regex word boundaries

Question

I am trying to write a regex to match a string with the START of ALL words in a string.

It would be used to lookup a city via the name or the zipcode (in a city and zipcode concatenated string) when the user types one character (or string).

The problem is that it also matches the END of the city (which is unwanted...) I am using the following regex :

patt = new RegExp('\\b' + searchchar, 'i')

eg :

• teststring : "Sint-niklaas (9100)"
• char typed : 's'
  => unwanted matches S int-niklaa s (9100) <== problem :-)
  => wanted matches S int-niklaas (9100)
  
  
• char typed : '9'
  => wanted matches Sint-niklaas ( 9 100) <== ok

Answer 1

I think you are looking for that:

patt = new RegExp('^(?:[^(]+\\()?' + searchchar, 'mi');

^ is an anchor for the start of the string line (with the m modifier). When you enter a letter, since there are only digits inside parenthesis, the optional non-capturing group fails and the letter can only be at the start. When a digit is entered, the optional group succeeds.