Shell Programming and Scripting

Hi,

I'm using a .ksh script to split one file into multible files by checking for the prefix per line. It works perfekt (thanks again for anyone involved in helping me with that ), but I want to remove the prefix per line too. Means only the line information itself should remain in the splitted files.

Here the code used to do the split, depending on filename, and creating variables with the filename for every file created:

Code:

if [[ ${_ftpfile} = @(ORD*) || ${_ftpfile} = @(SHP*) ]]; then
awk '{close(f);f=$1}{print > f".TXT"}' $_ftpfile
set B*.TXT
i=1
c=$#
x=$((c+1))
echo "$x"
while [[ $x -gt $i ]] ; do
eval _ftpfile$i="$"$i
i=$((i+1))
done
backup_files
fi

A line in a file then looks like:

BDD000000928 00000681 DB00000681 419999 7 0020302

How can I modify my awk to remove the BDD prefix? The lines then should look like:

000000928 00000681 DB00000681 419999 7 0020302

Thanks!

Code:

awk '{close(f);f=$1}{sub("BDD","");print > f".TXT"}' $_ftpfile

ok thanks so far ... forgot to mention that I have 4 files with 4 different prefixes per line

so how about different prefixes, like BDD, BHD, BAD, BND?

all those starts with "B" n ends with "D" so might this work?

Code:

awk '{close(f);f=$1}{sub("^B[A-Za-z]D","");print > f".TXT"}' $_ftpfile

In this case yes, but as it should work with different files, which might then have other prefixes, it should more be a generic thing like to always cut off the first 3 chars per line. That would help best ...

Code:

awk '{close(f);f=$1}{sub("^[A-Z][A-Z][A-Z]","");print > f".TXT"}' $_ftpfile

this should work, u can use SED too using only a dot instead of each [A-Z].

Code:

sed 's/^\(...\)\(.*\)/echo "\2" >>\1.file/' file | sh