Need some help with awk

jumper_bl · 11-23-2007, 03:09 PM

I am working with data which is in this format:

Name1
Product1
Cost
Inventory #

Name2
Product2
Cost
Inventory #

What I want to do is conver that file into

Name1,Product1, Cost, Inventory
Name2,Product2, Cost, Inventory

How should I do this?

Any help is appreciated.

druuna · 11-23-2007, 03:33 PM

Hi,

Something like this: awk 'BEGIN { FS="\n" ; RS="" } { print $1","$2","$3","$4 }' infile

Code:

$ cat infile
Name1
Product1
Cost 1.00
Inventory # 1

Name2
Product2
Cost 2.00
Inventory # 20

Name3
Product3
Cost 3.00
Inventory # 300

Name4
Product4
Cost 4.00
Inventory # 400


$ awk 'BEGIN { FS="\n" ; RS="" } { print $1","$2","$3","$4 }' infile
Name1,Product1,Cost 1.00,Inventory # 1
Name2,Product2,Cost 2.00,Inventory # 20
Name3,Product3,Cost 3.00,Inventory # 300
Name4,Product4,Cost 4.00,Inventory # 400

You can change the field separator (FS) and/or the record separator (RS) and manipulate the way the data is read.

Hope this helps.

ilikejam · 11-23-2007, 03:48 PM

Not pure awk, but I'd do it this way:

Code:

sed '/^$/d' /path/to/file | awk 'BEGIN { RS="\nInventory #\n"; FS="\n" } { for (i=1; i<=NF; i++) { printf $i; if (i != NF) printf ", " }; printf "\n" }'

which gives

Code:

[0 dave@cronus ~]$ cat go
Name1
Product1
Cost
Inventory #

Name2
Product2
Cost
Inventory #
[0 dave@cronus ~]$ sed '/^$/d' go | awk 'BEGIN { RS="\nInventory #\n"; FS="\n" } { for (i=1; i<=NF; i++) { printf $i; if (i!=NF) printf ", " }; printf "\n" }'
Name1, Product1, Cost
Name2, Product2, Cost
[0 dave@cronus ~]$

Dave

druuna · 11-23-2007, 03:52 PM

Hi,

@ilikejam: inventory # should be part of the output.....

ilikejam · 11-23-2007, 03:56 PM

Oh yeah. Right you are.

Code:

sed '/^$/d' /path/to/file | awk 'BEGIN { RS="#\n"; FS="\n" } { for (i=1; i<=NF; i++) { printf $i; if (i!=NF) printf ", " }; printf "\n" }'

Tinkster · 11-23-2007, 04:49 PM

And another awk ...

Code:

awk 'BEGIN {FS="\n";OFS=",";RS=""}{$1=$1;print} ' test

ilikejam · 11-23-2007, 05:00 PM

Clever. Someone always has a shorter way of doing things than me.

Maybe I should change my username to 'verbose'...

druuna · 11-23-2007, 05:07 PM

@ilikejam: On what platform did you try out your commands? I'm asking because they don't work here. The output is one long line, no linefeed in between the individual lines:

Code:

$ sed '/^$/d' infile | awk 'BEGIN { RS="\nInventory #\n"; FS="\n" } { for (i=1; i<=NF; i++) { printf $i; if (i!=NF) printf ", " }; printf "\n" }'
Name1, Product1, Cost 1.00, Inventory # 1, Name2, Product2, Cost 2.00, Inventory # 20, Name3, Product3, Cost 3.00, Inventory # 300, Name4, Product4, Cost 4.00, Inventory # 400, 
$

ilikejam · 11-23-2007, 05:11 PM

That's 'cause you've got 'Inventory # 300' instead of just 'Inventory #'. It's not picking up the record separator properly.

It's not entirely clear how the OP's data is actually formatted.

Dave

jumper_bl · 11-23-2007, 05:20 PM

thanks that works..

Quote:

Originally Posted by druuna

Hi,

Something like this: awk 'BEGIN { FS="\n" ; RS="" } { print $1","$2","$3","$4 }' infile

Code:

$ cat infile
Name1
Product1
Cost 1.00
Inventory # 1

Name2
Product2
Cost 2.00
Inventory # 20

Name3
Product3
Cost 3.00
Inventory # 300

Name4
Product4
Cost 4.00
Inventory # 400


$ awk 'BEGIN { FS="\n" ; RS="" } { print $1","$2","$3","$4 }' infile
Name1,Product1,Cost 1.00,Inventory # 1
Name2,Product2,Cost 2.00,Inventory # 20
Name3,Product3,Cost 3.00,Inventory # 300
Name4,Product4,Cost 4.00,Inventory # 400

You can change the field separator (FS) and/or the record separator (RS) and manipulate the way the data is read.

Hope this helps.

druuna · 11-23-2007, 05:24 PM

Quote:

Originally Posted by ilikejam

That's 'cause you've got 'Inventory # 300' instead of just 'Inventory #'. It's not picking up the record separator properly.

It's not entirely clear how the OP's data is actually formatted.

Dave

True. On both accounts