Re: Speaking of counting...

Re: Speaking of counting...

[rsc]

#!/bin/rc

first=0
last=0
incr=0
switch($#*) {
case 0
	echo 'Usage: $0 [first [incr]] last'
	exit 1
case 1
	last=$1
case 2
	first=$1
	last=$2
case 3
	first=$1
	incr=$2
	last=$3
}

awk -v 'first='$first -v 'last='$last -v 'incr='$incr 'BEGIN{
	if(incr == 0)
		incr = 1;
	if(first == 0)
		first = 1;
	for(i = first; i <= last; i += incr)
		print i;
}'

------ forwarded message follows ------

From hawkwind.utcs.toronto.edu!rc-owner Fri May  2 19:49:49 EDT 1997
Received: from corona.research.att.com by plan9; Fri May  2 19:49:49 EDT 1997
Received: from research.att.com (research.research.att.com [135.205.32.20]) by corona.research.att.com (8.7.5/8.7) with SMTP id TAA29692 for <rsc@corona.research.att.com>; Fri, 2 May 1997 19:50:01 -0400 (EDT)
Received: from hawkwind.utcs.utoronto.ca ([128.100.102.51]) by research; Fri May  2 19:48:54 EDT 1997
Received: from mail.clark.net ([168.143.0.10]) by hawkwind.utcs.utoronto.ca with SMTP id <24662>; Fri, 2 May 1997 19:45:47 -0400
Received: from clark.net (culliton@explorer.clark.net [168.143.0.7]) by mail.clark.net (8.8.5/8.6.5) with ESMTP id SAA17772; Fri, 2 May 1997 18:08:51 -0400 (EDT)
From: Tom Culliton <clark.net!culliton>
Received: (from culliton@localhost) by clark.net (8.8.5/8.7.1) id SAA16770; Fri, 2 May 1997 18:09:14 -0400 (EDT)
Date: 	Fri, 2 May 1997 18:09:14 -0400
Message-Id: <199705022209.SAA16770@clark.net>
To: hawkwind.utcs.toronto.edu!rc
Cc: LoVerso.Southborough.MA.US!john, osf.org!rsalz
Subject: Re: Speaking of counting...

So far jot and seq have been mentioned, (every machine I have available
says "jot not found" though... and ditto seq) but no specifics,

I have been accused of wanting to reinventing the wheel, (exactly what
I was trying to avoid)

Scott has shared some perl code, (Euw! The evil anit-rc!) ;-) (Thank
you.)

I swear I've seen this in rc/awk before...

Re: Speaking of counting...

[Byron Rakitzis]

Here's my version of seq:

#!/bin/rc
switch ($#*) {
case 1
        lo=1; hi=$1; incr=1
case 2
        lo=$1; hi=$2; incr=1
case 3
        lo=$1; hi=$2; incr=$3;
case *
        echo 'usage: seq [ lo ] hi [ incr ]' >[1=2]; exit 1
}

awk </dev/null 'END {
        if ('$incr' > 0) {
                for (i = '$lo'; i <= '$hi'; i += '$incr')
                        print i
        } else {
                for (i = '$lo'; i >= '$hi'; i += '$incr')
                        print i
        }
}'

It's possible that I got this from Paul Haahr (or someone else for
that matter --- I know my input-less awk scripts do all their work
in the BEGIN clause), and I don't have my plan9 manuals with me in
Europe so I have no idea if it matches up with plan9 seq.

Re: Speaking of counting...

[Tom Culliton]

So far jot and seq have been mentioned, (every machine I have available
says "jot not found" though... and ditto seq) but no specifics,

I have been accused of wanting to reinventing the wheel, (exactly what
I was trying to avoid)

Scott has shared some perl code, (Euw! The evil anit-rc!) ;-) (Thank
you.)

I swear I've seen this in rc/awk before...

Re: Speaking of counting...

[John Robert LoVerso]

jot has been in /usr/bin since 4.4BSD.

John

Re: Speaking of counting...

[Scott Schwartz]

mycroft@gnu.ai.mit.edu (Charles M. Hannum) writes:
| If you're going to do this, please at least use the same interface as
| something that already exists -- for example, the BSD jot(1) program.

Well, as long as rc ships with Plan 9, but not with BSD, we should
call it what Plan 9 does. :-)

Re: Speaking of counting...

[Rich Salz]

I remember an old bsd4.2 "contrib" command called jot that did this
kind of thing.

Re: Speaking of counting...

[Charles M. Hannum]

Tom Culliton <culliton@clark.net> writes:

> 
> Does anyone have a nice little generic range function for rc?

If you're going to do this, please at least use the same interface as
something that already exists -- for example, the BSD jot(1) program.

Re: Speaking of counting...

[Scott Schwartz]

Tom Culliton <culliton@clark.net> writes:
| Does anyone have a nice little generic range function for rc? 

In Plan 9 it's named "seq".  I've been using this approximation:

#!/bin/perl
require 'getopts.pl';

$opt_f = "%g\n";
do Getopts('f:');

if ($#ARGV == 2) {
  $start = $ARGV[0]; $step = $ARGV[1]; $end = $ARGV[2];
} elsif ($#ARGV == 1) {
  $start = $ARGV[0]; $step = 1; $end = $ARGV[1];
} elsif ($#ARGV == 0) {
  $start = 1; $step = 1; $end = $ARGV[0];
} else {
  die "usage: [-f fmt] [start [step]] end\n";
}


$opt_f =~ s/["`\$\@]/\\$&/g;
for ($n = $start; $n <= $end; $n += $step) {
  eval 'printf "' . $opt_f . '", $n';
}

Speaking of counting...

[Tom Culliton]

Does anyone have a nice little generic range function for rc?  The
type of thing that lets generate a list something like the indices
from a fortran or basic for statement.  (I'm actually modelling this
after the one in python.)

1 argument means	for (i = 0; i < $1; i += 1)
2 arguments mean	for (i = $1; i < $2; i += 1)
3 arguments mean	for (i = $1;
			     ($3 > 0 && i < $2) || ($3 > 0 && i > $2;
			     i += $3)

For example...

	# echo 99 down to but not including 0
	for (i in `{range 99 0 -1}) echo $i
	# echo 0 up to but not including 10
	for (i in `{range 0 10}) echo $i
	# echo 0 up to but not including 10 stepping 2
	for (i in `{range 0 10 2}) echo $i
	# echo 0 up to but not including 10
	for (i in `{range 10}) echo $i

I could whip something up using awk, but someone must already have one!

Tom