awk.md

AWK

Processing some files containing multiline xml-s inside

cat some.data.with.xmls.log \
 | tr -d '\n' \
 | sed 's/Case1\/Case2 request: /\n/g;s/Type1RQ>/Type1RQ>\n/g' \
 | awk '(/Type1RQ/ && /Type>Case1</) {print > "case1.txt"} (/Type1RQ/ && /Type>Case2</) {print > "case2.txt"}'

Removing end-of-line example Note: removes EOL from all lines except the ones where pattern '</ATQ>#&#' is seen

input="$1"
output="$2"

if [[ -f "${input}" ]];then
 echo "processing [${input}] to ${output}"
 zcat ${input} \
 | awk '/<\/ATQ>#&#/ {print; next} {printf "%s ", $0}' \
 | sed 's/#&#$//;s/^.*#&#//;s/> *</></g' \
 | head -n 100000 \
 > ${output}
else
 echo "missing input file [${input}]"
fi

Calculating average for some column

awk -F ',' 'BEGIN {sum=0}; {sum=sum+$2}; END {print sum/NR}' some-file.csv

Calculating difference

#!/bin/bash

#compares two columns
myFile=$1

#calculate diff
awk '
{
 if ($1!=0){
  printf "|%.2f|%.2f|%.2f|\n",$1,$2,(($2-$1)/$1);
 }else{
  printf "|%.2f|%.2f|n/a|\n",$1,$2;
 }
}' ${myFile}