Shell Programming and Scripting

Hi,

I want to extract some part of a pattern that matches my requirement in a string with PERL. A case in point is a string like:
$eqtst="abh nmae res = 10 s abh nmae req = 10 s";
from which I want the words preceding the "=" symbol.
Previously I was assured that there would be only 2 such "=" symbols in a string and so I used sthg like:
$eqtst =~ /\s+(\w+)\s*=.*\s+(\w+)\s*=/
the "()" extracted the words "res", "req" into variables $1, $2 respectively. And life was good .
Now however the number of "=" symbols arent limited (might exceed 2 in a string). How do I go about extracting the words which precede "=" now?

Thanks,
Abhishek

Can't you run the regexp inside a loop and catch all of them?

I could. But then
(1) I need to modify the regexp I am now using to catch one such occurence (it catches 2 occurences now). For every loop I need to chop off the part of the string I have already seen, so that the same word doesnt get assigned to $1.
(2) This is PERL Isnt there a one-liner somewhere?

Thanks,
Abhishek

regular expressions, though powerful, are not always the solution to a problem.

Code:

$test="abh nmae res = 10 s abh nmae req = 10 s";
@a = split("=",$test);
print $a[0],$a[1];

Not really understand why you don't use a loop. In fact, it is far simpler than you think:

Code:

#!/usr/bin/perl -w

$test = "abh name res = 10 s abh nmae req = 10 s";
while ($test =~ /([^\s]+)\s*=\s*[^\s]+/g) {
	print ">> $1\n";
}

Which captures "res" and "req". And this will handle any number of =.