UNIX for Dummies Questions & Answers

Hello

I have an input that looks like this:

d:/bla1/bla2/bla3/bla1.txt
d:/bla1/bla2/bla4/bla2.txt
d:/bla1/bla2/bla5/bla1.txt
d:/bla1/bla2/bla6/bla2.txt
d:/bla1/bla2/bla7/bla1.txt

and i need the output to looks like this:

d:/bla1/bla2/bla3/bla1_bla3.txt
d:/bla1/bla2/bla4/bla2_bla4.txt
d:/bla1/bla2/bla5/bla1_bla5.txt
d:/bla1/bla2/bla6/bla2_bla6.txt
d:/bla1/bla2/bla7/bla1_bla7.txt

How do i find the string located between d:/bla1/bla2/ and /bla1.txt? also how do i add that string before .txt?

Can anyone help me on this?

Code:

while read mLine
do
  set -- `echo $mLine | sed 's;/; ;g'`
  mFName=`echo $5 | sed "s/\./_$4./"`
  echo $1'/'$2'/'$3'/'$4'/'$mFName
done < input_file

assuming all other data are similar in format as provided:

Code:

awk 'BEGIN{FS="/"}
     { gsub(".txt", "_"$4".txt");print }
' "file"

Thank you both

damn, i still have alooooot to learn

Code:

sed 's/\/\(.*\)\/\(.*\)\/\(.*\)\/\(.*\)\(.txt\)/\/\1\/\2\/\3\/\4_\3\5/' filename

Another 'sed' version with fewer characters:

Code:

sed 's;\(.*/\)\(.*\)/\(.*\).txt;\1\2/\3_\2.txt;' input_file