// Copyright 2022 The Chromium Authors // Use of this source code is governed by a BSD-style license that can be // found in the LICENSE file. #ifndef THIRD_PARTY_BLINK_RENDERER_CORE_LOADER_URL_MATCHER_H_ #define THIRD_PARTY_BLINK_RENDERER_CORE_LOADER_URL_MATCHER_H_ #include <string_view> #include "third_party/blink/renderer/core/core_export.h" #include "third_party/blink/renderer/platform/weborigin/kurl.h" #include "third_party/blink/renderer/platform/weborigin/security_origin.h" #include "third_party/blink/renderer/platform/wtf/text/wtf_string.h" #include "third_party/blink/renderer/platform/wtf/vector.h" namespace blink { // UrlMatcher is a class to manage the list of URLs stored in the field trial // param. As the original data from the field trial params is delivered as a // special format string, this class parses and formats it, and stores the list. // // The expected param format is a comma separated string, and each string is // separated by the vertical bar. The left side of the vertical bar is a // host name, and the right side is a part of path or search params. // // The string is something like // "https://test.exmaple|/foo,http://another.test.example|?foo=bar,https:://yet.another.test.example" // Then the UrlMatcher will parse it to the formatted list like // [ // ["https://test.example", "/foo"], // ["http://another.test.example", "foo=bar"], // ["https:://yet.another.test.example", ""] // ] // Based on the above list, UrlMatcher::Match() checks 1) if the given url is a // same origin or not, 2) if it's a same origin, check the second value in the // list item. If it's an empty string, that means origin-level url matching. If // it has a string, check the path string and query string in the given url // contain it or not. class CORE_EXPORT UrlMatcher final { … }; } // namespace blink #endif // THIRD_PARTY_BLINK_RENDERER_CORE_LOADER_URL_MATCHER_H_