topic Re: awk vs. perl ... print $1? in Operating System - Linux

awk vs. perl ... print $1?

A. Daniel King_1 — Mon, 23 Jun 2003 21:36:38 GMT

Hi, perl gurus.

If someone could point me to the best/shortest way to convert a simple awk script to perl, I will relegate awk to the mental dustbin of seldom-used commands.

$awk -F: '{ print $1 }' /etc/passwd

a2p returns something absolutely ghastly for this one-liner. Is there a way, oh perl gurus, to repeat the elegance of the above awk solution in perl?

Thanks.

Re: awk vs. perl ... print $1?

Geoff Wild — Mon, 23 Jun 2003 22:07:01 GMT

Hmmnnn...this may help a bit:

# Sort /etc/passwd by user name.

$sort = Sort::Records->
new([width => 10,
split => [':', 0]]);
@pw = $sort->sort(`cat /etc/passwd`);

# Sort /etc/passwd by user ID.

$sort = Sort::Records->
new([type => 'int',
split => [':', 2]]);
@pw = $sort->sort(`cat /etc/passwd`);

Rgds...Geoff

Re: awk vs. perl ... print $1?

Donny Jekels — Mon, 23 Jun 2003 22:49:53 GMT

well belive it or not your perl distribution should have come with awk2perl, and a sed2perl convertor.

do a find for these scripts.

live free or die

Re: awk vs. perl ... print $1?

Mike Stroyan — Mon, 23 Jun 2003 22:52:52 GMT

I'm not a perl guru, but ...
I want to play too! ;-)

perl -anF: -e 'print @F[0],"\n"' /etc/passwd

Of course, I would probably really use

cut -d: -f1 /etc/passwd

Re: awk vs. perl ... print $1?

H.Merijn Brand (procura — Mon, 23 Jun 2003 22:54:27 GMT

# perl -aF: -nle'print $F[0]' /etc/passwd

is exactly the same as one-liner

-a autosplit to @F
-F: split on : instead of whitespace
-n don't print for every line
-l print newlines after every print (not for printf's)
-e use perl expression

$F[0] is the first field, perls lists/arrays are indexed starting from 0

Enjoy, have FUN! H.Merijn

Re: awk vs. perl ... print $1?

Donny Jekels — Mon, 23 Jun 2003 22:56:22 GMT

here it is.

http://www.perl.com/doc/manual/html/x2p/a2p.html

I used it to convert over 100 awk scripts and ran into only but a handfull errors. which needed only a little editing.

Re: awk vs. perl ... print $1?

Donny Jekels — Mon, 23 Jun 2003 23:00:18 GMT

and here is the link to sed to perl.

http://www.ictp.trieste.it/texi/perl/perl_72.html

hey sorry for the earlier post. about awk2perl and sed2perl.

its a2p and s2p, which are free translators and free to download.

good luck with your translation expidition.

live free or die
Donny

Re: awk vs. perl ... print $1?

H.Merijn Brand (procura — Tue, 24 Jun 2003 06:55:35 GMT

Donny, why do you litter this thread with exactly what he does *NOT* want. Read the original question again please.

Mike, are you using perl6? Though @F[0] is valid use in perl5, it is an array slice of only one element, it's use in your context is highly obfuscating.

Enjoy, have FUN! H.Merijn

Re: awk vs. perl ... print $1?

A. Daniel King_1 — Tue, 24 Jun 2003 13:03:23 GMT

Thanks, everyone for your posts.

Once again, Procura, you have come through.

I'm not certain I would have caught the obfuscation with @F[0]!

Re: awk vs. perl ... print $1?

Steve Post — Wed, 25 Jun 2003 11:09:12 GMT

an AWK fyi here.
you forgot the //.
awk -F: '// {print $1 }' /etc/passwd

and what about cut?
cat /etc/passwd | cut -d: -f1

steve

Re: awk vs. perl ... print $1?

A. Daniel King_1 — Mon, 14 Jul 2003 15:54:12 GMT

Steve,

Could you clarify the meaning of the //?

Thanks.

Re: awk vs. perl ... print $1?

Caesar_3 — Mon, 14 Jul 2003 19:09:26 GMT

Hello!

perl -aF: -nel 'print $F[0]'/etc/passwd
Should work.

Caesar

Re: awk vs. perl ... print $1?

Steve Post — Thu, 17 Jul 2003 13:25:07 GMT

I thought this message had already run its course. Sorry. About that.

On the "//" in the Awk command.
The // means "for every line."

You didn't have // in your awk script. I don't know what it was doing.

Normally I use awk this way.

cat file | awk ' { action to do}'

You gave awk an action to do, but did not select any lines?

other selections.
if field 2 is not billy, print the line.
awk '$2 !~ /billy/ { print $0 }'

if the line begins with hoohay, print field 3.
awk '/^hoohay/ { print $3 }'

print something at the start and end.
awk 'BEGIN { print "this is the start.\n" }
// { printf("here is a line %s\n",$0) }
END { print "this is the end." }'

steve.

Re: awk vs. perl ... print $1?

Tim D Fulford — Thu, 17 Jul 2003 13:34:10 GMT

Hi

// is implied so...
awk '{print $1}' file
is the same as
awk '//{print $1}'

I never use // & my awk works fine!!!!

Tim

Re: awk vs. perl ... print $1?

Tim D Fulford — Thu, 17 Jul 2003 13:34:34 GMT

Hi

// is implied so...
awk '{print $1}' file
is the same as
awk '//{print $1}' file

I never use // & my awk works fine!!!!

Tim

Re: awk vs. perl ... print $1?

Steve Post — Thu, 17 Jul 2003 13:40:11 GMT

ok Tim. I bet you're right. But I'm not the author. I can't see what version of awk he's using.
Maybe // does not have to be there.
I just put it in out of habit. In any case, you see it means "grab all lines."
steve

Re: awk vs. perl ... print $1?

Brian Burroughs — Wed, 29 Mar 2006 07:33:59 GMT

Old postings I know but...
Try this! Much more readable.

cat /etc/passwd | perl -anF: -e 'print $F[0],"\n"' -

Substitute the 0 for whichever column you want.

Re: awk vs. perl ... print $1?

Stuart Browne — Sat, 01 Apr 2006 17:33:51 GMT

Ugh. Actually, that is basically what procura and Mike's posts said.

During one of the forum upgrades however, all ['s got translated into their html [ values, and ]'s into ].

Re: awk vs. perl ... print $1?

Brian Burroughs — Mon, 03 Apr 2006 01:25:57 GMT

LOL

Here's me thinking I was being clever.

Re: awk vs. perl ... print $1?

A. Daniel King_1 — Mon, 22 May 2006 13:09:56 GMT

Not quite as elegant, but probably more extensible. Thanks, all.