using ranges (#942)

Author: lemire

src/checkers.cpp

@@ -56,8 +56,8 @@ ada_really_inline constexpr bool is_ipv4(std::string_view view) noexcept {
   }
   // We have 0x followed by some characters, we need to check that they are
   // hexadecimals.
-  return std::all_of(view.begin() + 2, view.end(),
-                     ada::unicode::is_lowercase_hex);
+  view.remove_prefix(2);
+  return std::ranges::all_of(view, ada::unicode::is_lowercase_hex);
 }
 
 // for use with path_signature, we include all characters that need percent

src/parser.cpp

@@ -1,6 +1,7 @@
 #include "ada/parser-inl.h"
 
 #include <limits>
+#include <ranges>
 
 #include "ada/character_sets-inl.h"
 #include "ada/common_defs.h"
@@ -624,7 +625,7 @@ result_type parse_url_impl(std::string_view user_input,
         // to optimize it.
         if (view.ends_with(' ')) {
           std::string modified_view =
-              std::string(view.begin(), view.end() - 1) + "%20";
+              std::string(view.substr(0, view.size() - 1)) + "%20";
           url.update_base_pathname(unicode::percent_encode(
               modified_view, character_sets::C0_CONTROL_PERCENT_ENCODE));
         } else {

src/unicode.cpp

@@ -15,6 +15,8 @@ ADA_POP_DISABLE_WARNINGS
 #include <emmintrin.h>
 #endif
 
+#include <ranges>
+
 namespace ada::unicode {
 
 constexpr bool is_tabs_or_newline(char c) noexcept {
@@ -55,8 +57,7 @@ ada_really_inline bool has_tabs_or_newline(
     std::string_view user_input) noexcept {
   // first check for short strings in which case we do it naively.
   if (user_input.size() < 16) {  // slow path
-    return std::any_of(user_input.begin(), user_input.end(),
-                       is_tabs_or_newline);
+    return std::ranges::any_of(user_input, is_tabs_or_newline);
   }
   // fast path for long strings (expected to be common)
   size_t i = 0;
@@ -94,8 +95,7 @@ ada_really_inline bool has_tabs_or_newline(
     std::string_view user_input) noexcept {
   // first check for short strings in which case we do it naively.
   if (user_input.size() < 16) {  // slow path
-    return std::any_of(user_input.begin(), user_input.end(),
-                       is_tabs_or_newline);
+    return std::ranges::any_of(user_input, is_tabs_or_newline);
   }
   // fast path for long strings (expected to be common)
   size_t i = 0;
@@ -426,10 +426,9 @@ bool percent_encode(const std::string_view input, const uint8_t character_set[],
                     std::string& out) {
   ada_log("percent_encode ", input, " to output string while ",
           append ? "appending" : "overwriting");
-  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);
+  });
   ada_log("percent_encode done checking, moved to ",
           std::distance(input.begin(), pointer));
 

src/url.cpp

@@ -4,15 +4,15 @@
 
 #include <numeric>
 #include <algorithm>
+#include <ranges>
 #include <string>
 #include <string_view>
 
 namespace ada {
 
 bool url::parse_opaque_host(std::string_view input) {
   ada_log("parse_opaque_host ", input, " [", input.size(), " bytes]");
-  if (std::ranges::any_of(input.begin(), input.end(),
-                          ada::unicode::is_forbidden_host_code_point)) {
+  if (std::ranges::any_of(input, ada::unicode::is_forbidden_host_code_point)) {
     return is_valid = false;
   }
 

src/url_aggregator.cpp

@@ -8,6 +8,7 @@
 #include "ada/url_aggregator.h"
 #include "ada/url_aggregator-inl.h"
 
+#include <ranges>
 #include <string>
 #include <string_view>
 
@@ -227,7 +228,7 @@ bool url_aggregator::set_protocol(const std::string_view input) {
 
   if (pointer != view.end() && *pointer == ':') {
     return parse_scheme_with_colon<true>(
-        std::string_view(view.data(), pointer - view.begin() + 1));
+        view.substr(0, pointer - view.begin() + 1));
   }
   return false;
 }
@@ -489,8 +490,8 @@ ada_really_inline bool url_aggregator::parse_host(std::string_view input) {
   ada_log("parse_host to_ascii succeeded ", *host, " [", host->size(),
           " bytes]");
 
-  if (std::any_of(host.value().begin(), host.value().end(),
-                  ada::unicode::is_forbidden_domain_code_point)) {
+  if (std::ranges::any_of(host.value(),
+                          ada::unicode::is_forbidden_domain_code_point)) {
     return is_valid = false;
   }
 
@@ -1182,8 +1183,7 @@ bool url_aggregator::parse_opaque_host(std::string_view input) {
   ada_log("parse_opaque_host ", input, " [", input.size(), " bytes]");
   ADA_ASSERT_TRUE(validate());
   ADA_ASSERT_TRUE(!helpers::overlaps(input, buffer));
-  if (std::any_of(input.begin(), input.end(),
-                  ada::unicode::is_forbidden_host_code_point)) {
+  if (std::ranges::any_of(input, ada::unicode::is_forbidden_host_code_point)) {
     return is_valid = false;
   }