diff --git a/deps/ntlmclient/utf8.h b/deps/ntlmclient/utf8.h
index a26ae85c3..5f02b5555 100644
--- a/deps/ntlmclient/utf8.h
+++ b/deps/ntlmclient/utf8.h
@@ -1,30 +1,30 @@
-// The latest version of this library is available on GitHub;
-// https://github.com/sheredom/utf8.h
+/* The latest version of this library is available on GitHub;
+ * https://github.com/sheredom/utf8.h */
 
-// This is free and unencumbered software released into the public domain.
-//
-// Anyone is free to copy, modify, publish, use, compile, sell, or
-// distribute this software, either in source code form or as a compiled
-// binary, for any purpose, commercial or non-commercial, and by any
-// means.
-//
-// In jurisdictions that recognize copyright laws, the author or authors
-// of this software dedicate any and all copyright interest in the
-// software to the public domain. We make this dedication for the benefit
-// of the public at large and to the detriment of our heirs and
-// successors. We intend this dedication to be an overt act of
-// relinquishment in perpetuity of all present and future rights to this
-// software under copyright law.
-//
-// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
-// EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
-// MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
-// IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR
-// OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
-// ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
-// OTHER DEALINGS IN THE SOFTWARE.
-//
-// For more information, please refer to <http://unlicense.org/>
+/* This is free and unencumbered software released into the public domain.
+ *
+ * Anyone is free to copy, modify, publish, use, compile, sell, or
+ * distribute this software, either in source code form or as a compiled
+ * binary, for any purpose, commercial or non-commercial, and by any
+ * means.
+ *
+ * In jurisdictions that recognize copyright laws, the author or authors
+ * of this software dedicate any and all copyright interest in the
+ * software to the public domain. We make this dedication for the benefit
+ * of the public at large and to the detriment of our heirs and
+ * successors. We intend this dedication to be an overt act of
+ * relinquishment in perpetuity of all present and future rights to this
+ * software under copyright law.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+ * IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * For more information, please refer to <http://unlicense.org/> */
 
 #ifndef SHEREDOM_UTF8_H_INCLUDED
 #define SHEREDOM_UTF8_H_INCLUDED
@@ -32,7 +32,14 @@
 #if defined(_MSC_VER)
 #pragma warning(push)
 
-// disable 'bytes padding added after construct' warning
+/* disable warning: no function prototype given: converting '()' to '(void)' */
+#pragma warning(disable : 4255)
+
+/* disable warning: '__cplusplus' is not defined as a preprocessor macro,
+ * replacing with '0' for '#if/#elif' */
+#pragma warning(disable : 4668)
+
+/* disable warning: bytes padding added after construct */
 #pragma warning(disable : 4820)
 #endif
 
@@ -43,7 +50,7 @@
 #pragma warning(pop)
 #endif
 
-#if defined(_MSC_VER)
+#if defined(_MSC_VER) && (_MSC_VER < 1920)
 typedef __int32 utf8_int32_t;
 #else
 #include <stdint.h>
@@ -54,24 +61,39 @@ typedef int32_t utf8_int32_t;
 #pragma clang diagnostic push
 #pragma clang diagnostic ignored "-Wold-style-cast"
 #pragma clang diagnostic ignored "-Wcast-qual"
+
+#if __has_warning("-Wunsafe-buffer-usage")
+#pragma clang diagnostic ignored "-Wunsafe-buffer-usage"
+#endif
 #endif
 
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-#if defined(__clang__) || defined(__GNUC__)
-#define utf8_nonnull __attribute__((nonnull))
-#define utf8_pure __attribute__((pure))
-#define utf8_restrict __restrict__
-#define utf8_weak __attribute__((weak))
-#elif defined(_MSC_VER)
+#if defined(__TINYC__)
+#define UTF8_ATTRIBUTE(a) __attribute((a))
+#else
+#define UTF8_ATTRIBUTE(a) __attribute__((a))
+#endif
+
+#if defined(_MSC_VER)
 #define utf8_nonnull
 #define utf8_pure
 #define utf8_restrict __restrict
 #define utf8_weak __inline
+#elif defined(__clang__) || defined(__GNUC__)
+#define utf8_nonnull UTF8_ATTRIBUTE(nonnull)
+#define utf8_pure UTF8_ATTRIBUTE(pure)
+#define utf8_restrict __restrict__
+#define utf8_weak UTF8_ATTRIBUTE(weak)
+#elif defined(__TINYC__)
+#define utf8_nonnull UTF8_ATTRIBUTE(nonnull)
+#define utf8_pure UTF8_ATTRIBUTE(pure)
+#define utf8_restrict
+#define utf8_weak UTF8_ATTRIBUTE(weak)
 #else
-#error Non clang, non gcc, non MSVC compiler found!
+#error Non clang, non gcc, non MSVC, non tcc compiler found!
 #endif
 
 #ifdef __cplusplus
@@ -80,385 +102,475 @@ extern "C" {
 #define utf8_null 0
 #endif
 
-// Return less than 0, 0, greater than 0 if src1 < src2, src1 == src2, src1 >
-// src2 respectively, case insensitive.
-utf8_nonnull utf8_pure utf8_weak int utf8casecmp(const void *src1,
-                                                 const void *src2);
+#if (defined(__cplusplus) && __cplusplus >= 201402L)
+#define utf8_constexpr14 constexpr
+#define utf8_constexpr14_impl constexpr
+#else
+/* constexpr and weak are incompatible. so only enable one of them */
+#define utf8_constexpr14 utf8_weak
+#define utf8_constexpr14_impl
+#endif
 
-// Append the utf8 string src onto the utf8 string dst.
-utf8_nonnull utf8_weak void *utf8cat(void *utf8_restrict dst,
-                                     const void *utf8_restrict src);
+#if defined(__cplusplus) && __cplusplus >= 202002L
+using utf8_int8_t = char8_t; /* Introduced in C++20 */
+#else
+typedef char utf8_int8_t;
+#endif
 
-// Find the first match of the utf8 codepoint chr in the utf8 string src.
-utf8_nonnull utf8_pure utf8_weak void *utf8chr(const void *src,
-                                               utf8_int32_t chr);
+/* Return less than 0, 0, greater than 0 if src1 < src2, src1 == src2, src1 >
+ * src2 respectively, case insensitive. */
+utf8_constexpr14 utf8_nonnull utf8_pure int
+utf8casecmp(const utf8_int8_t *src1, const utf8_int8_t *src2);
 
-// Return less than 0, 0, greater than 0 if src1 < src2,
-// src1 == src2, src1 > src2 respectively.
-utf8_nonnull utf8_pure utf8_weak int utf8cmp(const void *src1,
-                                             const void *src2);
+/* Append the utf8 string src onto the utf8 string dst. */
+utf8_nonnull utf8_weak utf8_int8_t *
+utf8cat(utf8_int8_t *utf8_restrict dst, const utf8_int8_t *utf8_restrict src);
 
-// Copy the utf8 string src onto the memory allocated in dst.
-utf8_nonnull utf8_weak void *utf8cpy(void *utf8_restrict dst,
-                                     const void *utf8_restrict src);
+/* Find the first match of the utf8 codepoint chr in the utf8 string src. */
+utf8_constexpr14 utf8_nonnull utf8_pure utf8_int8_t *
+utf8chr(const utf8_int8_t *src, utf8_int32_t chr);
 
-// Number of utf8 codepoints in the utf8 string src that consists entirely
-// of utf8 codepoints not from the utf8 string reject.
-utf8_nonnull utf8_pure utf8_weak size_t utf8cspn(const void *src,
-                                                 const void *reject);
+/* Return less than 0, 0, greater than 0 if src1 < src2,
+ * src1 == src2, src1 > src2 respectively. */
+utf8_constexpr14 utf8_nonnull utf8_pure int utf8cmp(const utf8_int8_t *src1,
+                                                    const utf8_int8_t *src2);
 
-// Duplicate the utf8 string src by getting its size, malloc'ing a new buffer
-// copying over the data, and returning that. Or 0 if malloc failed.
-utf8_nonnull utf8_weak void *utf8dup(const void *src);
+/* Copy the utf8 string src onto the memory allocated in dst. */
+utf8_nonnull utf8_weak utf8_int8_t *
+utf8cpy(utf8_int8_t *utf8_restrict dst, const utf8_int8_t *utf8_restrict src);
 
-// Number of utf8 codepoints in the utf8 string str,
-// excluding the null terminating byte.
-utf8_nonnull utf8_pure utf8_weak size_t utf8len(const void *str);
+/* Number of utf8 codepoints in the utf8 string src that consists entirely
+ * of utf8 codepoints not from the utf8 string reject. */
+utf8_constexpr14 utf8_nonnull utf8_pure size_t
+utf8cspn(const utf8_int8_t *src, const utf8_int8_t *reject);
 
-// Return less than 0, 0, greater than 0 if src1 < src2, src1 == src2, src1 >
-// src2 respectively, case insensitive. Checking at most n bytes of each utf8
-// string.
-utf8_nonnull utf8_pure utf8_weak int utf8ncasecmp(const void *src1,
-                                                  const void *src2, size_t n);
+/* Duplicate the utf8 string src by getting its size, malloc'ing a new buffer
+ * copying over the data, and returning that. Or 0 if malloc failed. */
+utf8_weak utf8_int8_t *utf8dup(const utf8_int8_t *src);
 
-// Append the utf8 string src onto the utf8 string dst,
-// writing at most n+1 bytes. Can produce an invalid utf8
-// string if n falls partway through a utf8 codepoint.
-utf8_nonnull utf8_weak void *utf8ncat(void *utf8_restrict dst,
-                                      const void *utf8_restrict src, size_t n);
+/* Number of utf8 codepoints in the utf8 string str,
+ * excluding the null terminating byte. */
+utf8_constexpr14 utf8_nonnull utf8_pure size_t utf8len(const utf8_int8_t *str);
 
-// Return less than 0, 0, greater than 0 if src1 < src2,
-// src1 == src2, src1 > src2 respectively. Checking at most n
-// bytes of each utf8 string.
-utf8_nonnull utf8_pure utf8_weak int utf8ncmp(const void *src1,
-                                              const void *src2, size_t n);
+/* Similar to utf8len, except that only at most n bytes of src are looked. */
+utf8_constexpr14 utf8_nonnull utf8_pure size_t utf8nlen(const utf8_int8_t *str,
+                                                        size_t n);
 
-// Copy the utf8 string src onto the memory allocated in dst.
-// Copies at most n bytes. If there is no terminating null byte in
-// the first n bytes of src, the string placed into dst will not be
-// null-terminated. If the size (in bytes) of src is less than n,
-// extra null terminating bytes are appended to dst such that at
-// total of n bytes are written. Can produce an invalid utf8
-// string if n falls partway through a utf8 codepoint.
-utf8_nonnull utf8_weak void *utf8ncpy(void *utf8_restrict dst,
-                                      const void *utf8_restrict src, size_t n);
+/* Return less than 0, 0, greater than 0 if src1 < src2, src1 == src2, src1 >
+ * src2 respectively, case insensitive. Checking at most n bytes of each utf8
+ * string. */
+utf8_constexpr14 utf8_nonnull utf8_pure int
+utf8ncasecmp(const utf8_int8_t *src1, const utf8_int8_t *src2, size_t n);
 
-// Similar to utf8dup, except that at most n bytes of src are copied. If src is
-// longer than n, only n bytes are copied and a null byte is added.
-//
-// Returns a new string if successful, 0 otherwise
-utf8_nonnull utf8_weak void *utf8ndup(const void *src, size_t n);
+/* Append the utf8 string src onto the utf8 string dst,
+ * writing at most n+1 bytes. Can produce an invalid utf8
+ * string if n falls partway through a utf8 codepoint. */
+utf8_nonnull utf8_weak utf8_int8_t *
+utf8ncat(utf8_int8_t *utf8_restrict dst, const utf8_int8_t *utf8_restrict src,
+         size_t n);
 
-// Locates the first occurence in the utf8 string str of any byte in the
-// utf8 string accept, or 0 if no match was found.
-utf8_nonnull utf8_pure utf8_weak void *utf8pbrk(const void *str,
-                                                const void *accept);
+/* Return less than 0, 0, greater than 0 if src1 < src2,
+ * src1 == src2, src1 > src2 respectively. Checking at most n
+ * bytes of each utf8 string. */
+utf8_constexpr14 utf8_nonnull utf8_pure int
+utf8ncmp(const utf8_int8_t *src1, const utf8_int8_t *src2, size_t n);
 
