more c++20 changes

include/ada/helpers.h

@@ -12,6 +12,10 @@
 #include <string_view>
 #include <optional>
 
+#if ADA_DEVELOPMENT_CHECKS
+#include <iostream>
+#endif // ADA_DEVELOPMENT_CHECKS
+
 /**
  * These functions are not part of our public API and may
  * change at any time.
@@ -128,7 +132,7 @@ ada_really_inline void resize(std::string_view& input, size_t pos) noexcept;
  * and whether a colon was found outside brackets. Used by the host parser.
  */
 ada_really_inline std::pair<size_t, bool> get_host_delimiter_location(
- const bool is_special, std::string_view& view) noexcept;
+ bool is_special, std::string_view& view) noexcept;
 
 /**
  * @private

include/ada/log.h

@@ -7,12 +7,15 @@
 #define ADA_LOG_H
 #include "ada/common_defs.h"
 
-#include <iostream>
 // To enable logging, set ADA_LOGGING to 1:
 #ifndef ADA_LOGGING
 #define ADA_LOGGING 0
 #endif
 
+#if ADA_LOGGING
+#include <iostream>
+#endif // ADA_LOGGING
+
 namespace ada {
 
 /**

include/ada/scheme.h

@@ -8,7 +8,6 @@
 #include "ada/common_defs.h"
 
 #include <array>
-#include <optional>
 #include <string>
 
 /**

include/ada/serializers.h

@@ -8,7 +8,6 @@
 #include "ada/common_defs.h"
 
 #include <array>
-#include <optional>
 #include <string>
 
 /**

include/ada/unicode-inl.h

@@ -19,8 +19,7 @@ namespace ada::unicode {
 ada_really_inline size_t percent_encode_index(const std::string_view input,
  const uint8_t character_set[]) {
  return std::distance(
- input.begin(),
- std::find_if(input.begin(), input.end(), [character_set](const char c) {
+ input.begin(), std::ranges::find_if(input, [character_set](const char c) {
  return character_sets::bit_at(character_set, c);
  }));
 }

include/ada/url-inl.h

@@ -133,7 +133,7 @@ inline void url::update_base_search(std::string_view input,
 }
 
 inline void url::update_base_search(std::optional<std::string> input) {
- query = input;
+ query = std::move(input);
 }
 
 inline void url::update_base_pathname(const std::string_view input) {
@@ -232,7 +232,7 @@ ada_really_inline size_t url::parse_port(std::string_view view,
  return 0;
  }
  ada_log("parse_port: ", parsed_port);
- const size_t consumed = size_t(r.ptr - view.data());
+ const auto consumed = size_t(r.ptr - view.data());
  ada_log("parse_port: consumed ", consumed);
  if (check_trailing_content) {
  is_valid &=
@@ -245,9 +245,8 @@ ada_really_inline size_t url::parse_port(std::string_view view,
  auto default_port = scheme_default_port();
  bool is_port_valid = (default_port == 0 && parsed_port == 0) ||
  (default_port != parsed_port);
- port = (r.ec == std::errc() && is_port_valid)
- ? std::optional<uint16_t>(parsed_port)
- : std::nullopt;
+ port = (r.ec == std::errc() && is_port_valid) ? std::optional(parsed_port)
+ : std::nullopt;
  }
  return consumed;
 }

include/ada/url_search_params-inl.h

@@ -193,7 +193,7 @@ inline url_search_params_entries_iter url_search_params::get_entries() {
 }
 
 template <typename T, url_search_params_iter_type Type>
-inline bool url_search_params_iter<T, Type>::has_next() {
+inline bool url_search_params_iter<T, Type>::has_next() const {
  return pos < params.params.size();
 }
 

include/ada/url_search_params.h

@@ -42,7 +42,9 @@ struct url_search_params {
  * @see
  * https://github.com/web-platform-tests/wpt/blob/master/url/urlsearchparams-constructor.any.js
  */
- url_search_params(const std::string_view input) { initialize(input); }
+ explicit url_search_params(const std::string_view input) {
+ initialize(input);
+ }
 
  url_search_params(const url_search_params &u) = default;
  url_search_params(url_search_params &&u) noexcept = default;
@@ -172,7 +174,7 @@ struct url_search_params_iter {
  */
  inline std::optional<T> next();
 
