range check in awk

[Kris_5]

Hi Folks,

Is there any function (or piece of program developed by guru's like you) available for range check of the given value. For example the value of "a" should be between 0 and 45. I can do the check whether it is a interger or charcter with [0-9]+ or [a-z] string match. Thanks in adv.

Kris

[Kris_5]

Hi,

I forgot to mention the value of a=10,20 or a=12-40 or a=10,20-40. I really don't want to use split function for this. Thanks,

Kris

[Mark Greene_1]

There is an "if" function that does numeric comparisions:
{
if ( a > 0 )
if ( a < 46 )
[do stuff with a here]
}

--
mark

[Ian Dennison_1]

Test file contains a number, followed by that letter of the alphabet,...

cat /tmp/a |awk 'BEGIN { FS=" " } ( $1 > 9 && $1 < 13 ) || ( $1 == 2 ) || ( $1 == 4 ) { print $0 }'

This prints 10 - 12, 2 and 4.

Use BEDMAS (Brackets, Exponents, Division, Multiplication, Addition & Subtraction) to formulate the conditional.

&& = AND
|| = OR

I assume that the use of 'awk' in the title was correct?

Share and Enjoy! Ian

[Praveen Bezawada]

HI
If i understand your requirement correctly you can do
if a=10,20
echo a | awk -F, '{ if ( $1 > 0 ) && ( $1 < 45 ) && ( $2 > 0 ) && ( $2 < 45 ) {
print $0
}

...BPK...

[Rodney Hills]

I gather that you just want to validate the sequence of ranges are valid. Since you don't want to use awk's split function for parsing the string, then I would recommend using perl.

Here is an example-

$mins="10,20-30,50";
$ok=1;
while($mins=~/[,]?(\d+)((-)(\d+))?/g) {
$ok=0 if $1 < 0 or $1 > 45;
next unless $4;
$ok=0 if $4 < 0 or $4 > 45;
}
print $ok;

This will print 1 if $mins has a valid range of numbers, otherwise it will print 0.

-- Rod Hills

[Jeff Machols]

Make sure when you use FS you are using FS="[-,]" so you get ranges and individual integers.

[Simon Abbott]

Hello Kris,

I see you don't want to use the split function but in this instance where you need to sub-divide the , seperated fields into - seperated fields, split really is the simplest way - I don't see a better alternative. Assuming that you want to test each comma seperated range against your >=0 and <=45 criteria the following code should work fine. I wrote this into a shell script for simplicity:

echo $1 | awk '

{
output = ""
x = split($0, cs, ",")
for (a = 1; a <= x; ++a)
{
y = split(cs[a], range, "-")
if ( range[1] <=45 && range[y] >= 0 )
output = "OK"

}

print output

}'

Where $1 is your comma seperated string, such as 3,10-36,47 etc

Regards,

Simon.

[Robin Wakefield]

Hi Kris,

This is an awk script that doesn't use split:

BEGIN{for (i=0;i<10;i++){numbers[i]=1};j=0;r=0;number=0;n=0}
{
for (i=1;i<=length(range);i++)
{
char=substr(range,i,1)
if (char in numbers)
{
number=number*10^(n++)+char
if (i continue
}
if (char == "-")
{
r=1
start=number
}
else
{
if (char == "," || i == length(range))
{
if (r == 1)
{
r=0
for (num=start;num narray[j++]=num
}
narray[j++]=number
}
}
number=0
n=0
}
for (i=0;i {
if ( narray[i] == $0 )
{
print "In range"
exit
}
}
print "Not in range"
}




echo 25 | awk -f /tmp/range.awk range=1-5,7,10-20
Not in range

# echo 15 | awk -f /tmp/range.awk range=1-5,7,10-20
In range

Rgds, Robin