Question

It’s been a while since I’ve done any serious regular expressions and It’s taking me ages to do something that should be quite simple.

It’s quite simple in that I’m delimiting the action and arguments.

I only need four named capture groups, ACTION, FROM, TO or FOUR.

I just need extract arguments from a string, these are some example of the string I’m trying to match against.

Input string: someaction from today until tomorrow

Captures: (ACTION=someaction, FROM=’today’, UNTIL=’tomorrow’, FOR=null)

Input String: someaction now + 3 until 12/12

Captures: (ACTION=someaction, FROM=’now + 3′, UNTIL=’12/12′, FOR=null)

Input string: someaction from tomorrow for 2 days

Captures: (ACTION=someaction, FROM=’tomorrow’, UNTIL=null, FOR=’2 days’)

Input string: someaction today for 6 hours

Captures: (ACTION=someaction, FROM=’today’, UNTIL=null, FOR=’6 hours’)

ACTION will always be the first whitespace delimited string. ‘someaction’ from my example.
Next will be either the word “from” or the value for “from” delimited by the words “until” or “for”.
Then will always follow the words, “until” or “for”.
Finally the named capture for “until” or “for” for the rest of the string.

I haven’t posted what I’ve got as it’s a complete mess. It doesn’t feel complicated what I’m trying to do, can anyone help give me some pointers.

Thanks,

Sam

Answer 1

As you don’t say what language you’re working with, here is a perl script that does the job:

#!/usr/bin/perl
use strict;
use warnings;
use Data::Dump qw(dump);

while(<DATA>) {
    my @capt = $_ =~ /^(\S+)\s+(?:from\s+)?(.*?)\s+(?:until\s+(.*$)|for\s+(.*$))/;
    dump @capt;
}

__DATA__
someaction from today until tomorrow
someaction now + 3 until 12/12
someaction from tomorrow for 2 days
someaction today for 6 hours

output:

("someaction", "today", "tomorrow", undef)
("someaction", "now + 3", "12/12", undef)
("someaction", "tomorrow", undef, "2 days")
("someaction", "today", undef, "6 hours")

It should be converted easily to another language.

explanation:

/               : regex delimiter
^               : start of the string
  (\S+)         : first capture: 1 or more non space char
  \s+           : 1 or more spaces
  (?:from\s+)?  : is there from litteral
  (.*?)         : second capture: any number of any char not greedy
  \s+           : some spaces
  (?:           : non cature group
    until\s+    : until + spaces
    (.*$)       : third capture group: everything until the end
    |           : OR
    for\s+      : for + spaces
    (.*$)       : fourth capture group: everything until the end
  )             : end of non capture group
/               : regex delimiter