-// Find the last match of the utf8 codepoint chr in the utf8 string src.
-utf8_nonnull utf8_pure utf8_weak void *utf8rchr(const void *src, int chr);
+/* Copy the utf8 string src onto the memory allocated in dst.
+ * Copies at most n bytes. If n falls partway through a utf8
+ * codepoint, or if dst doesn't have enough room for a null
+ * terminator, the final string will be cut short to preserve
+ * utf8 validity. */
 
-// Number of bytes in the utf8 string str,
-// including the null terminating byte.
-utf8_nonnull utf8_pure utf8_weak size_t utf8size(const void *str);
+utf8_nonnull utf8_weak utf8_int8_t *
+utf8ncpy(utf8_int8_t *utf8_restrict dst, const utf8_int8_t *utf8_restrict src,
+         size_t n);
 
-// Number of utf8 codepoints in the utf8 string src that consists entirely
-// of utf8 codepoints from the utf8 string accept.
-utf8_nonnull utf8_pure utf8_weak size_t utf8spn(const void *src,
-                                                const void *accept);
+/* Similar to utf8dup, except that at most n bytes of src are copied. If src is
+ * longer than n, only n bytes are copied and a null byte is added.
+ *
+ * Returns a new string if successful, 0 otherwise */
+utf8_weak utf8_int8_t *utf8ndup(const utf8_int8_t *src, size_t n);
 
-// The position of the utf8 string needle in the utf8 string haystack.
-utf8_nonnull utf8_pure utf8_weak void *utf8str(const void *haystack,
-                                               const void *needle);
+/* Locates the first occurrence in the utf8 string str of any byte in the
+ * utf8 string accept, or 0 if no match was found. */
+utf8_constexpr14 utf8_nonnull utf8_pure utf8_int8_t *
+utf8pbrk(const utf8_int8_t *str, const utf8_int8_t *accept);
 
-// The position of the utf8 string needle in the utf8 string haystack, case
-// insensitive.
-utf8_nonnull utf8_pure utf8_weak void *utf8casestr(const void *haystack,
-                                                   const void *needle);
+/* Find the last match of the utf8 codepoint chr in the utf8 string src. */
+utf8_constexpr14 utf8_nonnull utf8_pure utf8_int8_t *
+utf8rchr(const utf8_int8_t *src, int chr);
 
-// Return 0 on success, or the position of the invalid
-// utf8 codepoint on failure.
-utf8_nonnull utf8_pure utf8_weak void *utf8valid(const void *str);
+/* Number of bytes in the utf8 string str,
+ * including the null terminating byte. */
+utf8_constexpr14 utf8_nonnull utf8_pure size_t utf8size(const utf8_int8_t *str);
 
-// Sets out_codepoint to the next utf8 codepoint in str, and returns the address
-// of the utf8 codepoint after the current one in str.
-utf8_nonnull utf8_weak void *
-utf8codepoint(const void *utf8_restrict str,
+/* Similar to utf8size, except that the null terminating byte is excluded. */
+utf8_constexpr14 utf8_nonnull utf8_pure size_t
+utf8size_lazy(const utf8_int8_t *str);
+
+/* Similar to utf8size, except that only at most n bytes of src are looked and
+ * the null terminating byte is excluded. */
+utf8_constexpr14 utf8_nonnull utf8_pure size_t
+utf8nsize_lazy(const utf8_int8_t *str, size_t n);
+
+/* Number of utf8 codepoints in the utf8 string src that consists entirely
+ * of utf8 codepoints from the utf8 string accept. */
+utf8_constexpr14 utf8_nonnull utf8_pure size_t
+utf8spn(const utf8_int8_t *src, const utf8_int8_t *accept);
+
+/* The position of the utf8 string needle in the utf8 string haystack. */
+utf8_constexpr14 utf8_nonnull utf8_pure utf8_int8_t *
+utf8str(const utf8_int8_t *haystack, const utf8_int8_t *needle);
+
+/* The position of the utf8 string needle in the utf8 string haystack, case
+ * insensitive. */
+utf8_constexpr14 utf8_nonnull utf8_pure utf8_int8_t *
+utf8casestr(const utf8_int8_t *haystack, const utf8_int8_t *needle);
+
+/* Return 0 on success, or the position of the invalid
+ * utf8 codepoint on failure. */
+utf8_constexpr14 utf8_nonnull utf8_pure utf8_int8_t *
+utf8valid(const utf8_int8_t *str);
+
+/* Similar to utf8valid, except that only at most n bytes of src are looked. */
+utf8_constexpr14 utf8_nonnull utf8_pure utf8_int8_t *
+utf8nvalid(const utf8_int8_t *str, size_t n);
+
+/* Given a null-terminated string, makes the string valid by replacing invalid
+ * codepoints with a 1-byte replacement. Returns 0 on success. */
+utf8_nonnull utf8_weak int utf8makevalid(utf8_int8_t *str,
+                                         const utf8_int32_t replacement);
+
+/* Sets out_codepoint to the current utf8 codepoint in str, and returns the
+ * address of the next utf8 codepoint after the current one in str. */
+utf8_constexpr14 utf8_nonnull utf8_int8_t *
+utf8codepoint(const utf8_int8_t *utf8_restrict str,
               utf8_int32_t *utf8_restrict out_codepoint);
 
-// Returns the size of the given codepoint in bytes.
-utf8_weak size_t utf8codepointsize(utf8_int32_t chr);
+/* Calculates the size of the next utf8 codepoint in str. */
+utf8_constexpr14 utf8_nonnull size_t
+utf8codepointcalcsize(const utf8_int8_t *str);
 
-// Write a codepoint to the given string, and return the address to the next
-// place after the written codepoint. Pass how many bytes left in the buffer to
-// n. If there is not enough space for the codepoint, this function returns
-// null.
-utf8_nonnull utf8_weak void *utf8catcodepoint(void *utf8_restrict str,
-                                              utf8_int32_t chr, size_t n);
+/* Returns the size of the given codepoint in bytes. */
+utf8_constexpr14 size_t utf8codepointsize(utf8_int32_t chr);
 
-// Returns 1 if the given character is lowercase, or 0 if it is not.
-utf8_weak int utf8islower(utf8_int32_t chr);
+/* Write a codepoint to the given string, and return the address to the next
+ * place after the written codepoint. Pass how many bytes left in the buffer to
+ * n. If there is not enough space for the codepoint, this function returns
+ * null. */
+utf8_nonnull utf8_weak utf8_int8_t *
+utf8catcodepoint(utf8_int8_t *str, utf8_int32_t chr, size_t n);
 
-// Returns 1 if the given character is uppercase, or 0 if it is not.
-utf8_weak int utf8isupper(utf8_int32_t chr);
+/* Returns 1 if the given character is lowercase, or 0 if it is not. */
+utf8_constexpr14 int utf8islower(utf8_int32_t chr);
 
-// Transform the given string into all lowercase codepoints.
-utf8_nonnull utf8_weak void utf8lwr(void *utf8_restrict str);
+/* Returns 1 if the given character is uppercase, or 0 if it is not. */
+utf8_constexpr14 int utf8isupper(utf8_int32_t chr);
 
-// Transform the given string into all uppercase codepoints.
-utf8_nonnull utf8_weak void utf8upr(void *utf8_restrict str);
+/* Transform the given string into all lowercase codepoints. */
+utf8_nonnull utf8_weak void utf8lwr(utf8_int8_t *utf8_restrict str);
 
-// Make a codepoint lower case if possible.
-utf8_weak utf8_int32_t utf8lwrcodepoint(utf8_int32_t cp);
+/* Transform the given string into all uppercase codepoints. */
+utf8_nonnull utf8_weak void utf8upr(utf8_int8_t *utf8_restrict str);
 
-// Make a codepoint upper case if possible.
-utf8_weak utf8_int32_t utf8uprcodepoint(utf8_int32_t cp);
+/* Make a codepoint lower case if possible. */
+utf8_constexpr14 utf8_int32_t utf8lwrcodepoint(utf8_int32_t cp);
+
+/* Make a codepoint upper case if possible. */
+utf8_constexpr14 utf8_int32_t utf8uprcodepoint(utf8_int32_t cp);
+
+/* Sets out_codepoint to the current utf8 codepoint in str, and returns the
+ * address of the previous utf8 codepoint before the current one in str. */
+utf8_constexpr14 utf8_nonnull utf8_int8_t *
+utf8rcodepoint(const utf8_int8_t *utf8_restrict str,
+               utf8_int32_t *utf8_restrict out_codepoint);
+
+/* Duplicate the utf8 string src by getting its size, calling alloc_func_ptr to
+ * copy over data to a new buffer, and returning that. Or 0 if alloc_func_ptr
+ * returned null. */
+utf8_weak utf8_int8_t *utf8dup_ex(const utf8_int8_t *src,
+                                  utf8_int8_t *(*alloc_func_ptr)(utf8_int8_t *,
+                                                                 size_t),
+                                  utf8_int8_t *user_data);
+
+/* Similar to utf8dup, except that at most n bytes of src are copied. If src is
+ * longer than n, only n bytes are copied and a null byte is added.
+ *
+ * Returns a new string if successful, 0 otherwise. */
+utf8_weak utf8_int8_t *utf8ndup_ex(const utf8_int8_t *src, size_t n,
+                                   utf8_int8_t *(*alloc_func_ptr)(utf8_int8_t *,
+                                                                  size_t),
+                                   utf8_int8_t *user_data);
 
 #undef utf8_weak
 #undef utf8_pure
 #undef utf8_nonnull
 
-int utf8casecmp(const void *src1, const void *src2) {
-  utf8_int32_t src1_cp, src2_cp, src1_orig_cp, src2_orig_cp;
+utf8_constexpr14_impl int utf8casecmp(const utf8_int8_t *src1,
+                                      const utf8_int8_t *src2) {
+  utf8_int32_t src1_lwr_cp = 0, src2_lwr_cp = 0, src1_upr_cp = 0,
+               src2_upr_cp = 0, src1_orig_cp = 0, src2_orig_cp = 0;
 
   for (;;) {
-    src1 = utf8codepoint(src1, &src1_cp);
-    src2 = utf8codepoint(src2, &src2_cp);
+    src1 = utf8codepoint(src1, &src1_orig_cp);
+    src2 = utf8codepoint(src2, &src2_orig_cp);
 
-    // Take a copy of src1 & src2
-    src1_orig_cp = src1_cp;
-    src2_orig_cp = src2_cp;
+    /* lower the srcs if required */
+    src1_lwr_cp = utf8lwrcodepoint(src1_orig_cp);
+    src2_lwr_cp = utf8lwrcodepoint(src2_orig_cp);
 
-    // Lower the srcs if required
-    src1_cp = utf8lwrcodepoint(src1_cp);
-    src2_cp = utf8lwrcodepoint(src2_cp);
+    /* lower the srcs if required */
+    src1_upr_cp = utf8uprcodepoint(src1_orig_cp);
+    src2_upr_cp = utf8uprcodepoint(src2_orig_cp);
 
-    // Check if the lowered codepoints match
+    /* check if the lowered codepoints match */
     if ((0 == src1_orig_cp) && (0 == src2_orig_cp)) {
       return 0;
-    } else if (src1_cp == src2_cp) {
+    } else if ((src1_lwr_cp == src2_lwr_cp) || (src1_upr_cp == src2_upr_cp)) {
       continue;
     }
 
-    // If they don't match, then we return which of the original's are less
-    if (src1_orig_cp < src2_orig_cp) {
-      return -1;
-    } else if (src1_orig_cp > src2_orig_cp) {
-      return 1;
-    }
+    /* if they don't match, then we return the difference between the characters
+     */
+    return src1_lwr_cp - src2_lwr_cp;
   }
 }
 
-void *utf8cat(void *utf8_restrict dst, const void *utf8_restrict src) {
-  char *d = (char *)dst;
-  const char *s = (const char *)src;
-
-  // find the null terminating byte in dst
+utf8_int8_t *utf8cat(utf8_int8_t *utf8_restrict dst,
+                     const utf8_int8_t *utf8_restrict src) {
+  utf8_int8_t *d = dst;
+  /* find the null terminating byte in dst */
   while ('\0' != *d) {
     d++;
   }
 
-  // overwriting the null terminating byte in dst, append src byte-by-byte
-  while ('\0' != *s) {
-    *d++ = *s++;
+  /* overwriting the null terminating byte in dst, append src byte-by-byte */
+  while ('\0' != *src) {
+    *d++ = *src++;
   }
 
