chromium/extensions/browser/api/declarative_net_request/regex_rules_matcher.h

// Copyright 2019 The Chromium Authors
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#ifndef EXTENSIONS_BROWSER_API_DECLARATIVE_NET_REQUEST_REGEX_RULES_MATCHER_H_
#define EXTENSIONS_BROWSER_API_DECLARATIVE_NET_REQUEST_REGEX_RULES_MATCHER_H_

#include <memory>

#include "base/memory/raw_ptr.h"
#include "base/substring_set_matcher/substring_set_matcher.h"
#include "extensions/browser/api/declarative_net_request/constants.h"
#include "extensions/browser/api/declarative_net_request/ruleset_matcher_base.h"
#include "third_party/re2/src/re2/filtered_re2.h"

namespace extensions::declarative_net_request {

// Structure to hold a RegexRule together with its corresponding compiled
// re2::Re2 object.
struct RegexRuleInfo { … };

// RegexRulesMatcher deals with matching of regular expression rules. It is an
// implementation detail of RulesetMatcher. This uses the FilteredRE2 class from
// the re2 library to achieve fast matching of a set of declarative regex rules
// against a request. How this works:
//
// Initialization:
// 1. During initialization, we add each regex to the FilteredRE2 class.
// 2. We compile the FilteredRE2 object which returns us a set of substrings.
//    These are added to `substring_matcher_` for use in #3 below.
//
// Matching
// 3. Given a request url, we find the set of strings from #2. that are
//    substrings of the request url. This uses the
//    url_matcher::SubstringSetMatcher class which internally uses the
//    Aho-Corasick algorithm.
// 4. Given the list of matched strings from #3, FilteredRE2 returns the list
//    of regexes (rules) that might potentially match. To reduce the number of
//    regexes that need to be matched (since it's expensive), we prune the list
//    even further by checking if the rule metadata matches the request.
// 5. Given the list of potentially matching rules, we finally match the actual
//    regexes against the request url, as required.
class RegexRulesMatcher final : public RulesetMatcherBase { … };

}  // namespace extensions::declarative_net_request

#endif  // EXTENSIONS_BROWSER_API_DECLARATIVE_NET_REQUEST_REGEX_RULES_MATCHER_H_