如何匹配包含X和Y之间的数字的行？

Question

In the large file, I have lines of the form SIgnal_Catch[0] , where the number in braces ranges from 0 to 223345. I want to extract the lines where the number is between 2239 and 13569.

I tried this:

print "$line" if $line =~ /SIgnal_Catch\[2239-13569]/;

but it didn't match the lines I want. I also tried this:

print "$line" if $line =~ /SIgnal_Catch\[\d+]/;

but it matches too much. How can I do this? For this specific problem, I cannot use $1 for the range.

Answer 1

Don't try to make regex patterns do too much. You're writing a Perl program, so use Perl

This will do as you ask

if ( /SIgnal_Catch\[(\d+)\]/ and $1 >= 2239 and $1 <= 13569 ) {
    ...
}

Answer 2

Since you've said you can't use $1 or $& , here is the regex string I made. It works, but it's not pretty:

$line =~ /SIgnal_Catch\[\b(2239|22[456789]\d|2[3-9]\d{2}|[3-9]\d{3}|(10|11|12)\d{3}|13[0-4]\d{2}|135[0-6]\d)\b\]/;

Like I said in my comments, you basically have to think how the numbers can be represented in a string, and format all the alternatives. I tested this string on regex101 and it works. But if there is even a slight chance that this kind of regex may pop up again in the future, it's worth thinking about whether you can modify the tool (cfr Borodon's response, or my own comments to the original question) rather than just adding more regexes like this one.

NB: If you do test it on the regex101 web site, and you have several numbers in the Test String box, don't forget to add the global modifier g to the regex, as it will otherwise just find the first.

Answer 3

if ( $line =~ /.*SIgnal_Catch\[([1-9][0-9]{3,4})\].*/ ) {

    if ( $1 >= 2239  &&  $1 <= 13569 ) {
       print $line . "\n";
    }

}

Doing what you want with just regex would be, um, really hairy. Try something like the above.

Answer 4

As AntonH mentioned you can use the following below:

$_ =~ /SIgnal_Catch\[(\d+)\]/;

if ( $1 >= 2239 && $1 <= 13569 )
{
    print $_;
}

Your regex above will not work and your match is missing the "[" and "]".

Hope this helps.