-  // write out a new null terminating byte into dst
+  /* write out a new null terminating byte into dst */
   *d = '\0';
 
   return dst;
 }
 
-void *utf8chr(const void *src, utf8_int32_t chr) {
-  char c[5] = {'\0', '\0', '\0', '\0', '\0'};
+utf8_constexpr14_impl utf8_int8_t *utf8chr(const utf8_int8_t *src,
+                                           utf8_int32_t chr) {
+  utf8_int8_t c[5] = {'\0', '\0', '\0', '\0', '\0'};
 
   if (0 == chr) {
-    // being asked to return position of null terminating byte, so
-    // just run s to the end, and return!
-    const char *s = (const char *)src;
-    while ('\0' != *s) {
-      s++;
+    /* being asked to return position of null terminating byte, so
+     * just run s to the end, and return! */
+    while ('\0' != *src) {
+      src++;
     }
-    return (void *)s;
+    return (utf8_int8_t *)src;
   } else if (0 == ((utf8_int32_t)0xffffff80 & chr)) {
-    // 1-byte/7-bit ascii
-    // (0b0xxxxxxx)
-    c[0] = (char)chr;
+    /* 1-byte/7-bit ascii
+     * (0b0xxxxxxx) */
+    c[0] = (utf8_int8_t)chr;
   } else if (0 == ((utf8_int32_t)0xfffff800 & chr)) {
-    // 2-byte/11-bit utf8 code point
-    // (0b110xxxxx 0b10xxxxxx)
-    c[0] = 0xc0 | (char)(chr >> 6);
-    c[1] = 0x80 | (char)(chr & 0x3f);
+    /* 2-byte/11-bit utf8 code point
+     * (0b110xxxxx 0b10xxxxxx) */
+    c[0] = (utf8_int8_t)(0xc0 | (utf8_int8_t)(chr >> 6));
+    c[1] = (utf8_int8_t)(0x80 | (utf8_int8_t)(chr & 0x3f));
   } else if (0 == ((utf8_int32_t)0xffff0000 & chr)) {
-    // 3-byte/16-bit utf8 code point
-    // (0b1110xxxx 0b10xxxxxx 0b10xxxxxx)
-    c[0] = 0xe0 | (char)(chr >> 12);
-    c[1] = 0x80 | (char)((chr >> 6) & 0x3f);
-    c[2] = 0x80 | (char)(chr & 0x3f);
-  } else { // if (0 == ((int)0xffe00000 & chr)) {
-    // 4-byte/21-bit utf8 code point
-    // (0b11110xxx 0b10xxxxxx 0b10xxxxxx 0b10xxxxxx)
-    c[0] = 0xf0 | (char)(chr >> 18);
-    c[1] = 0x80 | (char)((chr >> 12) & 0x3f);
-    c[2] = 0x80 | (char)((chr >> 6) & 0x3f);
-    c[3] = 0x80 | (char)(chr & 0x3f);
+    /* 3-byte/16-bit utf8 code point
+     * (0b1110xxxx 0b10xxxxxx 0b10xxxxxx) */
+    c[0] = (utf8_int8_t)(0xe0 | (utf8_int8_t)(chr >> 12));
+    c[1] = (utf8_int8_t)(0x80 | (utf8_int8_t)((chr >> 6) & 0x3f));
+    c[2] = (utf8_int8_t)(0x80 | (utf8_int8_t)(chr & 0x3f));
+  } else { /* if (0 == ((int)0xffe00000 & chr)) { */
+    /* 4-byte/21-bit utf8 code point
+     * (0b11110xxx 0b10xxxxxx 0b10xxxxxx 0b10xxxxxx) */
+    c[0] = (utf8_int8_t)(0xf0 | (utf8_int8_t)(chr >> 18));
+    c[1] = (utf8_int8_t)(0x80 | (utf8_int8_t)((chr >> 12) & 0x3f));
+    c[2] = (utf8_int8_t)(0x80 | (utf8_int8_t)((chr >> 6) & 0x3f));
+    c[3] = (utf8_int8_t)(0x80 | (utf8_int8_t)(chr & 0x3f));
   }
 
-  // we've made c into a 2 utf8 codepoint string, one for the chr we are
-  // seeking, another for the null terminating byte. Now use utf8str to
-  // search
+  /* we've made c into a 2 utf8 codepoint string, one for the chr we are
+   * seeking, another for the null terminating byte. Now use utf8str to
+   * search */
   return utf8str(src, c);
 }
 
-int utf8cmp(const void *src1, const void *src2) {
-  const unsigned char *s1 = (const unsigned char *)src1;
-  const unsigned char *s2 = (const unsigned char *)src2;
-
-  while (('\0' != *s1) || ('\0' != *s2)) {
-    if (*s1 < *s2) {
+utf8_constexpr14_impl int utf8cmp(const utf8_int8_t *src1,
+                                  const utf8_int8_t *src2) {
+  while (('\0' != *src1) || ('\0' != *src2)) {
+    if (*src1 < *src2) {
       return -1;
-    } else if (*s1 > *s2) {
+    } else if (*src1 > *src2) {
       return 1;
     }
 
-    s1++;
-    s2++;
+    src1++;
+    src2++;
   }
 
-  // both utf8 strings matched
+  /* both utf8 strings matched */
   return 0;
 }
 
-int utf8coll(const void *src1, const void *src2);
+utf8_constexpr14_impl int utf8coll(const utf8_int8_t *src1,
+                                   const utf8_int8_t *src2);
 
-void *utf8cpy(void *utf8_restrict dst, const void *utf8_restrict src) {
-  char *d = (char *)dst;
-  const char *s = (const char *)src;
+utf8_int8_t *utf8cpy(utf8_int8_t *utf8_restrict dst,
+                     const utf8_int8_t *utf8_restrict src) {
+  utf8_int8_t *d = dst;
 
-  // overwriting anything previously in dst, write byte-by-byte
-  // from src
-  while ('\0' != *s) {
-    *d++ = *s++;
+  /* overwriting anything previously in dst, write byte-by-byte
+   * from src */
+  while ('\0' != *src) {
+    *d++ = *src++;
   }
 
-  // append null terminating byte
+  /* append null terminating byte */
   *d = '\0';
 
   return dst;
 }
 
-size_t utf8cspn(const void *src, const void *reject) {
-  const char *s = (const char *)src;
+utf8_constexpr14_impl size_t utf8cspn(const utf8_int8_t *src,
+                                      const utf8_int8_t *reject) {
   size_t chars = 0;
 
-  while ('\0' != *s) {
-    const char *r = (const char *)reject;
+  while ('\0' != *src) {
+    const utf8_int8_t *r = reject;
     size_t offset = 0;
 
     while ('\0' != *r) {
-      // checking that if *r is the start of a utf8 codepoint
-      // (it is not 0b10xxxxxx) and we have successfully matched
-      // a previous character (0 < offset) - we found a match
+      /* checking that if *r is the start of a utf8 codepoint
+       * (it is not 0b10xxxxxx) and we have successfully matched
+       * a previous character (0 < offset) - we found a match */
       if ((0x80 != (0xc0 & *r)) && (0 < offset)) {
         return chars;
       } else {
-        if (*r == s[offset]) {
-          // part of a utf8 codepoint matched, so move our checking
-          // onwards to the next byte
+        if (*r == src[offset]) {
+          /* part of a utf8 codepoint matched, so move our checking
+           * onwards to the next byte */
           offset++;
           r++;
         } else {
-          // r could be in the middle of an unmatching utf8 code point,
-          // so we need to march it on to the next character beginning,
+          /* r could be in the middle of an unmatching utf8 code point,
+           * so we need to march it on to the next character beginning, */
 
           do {
             r++;
           } while (0x80 == (0xc0 & *r));
 
-          // reset offset too as we found a mismatch
+          /* reset offset too as we found a mismatch */
           offset = 0;
         }
       }
     }
 
-    // the current utf8 codepoint in src did not match reject, but src
-    // could have been partway through a utf8 codepoint, so we need to
-    // march it onto the next utf8 codepoint starting byte
+    /* found a match at the end of *r, so didn't get a chance to test it */
+    if (0 < offset) {
+      return chars;
+    }
+
+    /* the current utf8 codepoint in src did not match reject, but src
+     * could have been partway through a utf8 codepoint, so we need to
+     * march it onto the next utf8 codepoint starting byte */
     do {
-      s++;
-    } while ((0x80 == (0xc0 & *s)));
+      src++;
+    } while ((0x80 == (0xc0 & *src)));
     chars++;
   }
 
   return chars;
 }
 
-size_t utf8size(const void *str);
+utf8_int8_t *utf8dup(const utf8_int8_t *src) {
+  return utf8dup_ex(src, utf8_null, utf8_null);
+}
 
-void *utf8dup(const void *src) {
-  const char *s = (const char *)src;
-  char *n = utf8_null;
+utf8_int8_t *utf8dup_ex(const utf8_int8_t *src,
+                        utf8_int8_t *(*alloc_func_ptr)(utf8_int8_t *, size_t),
+                        utf8_int8_t *user_data) {
+  utf8_int8_t *n = utf8_null;
 
-  // figure out how many bytes (including the terminator) we need to copy first
+  /* figure out how many bytes (including the terminator) we need to copy first
+   */
   size_t bytes = utf8size(src);
 
-  n = (char *)malloc(bytes);
+  if (alloc_func_ptr) {
+    n = alloc_func_ptr(user_data, bytes);
+  } else {
+#if !defined(UTF8_NO_STD_MALLOC)
+    n = (utf8_int8_t *)malloc(bytes);
+#else
+    return utf8_null;
+#endif
+  }
 
   if (utf8_null == n) {
-    // out of memory so we bail
+    /* out of memory so we bail */
     return utf8_null;
   } else {
     bytes = 0;
 
-    // copy src byte-by-byte into our new utf8 string
-    while ('\0' != s[bytes]) {
-      n[bytes] = s[bytes];
+    /* copy src byte-by-byte into our new utf8 string */
+    while ('\0' != src[bytes]) {
+      n[bytes] = src[bytes];
       bytes++;
     }
 
-    // append null terminating byte
+    /* append null terminating byte */
     n[bytes] = '\0';
     return n;
   }
 }
 
-void *utf8fry(const void *str);
+utf8_constexpr14_impl utf8_int8_t *utf8fry(const utf8_int8_t *str);
 
-size_t utf8len(const void *str) {
-  const unsigned char *s = (const unsigned char *)str;
+utf8_constexpr14_impl size_t utf8len(const utf8_int8_t *str) {
+  return utf8nlen(str, SIZE_MAX);
+}
+
+utf8_constexpr14_impl size_t utf8nlen(const utf8_int8_t *str, size_t n) {
+  const utf8_int8_t *t = str;
   size_t length = 0;
 
-  while ('\0' != *s) {
-    if (0xf0 == (0xf8 & *s)) {
-      // 4-byte utf8 code point (began with 0b11110xxx)
-      s += 4;
-    } else if (0xe0 == (0xf0 & *s)) {
-      // 3-byte utf8 code point (began with 0b1110xxxx)
-      s += 3;
-    } else if (0xc0 == (0xe0 & *s)) {
-      // 2-byte utf8 code point (began with 0b110xxxxx)
-      s += 2;
-    } else { // if (0x00 == (0x80 & *s)) {
-      // 1-byte ascii (began with 0b0xxxxxxx)
-      s += 1;
+  while ((size_t)(str - t) < n && '\0' != *str) {
+    if (0xf0 == (0xf8 & *str)) {
+      /* 4-byte utf8 code point (began with 0b11110xxx) */
+      str += 4;
+    } else if (0xe0 == (0xf0 & *str)) {
+      /* 3-byte utf8 code point (began with 0b1110xxxx) */
+      str += 3;
+    } else if (0xc0 == (0xe0 & *str)) {
+      /* 2-byte utf8 code point (began with 0b110xxxxx) */
+      str += 2;
+    } else { /* if (0x00 == (0x80 & *s)) { */
+      /* 1-byte ascii (began with 0b0xxxxxxx) */
+      str += 1;
     }
 
-    // no matter the bytes we marched s forward by, it was
-    // only 1 utf8 codepoint
+    /* no matter the bytes we marched s forward by, it was
+     * only 1 utf8 codepoint */
     length++;
   }
 
