hubert depesz lubaczewski wrote:
> On Thu, Feb 05, 2009 at 02:31:24PM +0100, Ivan Voras wrote:
>> The problem is how to do it efficiently. I see there isn't a built-in
>> function that counts character occurrences so I'd have to write it
>> myself. An additional constraint is that it must be implemented with
>> built-in capabilities, i.e. SQL and plpsql languages. I can do it the
>> brute force way, looping over the string and processing one by one
>> character with substring(), but is there a faster way?
>
> # select length(regexp_replace('/some/string/with/slashes', '[^/]+', '', 'g'));
> length
> --------
> 4
> (1 row)
Thank you (and Heikki), I had a feeling I was missing an approach.
