autostart_wgs_qc

#!/bin/sh

POLL_INTERVAL=900      # seconds - i.e. 15 minutes 

function read_answer_with_default() {
   read answer
   if [ -z "$answer" ]; then
      answer=$@
   fi
}

function get_opts() {
   help_text="
usage example (currently run in a screen session) :\n
autostart_wgs_qc 230307_A01439_0152_AHVWF2DRX2 | tee /dataset/2024_illumina_sequencing_d/scratch/postprocessing/autostart_logs/230623_A01439_0188_AHFFGNDRX3.log \n
"
   while getopts ":h" opt; do
   case $opt in
       h)
         echo -e $help_text
         exit 0
         ;;
       \?)
         echo "Invalid option: -$OPTARG" >&2
         exit 1
         ;;
       :)
         echo "Option -$OPTARG requires an argument." >&2
         exit 1
         ;;
     esac
   done

   shift $((OPTIND-1))
   
   if [ -z "$1" ]; then
      echo "must supply a run name "
      exit 1
   fi

   ARGRUN=$1
   get_run_folder
}


function get_pipeline() {
   wgs_version=$1
   RUN=$2 

   export WGS_PRISM_BIN=~/wgs_prism2

   cd $WGS_PRISM_BIN

   echo "running wgs qc pipeline version $wgs_version"
   is_alpha=`echo $wgs_version | grep alpha`
   if [ $is_alpha ]; then 
      #git checkout -q $wgs_version 
      echo
      if [ $? != 0 ]; then
         echo "unable to checkout pipeline version $wgs_version"
         exit 1
      fi
      echo "running exec ./_run_wgs_qc -r $RUN $wgs_version "
      #send_mail
      exec ./_run_wgs_qc -r $RUN $wgs_version 
   else
      echo "(no checkout for alpha versions, just running in current branch)"
      echo "running exec ./_run_wgs_qc -r $RUN $wgs_version "
      send_mail
      exec ./_run_wgs_qc -r $RUN $wgs_version 
   fi
   
}

function send_mail() {
   echo "sending mail to vanstijnt , mccullocha, bairdh, perrybe, andersonr, andrewsa, henryh, frenchm, hicklandm "
   echo "(please check results carefully, non-GBS flowcells often need custom processing, mostly involving customising the sample-sheet)" | mutt -s "FYI - looks like upload of (non GBS) $RUN is complete so auto-starting processing" vanstijnt , mccullocha, bairdh, perrybe, andersonr, andrewsa, henryh, frenchm, hicklandm 
}


function get_run_folder() {
   #try novaseq
   RUN_FOLDER=/dataset/2024_illumina_sequencing_d/active/$ARGRUN
   if [ ! -d $RUN_FOLDER ]; then
      # try iseq
      RUN_FOLDER=/dataset/2024_illumina_sequencing_d/active/$ARGRUN
      if [ ! -d $RUN_FOLDER ]; then 
         # try miseq
         RUN_FOLDER=/dataset/2024_illumina_sequencing_d/active/$ARGRUN
         if [ ! -d $RUN_FOLDER ]; then
            echo "Sorry could not find run folder for $ARGRUN ! - giving up"
            exit 1
         fi
      fi
   fi
}
   
function get_landmark() {
   RUN=$1
   landmark=""
   if [ -f $RUN_FOLDER/RTAComplete.txt ]; then
      landmark=$RUN_FOLDER/RTAComplete.txt 
   fi
}

function get_digest() {
   RUN=$1
   digest=`ls -lR $RUN_FOLDER/ | md5sum -b `
   echo $digest
}


function poll_for() {
   RUN=$1 

   # polls for the run landmark 
   poll_count=0
   while [ 1 ]; do
      get_landmark $RUN
      get_digest $RUN
      if [ ! -z "$landmark" ]; then
         if [ $poll_count == 0 ]; then
            echo "*** warning - landmark $landmark is already there ! *** "
            echo "(final digest : "
            get_digest $RUN
            echo ")"
            get_pipeline 1.0.4alpha $RUN
         else
            echo "found landmark ($landmark) - starting processing"
            echo "(final digest : "
            get_digest $RUN
            echo ")"
            get_pipeline 1.0.4alpha $RUN
         fi
      else
         let poll_count=${poll_count}+1
         echo "sleeping for $POLL_INTERVAL (poll count = $poll_count )"
         sleep $POLL_INTERVAL
      fi
   done
}

get_opts "$@"
poll_for $ARGRUN