use std_regex_provider everywhere

Author: anonrig

include/ada/url_pattern-inl.h

@@ -38,7 +38,7 @@ std::string url_pattern_component<regex_provider>::to_string() const {
 template <url_pattern_regex::regex_concept regex_provider>
 url_pattern_component_result
 url_pattern_component<regex_provider>::create_component_match_result(
-    std::string_view input, const std::smatch& exec_result) {
+    std::string_view input, const std::vector<std::string>& exec_result) {
   // Let result be a new URLPatternComponentResult.
   // Set result["input"] to input.
   // Let groups be a record<USVString, (USVString or undefined)>.
@@ -60,11 +60,9 @@ url_pattern_component<regex_provider>::create_component_match_result(
     // Let name be component’s group name list[index - 1].
     // Let value be Get(execResult, ToString(index)).
     // Set groups[name] to value.
-    auto exec = exec_result[index];
-    if (!exec.matched) continue;
     result.groups.insert({
         group_name_list[group_index],
-        exec.str(),
+        exec_result[index],
     });
 
     group_index++;

include/ada/url_pattern.h

@@ -9,7 +9,6 @@
 #include "ada/expected.h"
 #include "ada/url_pattern_regex.h"
 
-#include <regex>
 #include <string>
 #include <unordered_map>
 #include <variant>
@@ -232,7 +231,7 @@ class url_pattern_component {
 
   // @see https://urlpattern.spec.whatwg.org/#create-a-component-match-result
   url_pattern_component_result create_component_match_result(
-      std::string_view input, const std::smatch& exec_result);
+      std::string_view input, const std::vector<std::string>& exec_result);
 
   std::string to_string() const;
 

include/ada/url_pattern_regex.h

@@ -23,12 +23,12 @@ concept regex_concept = requires(T t, std::string_view pattern,
 
   // Function to perform regex search
   {
-    t.regex_search(input, std::declval<typename T::regex_type&>())
-  } -> std::same_as<std::vector<std::string>>;
+    T::regex_search(input, std::declval<typename T::regex_type&>())
+  } -> std::same_as<std::optional<std::vector<std::string>>>;
 
   // Function to match regex pattern
   {
-    t.regex_match(input, std::declval<typename T::regex_type&>())
+    T::regex_match(input, std::declval<typename T::regex_type&>())
   } -> std::same_as<bool>;
 
   // Copy constructor
@@ -44,9 +44,9 @@ class std_regex_provider {
   using regex_type = std::regex;
   static std::optional<regex_type> create_instance(std::string_view pattern,
                                                    bool ignore_case);
-  std::vector<std::string> regex_search(std::string_view input,
-                                        const regex_type& pattern);
-  bool regex_match(std::string_view input, const regex_type& pattern);
+  static std::optional<std::vector<std::string>> regex_search(
+      std::string_view input, const regex_type& pattern);
+  static bool regex_match(std::string_view input, const regex_type& pattern);
 };
 
 }  // namespace ada::url_pattern_regex

src/url_pattern.cpp

@@ -2,7 +2,6 @@
 
 #include <algorithm>
 #include <optional>
-#include <regex>
 #include <string>
 
 namespace ada {
@@ -678,60 +677,45 @@ result<std::optional<url_pattern_result>> url_pattern<regex_provider>::match(
     }
   }
 
-  auto regex_flags = std::regex_constants::match_any;
-
   // Let protocolExecResult be RegExpBuiltinExec(urlPattern’s protocol
   // component's regular expression, protocol).
-  std::smatch protocol_exec_result_value;
   auto protocol_exec_result =
-      std::regex_search(protocol, protocol_exec_result_value,
-                        protocol_component.regexp, regex_flags);
+      regex_provider::regex_search(protocol, protocol_component.regexp);
 
   // Let usernameExecResult be RegExpBuiltinExec(urlPattern’s username
   // component's regular expression, username).
-  std::smatch username_exec_result_value;
   auto username_exec_result =
-      std::regex_search(username, username_exec_result_value,
-                        username_component.regexp, regex_flags);
+      regex_provider::regex_search(username, username_component.regexp);
 
   // Let passwordExecResult be RegExpBuiltinExec(urlPattern’s password
   // component's regular expression, password).
-  std::smatch password_exec_result_value;
   auto password_exec_result =
-      std::regex_search(password, password_exec_result_value,
-                        password_component.regexp, regex_flags);
+      regex_provider::regex_search(password, password_component.regexp);
 
   // Let hostnameExecResult be RegExpBuiltinExec(urlPattern’s hostname
   // component's regular expression, hostname).
-  std::smatch hostname_exec_result_value;
   auto hostname_exec_result =
-      std::regex_search(hostname, hostname_exec_result_value,
-                        hostname_component.regexp, regex_flags);
+      regex_provider::regex_search(hostname, hostname_component.regexp);
 
   // Let portExecResult be RegExpBuiltinExec(urlPattern’s port component's
   // regular expression, port).
-  std::smatch port_exec_result_value;
-  auto port_exec_result = std::regex_search(port, port_exec_result_value,
-                                            port_component.regexp, regex_flags);
+  auto port_exec_result =
+      regex_provider::regex_search(port, port_component.regexp);
 