- inline bool has_next();
+ inline bool has_next() const;
 
  private:
  static url_search_params EMPTY;

src/checkers.cpp

@@ -11,15 +11,14 @@ ada_really_inline ada_constexpr bool is_ipv4(std::string_view view) noexcept {
  // with 'x' or a lowercase hex character.
  // Most of the time, this will be false so this simple check will save a lot
  // of effort.
- char last_char = view.back();
  // If the address ends with a dot, we need to prune it (special case).
- if (last_char == '.') {
+ if (view.ends_with('.')) {
  view.remove_suffix(1);
  if (view.empty()) {
  return false;
  }
- last_char = view.back();
  }
+ char last_char = view.back();
  bool possible_ipv4 = (last_char >= '0' && last_char <= '9') ||
  (last_char >= 'a' && last_char <= 'f') ||
  last_char == 'x';
@@ -35,7 +34,7 @@ ada_really_inline ada_constexpr bool is_ipv4(std::string_view view) noexcept {
  /** Optimization opportunity: we have basically identified the last number of
  the ipv4 if we return true here. We might as well parse it and have at
  least one number parsed when we get to parse_ipv4. */
- if (std::all_of(view.begin(), view.end(), ada::checkers::is_digit)) {
+ if (std::ranges::all_of(view, ada::checkers::is_digit)) {
  return true;
  }
  // It could be hex (0x), but not if there is a single character.

src/helpers.cpp

@@ -96,13 +96,11 @@ ada_really_inline std::optional<std::string_view> prune_hash(
 
 ada_really_inline bool shorten_path(std::string& path,
  ada::scheme::type type) noexcept {
- size_t first_delimiter = path.find_first_of('/', 1);
-
  // Let path be url's path.
  // If url's scheme is "file", path's size is 1, and path[0] is a normalized
  // Windows drive letter, then return.
  if (type == ada::scheme::type::FILE &&
- first_delimiter == std::string_view::npos && !path.empty()) {
+ path.find('/', 1) == std::string_view::npos && !path.empty()) {
  if (checkers::is_normalized_windows_drive_letter(
  helpers::substring(path, 1))) {
  return false;
@@ -121,13 +119,11 @@ ada_really_inline bool shorten_path(std::string& path,
 
 ada_really_inline bool shorten_path(std::string_view& path,
  ada::scheme::type type) noexcept {
- size_t first_delimiter = path.find_first_of('/', 1);
-
  // Let path be url's path.
  // If url's scheme is "file", path's size is 1, and path[0] is a normalized
  // Windows drive letter, then return.
  if (type == ada::scheme::type::FILE &&
- first_delimiter == std::string_view::npos && !path.empty()) {
+ path.find('/', 1) == std::string_view::npos && !path.empty()) {
  if (checkers::is_normalized_windows_drive_letter(
  helpers::substring(path, 1))) {
  return false;
@@ -150,11 +146,7 @@ ada_really_inline void remove_ascii_tab_or_newline(
  std::string& input) noexcept {
  // if this ever becomes a performance issue, we could use an approach similar
  // to has_tabs_or_newline
- input.erase(std::remove_if(input.begin(), input.end(),
- [](char c) {
- return ada::unicode::is_ascii_tab_or_newline(c);
- }),
- input.end());
+ std::erase_if(input, ada::unicode::is_ascii_tab_or_newline);
 }
 
 ada_really_inline constexpr std::string_view substring(std::string_view input,

src/parser.cpp

@@ -238,10 +238,9 @@ result_type parse_url_impl(std::string_view user_input,
  // TODO: We could do various processing early on, using a single pass
  // over the string to collect information about it, e.g., telling us
  // whether there is a @ and if so, where (or how many).
- const bool contains_ampersand =
- (url_data.find('@', input_position) != std::string_view::npos);
 
