Shell Script that sorts data with number beginning on each line.

sunksullen · 05-08-2007, 08:58 PM

I need a shell script that sorts data alphabetically, and adds a number that begins on each line. I've seen it done a long time ago and I believe it was with the "sort" command..or maybe even wc -l. I've been looking everywhere and can't seem to find the solution.

Basically I need it to make a text file like this:

purple
yellow
red
blue
green

into this:

1:blue
2:green
3

urple
4:red
5:yellow

Any help is greatly appreciated! I'm making a large database and like how numbers organize each listing in each category. ANY solution or suggestion would help greatly. Thanks!

ghostdog74 · 05-08-2007, 09:44 PM

if you have GNU awk:

Code:

awk ' { arr[c++] = $0}
END {
	n = asort(arr,dest)
	for (i=1;i<=n;i++) print i":"dest[i]
} ' "file"

output:

Code:

# ./test.sh
1:blue
2:green
3:purple
4:red
5:yellow

twantrd · 05-08-2007, 09:56 PM

Or in perl:

Code:

#!/usr/bin/perl

my $filename="/tmp/test/file";
`cat $filename | sort > /tmp/test/sortedfile`;
open(FILE, "/tmp/test/sortedfile") || die "Could not open /tmp/test/sortedfile: $!\n";
$i=0;
while (<FILE>)
{
  print ++$i, ": $_";
}
close(FILE);

twantrd@twantrd:/tmp/test$ ./sort.pl
1: blue
2: green
3: purple
4: red
5: yellow

However, ghostdog's script looks better

.

-twantrd

sunksullen · 05-08-2007, 09:59 PM

THANKS!! It works wonderfully.
I'm still curious if its possible to get the output another way using sort.

I'm just opening the book "sed & awk" by oreilly 1990. Do you think its to old? Is GNUawk better than awk? Also how did the code you gave me tell the shell that it was gnu and not standard awk? Thanks again.

Quote:

Originally Posted by ghostdog74

if you have GNU awk:

Code:

awk ' { arr[c++] = $0}
END {
	n = asort(arr,dest)
	for (i=1;i<=n;i++) print i":"dest[i]
} ' "file"

output:

Code:

# ./test.sh
1:blue
2:green
3:purple
4:red
5:yellow

ghostdog74 · 05-08-2007, 10:25 PM

Quote:

Originally Posted by sunksullen

THANKS!! It works wonderfully.
I'm still curious if its possible to get the output another way using sort.

Code:

awk '
{arr[c++] = $0}
END {
   for (word in arr) {
	print arr[word] | "sort"
   }
   close("sort")
}' "file"

Quote:

I'm just opening the book "sed & awk" by oreilly 1990. Do you think its to old?

If you ask me, no, its not. you can still learn the foundations.

Quote:

Is GNUawk better than awk?

To me, yes. As it has other useful features, like the inbuilt asort(), asorti() functions. plus others.

Quote:

Also how did the code you gave me tell the shell that it was gnu and not standard awk? Thanks again.

if you encounter error using the asort(), then most prob your awk version is not gawk. you can check your version , awk --version. Here's mine

Code:

awk --version
GNU Awk 3.1.5
Copyright (C) 1989, 1991-2005 Free Software Foundation.

Valdis Grinbergs · 05-08-2007, 10:41 PM

Odds are, the awk you have is GNU awk.
Try entering awk --version and check the output. Mine says "GNU Awk 3.1.5..."

Your book is probably somewhat outdated; according to O'Reilly, their second edition (1997) was updated for The POSIX standard for sed and awk that was published after your 1990 book. I recommend you try using your book anyway and supplement it with online tutorials (one good one is http://www.grymoire.com/Unix/Awk.html, but there are others).

I recommend you use awk or perl because, once you know them, they can do much more for you than the bash shell. However, if you really want to use sort, you could do:
sort inputfile | cat -n | tr -s '\t' ' '
which would give you something close to what you asked for:
1 blue
2 green
3 purple
4 red
5 yellow

ghostdog74 · 05-08-2007, 11:15 PM

Quote:

Originally Posted by twantrd

Or in perl:

Code:

#!/usr/bin/perl

my $filename="/tmp/test/file";
`cat $filename | sort > /tmp/test/sortedfile`;
open(FILE, "/tmp/test/sortedfile") || die "Could not open /tmp/test/sortedfile: $!\n";
$i=0;
while (<FILE>)
{
  print ++$i, ": $_";
}
close(FILE);

twantrd@twantrd:/tmp/test$ ./sort.pl
1: blue
2: green
3: purple
4: red
5: yellow

However, ghostdog's script looks better

.

-twantrd

Perl has internal sort function, so i think you could use that too, so there's no need to use external shell commands.

Code:

$file="file";
open(FH, "<$file") or die "Cannot open ";
# or @data=<FH>
while ( my $line = <FH> ) {
	chomp($line);
	push(@data,$line );
}
@sorted = sort @data;
foreach my $i (@sorted) {
	print $i ."\n";
}

bigearsbilly · 05-09-2007, 02:50 AM

sort file | cat -n

Code:

billym.primadtpdev>sort 1|cat -n         
     1  blue
     2  green
     3  purple
     4  red
     5  yellow

livetoday · 05-09-2007, 05:42 AM

Thanks everbody....I learned the basics of array in awk and also the "cat -n " command....I used to do the same thing using nl command piping with sort, however, this does not number the spaces in the file.....

sunksullen · 05-09-2007, 12:13 PM

Thanks so much everyone. All my questions were answered perfectly.

elsheikhmh · 05-09-2007, 03:11 PM

a nice hack to the problem that nl is not using ":" is to first sort, number, then use awk with very simple action:

Code:

$ sort data | nl | awk '{print $1 ": " $2}'

OR with cat -n

Code:

$ sort data | cat -n | awk '{print $1 ": " $2}'

cheers,
-Mustafa

elsheikhmh · 05-09-2007, 03:28 PM

Sorry, but I'm just keeping my scripting skills away from rust

Sort, then count the lines, then loop and print the index along with the specific line using head & tail.

Code:

sort data > data.tmp
COUNT=`cat data.tmp | wc -l`
for i in `seq 1 $COUNT` 
do
echo $i: `head -$i data.tmp | tail -1`
done

-Mustafa

adam_blackice · 05-09-2007, 03:35 PM

thanx "darsh" for your nice trick

#your friend tantawy