UNIX for Dummies Questions & Answers

I have a source file (sample below) that I'm trying to standarize into a text file with pipe delimited currently I'm looking at the file as space delimited....

I have one issue that I don't know how to handle - in the second set of " " is a filename along with the path - most cases that filename does not have any spaces - and I want to look at that field as 1 and not 5.

Source Record:
May 29 12:09:02 2011 28061] [ovis-ftp] OK DOWNLOAD: "11.111.22.33", "/ovis_ftp_test_file_DO_NOT_REMOVE.txt", 312 bytes, 32.83Kbyte/sec
May 29 09:09:06 2011 2578] [tukbr] OK UPLOAD: "222.22.111.1", "/PHX Volume Movemtent Reports 05.29.11.txt", 137149 bytes, 64.43Kbyte/sec

Target Record:
May|29|12:09:02|2011|28061]|[ovis-ftp]|OK|DOWNLOAD:|"11.111.22.33",|"/ovis_ftp_test_file_DO_NOT_REMOVE.txt",|312|bytes,| 32.83Kbyte/sec
May|29|09:09:06|2011|2578]|[tukbr]|OK|UPLOAD:|"222.22.111.1",|"/PHX|Volume|Movemtent|Reports|05.29.11.txt",|137149 bytes,|64.43Kbyte/sec

Any suggestions would be greatly appreciated.
Thanks

could I suggest using a comma-delimiter to parse out the first four fields, and then apply split() against each field according to taste? This approach would allow you to apply a loop that can accommodate the variable contents in your $2 field...as well as the fixed layouts in fields $1, $3 and $4.

Hmmm. You can't tell awk to split only into n fields. You could do this in awk but it'd be a mess of splitting on different variables and substitutions and might still have conditions that break it. But it's easy enough in shell:

Code:

while IFS=",: " read A B C D E F G H I J
do
        echo "$A|$B|$C|$D|$E|$F|$G|$H|$I|$J"
done < infile

shell has a FS-like variable called IFS, the read builtin will split on any character in that string.

If it runs out of variables to store in, it'll store everything remaining into the last variable, field separators included.