- if (!contains_ampersand) {
+ // Check if url data contains an @.
+ if (url_data.find('@', input_position) == std::string_view::npos) {
  state = ada::state::HOST;
  break;
  }
@@ -253,12 +252,12 @@ result_type parse_url_impl(std::string_view user_input,
  * --------^
  */
  do {
- std::string_view view = helpers::substring(url_data, input_position);
+ std::string_view view = url_data.substr(input_position);
  // The delimiters are @, /, ? \\.
  size_t location =
  url.is_special() ? helpers::find_authority_delimiter_special(view)
  : helpers::find_authority_delimiter(view);
- std::string_view authority_view(view.data(), location);
+ std::string_view authority_view = view.substr(0, location);
  size_t end_of_authority = input_position + authority_view.size();
  // If c is U+0040 (@), then:
  if ((end_of_authority != input_size) &&
@@ -362,8 +361,7 @@ result_type parse_url_impl(std::string_view user_input,
  // If c is U+002F (/) and remaining starts with U+002F (/),
  // then set state to special authority ignore slashes state and increase
  // pointer by 1.
- std::string_view view = helpers::substring(url_data, input_position);
- if (view.starts_with("//")) {
+ if (url_data.substr(input_position, 2) == "//") {
  state = ada::state::SPECIAL_AUTHORITY_IGNORE_SLASHES;
  input_position += 2;
  } else {
@@ -458,7 +456,7 @@ result_type parse_url_impl(std::string_view user_input,
  } else {
  std::string_view path = url.get_pathname();
  if (helpers::shorten_path(path, url.type)) {
- url.update_base_pathname(std::string(path));
+ url.update_base_pathname(std::move(std::string(path)));
  }
  }
  // Set state to path state and decrease pointer by 1.
@@ -519,8 +517,7 @@ result_type parse_url_impl(std::string_view user_input,
  // If c is U+002F (/) and remaining starts with U+002F (/),
  // then set state to special authority ignore slashes state and increase
  // pointer by 1.
- std::string_view view = helpers::substring(url_data, input_position);
- if (view.starts_with("//")) {
+ if (url_data.substr(input_position, 2) == "//") {
  input_position += 2;
  }
 
@@ -553,7 +550,7 @@ result_type parse_url_impl(std::string_view user_input,
 
  // Percent-encode after encoding, with encoding, buffer, and
  // queryPercentEncodeSet, and append the result to url's query.
- url.update_base_search(helpers::substring(url_data, input_position),
+ url.update_base_search(url_data.substr(input_position),
  query_percent_encode_set);
  ada_log("QUERY update_base_search completed ");
  if (fragment.has_value()) {
@@ -565,8 +562,7 @@ result_type parse_url_impl(std::string_view user_input,
  case ada::state::HOST: {
  ada_log("HOST ", helpers::substring(url_data, input_position));
 
