Question

I have a regex problem which bugs me and have no clue how to solve it.

I have an input field with a text and I like to extract certain values out of it.
I would like to extract a title, description, a price and a special price.

Examples for the input:

• everything what is plain text is concerned as title.
• everything within within hashes (#description goes here#) is considered the description.
• $23.49 is considered as price and %$19.99 would match the special price.

The CoffeeScript pattern I’m using:

 pattern = ///
  ([^$]+)
  (#(.+?)#+)
  ([\$]\d+\. \d+)
  ([\%\$]\d+\. \d+)
  ///
  params = [title,description,oldPrice,newPrice]=input_txt.match(pattern)[1..4]

It does not work. It should work if I enter all values in the given sequence and I also have to provide a the asked substring.

What I would like to have is the ability to extract the sequments if the are provided (so optional) and no matter of the sequence…
How can I extract optional sequences of an string…
EDIT///
I provide some examples

exmp1:

Kindle #Amazon's ebook reader# $79.00

this should be extracted as

title:Kindle 
description: Amazon's ebook reader 
oldPrice:$79.00

exmp2:

Nike Sneaker's $109.00 %$89.00

this should be extracted as

title:Nikes Sneaker's 
oldPrice:$109.00 
newPrice:$89.00

exmp3:

$100.00 Just dance 3 #for XBox# 

this should be extracted to

title: Just dance 3 
description: for XBox 
oldPrice:$100.00

Any help would be great …

Answer 1

The nature of regular grammars makes it hard to solve your problem. As a work around the simplest solution would be to just execute your regex 4 times:

1. Match /#(.+?)#+/ and remove the result string (string replace) from the original
2. Match /[\%\$]\d+. \d+/ and remove the result string from the original
3. Match /[\$]\d+. \d+/ and… you get the pattern
4. Now what remains in the original is the the title.