+  if ((size_t)(str - t) > n) {
+    length--;
+  }
   return length;
 }
 
-int utf8ncasecmp(const void *src1, const void *src2, size_t n) {
-  utf8_int32_t src1_cp, src2_cp, src1_orig_cp, src2_orig_cp;
+utf8_constexpr14_impl int utf8ncasecmp(const utf8_int8_t *src1,
+                                       const utf8_int8_t *src2, size_t n) {
+  utf8_int32_t src1_lwr_cp = 0, src2_lwr_cp = 0, src1_upr_cp = 0,
+               src2_upr_cp = 0, src1_orig_cp = 0, src2_orig_cp = 0;
 
   do {
-    const unsigned char *const s1 = (const unsigned char *)src1;
-    const unsigned char *const s2 = (const unsigned char *)src2;
+    const utf8_int8_t *const s1 = src1;
+    const utf8_int8_t *const s2 = src2;
 
-    // first check that we have enough bytes left in n to contain an entire
-    // codepoint
+    /* first check that we have enough bytes left in n to contain an entire
+     * codepoint */
     if (0 == n) {
       return 0;
     }
@@ -467,10 +579,8 @@ int utf8ncasecmp(const void *src1, const void *src2, size_t n) {
       const utf8_int32_t c1 = (0xe0 & *s1);
       const utf8_int32_t c2 = (0xe0 & *s2);
 
-      if (c1 < c2) {
-        return -1;
-      } else if (c1 > c2) {
-        return 1;
+      if (c1 != c2) {
+        return c1 - c2;
       } else {
         return 0;
       }
@@ -480,10 +590,8 @@ int utf8ncasecmp(const void *src1, const void *src2, size_t n) {
       const utf8_int32_t c1 = (0xf0 & *s1);
       const utf8_int32_t c2 = (0xf0 & *s2);
 
-      if (c1 < c2) {
-        return -1;
-      } else if (c1 > c2) {
-        return 1;
+      if (c1 != c2) {
+        return c1 - c2;
       } else {
         return 0;
       }
@@ -493,307 +601,343 @@ int utf8ncasecmp(const void *src1, const void *src2, size_t n) {
       const utf8_int32_t c1 = (0xf8 & *s1);
       const utf8_int32_t c2 = (0xf8 & *s2);
 
-      if (c1 < c2) {
-        return -1;
-      } else if (c1 > c2) {
-        return 1;
+      if (c1 != c2) {
+        return c1 - c2;
       } else {
         return 0;
       }
     }
 
-    src1 = utf8codepoint(src1, &src1_cp);
-    src2 = utf8codepoint(src2, &src2_cp);
-    n -= utf8codepointsize(src1_cp);
+    src1 = utf8codepoint(src1, &src1_orig_cp);
+    src2 = utf8codepoint(src2, &src2_orig_cp);
+    n -= utf8codepointsize(src1_orig_cp);
 
-    // Take a copy of src1 & src2
-    src1_orig_cp = src1_cp;
-    src2_orig_cp = src2_cp;
+    src1_lwr_cp = utf8lwrcodepoint(src1_orig_cp);
+    src2_lwr_cp = utf8lwrcodepoint(src2_orig_cp);
 
-    // Lower srcs if required
-    src1_cp = utf8lwrcodepoint(src1_cp);
-    src2_cp = utf8lwrcodepoint(src2_cp);
+    src1_upr_cp = utf8uprcodepoint(src1_orig_cp);
+    src2_upr_cp = utf8uprcodepoint(src2_orig_cp);
 
-    // Check if the lowered codepoints match
+    /* check if the lowered codepoints match */
     if ((0 == src1_orig_cp) && (0 == src2_orig_cp)) {
       return 0;
-    } else if (src1_cp == src2_cp) {
+    } else if ((src1_lwr_cp == src2_lwr_cp) || (src1_upr_cp == src2_upr_cp)) {
       continue;
     }
 
-    // If they don't match, then we return which of the original's are less
-    if (src1_orig_cp < src2_orig_cp) {
-      return -1;
-    } else if (src1_orig_cp > src2_orig_cp) {
-      return 1;
-    }
+    /* if they don't match, then we return the difference between the characters
+     */
+    return src1_lwr_cp - src2_lwr_cp;
   } while (0 < n);
 
-  // both utf8 strings matched
+  /* both utf8 strings matched */
   return 0;
 }
 
-void *utf8ncat(void *utf8_restrict dst, const void *utf8_restrict src,
-               size_t n) {
-  char *d = (char *)dst;
-  const char *s = (const char *)src;
+utf8_int8_t *utf8ncat(utf8_int8_t *utf8_restrict dst,
+                      const utf8_int8_t *utf8_restrict src, size_t n) {
+  utf8_int8_t *d = dst;
 
-  // find the null terminating byte in dst
+  /* find the null terminating byte in dst */
   while ('\0' != *d) {
     d++;
   }
 
-  // overwriting the null terminating byte in dst, append src byte-by-byte
-  // stopping if we run out of space
-  do {
-    *d++ = *s++;
-  } while (('\0' != *s) && (0 != --n));
+  /* overwriting the null terminating byte in dst, append src byte-by-byte
+   * stopping if we run out of space */
+  while (('\0' != *src) && (0 != n--)) {
+    *d++ = *src++;
+  }
 
-  // write out a new null terminating byte into dst
+  /* write out a new null terminating byte into dst */
   *d = '\0';
 
   return dst;
 }
 
-int utf8ncmp(const void *src1, const void *src2, size_t n) {
-  const unsigned char *s1 = (const unsigned char *)src1;
-  const unsigned char *s2 = (const unsigned char *)src2;
-
-  while ((('\0' != *s1) || ('\0' != *s2)) && (0 != n--)) {
-    if (*s1 < *s2) {
+utf8_constexpr14_impl int utf8ncmp(const utf8_int8_t *src1,
+                                   const utf8_int8_t *src2, size_t n) {
+  while ((0 != n--) && (('\0' != *src1) || ('\0' != *src2))) {
+    if (*src1 < *src2) {
       return -1;
-    } else if (*s1 > *s2) {
+    } else if (*src1 > *src2) {
       return 1;
     }
 
-    s1++;
-    s2++;
+    src1++;
+    src2++;
   }
 
-  // both utf8 strings matched
+  /* both utf8 strings matched */
   return 0;
 }
 
-void *utf8ncpy(void *utf8_restrict dst, const void *utf8_restrict src,
-               size_t n) {
-  char *d = (char *)dst;
-  const char *s = (const char *)src;
+utf8_int8_t *utf8ncpy(utf8_int8_t *utf8_restrict dst,
+                      const utf8_int8_t *utf8_restrict src, size_t n) {
+  utf8_int8_t *d = dst;
+  size_t index = 0, check_index = 0;
 
-  // overwriting anything previously in dst, write byte-by-byte
-  // from src
-  do {
-    *d++ = *s++;
-  } while (('\0' != *s) && (0 != --n));
+  if (n == 0) {
+    return dst;
+  }
 
-  // append null terminating byte
-  while (0 != n) {
-    *d++ = '\0';
-    n--;
+  /* overwriting anything previously in dst, write byte-by-byte
+   * from src */
+  for (index = 0; index < n; index++) {
+    d[index] = src[index];
+    if ('\0' == src[index]) {
+      break;
+    }
+  }
+
+  for (check_index = index - 1;
+       check_index > 0 && 0x80 == (0xc0 & d[check_index]); check_index--) {
+    /* just moving the index */
+  }
+
+  if (check_index < index &&
+      ((index - check_index) < utf8codepointcalcsize(&d[check_index]) ||
+       (index - check_index) == n)) {
+    index = check_index;
+  }
+
+  /* append null terminating byte */
+  for (; index < n; index++) {
+    d[index] = 0;
   }
 
   return dst;
 }
 
-void *utf8ndup(const void *src, size_t n) {
-  const char *s = (const char *)src;
-  char *c = utf8_null;
+utf8_int8_t *utf8ndup(const utf8_int8_t *src, size_t n) {
+  return utf8ndup_ex(src, n, utf8_null, utf8_null);
+}
+
+utf8_int8_t *utf8ndup_ex(const utf8_int8_t *src, size_t n,
+                         utf8_int8_t *(*alloc_func_ptr)(utf8_int8_t *, size_t),
+                         utf8_int8_t *user_data) {
+  utf8_int8_t *c = utf8_null;
   size_t bytes = 0;
 
-  // Find the end of the string or stop when n is reached
-  while ('\0' != s[bytes] && bytes < n) {
+  /* Find the end of the string or stop when n is reached */
+  while ('\0' != src[bytes] && bytes < n) {
     bytes++;
   }
 
-  // In case bytes is actually less than n, we need to set it
-  // to be used later in the copy byte by byte.
+  /* In case bytes is actually less than n, we need to set it
+   * to be used later in the copy byte by byte. */
   n = bytes;
 
-  c = (char *)malloc(bytes + 1);
+  if (alloc_func_ptr) {
+    c = alloc_func_ptr(user_data, bytes + 1);
+  } else {
+#if !defined(UTF8_NO_STD_MALLOC)
+    c = (utf8_int8_t *)malloc(bytes + 1);
+#else
+    c = utf8_null;
+#endif
+  }
+
   if (utf8_null == c) {
-    // out of memory so we bail
+    /* out of memory so we bail */
     return utf8_null;
   }
 
   bytes = 0;
 
-  // copy src byte-by-byte into our new utf8 string
-  while ('\0' != s[bytes] && bytes < n) {
-    c[bytes] = s[bytes];
+  /* copy src byte-by-byte into our new utf8 string */
+  while ('\0' != src[bytes] && bytes < n) {
+    c[bytes] = src[bytes];
     bytes++;
   }
 
-  // append null terminating byte
+  /* append null terminating byte */
   c[bytes] = '\0';
   return c;
 }
 
-void *utf8rchr(const void *src, int chr) {
-  const char *s = (const char *)src;
-  const char *match = utf8_null;
-  char c[5] = {'\0', '\0', '\0', '\0', '\0'};
+utf8_constexpr14_impl utf8_int8_t *utf8rchr(const utf8_int8_t *src, int chr) {
+
+  utf8_int8_t *match = utf8_null;
+  utf8_int8_t c[5] = {'\0', '\0', '\0', '\0', '\0'};
 
   if (0 == chr) {
-    // being asked to return position of null terminating byte, so
-    // just run s to the end, and return!
-    while ('\0' != *s) {
-      s++;
+    /* being asked to return position of null terminating byte, so
+     * just run s to the end, and return! */
+    while ('\0' != *src) {
+      src++;
     }
-    return (void *)s;
+    return (utf8_int8_t *)src;
   } else if (0 == ((int)0xffffff80 & chr)) {
-    // 1-byte/7-bit ascii
-    // (0b0xxxxxxx)
-    c[0] = (char)chr;
+    /* 1-byte/7-bit ascii
+     * (0b0xxxxxxx) */
+    c[0] = (utf8_int8_t)chr;
   } else if (0 == ((int)0xfffff800 & chr)) {
-    // 2-byte/11-bit utf8 code point
-    // (0b110xxxxx 0b10xxxxxx)
-    c[0] = 0xc0 | (char)(chr >> 6);
-    c[1] = 0x80 | (char)(chr & 0x3f);
+    /* 2-byte/11-bit utf8 code point
+     * (0b110xxxxx 0b10xxxxxx) */
+    c[0] = (utf8_int8_t)(0xc0 | (utf8_int8_t)(chr >> 6));
+    c[1] = (utf8_int8_t)(0x80 | (utf8_int8_t)(chr & 0x3f));
   } else if (0 == ((int)0xffff0000 & chr)) {
-    // 3-byte/16-bit utf8 code point
-    // (0b1110xxxx 0b10xxxxxx 0b10xxxxxx)
-    c[0] = 0xe0 | (char)(chr >> 12);
-    c[1] = 0x80 | (char)((chr >> 6) & 0x3f);
-    c[2] = 0x80 | (char)(chr & 0x3f);
-  } else { // if (0 == ((int)0xffe00000 & chr)) {
-    // 4-byte/21-bit utf8 code point
-    // (0b11110xxx 0b10xxxxxx 0b10xxxxxx 0b10xxxxxx)
-    c[0] = 0xf0 | (char)(chr >> 18);
-    c[1] = 0x80 | (char)((chr >> 12) & 0x3f);
-    c[2] = 0x80 | (char)((chr >> 6) & 0x3f);
-    c[3] = 0x80 | (char)(chr & 0x3f);
+    /* 3-byte/16-bit utf8 code point
+     * (0b1110xxxx 0b10xxxxxx 0b10xxxxxx) */
+    c[0] = (utf8_int8_t)(0xe0 | (utf8_int8_t)(chr >> 12));
+    c[1] = (utf8_int8_t)(0x80 | (utf8_int8_t)((chr >> 6) & 0x3f));
+    c[2] = (utf8_int8_t)(0x80 | (utf8_int8_t)(chr & 0x3f));
+  } else { /* if (0 == ((int)0xffe00000 & chr)) { */
+    /* 4-byte/21-bit utf8 code point
+     * (0b11110xxx 0b10xxxxxx 0b10xxxxxx 0b10xxxxxx) */
+    c[0] = (utf8_int8_t)(0xf0 | (utf8_int8_t)(chr >> 18));
+    c[1] = (utf8_int8_t)(0x80 | (utf8_int8_t)((chr >> 12) & 0x3f));
+    c[2] = (utf8_int8_t)(0x80 | (utf8_int8_t)((chr >> 6) & 0x3f));
+    c[3] = (utf8_int8_t)(0x80 | (utf8_int8_t)(chr & 0x3f));
   }
 
