background_parallel.bash

e='echo -e'

or col in $cols; do


      	# First we get the column we want to process with an SQL query
      	# sed 1d supresses the column name
	$e "$(date):\tgetting column $col ... " >> $logfile
	column=$($sqlget "select \`$col\` from \`$db\`.\`$table\`"| sed 1d);


      	# How can I parallelize the following three (well, 6) operations?  An & after would background them, but
      	# I need to wait until they are all finished to write the output (CSV) line to $columnfile

      	# (1)
      	$e "$(date):\textracting charset from \`$db\`.\`$table\` column $col" >> $logfile
      	charset=$(echo "$column" | awk '{for(i=1;i<=NF;i++)if(!a[$i]++)printf $i}' FS="");
	charset=$(echo $charset | sed 's/\"/\"\"/g');
     
      	# (2)
	$e "$(date):\textracting number of characters from \`$db\`.\`$table\` column $col" >> $logfile
	charcount_min=$($e "$column" |  awk '{print length}' | awk 'BEGIN{a=1000}{if ($1<0+a) a=$1} END{print a}' )
	charcount_max=$($e "$column" |  awk '{print length}' | awk 'BEGIN{a=   0}{if ($1>0+a) a=$1} END{print a}' )
      
      	# (3)
      	$e "$(date):\textracting number of periods from \`$db\`.\`$table\` column $col" >> $logfile
	num_period_min=$($e "$column" |  sed 's/[^.]//g' | awk '{print length}' | awk 'BEGIN{a=1000}{if ($1<0+a) a=$1} END{print a}' )
	num_period_max=$($e "$column" |  sed 's/[^.]//g' | awk '{print length}' | awk 'BEGIN{a=   0}{if ($1>0+a) a=$1} END{print a}' )

	# once (1), (2), (3) are done, we can do the following:
	$e "$db,$table,$col,\"$charset\",$charcount_min,$charcount_max,$num_period_min,$num_period_max"  >> "$columnfile";
      
done;