item preprocessing allows extracting content, based on regexp. a way to extract the number of matches is desired.

for example, the apache status page has scoreboard like this :

Scoreboard: WWWWWW..KW..KL..................................................................................................................................................................................................................................................

here we would have to count the number of occurrences of each letter to know how many processes are in each state.

a vague idea how this could work - a regexp like this :

Scoreboard: (W)

would have either some syntax for the output field, or there would be a choice between "output" and "number of matches". the latter would allow to specify which capture group to use. in this example, it would return 7.