Also, by the look of it, the current implementation in Min is very inefficient: for every network request, all the filters will be evaluated to check for a match.
Statistically, on average the majority of network requests are no match, and in the current implementation, no match means having iterated through all the filters, thus a significant overhead added to every single network request.
Furthermore, the evaluation of wildcard-based filters is itself quite inefficient: every time a wildcard-based filter is evaluated, an array of substrings is created and discarded.
Also, by the look of it, the current implementation in Min is very inefficient: for every network request, all the filters will be evaluated to check for a match.
Statistically, on average the majority of network requests are no match, and in the current implementation, no match means having iterated through all the filters, thus a significant overhead added to every single network request.
Furthermore, the evaluation of wildcard-based filters is itself quite inefficient: every time a wildcard-based filter is evaluated, an array of substrings is created and discarded.
[1] https://github.com/PalmerAL/min/blob/master/ext/abp-filter-p...