   // Let pathnameExecResult be RegExpBuiltinExec(urlPattern’s pathname
   // component's regular expression, pathname).
-  std::smatch pathname_exec_result_value;
   auto pathname_exec_result =
-      std::regex_search(pathname, pathname_exec_result_value,
-                        pathname_component.regexp, regex_flags);
+      regex_provider::regex_search(pathname, pathname_component.regexp);
 
   // Let searchExecResult be RegExpBuiltinExec(urlPattern’s search component's
   // regular expression, search).
-  std::smatch search_exec_result_value;
-  auto search_exec_result = std::regex_search(
-      search, search_exec_result_value, search_component.regexp, regex_flags);
+  auto search_exec_result =
+      regex_provider::regex_search(search, search_component.regexp);
 
   // Let hashExecResult be RegExpBuiltinExec(urlPattern’s hash component's
   // regular expression, hash).
-  std::smatch hash_exec_result_value;
-  auto hash_exec_result = std::regex_search(hash, hash_exec_result_value,
-                                            hash_component.regexp, regex_flags);
+  auto hash_exec_result =
+      regex_provider::regex_search(hash, hash_component.regexp);
 
   // If protocolExecResult, usernameExecResult, passwordExecResult,
   // hostnameExecResult, portExecResult, pathnameExecResult, searchExecResult,
@@ -749,42 +733,42 @@ result<std::optional<url_pattern_result>> url_pattern<regex_provider>::match(
   // Set result["protocol"] to the result of creating a component match result
   // given urlPattern’s protocol component, protocol, and protocolExecResult.
   result.protocol = protocol_component.create_component_match_result(
-      protocol, protocol_exec_result_value);
+      protocol, *protocol_exec_result);
 
   // Set result["username"] to the result of creating a component match result
   // given urlPattern’s username component, username, and usernameExecResult.
   result.username = username_component.create_component_match_result(
-      username, username_exec_result_value);
+      username, *username_exec_result);
 
   // Set result["password"] to the result of creating a component match result
   // given urlPattern’s password component, password, and passwordExecResult.
   result.password = password_component.create_component_match_result(
-      password, password_exec_result_value);
+      password, *password_exec_result);
 
   // Set result["hostname"] to the result of creating a component match result
   // given urlPattern’s hostname component, hostname, and hostnameExecResult.
   result.hostname = hostname_component.create_component_match_result(
-      hostname, hostname_exec_result_value);
+      hostname, *hostname_exec_result);
 
   // Set result["port"] to the result of creating a component match result given
   // urlPattern’s port component, port, and portExecResult.
-  result.port = port_component.create_component_match_result(
-      port, port_exec_result_value);
+  result.port =
+      port_component.create_component_match_result(port, *port_exec_result);
 
   // Set result["pathname"] to the result of creating a component match result
   // given urlPattern’s pathname component, pathname, and pathnameExecResult.
   result.pathname = pathname_component.create_component_match_result(
-      pathname, pathname_exec_result_value);
+      pathname, *pathname_exec_result);
 
   // Set result["search"] to the result of creating a component match result
   // given urlPattern’s search component, search, and searchExecResult.
   result.search = search_component.create_component_match_result(
-      search, search_exec_result_value);
+      search, *search_exec_result);
 
   // Set result["hash"] to the result of creating a component match result given
   // urlPattern’s hash component, hash, and hashExecResult.
-  result.hash = hash_component.create_component_match_result(
-      hash, hash_exec_result_value);
+  result.hash =
+      hash_component.create_component_match_result(hash, *hash_exec_result);
 
   return result;
 }

src/url_pattern_helpers.cpp

@@ -193,10 +193,11 @@ template <url_pattern_regex::regex_concept regex_provider>
 bool protocol_component_matches_special_scheme(
     url_pattern_component<regex_provider>& component) {
   auto regex = component.regexp;
-  // TODO: Use provider.regex_match
-  return std::regex_match("http", regex) || std::regex_match("https", regex) ||
-         std::regex_match("ws", regex) || std::regex_match("wss", regex) ||
-         std::regex_match("ftp", regex);
+  return regex_provider::regex_match("http", regex) ||
+         regex_provider::regex_match("https", regex) ||
+         regex_provider::regex_match("ws", regex) ||
+         regex_provider::regex_match("wss", regex) ||
+         regex_provider::regex_match("ftp", regex);
 }
 
 template <url_pattern_regex::regex_concept regex_provider>

src/url_pattern_regex.cpp

@@ -19,17 +19,22 @@ std::optional<std::regex> std_regex_provider::create_instance(
   }
 }
 
-std::vector<std::string> std_regex_provider::regex_search(
+std::optional<std::vector<std::string>> std_regex_provider::regex_search(
     std::string_view input, const std::regex& pattern) {
-  std::vector<std::string> matches;
   std::string input_str(
       input.begin(),
       input.end());  // Convert string_view to string for regex_search
   std::smatch match_result;
-  while (std::regex_search(input_str, match_result, pattern,
-                           std::regex_constants::match_any)) {
-    matches.push_back(match_result.str());
-    input_str = match_result.suffix().str();
+  if (!std::regex_search(input_str, match_result, pattern,
+                         std::regex_constants::match_any)) {
+    return std::nullopt;
+  }
+  std::vector<std::string> matches;
+  matches.reserve(match_result.size() - 1);
+  for (const auto& match : match_result) {
+    if (match.matched) {
+      matches.push_back(match.str());
+    }
   }
   return matches;
 }