-  // we've created a 2 utf8 codepoint string in c that is
-  // the utf8 character asked for by chr, and a null
-  // terminating byte
+  /* we've created a 2 utf8 codepoint string in c that is
+   * the utf8 character asked for by chr, and a null
+   * terminating byte */
 
-  while ('\0' != *s) {
+  while ('\0' != *src) {
     size_t offset = 0;
 
-    while (s[offset] == c[offset]) {
+    while ((src[offset] == c[offset]) && ('\0' != src[offset])) {
       offset++;
     }
 
     if ('\0' == c[offset]) {
-      // we found a matching utf8 code point
-      match = s;
-      s += offset;
-    } else {
-      s += offset;
+      /* we found a matching utf8 code point */
+      match = (utf8_int8_t *)src;
+      src += offset;
 
-      // need to march s along to next utf8 codepoint start
-      // (the next byte that doesn't match 0b10xxxxxx)
-      if ('\0' != *s) {
+      if ('\0' == *src) {
+        break;
+      }
+    } else {
+      src += offset;
+
+      /* need to march s along to next utf8 codepoint start
+       * (the next byte that doesn't match 0b10xxxxxx) */
+      if ('\0' != *src) {
         do {
-          s++;
-        } while (0x80 == (0xc0 & *s));
+          src++;
+        } while (0x80 == (0xc0 & *src));
       }
     }
   }
 
-  // return the last match we found (or 0 if no match was found)
-  return (void *)match;
+  /* return the last match we found (or 0 if no match was found) */
+  return match;
 }
 
-void *utf8pbrk(const void *str, const void *accept) {
-  const char *s = (const char *)str;
-
-  while ('\0' != *s) {
-    const char *a = (const char *)accept;
+utf8_constexpr14_impl utf8_int8_t *utf8pbrk(const utf8_int8_t *str,
+                                            const utf8_int8_t *accept) {
+  while ('\0' != *str) {
+    const utf8_int8_t *a = accept;
     size_t offset = 0;
 
     while ('\0' != *a) {
-      // checking that if *a is the start of a utf8 codepoint
-      // (it is not 0b10xxxxxx) and we have successfully matched
-      // a previous character (0 < offset) - we found a match
+      /* checking that if *a is the start of a utf8 codepoint
+       * (it is not 0b10xxxxxx) and we have successfully matched
+       * a previous character (0 < offset) - we found a match */
       if ((0x80 != (0xc0 & *a)) && (0 < offset)) {
-        return (void *)s;
+        return (utf8_int8_t *)str;
       } else {
-        if (*a == s[offset]) {
-          // part of a utf8 codepoint matched, so move our checking
-          // onwards to the next byte
+        if (*a == str[offset]) {
+          /* part of a utf8 codepoint matched, so move our checking
+           * onwards to the next byte */
           offset++;
           a++;
         } else {
-          // r could be in the middle of an unmatching utf8 code point,
-          // so we need to march it on to the next character beginning,
+          /* r could be in the middle of an unmatching utf8 code point,
+           * so we need to march it on to the next character beginning, */
 
           do {
             a++;
           } while (0x80 == (0xc0 & *a));
 
-          // reset offset too as we found a mismatch
+          /* reset offset too as we found a mismatch */
           offset = 0;
         }
       }
     }
 
-    // we found a match on the last utf8 codepoint
+    /* we found a match on the last utf8 codepoint */
     if (0 < offset) {
-      return (void *)s;
+      return (utf8_int8_t *)str;
     }
 
-    // the current utf8 codepoint in src did not match accept, but src
-    // could have been partway through a utf8 codepoint, so we need to
-    // march it onto the next utf8 codepoint starting byte
+    /* the current utf8 codepoint in src did not match accept, but src
+     * could have been partway through a utf8 codepoint, so we need to
+     * march it onto the next utf8 codepoint starting byte */
     do {
-      s++;
-    } while ((0x80 == (0xc0 & *s)));
+      str++;
+    } while ((0x80 == (0xc0 & *str)));
   }
 
   return utf8_null;
 }
 
-size_t utf8size(const void *str) {
-  const char *s = (const char *)str;
+utf8_constexpr14_impl size_t utf8size(const utf8_int8_t *str) {
+  return utf8size_lazy(str) + 1;
+}
+
+utf8_constexpr14_impl size_t utf8size_lazy(const utf8_int8_t *str) {
+  return utf8nsize_lazy(str, SIZE_MAX);
+}
+
+utf8_constexpr14_impl size_t utf8nsize_lazy(const utf8_int8_t *str, size_t n) {
   size_t size = 0;
-  while ('\0' != s[size]) {
+  while (size < n && '\0' != str[size]) {
     size++;
   }
-
-  // we are including the null terminating byte in the size calculation
-  size++;
   return size;
 }
 
-size_t utf8spn(const void *src, const void *accept) {
-  const char *s = (const char *)src;
+utf8_constexpr14_impl size_t utf8spn(const utf8_int8_t *src,
+                                     const utf8_int8_t *accept) {
   size_t chars = 0;
 
-  while ('\0' != *s) {
-    const char *a = (const char *)accept;
+  while ('\0' != *src) {
+    const utf8_int8_t *a = accept;
     size_t offset = 0;
 
     while ('\0' != *a) {
-      // checking that if *r is the start of a utf8 codepoint
-      // (it is not 0b10xxxxxx) and we have successfully matched
-      // a previous character (0 < offset) - we found a match
+      /* checking that if *r is the start of a utf8 codepoint
+       * (it is not 0b10xxxxxx) and we have successfully matched
+       * a previous character (0 < offset) - we found a match */
       if ((0x80 != (0xc0 & *a)) && (0 < offset)) {
-        // found a match, so increment the number of utf8 codepoints
-        // that have matched and stop checking whether any other utf8
-        // codepoints in a match
+        /* found a match, so increment the number of utf8 codepoints
+         * that have matched and stop checking whether any other utf8
+         * codepoints in a match */
         chars++;
-        s += offset;
+        src += offset;
+        offset = 0;
         break;
       } else {
-        if (*a == s[offset]) {
+        if (*a == src[offset]) {
           offset++;
           a++;
         } else {
-          // a could be in the middle of an unmatching utf8 codepoint,
-          // so we need to march it on to the next character beginning,
+          /* a could be in the middle of an unmatching utf8 codepoint,
+           * so we need to march it on to the next character beginning, */
           do {
             a++;
           } while (0x80 == (0xc0 & *a));
 
-          // reset offset too as we found a mismatch
+          /* reset offset too as we found a mismatch */
           offset = 0;
         }
       }
     }
 
-    // if a got to its terminating null byte, then we didn't find a match.
-    // Return the current number of matched utf8 codepoints
+    /* found a match at the end of *a, so didn't get a chance to test it */
+    if (0 < offset) {
+      chars++;
+      src += offset;
+      continue;
+    }
+
+    /* if a got to its terminating null byte, then we didn't find a match.
+     * Return the current number of matched utf8 codepoints */
     if ('\0' == *a) {
       return chars;
     }
@@ -802,302 +946,405 @@ size_t utf8spn(const void *src, const void *accept) {
   return chars;
 }
 
-void *utf8str(const void *haystack, const void *needle) {
-  const char *h = (const char *)haystack;
+utf8_constexpr14_impl utf8_int8_t *utf8str(const utf8_int8_t *haystack,
+                                           const utf8_int8_t *needle) {
+  utf8_int32_t throwaway_codepoint = 0;
 
-  // if needle has no utf8 codepoints before the null terminating
-  // byte then return haystack
-  if ('\0' == *((const char *)needle)) {
-    return (void *)haystack;
+  /* if needle has no utf8 codepoints before the null terminating
+   * byte then return haystack */
+  if ('\0' == *needle) {
+    return (utf8_int8_t *)haystack;
   }
 
-  while ('\0' != *h) {
-    const char *maybeMatch = h;
-    const char *n = (const char *)needle;
+  while ('\0' != *haystack) {
+    const utf8_int8_t *maybeMatch = haystack;
+    const utf8_int8_t *n = needle;
 
-    while (*h == *n && (*h != '\0' && *n != '\0')) {
+    while (*haystack == *n && (*haystack != '\0' && *n != '\0')) {
       n++;
-      h++;
+      haystack++;
     }
 
     if ('\0' == *n) {
-      // we found the whole utf8 string for needle in haystack at
-      // maybeMatch, so return it
-      return (void *)maybeMatch;
+      /* we found the whole utf8 string for needle in haystack at
+       * maybeMatch, so return it */
+      return (utf8_int8_t *)maybeMatch;
     } else {
-      // h could be in the middle of an unmatching utf8 codepoint,
-      // so we need to march it on to the next character beginning,
-      if ('\0' != *h) {
-        do {
-          h++;
-        } while (0x80 == (0xc0 & *h));
-      }
+      /* h could be in the middle of an unmatching utf8 codepoint,
+       * so we need to march it on to the next character beginning
+       * starting from the current character */
+      haystack = utf8codepoint(maybeMatch, &throwaway_codepoint);
     }
   }
 
-  // no match
+  /* no match */
   return utf8_null;
 }
 
-void *utf8casestr(const void *haystack, const void *needle) {
-  const void *h = haystack;
-
-  // if needle has no utf8 codepoints before the null terminating
-  // byte then return haystack
-  if ('\0' == *((const char *)needle)) {
-    return (void *)haystack;
+utf8_constexpr14_impl utf8_int8_t *utf8casestr(const utf8_int8_t *haystack,
+                                               const utf8_int8_t *needle) {
+  /* if needle has no utf8 codepoints before the null terminating
+   * byte then return haystack */
+  if ('\0' == *needle) {
+    return (utf8_int8_t *)haystack;
   }
 
   for (;;) {
-    const void *maybeMatch = h;
-    const void *n = needle;
-    utf8_int32_t h_cp, n_cp;
+    const utf8_int8_t *maybeMatch = haystack;
+    const utf8_int8_t *n = needle;
+    utf8_int32_t h_cp = 0, n_cp = 0;
 
-    h = utf8codepoint(h, &h_cp);
+    /* Get the next code point and track it */
+    const utf8_int8_t *nextH = haystack = utf8codepoint(haystack, &h_cp);
     n = utf8codepoint(n, &n_cp);
 
     while ((0 != h_cp) && (0 != n_cp)) {
       h_cp = utf8lwrcodepoint(h_cp);
       n_cp = utf8lwrcodepoint(n_cp);
 
-      // if we find a mismatch, bail out!
+      /* if we find a mismatch, bail out! */
       if (h_cp != n_cp) {
         break;
       }
 
-      h = utf8codepoint(h, &h_cp);
+      haystack = utf8codepoint(haystack, &h_cp);
       n = utf8codepoint(n, &n_cp);
     }
 
     if (0 == n_cp) {
-      // we found the whole utf8 string for needle in haystack at
-      // maybeMatch, so return it
-      return (void *)maybeMatch;
+      /* we found the whole utf8 string for needle in haystack at
+       * maybeMatch, so return it */
+      return (utf8_int8_t *)maybeMatch;
     }
 
     if (0 == h_cp) {
-      // no match
+      /* no match */
       return utf8_null;
     }
+
+    /* Roll back to the next code point in the haystack to test */
+    haystack = nextH;
   }
 }
 
