Finding volume and item count using regular expressions

I am currently building a JavaScript web scraper for a grocery store that processes a title of a product and then returns the item count, volume and price per litre of a product. Most of the product titles look something like this:

Coca cola (vanilla flavour) 12 x 330 mL

In order to obtain meta data about this product, I have written a Regular Expression. It will look for look for a word boundary followed by a 1 or 2 digit number, whitespace, the string 'x', another whitespace and finally a 1, 2 or 3 digit number:

const filter = new RegExp(/\b\d{1,2}\sx\s\d{1,3}/);

I then test each result for a match with the Regular Expression and then calculate the item count, item volume, volume in litres and then the price per litre.

  if (result.title.match(filter)) {
     result.itemCount = parseInt(result.title.match(/\d{1}\s/));
     result.itemVolume = parseInt(result.title.match(/\d{2,3}\s/));
     result.litreVolume = (result.itemCount * result.itemVolume) / 1000;
     result.pricePerLitre = +(result.price / result.litreVolume).toFixed(2);
  } else {
     result.itemCount = 1;
     result.itemVolume = parseInt(result.title.match(/\d{2,3}\s/));
     result.litreVolume = result.itemVolume / 1000;
     result.pricePerLitre = +(result.price / result.litreVolume).toFixed(2);
  }

90% of the results look good, but sometimes I get unexpected results. For example:

• an item count of NaN, which may have to do with the fact that some titles contain several more numbers (Coca Cola (4-Way) 12 x 330 mL))
• a volume of Infinity
• a price per litre that is way too high

Clearly I am doing something wrong with my approach to calculating the desired meta data. What would be a better way of doing calculations with RegEx? Am I missing something that would make my calculations less prone to errors?

If i understand correctly filter \b\d{1,2}\sx\s\d{1,3} works, but your sub filters do not (\d{1}\s)...

I only used to using regex in c# but, i saw you could use groups in java also.
change your pattern to (\b\d{1,2})\sx\s(\d{1,3}). When you put brackets in your regex, that part becomes a group that you can acces afterwards.

As i said, i haven't used java in a few years, but i picked this code snippet from the web. It shows how to use groups in java. As pattern you should use the (\b\d{1,2})\sx\s(\d{1,3}). If it is the same as in c# group(0) is the whole result, group(1) is your first actual group, group(2) is the second.

// Create a Pattern object
Pattern r = Pattern.compile(pattern);

// Now create matcher object.
Matcher m = r.matcher(line);
      
if (m.find( )) {
System.out.println("Found value: " + m.group(0) );
System.out.println("Found value: " + m.group(1) );
}

I think you can write it with less code than stated above, but you get the picture