def __init__(self,
rules,
supported_options=None,
skip_unsupported_rules=True,
use_re2='auto',
max_mem=256 * 1024 * 1024,
rule_cls=AdblockRule):
if supported_options is None:
self.supported_options = rule_cls.BINARY_OPTIONS + ['domain']
else:
self.supported_options = supported_options
self.uses_re2 = _is_re2_supported() if use_re2 == 'auto' else use_re2
self.re2_max_mem = max_mem
self.rule_cls = rule_cls
self.skip_unsupported_rules = skip_unsupported_rules
_params = dict((opt, True) for opt in self.supported_options)
self.rules = [
r for r in (r if isinstance(r, rule_cls) else rule_cls(r)
for r in rules)
if (r.regex or r.options) and r.matching_supported(_params)
]
# "advanced" rules are rules with options,
# "basic" rules are rules without options
advanced_rules, basic_rules = split_data(self.rules,
lambda r: r.options)
# Rules with domain option are handled separately:
# if user passes a domain we can discard all rules which
# require another domain. So we build an index:
# {domain: [rules_which_require_it]}, and only check
# rules which require our domain. If a rule doesn't require any
# domain.
# TODO: what about ~rules? Should we match them earlier?
domain_required_rules, non_domain_rules = split_data(
advanced_rules, lambda r:
('domain' in r.options and any(r.options["domain"].values())))
# split rules into blacklists and whitelists
self.blacklist, self.whitelist = self._split_bw(basic_rules)
_combined = partial(_combined_regex,
use_re2=self.uses_re2,
max_mem=max_mem)
self.blacklist_re = _combined([r.regex for r in self.blacklist])
self.whitelist_re = _combined([r.regex for r in self.whitelist])
self.blacklist_with_options, self.whitelist_with_options = \
self._split_bw(non_domain_rules)
self.blacklist_require_domain, self.whitelist_require_domain = \
self._split_bw_domain(domain_required_rules)
def __init__(self, rules, supported_options=None, skip_unsupported_rules=True,
use_re2='auto', max_mem=256*1024*1024, rule_cls=AdblockRule):
if supported_options is None:
self.supported_options = rule_cls.BINARY_OPTIONS + ['domain']
else:
self.supported_options = supported_options
self.uses_re2 = _is_re2_supported() if use_re2 == 'auto' else use_re2
self.re2_max_mem = max_mem
self.rule_cls = rule_cls
self.skip_unsupported_rules = skip_unsupported_rules
_params = dict((opt, True) for opt in self.supported_options)
self.rules = [
r for r in (
r if isinstance(r, rule_cls) else rule_cls(r)
for r in rules
)
if r.regex and r.matching_supported(_params)
]
# "advanced" rules are rules with options,
# "basic" rules are rules without options
advanced_rules, basic_rules = split_data(self.rules, lambda r: r.options)
# Rules with domain option are handled separately:
# if user passes a domain we can discard all rules which
# require another domain. So we build an index:
# {domain: [rules_which_require_it]}, and only check
# rules which require our domain. If a rule doesn't require any
# domain.
# TODO: what about ~rules? Should we match them earlier?
domain_required_rules, non_domain_rules = split_data(
advanced_rules,
lambda r: (
'domain' in r.options
and any(r.options["domain"].values())
)
)
# split rules into blacklists and whitelists
self.blacklist, self.whitelist = self._split_bw(basic_rules)
_combined = partial(_combined_regex, use_re2=self.uses_re2, max_mem=max_mem)
self.blacklist_re = _combined([r.regex for r in self.blacklist])
self.whitelist_re = _combined([r.regex for r in self.whitelist])
self.blacklist_with_options, self.whitelist_with_options = \
self._split_bw(non_domain_rules)
self.blacklist_require_domain, self.whitelist_require_domain = \
self._split_bw_domain(domain_required_rules)
def __init__(self, rules, supported_options=None, skip_unsupported_rules=True,
use_re2='auto', max_mem=256*1024*1024, rule_cls=AdblockRule):
if supported_options is None:
self.supported_options = rule_cls.BINARY_OPTIONS + ['domain']
else:
self.supported_options = supported_options
self.uses_re2 = _is_re2_supported() if use_re2 == 'auto' else use_re2
self.re2_max_mem = max_mem
self.rule_cls = rule_cls
self.skip_unsupported_rules = skip_unsupported_rules
_params = dict((opt, True) for opt in self.supported_options)
self.rules = [
r for r in (self.rule_cls(r) for r in rules)
if r.regex and r.matching_supported(_params)
]
# "advanced" rules are rules with options,
# "basic" rules are rules without options
advanced_rules, basic_rules = split_data(self.rules, lambda r: r.options)
# rules with domain option are handled separately:
# we may discard most rules based on domain information,
# so parser builds an index for that later.
domain_rules, non_domain_rules = split_data(
advanced_rules,
lambda r: 'domain' in r.options
)
# split rules into blacklists and whitelists
self.blacklist, self.whitelist = self._split_bw(basic_rules)
_combined = partial(_combined_regex, use_re2=self.uses_re2, max_mem=max_mem)
self.blacklist_re = _combined([r.regex for r in self.blacklist])
self.whitelist_re = _combined([r.regex for r in self.whitelist])
self.blacklist_adv, self.whitelist_adv = self._split_bw(non_domain_rules)
self.blacklist_domains, self.whitelist_domains = self._split_bw_domain(domain_rules)
def _split_bw(cls, rules):
return split_data(rules, lambda r: not r.is_exception)
def _split_bw(cls, rules):
return split_data(rules, lambda r: not r.is_exception)