1) == 0)   {\n",
    "  md5.command <- \"md5sum\"\n",
    "} else {\n",
    "  md5.command <- \"md5\"\n",
    "}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1] \"ed902ade9655affa9b9e92c08be9f761  EPI-103_S27_L005_R1_001.fastq.gz initial\"\n",
      "[1] \"4ec237a381460fd7635b77b84bc530f1  EPI-103_S27_L005_R2_001.fastq.gz initial\"\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Parsed with column specification:\n",
      "cols(\n",
      "  X1 = col_character(),\n",
      "  X2 = col_character(),\n",
      "  X3 = col_character()\n",
      ")\n",
      "Parsed with column specification:\n",
      "cols(\n",
      "  X1 = col_character(),\n",
      "  X2 = col_character()\n",
      ")\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1] \"EPI-103_S27_L005_R1_001.fastq.gz copied sucessfully, MD5 for copied file is  ed902ade9655affa9b9e92c08be9f761\"\n",
      "[1] \"EPI-103_S27_L005_R2_001.fastq.gz copied sucessfully, MD5 for copied file is  4ec237a381460fd7635b77b84bc530f1\"\n"
     ]
    }
   ],
   "source": [
    "## Main Logic loop, First three if statements are checks to make sure the user has updated the script with directories\n",
    "## and the MD5 file to ensure correct operation. If any of the criteria are not met, script stops automatically.\n",
    "if(dataDir == \"Input Data Dir Here\" )   {\n",
    "  print(\"Update dataDir variable with where your data is\")\n",
    "\n",
    "} else if (owlDir == \"Input Owl Directory Here\")   {\n",
    "  print(\"Update owlDir with where you want the files copied to Owl\")\n",
    "\n",
    "} else if (facilityMD5FileName == \"File Name Here\")   {\n",
    "  print(\"Update Script with MD5 file provided by the facility\")\n",
    "\n",
    "## The else loop assumes that everything is correct and proceeds with the checking and copying process\n",
    "} else {\n",
    "  # Sets the working directory to dataDir\n",
    "  setwd(dataDir)\n",
    "  # Pulls in all of the files which match the .gz file extension. May want to add a user supplied option for compression\n",
    "  #schemes other than gzip.\n",
    "  filenames <- list.files(path = dataDir, pattern = \"*.gz\")\n",
    "\n",
    "  # Rums MD5 checks on all of the files, saving them to the external file chksum2.txt. This is just temporary and is removed\n",
    "  # during cleanup\n",
    "  for(i in 1: length(filenames))   {\n",
    "    tempMD5 <- system(paste0(md5.command, \" \", filenames[i]), intern = TRUE)\n",
    "    print(paste(tempMD5, \"initial\"))\n",
    "    system(paste0(\"echo \", tempMD5, \" >> chksum2.txt\"))\n",
    "  }\n",
    "  # reads in and formats the facility and local MD5 files, removing any NA spaces due to read_delim only using a single whitespace\n",
    "  # character to delimit, but most MD5 files seem to have two. Then names columns appropriately\n",
    "  facility.MD5s <- read_delim(paste0(dataDir,facilityMD5FileName), \n",
    "                              \"  \", escape_double = FALSE, col_names = FALSE, \n",
    "                              trim_ws = TRUE)\n",
    "  \n",
    "  facility.MD5s <- facility.MD5s[,!apply(is.na(facility.MD5s),2,all)]\n",
    "  colnames(facility.MD5s) <- c(\"md5\", \"name\")\n",
    "  \n",
    "  file.MD5s <- read_delim(paste0(dataDir,\"chksum2.txt\"), \n",
    "                          \"  \", escape_double = FALSE, col_names = FALSE, \n",
    "                          trim_ws = TRUE)\n",
    "  \n",
    "  file.MD5s <- file.MD5s[,!apply(is.na(file.MD5s),2,all)]\n",
    "  colnames(file.MD5s) <- c(\"md5\", \"name\")\n",
    "  \n",
    "  ## Logic loop for checking MD5s and initiationg copying if MD5s match.\n",
    "  for(i in 1:nrow(facility.MD5s))   {\n",
    "    setwd(owlDir)\n",
    "    ## Checks if the number of files match between chksum2.txt and the facility file. Stops script if they don't\n",
    "    if (nrow(facility.MD5s) != nrow(file.MD5s))   {\n",
    "      print(\"Number of Facility entries does not match number of files, check if all files are present\")\n",
    "      break\n",
    "    # This loop is for when MD5s match, and will first copy the file to the supplied Owl directory, then re-run\n",
    "    # an MD5, comparing it to the facility file again, and if that matches append the MD5 checksum to the existing\n",
    "    # MD5 file and add the file name to the readme.MD file in Owl. If it fails, then it prints that the copy has failed,\n",
    "    # removes the file from owl, and then stops the script\n",
    "    }else if(facility.MD5s$md5[which(facility.MD5s$name == file.MD5s$name[i])] == file.MD5s$md5[i]) {\n",
    "      system(paste0(\"scp \", dataDir, file.MD5s$name[i], \" \", owlDir))\n",
    "      tempMD5 <- substr(system(paste0(md5.command,\" \",owlDir, \"/\", filenames[i]), intern = TRUE),1 , 32)\n",
    "      if (facility.MD5s$md5[which(facility.MD5s$name == file.MD5s$name[i])] == tempMD5)   {\n",
    "        system(paste0(\"echo \", file.MD5s$name[i], \" >> readme.MD\"))\n",
    "        system(paste0(\"echo \", tempMD5, \"  \", file.MD5s$name[i] ,\">> checksum.MD5\"))\n",
    "        print(paste(file.MD5s$name[i], \"copied sucessfully, MD5 for copied file is \", file.MD5s$md5[i]))\n",
    "      }else   {\n",
    "        print(\"Copy Failure. Produced incorrect MD5\")\n",
    "        system(\"rm \", owlDir, \"/\", file.MD5s$name[i], intern = FALSE)\n",
    "        break\n",
    "      }\n",
    "    # This final if statement is for if the inital file checksum and facility checksums do not match, if that's\n",
    "    # the case then it prints that they've failed, with the file name, and saves the file name to a MD5Mismatch\n",
    "    # file for further consideration. This does not stop the loop however\n",
    "    } else if(facility.MD5s$md5[which(facility.MD5s$name == file.MD5s$name[i])] != file.MD5s$md5[i])   {\n",
    "      print(paste(\"MD5 mismatch between facility and copied file for file \", file.MD5s$name[i]))\n",
    "      system(paste0(\"echo \", file.MD5s$name[i], \" >> MD5Mismatch.txt\"))\n",
    "    }\n",
    "  }  \n",
    "setwd(dataDir)\n",
    "system(\"rm chksum2.txt\", intern = FALSE)\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "R",
   "language": "R",
   "name": "ir"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "3.3.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}