Currently if the number of cached classes is less than the number of cached responses, every response will be re-parsed and those classes will be compared against the cached classes to see which ones need to be written to disk. It would be better if we instead detected which classes need to be parsed to begin with, and only parse and write those to disk. This should also allow having classes which have no corresponding response cached.
Currently if the number of cached classes is less than the number of cached responses, every response will be re-parsed and those classes will be compared against the cached classes to see which ones need to be written to disk. It would be better if we instead detected which classes need to be parsed to begin with, and only parse and write those to disk. This should also allow having classes which have no corresponding response cached.