From 9e00ad7036c079c330cd103a21023b417a089f65 Mon Sep 17 00:00:00 2001
From: fallenoak <git@fallenoak.me>
Date: Sun, 22 Nov 2020 00:24:32 -0600
Subject: [PATCH] feat(unicode): add SUniSPutUTF8

---
 storm/Unicode.cpp | 54 +++++++++++++++++++++++++++++++++++++++++++++++
 storm/Unicode.hpp |  2 ++
 test/Unicode.cpp  | 29 +++++++++++++++++++++++++
 3 files changed, 85 insertions(+)

diff --git a/storm/Unicode.cpp b/storm/Unicode.cpp
index f5aa864..9c64d13 100644
--- a/storm/Unicode.cpp
+++ b/storm/Unicode.cpp
@@ -57,3 +57,57 @@ uint32_t SUniSGetUTF8(const uint8_t* strptr, int32_t* chars) {
 
     return value;
 }
+
+void SUniSPutUTF8(uint32_t c, char* strptr) {
+    if (!strptr) {
+        return;
+    }
+
+    auto curstr = strptr;
+    auto v3 = c;
+    char v4, v5, v6, v7;
+
+    if (c >= 0x80) {
+        if (c >= 0x800) {
+            if (c >= 0x10000) {
+                if (c >= 0x200000) {
+                    if (c >= 0x400000) {
+                        if (c >= 0x80000000) {
+                            *curstr = 0;
+                            return;
+                        }
+
+                        *strptr = (c >> 30) | 0xFC;
+                        curstr = strptr + 1;
+                        // TODO this seems likely to be bitwise right shift 24, not 8
+                        v7 = ((c >> 8) & 0x3F) | 0x80;
+                    } else {
+                        // TODO this seems likely to be bitwise right shift 24, not 8
+                        v7 = (c >> 8) | 0xF8;
+                    }
+
+                    *curstr++ = v7;
+                    v6 = ((c >> 18) & 0x3F) | 0x80;
+                } else {
+                    v6 = (c >> 18) | 0xF0;
+                }
+
+                *curstr++ = v6;
+                v5 = ((c >> 12) & 0x3F) | 0x80;
+            } else {
+                v5 = (c >> 12) | 0xE0;
+            }
+
+            *curstr++ = v5;
+            v4 = ((c >> 6) & 0x3F) | 0x80;
+        } else {
+            v4 = (c >> 6) | 0xC0;
+        }
+
+        *curstr++ = v4;
+        v3 = (c & 0x3F) | 0x80;
+    }
+
+    *curstr++ = v3;
+    *curstr = '\0';
+}
diff --git a/storm/Unicode.hpp b/storm/Unicode.hpp
index cdccd1d..50f703d 100644
--- a/storm/Unicode.hpp
+++ b/storm/Unicode.hpp
@@ -5,4 +5,6 @@
 
 uint32_t SUniSGetUTF8(const uint8_t* strptr, int32_t* chars);
 
+void SUniSPutUTF8(uint32_t c, char* strptr);
+
 #endif
diff --git a/test/Unicode.cpp b/test/Unicode.cpp
index e951c5c..3e61ec3 100644
--- a/test/Unicode.cpp
+++ b/test/Unicode.cpp
@@ -1,4 +1,5 @@
 #include "storm/Unicode.hpp"
+#include "storm/String.hpp"
 #include "test/Test.hpp"
 
 TEST_CASE("SUniSGetUTF8", "[unicode]") {
@@ -30,3 +31,31 @@ TEST_CASE("SUniSGetUTF8", "[unicode]") {
         REQUIRE(chars == 0);
     }
 }
+
+TEST_CASE("SUniSPutUTF8", "[unicode]") {
+    SECTION("writes ascii-range utf-8 first character") {
+        auto code = 'f';
+        char buffer[100] = { 0 };
+        SUniSPutUTF8(code, buffer);
+
+        REQUIRE(SStrLen(buffer) == 1);
+        REQUIRE(!SStrCmp(buffer, "f", SStrLen(buffer)));
+    }
+
+    SECTION("writes non-ascii-range utf-8 first character") {
+        auto code = 0x1F642;
+        char buffer[100] = { 0 };
+        SUniSPutUTF8(code, buffer);
+
+        REQUIRE(SStrLen(buffer) == 4);
+        REQUIRE(!SStrCmp(buffer, "\xF0\x9F\x99\x82", SStrLen(buffer)));
+    }
+
+    SECTION("writes null first character") {
+        auto code = '\0';
+        char buffer[100] = { 0 };
+        SUniSPutUTF8(code, buffer);
+
+        REQUIRE(SStrLen(buffer) == 0);
+    }
+}