-void *utf8valid(const void *str) {
-  const char *s = (const char *)str;
+utf8_constexpr14_impl utf8_int8_t *utf8valid(const utf8_int8_t *str) {
+  return utf8nvalid(str, SIZE_MAX);
+}
 
-  while ('\0' != *s) {
-    if (0xf0 == (0xf8 & *s)) {
-      // ensure each of the 3 following bytes in this 4-byte
-      // utf8 codepoint began with 0b10xxxxxx
-      if ((0x80 != (0xc0 & s[1])) || (0x80 != (0xc0 & s[2])) ||
-          (0x80 != (0xc0 & s[3]))) {
-        return (void *)s;
+utf8_constexpr14_impl utf8_int8_t *utf8nvalid(const utf8_int8_t *str,
+                                              size_t n) {
+  const utf8_int8_t *t = str;
+  size_t consumed = 0;
+
+  while ((void)(consumed = (size_t)(str - t)), consumed < n && '\0' != *str) {
+    const size_t remaining = n - consumed;
+
+    if (0xf0 == (0xf8 & *str)) {
+      /* ensure that there's 4 bytes or more remaining */
+      if (remaining < 4) {
+        return (utf8_int8_t *)str;
       }
 
-      // ensure that our utf8 codepoint ended after 4 bytes
-      if (0x80 == (0xc0 & s[4])) {
-        return (void *)s;
+      /* ensure each of the 3 following bytes in this 4-byte
+       * utf8 codepoint began with 0b10xxxxxx */
+      if ((0x80 != (0xc0 & str[1])) || (0x80 != (0xc0 & str[2])) ||
+          (0x80 != (0xc0 & str[3]))) {
+        return (utf8_int8_t *)str;
       }
 
-      // ensure that the top 5 bits of this 4-byte utf8
-      // codepoint were not 0, as then we could have used
-      // one of the smaller encodings
-      if ((0 == (0x07 & s[0])) && (0 == (0x30 & s[1]))) {
-        return (void *)s;
+      /* ensure that our utf8 codepoint ended after 4 bytes */
+      if ((remaining != 4) && (0x80 == (0xc0 & str[4]))) {
+        return (utf8_int8_t *)str;
       }
 
-      // 4-byte utf8 code point (began with 0b11110xxx)
-      s += 4;
-    } else if (0xe0 == (0xf0 & *s)) {
-      // ensure each of the 2 following bytes in this 3-byte
-      // utf8 codepoint began with 0b10xxxxxx
-      if ((0x80 != (0xc0 & s[1])) || (0x80 != (0xc0 & s[2]))) {
-        return (void *)s;
+      /* ensure that the top 5 bits of this 4-byte utf8
+       * codepoint were not 0, as then we could have used
+       * one of the smaller encodings */
+      if ((0 == (0x07 & str[0])) && (0 == (0x30 & str[1]))) {
+        return (utf8_int8_t *)str;
       }
 
-      // ensure that our utf8 codepoint ended after 3 bytes
-      if (0x80 == (0xc0 & s[3])) {
-        return (void *)s;
+      /* 4-byte utf8 code point (began with 0b11110xxx) */
+      str += 4;
+    } else if (0xe0 == (0xf0 & *str)) {
+      /* ensure that there's 3 bytes or more remaining */
+      if (remaining < 3) {
+        return (utf8_int8_t *)str;
       }
 
-      // ensure that the top 5 bits of this 3-byte utf8
-      // codepoint were not 0, as then we could have used
-      // one of the smaller encodings
-      if ((0 == (0x0f & s[0])) && (0 == (0x20 & s[1]))) {
-        return (void *)s;
+      /* ensure each of the 2 following bytes in this 3-byte
+       * utf8 codepoint began with 0b10xxxxxx */
+      if ((0x80 != (0xc0 & str[1])) || (0x80 != (0xc0 & str[2]))) {
+        return (utf8_int8_t *)str;
       }
 
-      // 3-byte utf8 code point (began with 0b1110xxxx)
-      s += 3;
-    } else if (0xc0 == (0xe0 & *s)) {
-      // ensure the 1 following byte in this 2-byte
-      // utf8 codepoint began with 0b10xxxxxx
-      if (0x80 != (0xc0 & s[1])) {
-        return (void *)s;
+      /* ensure that our utf8 codepoint ended after 3 bytes */
+      if ((remaining != 3) && (0x80 == (0xc0 & str[3]))) {
+        return (utf8_int8_t *)str;
       }
 
-      // ensure that our utf8 codepoint ended after 2 bytes
-      if (0x80 == (0xc0 & s[2])) {
-        return (void *)s;
+      /* ensure that the top 5 bits of this 3-byte utf8
+       * codepoint were not 0, as then we could have used
+       * one of the smaller encodings */
+      if ((0 == (0x0f & str[0])) && (0 == (0x20 & str[1]))) {
+        return (utf8_int8_t *)str;
       }
 
-      // ensure that the top 4 bits of this 2-byte utf8
-      // codepoint were not 0, as then we could have used
-      // one of the smaller encodings
-      if (0 == (0x1e & s[0])) {
-        return (void *)s;
+      /* 3-byte utf8 code point (began with 0b1110xxxx) */
+      str += 3;
+    } else if (0xc0 == (0xe0 & *str)) {
+      /* ensure that there's 2 bytes or more remaining */
+      if (remaining < 2) {
+        return (utf8_int8_t *)str;
       }
 
-      // 2-byte utf8 code point (began with 0b110xxxxx)
-      s += 2;
-    } else if (0x00 == (0x80 & *s)) {
-      // 1-byte ascii (began with 0b0xxxxxxx)
-      s += 1;
+      /* ensure the 1 following byte in this 2-byte
+       * utf8 codepoint began with 0b10xxxxxx */
+      if (0x80 != (0xc0 & str[1])) {
+        return (utf8_int8_t *)str;
+      }
+
+      /* ensure that our utf8 codepoint ended after 2 bytes */
+      if ((remaining != 2) && (0x80 == (0xc0 & str[2]))) {
+        return (utf8_int8_t *)str;
+      }
+
+      /* ensure that the top 4 bits of this 2-byte utf8
+       * codepoint were not 0, as then we could have used
+       * one of the smaller encodings */
+      if (0 == (0x1e & str[0])) {
+        return (utf8_int8_t *)str;
+      }
+
+      /* 2-byte utf8 code point (began with 0b110xxxxx) */
+      str += 2;
+    } else if (0x00 == (0x80 & *str)) {
+      /* 1-byte ascii (began with 0b0xxxxxxx) */
+      str += 1;
     } else {
-      // we have an invalid 0b1xxxxxxx utf8 code point entry
-      return (void *)s;
+      /* we have an invalid 0b1xxxxxxx utf8 code point entry */
+      return (utf8_int8_t *)str;
     }
   }
 
   return utf8_null;
 }
 
-void *utf8codepoint(const void *utf8_restrict str,
-                    utf8_int32_t *utf8_restrict out_codepoint) {
-  const char *s = (const char *)str;
+int utf8makevalid(utf8_int8_t *str, const utf8_int32_t replacement) {
+  utf8_int8_t *read = str;
+  utf8_int8_t *write = read;
+  const utf8_int8_t r = (utf8_int8_t)replacement;
+  utf8_int32_t codepoint = 0;
 
-  if (0xf0 == (0xf8 & s[0])) {
-    // 4 byte utf8 codepoint
-    *out_codepoint = ((0x07 & s[0]) << 18) | ((0x3f & s[1]) << 12) |
-                     ((0x3f & s[2]) << 6) | (0x3f & s[3]);
-    s += 4;
-  } else if (0xe0 == (0xf0 & s[0])) {
-    // 3 byte utf8 codepoint
-    *out_codepoint =
-        ((0x0f & s[0]) << 12) | ((0x3f & s[1]) << 6) | (0x3f & s[2]);
-    s += 3;
-  } else if (0xc0 == (0xe0 & s[0])) {
-    // 2 byte utf8 codepoint
-    *out_codepoint = ((0x1f & s[0]) << 6) | (0x3f & s[1]);
-    s += 2;
-  } else {
-    // 1 byte utf8 codepoint otherwise
-    *out_codepoint = s[0];
-    s += 1;
+  if (replacement > 0x7f) {
+    return -1;
   }
 
-  return (void *)s;
+  while ('\0' != *read) {
+    if (0xf0 == (0xf8 & *read)) {
+      /* ensure each of the 3 following bytes in this 4-byte
+       * utf8 codepoint began with 0b10xxxxxx */
+      if ((0x80 != (0xc0 & read[1])) || (0x80 != (0xc0 & read[2])) ||
+          (0x80 != (0xc0 & read[3]))) {
+        *write++ = r;
+        read++;
+        continue;
+      }
+
+      /* 4-byte utf8 code point (began with 0b11110xxx) */
+      read = utf8codepoint(read, &codepoint);
+      write = utf8catcodepoint(write, codepoint, 4);
+    } else if (0xe0 == (0xf0 & *read)) {
+      /* ensure each of the 2 following bytes in this 3-byte
+       * utf8 codepoint began with 0b10xxxxxx */
+      if ((0x80 != (0xc0 & read[1])) || (0x80 != (0xc0 & read[2]))) {
+        *write++ = r;
+        read++;
+        continue;
+      }
+
+      /* 3-byte utf8 code point (began with 0b1110xxxx) */
+      read = utf8codepoint(read, &codepoint);
+      write = utf8catcodepoint(write, codepoint, 3);
+    } else if (0xc0 == (0xe0 & *read)) {
+      /* ensure the 1 following byte in this 2-byte
+       * utf8 codepoint began with 0b10xxxxxx */
+      if (0x80 != (0xc0 & read[1])) {
+        *write++ = r;
+        read++;
+        continue;
+      }
+
+      /* 2-byte utf8 code point (began with 0b110xxxxx) */
+      read = utf8codepoint(read, &codepoint);
+      write = utf8catcodepoint(write, codepoint, 2);
+    } else if (0x00 == (0x80 & *read)) {
+      /* 1-byte ascii (began with 0b0xxxxxxx) */
+      read = utf8codepoint(read, &codepoint);
+      write = utf8catcodepoint(write, codepoint, 1);
+    } else {
+      /* if we got here then we've got a dangling continuation (0b10xxxxxx) */
+      *write++ = r;
+      read++;
+      continue;
+    }
+  }
+
+  *write = '\0';
+
+  return 0;
 }
 
-size_t utf8codepointsize(utf8_int32_t chr) {
+utf8_constexpr14_impl utf8_int8_t *
+utf8codepoint(const utf8_int8_t *utf8_restrict str,
+              utf8_int32_t *utf8_restrict out_codepoint) {
+  if (0xf0 == (0xf8 & str[0])) {
+    /* 4 byte utf8 codepoint */
+    *out_codepoint = ((0x07 & str[0]) << 18) | ((0x3f & str[1]) << 12) |
+                     ((0x3f & str[2]) << 6) | (0x3f & str[3]);
+    str += 4;
+  } else if (0xe0 == (0xf0 & str[0])) {
+    /* 3 byte utf8 codepoint */
+    *out_codepoint =
+        ((0x0f & str[0]) << 12) | ((0x3f & str[1]) << 6) | (0x3f & str[2]);
+    str += 3;
+  } else if (0xc0 == (0xe0 & str[0])) {
+    /* 2 byte utf8 codepoint */
+    *out_codepoint = ((0x1f & str[0]) << 6) | (0x3f & str[1]);
+    str += 2;
+  } else {
+    /* 1 byte utf8 codepoint otherwise */
+    *out_codepoint = str[0];
+    str += 1;
+  }
+
+  return (utf8_int8_t *)str;
+}
+
+utf8_constexpr14_impl size_t utf8codepointcalcsize(const utf8_int8_t *str) {
+  if (0xf0 == (0xf8 & str[0])) {
+    /* 4 byte utf8 codepoint */
+    return 4;
+  } else if (0xe0 == (0xf0 & str[0])) {
+    /* 3 byte utf8 codepoint */
+    return 3;
+  } else if (0xc0 == (0xe0 & str[0])) {
+    /* 2 byte utf8 codepoint */
+    return 2;
+  }
+
+  /* 1 byte utf8 codepoint otherwise */
+  return 1;
+}
+
+utf8_constexpr14_impl size_t utf8codepointsize(utf8_int32_t chr) {
   if (0 == ((utf8_int32_t)0xffffff80 & chr)) {
     return 1;
   } else if (0 == ((utf8_int32_t)0xfffff800 & chr)) {
     return 2;
   } else if (0 == ((utf8_int32_t)0xffff0000 & chr)) {
     return 3;
-  } else { // if (0 == ((int)0xffe00000 & chr)) {
+  } else { /* if (0 == ((int)0xffe00000 & chr)) { */
     return 4;
   }
 }
 
