pull out num from str

icoder has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: pull out num from str by aaron_baugher (Curate) on Apr 23, 2012 at 07:50 UTC
First of all, a character class already means "one of these characters", so your pipe isn't doing what you think it is. It would match `\|123`. So that should be `[+-]`. It's also not necessary to escape the plus sign within a character class. Then, to say "zero or one of these," append ? to it, giving you `[+-]?` . You can also drop the `.` from the beginning and `.$` from the end, since they're implied anyway. That gets your regex down to `/([+-]?\d+)/` . Aaron B. My Woefully Neglected Blog, where I occasionally mention Perl.	[reply] [d/l] [select]
Re^2: pull out num from str by jwkrahn (Abbot) on Apr 23, 2012 at 08:45 UTC
You can also drop the . from the beginning* `"123 456" =~ /([+-]?\d+)/` and `"123 456" =~ /.([+-]?\d+)/` match two different things because `.` at the beginning is greedy.	[reply] [d/l] [select]
Re^2: pull out num from str by icoder (Novice) on Apr 23, 2012 at 09:14 UTC
Thanks . works exactly as expected	[reply]
Re^2: pull out num from str by icoder (Novice) on Apr 23, 2012 at 11:55 UTC
This works fine , Is there any disadvantages of using this regex,because ,others in this thread posted lenthy regexes which will do the same thing. please suggest	[reply]
Re^3: pull out num from str by aaron_baugher (Curate) on Apr 23, 2012 at 17:01 UTC
If it works with all the formats your data includes, that's what's important. As jwkrahn pointed out, by removing your leading `.`, I did change the effect of yours in any case where there is more than one matching number in a string: `$str = 'This sentence has two numbers, +13 and -15.'; $str =~ /([+-]?\d+)/; # matches +13, first match found $str =~ /.([+-]?\d+)/; # matches -15 because .* is greedy` [download] So which of those you use depends on whether you want to find the first number or the last number in a line. If your lines only have one number, it won't matter. Aaron B. My Woefully Neglected Blog, where I occasionally mention Perl.	[reply] [d/l] [select]
Re: pull out num from str by Anonymous Monk on Apr 23, 2012 at 07:10 UTC
Regexp::English `#!/usr/bin/perl -- use Regexp::English; print Regexp::English ->new ->remember ->multiple ->digit, "\n"; __END__ (?^:((?:\d+)))` [download]	[reply] [d/l]
Re^2: pull out num from str by icoder (Novice) on Apr 23, 2012 at 07:16 UTC
I can't use any packages here	[reply]
Re^3: pull out num from str by rovf (Priest) on Apr 23, 2012 at 08:42 UTC
I can't use any packages here Strange rule, but than simply paste the package code into your main package... -- Ronald Fischer <ynnor@mm.st>	[reply] [d/l]
Re: pull out num from str by JavaFan (Canon) on Apr 23, 2012 at 11:20 UTC
`$ perl -MRegexp::Common -E 'say $RE{num}{int}' (?:(?:[+-]?)(?:[0123456789]+))` [download]	[reply] [d/l]
Re^2: pull out num from str by icoder (Novice) on Apr 23, 2012 at 14:03 UTC
In what ways the above regex is better than this one ? `m/([+-]?\d+)/` suggestion needed	[reply] [d/l]
Re^3: pull out num from str by JavaFan (Canon) on Apr 23, 2012 at 14:16 UTC
Well, it's "better" in the sense that it doesn't match on a sequence of digits thrown together from a couple of handful of different scripts. But I posted it mainly to point you to the `Regexp::Common` module. Useful if you need patterns for common tasks, but you aren't able to write them yourself without asking on the internet.	[reply] [d/l]
Re^4: pull out num from str by icoder (Novice) on Apr 24, 2012 at 05:33 UTC
Re: pull out num from str by 2teez (Vicar) on Apr 23, 2012 at 09:12 UTC
On both strings the below regex works: `regex: m{.+?([+-]?[[:digit:]]+?)\s+?.+?};`	[reply] [d/l]