From 025b4c410577a10692c608e7dbad712655abb858 Mon Sep 17 00:00:00 2001
From: Shawn Landden <shawn@churchofgit.com>
Date: Sat, 31 Oct 2015 20:39:15 -0700
Subject: utf8.[ch]: use char32_t and char16_t instead of int, int32_t, int16_t

rework C11 utf8.[ch] to use char32_t instead of uint32_t when referring
to unicode chars, to make things more expressive.
---
 src/basic/utf8.c | 35 +++++++++++++++++++----------------
 1 file changed, 19 insertions(+), 16 deletions(-)

(limited to 'src/basic/utf8.c')

diff --git a/src/basic/utf8.c b/src/basic/utf8.c
index 7600d99903..cc041d5b34 100644
--- a/src/basic/utf8.c
+++ b/src/basic/utf8.c
@@ -54,7 +54,7 @@
 #include "utf8.h"
 #include "util.h"
 
-bool unichar_is_valid(uint32_t ch) {
+bool unichar_is_valid(char32_t ch) {
 
         if (ch >= 0x110000) /* End of unicode space */
                 return false;
@@ -68,7 +68,7 @@ bool unichar_is_valid(uint32_t ch) {
         return true;
 }
 
-static bool unichar_is_control(uint32_t ch) {
+static bool unichar_is_control(char32_t ch) {
 
         /*
           0 to ' '-1 is the C0 range.
@@ -104,8 +104,9 @@ static int utf8_encoded_expected_len(const char *str) {
 }
 
 /* decode one unicode char */
-int utf8_encoded_to_unichar(const char *str) {
-        int unichar, len, i;
+char32_t utf8_encoded_to_unichar(const char *str) {
+       char32_t unichar;
+       int len, i;
 
         assert(str);
 
@@ -113,31 +114,31 @@ int utf8_encoded_to_unichar(const char *str) {
 
         switch (len) {
         case 1:
-                return (int)str[0];
+                return (char32_t)str[0];
         case 2:
                 unichar = str[0] & 0x1f;
                 break;
         case 3:
-                unichar = (int)str[0] & 0x0f;
+                unichar = (char32_t)str[0] & 0x0f;
                 break;
         case 4:
-                unichar = (int)str[0] & 0x07;
+                unichar = (char32_t)str[0] & 0x07;
                 break;
         case 5:
-                unichar = (int)str[0] & 0x03;
+                unichar = (char32_t)str[0] & 0x03;
                 break;
         case 6:
-                unichar = (int)str[0] & 0x01;
+                unichar = (char32_t)str[0] & 0x01;
                 break;
         default:
                 return -EINVAL;
         }
 
         for (i = 1; i < len; i++) {
-                if (((int)str[i] & 0xc0) != 0x80)
+                if (((char32_t)str[i] & 0xc0) != 0x80)
                         return -EINVAL;
                 unichar <<= 6;
-                unichar |= (int)str[i] & 0x3f;
+                unichar |= (char32_t)str[i] & 0x3f;
         }
 
         return unichar;
@@ -149,7 +150,8 @@ bool utf8_is_printable_newline(const char* str, size_t length, bool newline) {
         assert(str);
 
         for (p = str; length;) {
-                int encoded_len, val;
+                int encoded_len;
+                char32_t val;
 
                 encoded_len = utf8_encoded_valid_unichar(p);
                 if (encoded_len < 0 ||
@@ -277,7 +279,7 @@ char *ascii_is_valid(const char *str) {
  * Returns: The length in bytes that the UTF-8 representation does or would
  *          occupy.
  */
-size_t utf8_encode_unichar(char *out_utf8, uint32_t g) {
+size_t utf8_encode_unichar(char *out_utf8, char32_t g) {
 
         if (g < (1 << 7)) {
                 if (out_utf8)
@@ -321,7 +323,7 @@ char *utf16_to_utf8(const void *s, size_t length) {
         t = r;
 
         while (f < (const uint8_t*) s + length) {
-                uint16_t w1, w2;
+                char16_t w1, w2;
 
                 /* see RFC 2781 section 2.2 */
 
@@ -329,7 +331,7 @@ char *utf16_to_utf8(const void *s, size_t length) {
                 f += 2;
 
                 if (!utf16_is_surrogate(w1)) {
-                        t += utf8_encode_unichar(t, w1);
+                        t += utf8_encode_unichar(t, (char32_t) w1);
 
                         continue;
                 }
@@ -373,7 +375,8 @@ static int utf8_unichar_to_encoded_len(int unichar) {
 
 /* validate one encoded unicode char and return its length */
 int utf8_encoded_valid_unichar(const char *str) {
-        int len, unichar, i;
+        int len, i;
+        char32_t unichar;
 
         assert(str);
 
-- 
cgit v1.2.3-54-g00ecf