-void *utf8catcodepoint(void *utf8_restrict str, utf8_int32_t chr, size_t n) {
-  char *s = (char *)str;
-
+utf8_int8_t *utf8catcodepoint(utf8_int8_t *str, utf8_int32_t chr, size_t n) {
   if (0 == ((utf8_int32_t)0xffffff80 & chr)) {
-    // 1-byte/7-bit ascii
-    // (0b0xxxxxxx)
+    /* 1-byte/7-bit ascii
+     * (0b0xxxxxxx) */
     if (n < 1) {
       return utf8_null;
     }
-    s[0] = (char)chr;
-    s += 1;
+    str[0] = (utf8_int8_t)chr;
+    str += 1;
   } else if (0 == ((utf8_int32_t)0xfffff800 & chr)) {
-    // 2-byte/11-bit utf8 code point
-    // (0b110xxxxx 0b10xxxxxx)
+    /* 2-byte/11-bit utf8 code point
+     * (0b110xxxxx 0b10xxxxxx) */
     if (n < 2) {
       return utf8_null;
     }
-    s[0] = 0xc0 | (char)(chr >> 6);
-    s[1] = 0x80 | (char)(chr & 0x3f);
-    s += 2;
+    str[0] = (utf8_int8_t)(0xc0 | (utf8_int8_t)((chr >> 6) & 0x1f));
+    str[1] = (utf8_int8_t)(0x80 | (utf8_int8_t)(chr & 0x3f));
+    str += 2;
   } else if (0 == ((utf8_int32_t)0xffff0000 & chr)) {
-    // 3-byte/16-bit utf8 code point
-    // (0b1110xxxx 0b10xxxxxx 0b10xxxxxx)
+    /* 3-byte/16-bit utf8 code point
+     * (0b1110xxxx 0b10xxxxxx 0b10xxxxxx) */
     if (n < 3) {
       return utf8_null;
     }
-    s[0] = 0xe0 | (char)(chr >> 12);
-    s[1] = 0x80 | (char)((chr >> 6) & 0x3f);
-    s[2] = 0x80 | (char)(chr & 0x3f);
-    s += 3;
-  } else { // if (0 == ((int)0xffe00000 & chr)) {
-    // 4-byte/21-bit utf8 code point
-    // (0b11110xxx 0b10xxxxxx 0b10xxxxxx 0b10xxxxxx)
+    str[0] = (utf8_int8_t)(0xe0 | (utf8_int8_t)((chr >> 12) & 0x0f));
+    str[1] = (utf8_int8_t)(0x80 | (utf8_int8_t)((chr >> 6) & 0x3f));
+    str[2] = (utf8_int8_t)(0x80 | (utf8_int8_t)(chr & 0x3f));
+    str += 3;
+  } else { /* if (0 == ((int)0xffe00000 & chr)) { */
+    /* 4-byte/21-bit utf8 code point
+     * (0b11110xxx 0b10xxxxxx 0b10xxxxxx 0b10xxxxxx) */
     if (n < 4) {
       return utf8_null;
     }
-    s[0] = 0xf0 | (char)(chr >> 18);
-    s[1] = 0x80 | (char)((chr >> 12) & 0x3f);
-    s[2] = 0x80 | (char)((chr >> 6) & 0x3f);
-    s[3] = 0x80 | (char)(chr & 0x3f);
-    s += 4;
+    str[0] = (utf8_int8_t)(0xf0 | (utf8_int8_t)((chr >> 18) & 0x07));
+    str[1] = (utf8_int8_t)(0x80 | (utf8_int8_t)((chr >> 12) & 0x3f));
+    str[2] = (utf8_int8_t)(0x80 | (utf8_int8_t)((chr >> 6) & 0x3f));
+    str[3] = (utf8_int8_t)(0x80 | (utf8_int8_t)(chr & 0x3f));
+    str += 4;
   }
 
-  return s;
+  return str;
 }
 
-int utf8islower(utf8_int32_t chr) { return chr != utf8uprcodepoint(chr); }
+utf8_constexpr14_impl int utf8islower(utf8_int32_t chr) {
+  return chr != utf8uprcodepoint(chr);
+}
 
-int utf8isupper(utf8_int32_t chr) { return chr != utf8lwrcodepoint(chr); }
+utf8_constexpr14_impl int utf8isupper(utf8_int32_t chr) {
+  return chr != utf8lwrcodepoint(chr);
+}
 
-void utf8lwr(void *utf8_restrict str) {
-  void *p, *pn;
-  utf8_int32_t cp;
-
-  p = (char *)str;
-  pn = utf8codepoint(p, &cp);
+void utf8lwr(utf8_int8_t *utf8_restrict str) {
+  utf8_int32_t cp = 0;
+  utf8_int8_t *pn = utf8codepoint(str, &cp);
 
   while (cp != 0) {
     const utf8_int32_t lwr_cp = utf8lwrcodepoint(cp);
     const size_t size = utf8codepointsize(lwr_cp);
 
     if (lwr_cp != cp) {
-      utf8catcodepoint(p, lwr_cp, size);
+      utf8catcodepoint(str, lwr_cp, size);
     }
 
-    p = pn;
-    pn = utf8codepoint(p, &cp);
+    str = pn;
+    pn = utf8codepoint(str, &cp);
   }
 }
 
-void utf8upr(void *utf8_restrict str) {
-  void *p, *pn;
-  utf8_int32_t cp;
-
-  p = (char *)str;
-  pn = utf8codepoint(p, &cp);
+void utf8upr(utf8_int8_t *utf8_restrict str) {
+  utf8_int32_t cp = 0;
+  utf8_int8_t *pn = utf8codepoint(str, &cp);
 
   while (cp != 0) {
     const utf8_int32_t lwr_cp = utf8uprcodepoint(cp);
     const size_t size = utf8codepointsize(lwr_cp);
 
     if (lwr_cp != cp) {
-      utf8catcodepoint(p, lwr_cp, size);
+      utf8catcodepoint(str, lwr_cp, size);
     }
 
-    p = pn;
-    pn = utf8codepoint(p, &cp);
+    str = pn;
+    pn = utf8codepoint(str, &cp);
   }
 }
 
-utf8_int32_t utf8lwrcodepoint(utf8_int32_t cp) {
+utf8_constexpr14_impl utf8_int32_t utf8lwrcodepoint(utf8_int32_t cp) {
   if (((0x0041 <= cp) && (0x005a >= cp)) ||
       ((0x00c0 <= cp) && (0x00d6 >= cp)) ||
       ((0x00d8 <= cp) && (0x00de >= cp)) ||
       ((0x0391 <= cp) && (0x03a1 >= cp)) ||
-      ((0x03a3 <= cp) && (0x03ab >= cp))) {
+      ((0x03a3 <= cp) && (0x03ab >= cp)) ||
+      ((0x0410 <= cp) && (0x042f >= cp))) {
     cp += 32;
+  } else if ((0x0400 <= cp) && (0x040f >= cp)) {
+    cp += 80;
   } else if (((0x0100 <= cp) && (0x012f >= cp)) ||
              ((0x0132 <= cp) && (0x0137 >= cp)) ||
              ((0x014a <= cp) && (0x0177 >= cp)) ||
@@ -1107,7 +1354,9 @@ utf8_int32_t utf8lwrcodepoint(utf8_int32_t cp) {
              ((0x01f8 <= cp) && (0x021f >= cp)) ||
              ((0x0222 <= cp) && (0x0233 >= cp)) ||
              ((0x0246 <= cp) && (0x024f >= cp)) ||
-             ((0x03d8 <= cp) && (0x03ef >= cp))) {
+             ((0x03d8 <= cp) && (0x03ef >= cp)) ||
+             ((0x0460 <= cp) && (0x0481 >= cp)) ||
+             ((0x048a <= cp) && (0x04ff >= cp))) {
     cp |= 0x1;
   } else if (((0x0139 <= cp) && (0x0148 >= cp)) ||
              ((0x0179 <= cp) && (0x017e >= cp)) ||
@@ -1118,62 +1367,147 @@ utf8_int32_t utf8lwrcodepoint(utf8_int32_t cp) {
     cp &= ~0x1;
   } else {
     switch (cp) {
-    default: break;
-    case 0x0178: cp = 0x00ff; break;
-    case 0x0243: cp = 0x0180; break;
-    case 0x018e: cp = 0x01dd; break;
-    case 0x023d: cp = 0x019a; break;
-    case 0x0220: cp = 0x019e; break;
-    case 0x01b7: cp = 0x0292; break;
-    case 0x01c4: cp = 0x01c6; break;
-    case 0x01c7: cp = 0x01c9; break;
-    case 0x01ca: cp = 0x01cc; break;
-    case 0x01f1: cp = 0x01f3; break;
-    case 0x01f7: cp = 0x01bf; break;
-    case 0x0187: cp = 0x0188; break;
-    case 0x018b: cp = 0x018c; break;
-    case 0x0191: cp = 0x0192; break;
-    case 0x0198: cp = 0x0199; break;
-    case 0x01a7: cp = 0x01a8; break;
-    case 0x01ac: cp = 0x01ad; break;
-    case 0x01af: cp = 0x01b0; break;
-    case 0x01b8: cp = 0x01b9; break;
-    case 0x01bc: cp = 0x01bd; break;
-    case 0x01f4: cp = 0x01f5; break;
-    case 0x023b: cp = 0x023c; break;
-    case 0x0241: cp = 0x0242; break;
-    case 0x03fd: cp = 0x037b; break;
-    case 0x03fe: cp = 0x037c; break;
-    case 0x03ff: cp = 0x037d; break;
-    case 0x037f: cp = 0x03f3; break;
-    case 0x0386: cp = 0x03ac; break;
-    case 0x0388: cp = 0x03ad; break;
-    case 0x0389: cp = 0x03ae; break;
-    case 0x038a: cp = 0x03af; break;
-    case 0x038c: cp = 0x03cc; break;
-    case 0x038e: cp = 0x03cd; break;
-    case 0x038f: cp = 0x03ce; break;
-    case 0x0370: cp = 0x0371; break;
-    case 0x0372: cp = 0x0373; break;
-    case 0x0376: cp = 0x0377; break;
-    case 0x03f4: cp = 0x03d1; break;
-    case 0x03cf: cp = 0x03d7; break;
-    case 0x03f9: cp = 0x03f2; break;
-    case 0x03f7: cp = 0x03f8; break;
-    case 0x03fa: cp = 0x03fb; break;
-    };
+    default:
+      break;
+    case 0x0178:
+      cp = 0x00ff;
+      break;
+    case 0x0243:
+      cp = 0x0180;
+      break;
+    case 0x018e:
+      cp = 0x01dd;
+      break;
+    case 0x023d:
+      cp = 0x019a;
+      break;
+    case 0x0220:
+      cp = 0x019e;
+      break;
+    case 0x01b7:
+      cp = 0x0292;
+      break;
+    case 0x01c4:
+      cp = 0x01c6;
+      break;
+    case 0x01c7:
+      cp = 0x01c9;
+      break;
+    case 0x01ca:
+      cp = 0x01cc;
+      break;
+    case 0x01f1:
+      cp = 0x01f3;
+      break;
+    case 0x01f7:
+      cp = 0x01bf;
+      break;
+    case 0x0187:
+      cp = 0x0188;
+      break;
+    case 0x018b:
+      cp = 0x018c;
+      break;
+    case 0x0191:
+      cp = 0x0192;
+      break;
+    case 0x0198:
+      cp = 0x0199;
+      break;
+    case 0x01a7:
+      cp = 0x01a8;
+      break;
+    case 0x01ac:
+      cp = 0x01ad;
+      break;
+    case 0x01b8:
+      cp = 0x01b9;
+      break;
+    case 0x01bc:
+      cp = 0x01bd;
+      break;
+    case 0x01f4:
+      cp = 0x01f5;
+      break;
+    case 0x023b:
+      cp = 0x023c;
+      break;
+    case 0x0241:
+      cp = 0x0242;
+      break;
+    case 0x03fd:
+      cp = 0x037b;
+      break;
+    case 0x03fe:
+      cp = 0x037c;
+      break;
+    case 0x03ff:
+      cp = 0x037d;
+      break;
+    case 0x037f:
+      cp = 0x03f3;
+      break;
+    case 0x0386:
+      cp = 0x03ac;
+      break;
+    case 0x0388:
+      cp = 0x03ad;
+      break;
+    case 0x0389:
+      cp = 0x03ae;
+      break;
+    case 0x038a:
+      cp = 0x03af;
+      break;
+    case 0x038c:
+      cp = 0x03cc;
+      break;
+    case 0x038e:
+      cp = 0x03cd;
+      break;
+    case 0x038f:
+      cp = 0x03ce;
+      break;
+    case 0x0370:
+      cp = 0x0371;
+      break;
+    case 0x0372:
+      cp = 0x0373;
+      break;
+    case 0x0376:
+      cp = 0x0377;
+      break;
+    case 0x03f4:
+      cp = 0x03b8;
+      break;
+    case 0x03cf:
+      cp = 0x03d7;
+      break;
+    case 0x03f9:
+      cp = 0x03f2;
+      break;
+    case 0x03f7:
+      cp = 0x03f8;
+      break;
+    case 0x03fa:
+      cp = 0x03fb;
+      break;
+    }
   }
 
