From a28dfdc394dad6e7aee165cd7dbe1ab18a87b42b Mon Sep 17 00:00:00 2001
From: jaime-m-p <>
Date: Fri, 17 May 2024 19:04:08 +0200
Subject: [PATCH] Using range_nfd instead of std::tuple

---
 scripts/gen-unicode-data.py | 2 +-
 unicode-data.cpp            | 2 +-
 unicode-data.h              | 8 +++++++-
 unicode.cpp                 | 9 ++++-----
 4 files changed, 13 insertions(+), 8 deletions(-)

diff --git a/scripts/gen-unicode-data.py b/scripts/gen-unicode-data.py
index 06fab5372..744873c2a 100644
--- a/scripts/gen-unicode-data.py
+++ b/scripts/gen-unicode-data.py
@@ -128,7 +128,7 @@ for tuple in table_uppercase:
     out("{0x%06X, 0x%06X}," % tuple)
 out("};\n")
 
-out("const std::vector<std::tuple<uint32_t, uint32_t, uint32_t>> unicode_ranges_nfd = {  // start, last, nfd")
+out("const std::vector<range_nfd> unicode_ranges_nfd = {  // start, last, nfd")
 for triple in ranges_nfd:
     out("{0x%06X, 0x%06X, 0x%06X}," % triple)
 out("};\n")
diff --git a/unicode-data.cpp b/unicode-data.cpp
index 007161f1b..d7c1c898d 100644
--- a/unicode-data.cpp
+++ b/unicode-data.cpp
@@ -5150,7 +5150,7 @@ const std::unordered_map<uint32_t, uint32_t> unicode_map_uppercase = {
 {0x01E943, 0x01E921},
 };
 
-const std::vector<std::tuple<uint32_t, uint32_t, uint32_t>> unicode_ranges_nfd = {  // start, last, nfd
+const std::vector<range_nfd> unicode_ranges_nfd = {  // start, last, nfd
 {0x000000, 0x000000, 0x000000},
 {0x0000C0, 0x0000C5, 0x000041},
 {0x0000C7, 0x0000C7, 0x000043},
diff --git a/unicode-data.h b/unicode-data.h
index ce26748d0..e27fe1770 100644
--- a/unicode-data.h
+++ b/unicode-data.h
@@ -5,10 +5,16 @@
 #include <unordered_map>
 #include <unordered_set>
 
+struct range_nfd {
+    uint32_t first;
+    uint32_t last;
+    uint32_t nfd;
+};
+
 static const uint32_t MAX_CODEPOINTS = 0x110000;
 
 extern const std::vector<std::pair<uint32_t, uint16_t>> unicode_ranges_flags;
 extern const std::unordered_set<uint32_t> unicode_set_whitespace;
 extern const std::unordered_map<uint32_t, uint32_t> unicode_map_lowercase;
 extern const std::unordered_map<uint32_t, uint32_t> unicode_map_uppercase;
-extern const std::vector<std::tuple<uint32_t, uint32_t, uint32_t>> unicode_ranges_nfd;
+extern const std::vector<range_nfd> unicode_ranges_nfd;
diff --git a/unicode.cpp b/unicode.cpp
index e47d1cc2e..056a4c741 100644
--- a/unicode.cpp
+++ b/unicode.cpp
@@ -135,7 +135,7 @@ static std::vector<codepoint_flags> unicode_cpt_flags_array() {
     }
 
     for (auto &range : unicode_ranges_nfd) {  // start, last, nfd
-        cpt_flags[std::get<2>(range)].is_nfd = true;
+        cpt_flags[range.nfd].is_nfd = true;
     }
 
     return cpt_flags;
@@ -580,15 +580,14 @@ std::string unicode_cpt_to_utf8(uint32_t cp) {
 }
 
 std::vector<uint32_t> unicode_cpts_normalize_nfd(const std::vector<uint32_t> & cpts) {
-    // unicode_ranges_nfd[i] -> tuple(first, last, nfd)
-    auto comp = +[] (const uint32_t cpt, const decltype(unicode_ranges_nfd)::value_type & triple) {
-        return cpt < std::get<0>(triple);
+    auto comp = [] (const uint32_t cpt, const range_nfd & range) {
+        return cpt < range.first;
     };
     std::vector<uint32_t> result(cpts.size());
     for (size_t i = 0; i < cpts.size(); ++i) {
         const uint32_t cpt = cpts[i];
         auto it = std::upper_bound(unicode_ranges_nfd.cbegin(), unicode_ranges_nfd.cend(), cpt, comp) - 1;
-        result[i] = (std::get<0>(*it) <= cpt && cpt <= std::get<1>(*it)) ? std::get<2>(*it) : cpt;
+        result[i] = (it->first <= cpt && cpt <= it->last) ? it->nfd : cpt;
     }
     return result;
 }