- std::string_view host_view =
- helpers::substring(url_data, input_position);
+ std::string_view host_view = url_data.substr(input_position);
  auto [location, found_colon] =
  helpers::get_host_delimiter_location(url.is_special(), host_view);
  input_position = (location != std::string_view::npos)
@@ -597,7 +593,7 @@ result_type parse_url_impl(std::string_view user_input,
  else {
  // If url is special and host_view is the empty string, validation
  // error, return failure.
- if (url.is_special() && host_view.empty()) {
+ if (host_view.empty() && url.is_special()) {
  url.is_valid = false;
  return url;
  }
@@ -620,7 +616,7 @@ result_type parse_url_impl(std::string_view user_input,
  }
  case ada::state::OPAQUE_PATH: {
  ada_log("OPAQUE_PATH ", helpers::substring(url_data, input_position));
- std::string_view view = helpers::substring(url_data, input_position);
+ std::string_view view = url_data.substr(input_position);
  // If c is U+003F (?), then set url's query to the empty string and
  // state to query state.
  size_t location = view.find('?');
@@ -640,10 +636,8 @@ result_type parse_url_impl(std::string_view user_input,
  }
  case ada::state::PORT: {
  ada_log("PORT ", helpers::substring(url_data, input_position));
- std::string_view port_view =
- helpers::substring(url_data, input_position);
- size_t consumed_bytes = url.parse_port(port_view, true);
- input_position += consumed_bytes;
+ std::string_view port_view = url_data.substr(input_position);
+ input_position += url.parse_port(port_view, true);
  if (!url.is_valid) {
  return url;
  }
@@ -698,8 +692,8 @@ result_type parse_url_impl(std::string_view user_input,
  break;
  }
  case ada::state::PATH: {
- std::string_view view = helpers::substring(url_data, input_position);
  ada_log("PATH ", helpers::substring(url_data, input_position));
+ std::string_view view = url_data.substr(input_position);
 
  // Most time, we do not need percent encoding.
  // Furthermore, we can immediately locate the '?'.
@@ -752,7 +746,7 @@ result_type parse_url_impl(std::string_view user_input,
  // url's path.
  if (!base_url->get_pathname().empty()) {
  if (!checkers::is_windows_drive_letter(
- helpers::substring(url_data, input_position))) {
+ url_data.substr(input_position))) {
  std::string_view first_base_url_path =
  base_url->get_pathname().substr(1);
  size_t loc = first_base_url_path.find('/');
@@ -780,8 +774,8 @@ result_type parse_url_impl(std::string_view user_input,
  break;
  }
  case ada::state::FILE_HOST: {
- std::string_view view = helpers::substring(url_data, input_position);
  ada_log("FILE_HOST ", helpers::substring(url_data, input_position));
+ std::string_view view = url_data.substr(input_position);
 
  size_t location = view.find_first_of("/\\?");
  std::string_view file_host_buffer(
@@ -827,8 +821,7 @@ result_type parse_url_impl(std::string_view user_input,
  }
  case ada::state::FILE: {
  ada_log("FILE ", helpers::substring(url_data, input_position));
- std::string_view file_view =
- helpers::substring(url_data, input_position);
+ std::string_view file_view = url_data.substr(input_position);
 
  url.set_protocol_as_file();
  if constexpr (result_type_is_ada_url) {
@@ -881,7 +874,7 @@ result_type parse_url_impl(std::string_view user_input,
  } else {
  std::string_view path = url.get_pathname();
  if (helpers::shorten_path(path, url.type)) {
- url.update_base_pathname(std::string(path));
+ url.update_base_pathname(std::move(std::string(path)));
  }
  }
  }

src/unicode.cpp

@@ -372,7 +372,6 @@ std::string percent_decode(const std::string_view input, size_t first_percent) {
  !is_ascii_hex_digit(pointer[2])))) {
  dest += ch;
  pointer++;
- continue;
  } else {
  unsigned a = convert_hex_to_binary(pointer[1]);
  unsigned b = convert_hex_to_binary(pointer[2]);
@@ -386,10 +385,9 @@ std::string percent_decode(const std::string_view input, size_t first_percent) {
 
 std::string percent_encode(const std::string_view input,
  const uint8_t character_set[]) {
- auto pointer =
- std::find_if(input.begin(), input.end(), [character_set](const char c) {
- return character_sets::bit_at(character_set, c);
- });
+ auto pointer = std::ranges::find_if(input, [character_set](const char c) {
+ return character_sets::bit_at(character_set, c);
+ });
  // Optimization: Don't iterate if percent encode is not required
  if (pointer == input.end()) {
  return std::string(input);