Hi Vinay,

Try with this script and it may solve your problem

awk '$1=="AAA" print { $0 ;}' Number.dat > AAA.dat

awk '$1=="BBB" print { $0 ;}' Number.dat > BBB.dat

Let me know if you need more details

Regards,
Siva.P
Bangalore

Syntax errors notwithstanding, that obviously reads the file multiple times. But yes, if you have trouble with sed, then awk is also an option.

awk -F '|' '$1 == "AAA" { print >"/path/to/AAA.dat"; next }
$1 == "BBB" { print > "/path/to/BBB.dat"; next }' numbers.dat

Hello Shiva,

Its giving an error :
awk: syntax error near line 1
awk: bailing out near line 1

Also as you have said,
awk '$1=="AAA" print { $0 ;}' Number.dat > AAA.dat

awk '$1=="BBB" print { $0 ;}' Number.dat > BBB.dat

If I am correct, need to give these scripts seperately.

I have some concerns,
I dont want to run thro' number.dat repeatedly. I am planning to make a script that allows to sort and move data to respective files without traversing thro' it again and again.

Kindly assist if possible,


Thanks and regards,
Vinay

Scroll back up to comment #17 if you missed that.

hi era, shiva,unix gurus,
I tried the awk present in comment #17.
awk -F '|' '$1 == "AAA" { print >"/path/to/AAA.dat"; next } $1 == "BBB" { print > "/path/to/BBB.dat"; next }' numbers.dat
Its giving an error.
awk: syntax error near line 1
awk: bailing out near line 1


I am not able to figure out what the error is...

You guys have made my day...
Now I have started going thro' awd basics and sed basics. Truly great concepts...

Please assist me if possible

Thanks and Regards,
Vinay

Perhaps at this point it would make sense to step back and tell us what platform you're on and whether you can figure out the versions of sed and awk you have at your disposal.

If you're on something like HP-UX or Irix and you have very old versions of both awk and sed, maybe a Perl script would be the best solution.

If you're on HP-UX or Solaris then there may be more recent versions of these commands if you look around for a bit. Search these forums for xpg4 and the name of your platform for some pointers.

Maybe you have a command nawk or mawk or gawk which would work better than just awk.

Here's a Perl script, for the heck of it.

perl -ne 'if (m/^(AAA|BBB)\|/) { open (H, ">>$1.dat"); print H; close H; }' numbers.dat