   return cp;
 }
 
-utf8_int32_t utf8uprcodepoint(utf8_int32_t cp) {
+utf8_constexpr14_impl utf8_int32_t utf8uprcodepoint(utf8_int32_t cp) {
   if (((0x0061 <= cp) && (0x007a >= cp)) ||
       ((0x00e0 <= cp) && (0x00f6 >= cp)) ||
       ((0x00f8 <= cp) && (0x00fe >= cp)) ||
       ((0x03b1 <= cp) && (0x03c1 >= cp)) ||
-      ((0x03c3 <= cp) && (0x03cb >= cp))) {
+      ((0x03c3 <= cp) && (0x03cb >= cp)) ||
+      ((0x0430 <= cp) && (0x044f >= cp))) {
     cp -= 32;
+  } else if ((0x0450 <= cp) && (0x045f >= cp)) {
+    cp -= 80;
   } else if (((0x0100 <= cp) && (0x012f >= cp)) ||
              ((0x0132 <= cp) && (0x0137 >= cp)) ||
              ((0x014a <= cp) && (0x0177 >= cp)) ||
@@ -1183,7 +1517,9 @@ utf8_int32_t utf8uprcodepoint(utf8_int32_t cp) {
              ((0x01f8 <= cp) && (0x021f >= cp)) ||
              ((0x0222 <= cp) && (0x0233 >= cp)) ||
              ((0x0246 <= cp) && (0x024f >= cp)) ||
-             ((0x03d8 <= cp) && (0x03ef >= cp))) {
+             ((0x03d8 <= cp) && (0x03ef >= cp)) ||
+             ((0x0460 <= cp) && (0x0481 >= cp)) ||
+             ((0x048a <= cp) && (0x04ff >= cp))) {
     cp &= ~0x1;
   } else if (((0x0139 <= cp) && (0x0148 >= cp)) ||
              ((0x0179 <= cp) && (0x017e >= cp)) ||
@@ -1194,64 +1530,175 @@ utf8_int32_t utf8uprcodepoint(utf8_int32_t cp) {
     cp |= 0x1;
   } else {
     switch (cp) {
-    default: break;
-    case 0x00ff: cp = 0x0178; break;
-    case 0x0180: cp = 0x0243; break;
-    case 0x01dd: cp = 0x018e; break;
-    case 0x019a: cp = 0x023d; break;
-    case 0x019e: cp = 0x0220; break;
-    case 0x0292: cp = 0x01b7; break;
-    case 0x01c6: cp = 0x01c4; break;
-    case 0x01c9: cp = 0x01c7; break;
-    case 0x01cc: cp = 0x01ca; break;
-    case 0x01f3: cp = 0x01f1; break;
-    case 0x01bf: cp = 0x01f7; break;
-    case 0x0188: cp = 0x0187; break;
-    case 0x018c: cp = 0x018b; break;
-    case 0x0192: cp = 0x0191; break;
-    case 0x0199: cp = 0x0198; break;
-    case 0x01a8: cp = 0x01a7; break;
-    case 0x01ad: cp = 0x01ac; break;
-    case 0x01b0: cp = 0x01af; break;
-    case 0x01b9: cp = 0x01b8; break;
-    case 0x01bd: cp = 0x01bc; break;
-    case 0x01f5: cp = 0x01f4; break;
-    case 0x023c: cp = 0x023b; break;
-    case 0x0242: cp = 0x0241; break;
-    case 0x037b: cp = 0x03fd; break;
-    case 0x037c: cp = 0x03fe; break;
-    case 0x037d: cp = 0x03ff; break;
-    case 0x03f3: cp = 0x037f; break;
-    case 0x03ac: cp = 0x0386; break;
-    case 0x03ad: cp = 0x0388; break;
-    case 0x03ae: cp = 0x0389; break;
-    case 0x03af: cp = 0x038a; break;
-    case 0x03cc: cp = 0x038c; break;
-    case 0x03cd: cp = 0x038e; break;
-    case 0x03ce: cp = 0x038f; break;
-    case 0x0371: cp = 0x0370; break;
-    case 0x0373: cp = 0x0372; break;
-    case 0x0377: cp = 0x0376; break;
-    case 0x03d1: cp = 0x03f4; break;
-    case 0x03d7: cp = 0x03cf; break;
-    case 0x03f2: cp = 0x03f9; break;
-    case 0x03f8: cp = 0x03f7; break;
-    case 0x03fb: cp = 0x03fa; break;
-    };
+    default:
+      break;
+    case 0x00ff:
+      cp = 0x0178;
+      break;
+    case 0x0180:
+      cp = 0x0243;
+      break;
+    case 0x01dd:
+      cp = 0x018e;
+      break;
+    case 0x019a:
+      cp = 0x023d;
+      break;
+    case 0x019e:
+      cp = 0x0220;
+      break;
+    case 0x0292:
+      cp = 0x01b7;
+      break;
+    case 0x01c6:
+      cp = 0x01c4;
+      break;
+    case 0x01c9:
+      cp = 0x01c7;
+      break;
+    case 0x01cc:
+      cp = 0x01ca;
+      break;
+    case 0x01f3:
+      cp = 0x01f1;
+      break;
+    case 0x01bf:
+      cp = 0x01f7;
+      break;
+    case 0x0188:
+      cp = 0x0187;
+      break;
+    case 0x018c:
+      cp = 0x018b;
+      break;
+    case 0x0192:
+      cp = 0x0191;
+      break;
+    case 0x0199:
+      cp = 0x0198;
+      break;
+    case 0x01a8:
+      cp = 0x01a7;
+      break;
+    case 0x01ad:
+      cp = 0x01ac;
+      break;
+    case 0x01b9:
+      cp = 0x01b8;
+      break;
+    case 0x01bd:
+      cp = 0x01bc;
+      break;
+    case 0x01f5:
+      cp = 0x01f4;
+      break;
+    case 0x023c:
+      cp = 0x023b;
+      break;
+    case 0x0242:
+      cp = 0x0241;
+      break;
+    case 0x037b:
+      cp = 0x03fd;
+      break;
+    case 0x037c:
+      cp = 0x03fe;
+      break;
+    case 0x037d:
+      cp = 0x03ff;
+      break;
+    case 0x03f3:
+      cp = 0x037f;
+      break;
+    case 0x03ac:
+      cp = 0x0386;
+      break;
+    case 0x03ad:
+      cp = 0x0388;
+      break;
+    case 0x03ae:
+      cp = 0x0389;
+      break;
+    case 0x03af:
+      cp = 0x038a;
+      break;
+    case 0x03cc:
+      cp = 0x038c;
+      break;
+    case 0x03cd:
+      cp = 0x038e;
+      break;
+    case 0x03ce:
+      cp = 0x038f;
+      break;
+    case 0x0371:
+      cp = 0x0370;
+      break;
+    case 0x0373:
+      cp = 0x0372;
+      break;
+    case 0x0377:
+      cp = 0x0376;
+      break;
+    case 0x03d1:
+      cp = 0x0398;
+      break;
+    case 0x03d7:
+      cp = 0x03cf;
+      break;
+    case 0x03f2:
+      cp = 0x03f9;
+      break;
+    case 0x03f8:
+      cp = 0x03f7;
+      break;
+    case 0x03fb:
+      cp = 0x03fa;
+      break;
+    }
   }
 
   return cp;
 }
 
+utf8_constexpr14_impl utf8_int8_t *
+utf8rcodepoint(const utf8_int8_t *utf8_restrict str,
+               utf8_int32_t *utf8_restrict out_codepoint) {
+  const utf8_int8_t *s = (const utf8_int8_t *)str;
+
+  if (0xf0 == (0xf8 & s[0])) {
+    /* 4 byte utf8 codepoint */
+    *out_codepoint = ((0x07 & s[0]) << 18) | ((0x3f & s[1]) << 12) |
+                     ((0x3f & s[2]) << 6) | (0x3f & s[3]);
+  } else if (0xe0 == (0xf0 & s[0])) {
+    /* 3 byte utf8 codepoint */
+    *out_codepoint =
+        ((0x0f & s[0]) << 12) | ((0x3f & s[1]) << 6) | (0x3f & s[2]);
+  } else if (0xc0 == (0xe0 & s[0])) {
+    /* 2 byte utf8 codepoint */
+    *out_codepoint = ((0x1f & s[0]) << 6) | (0x3f & s[1]);
+  } else {
+    /* 1 byte utf8 codepoint otherwise */
+    *out_codepoint = s[0];
+  }
+
+  do {
+    s--;
+  } while ((0 != (0x80 & s[0])) && (0x80 == (0xc0 & s[0])));
+
+  return (utf8_int8_t *)s;
+}
+
 #undef utf8_restrict
+#undef utf8_constexpr14
 #undef utf8_null
 
 #ifdef __cplusplus
-} // extern "C"
+} /* extern "C" */
 #endif
 
 #if defined(__clang__)
 #pragma clang diagnostic pop
 #endif
 
-#endif // SHEREDOM_UTF8_H_INCLUDED
+#endif /* SHEREDOM_UTF8_H_INCLUDED */
diff --git a/examples/CMakeLists.txt b/examples/CMakeLists.txt
index 8e38c7d4e..ad1a5deb6 100644
--- a/examples/CMakeLists.txt
+++ b/examples/CMakeLists.txt
@@ -4,6 +4,7 @@ file(GLOB SRC_EXAMPLES *.c *.h)
 
 add_executable(lg2 ${SRC_EXAMPLES})
 set_target_properties(lg2 PROPERTIES C_STANDARD 90)
+set_target_properties(lg2 PROPERTIES C_EXTENSIONS OFF)
 
 # Ensure that we do not use deprecated functions internally
 add_definitions(-DGIT_DEPRECATE_HARD)
diff --git a/fuzzers/CMakeLists.txt b/fuzzers/CMakeLists.txt
index 01f0f5199..2961b92c5 100644
--- a/fuzzers/CMakeLists.txt
+++ b/fuzzers/CMakeLists.txt
@@ -21,7 +21,7 @@ foreach(fuzz_target_src ${SRC_FUZZERS})
 
 	add_executable(${fuzz_target_name} ${${fuzz_target_name}_SOURCES})
 	set_target_properties(${fuzz_target_name} PROPERTIES C_STANDARD 90)
-
+	set_target_properties(${fuzz_target_name} PROPERTIES C_EXTENSIONS OFF)
 	target_include_directories(${fuzz_target_name} PRIVATE ${LIBGIT2_INCLUDES} ${LIBGIT2_DEPENDENCY_INCLUDES})
 	target_include_directories(${fuzz_target_name} SYSTEM PRIVATE ${LIBGIT2_SYSTEM_INCLUDES})
 
diff --git a/include/git2/stdint.h b/include/git2/stdint.h
index 6950427d2..4b235c73f 100644
--- a/include/git2/stdint.h
+++ b/include/git2/stdint.h
@@ -1,37 +1,37 @@
-// ISO C9x  compliant stdint.h for Microsoft Visual Studio
-// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124
-//
-//  Copyright (c) 2006-2008 Alexander Chemeris
-//
-// Redistribution and use in source and binary forms, with or without
-// modification, are permitted provided that the following conditions are met:
-//
-//   1. Redistributions of source code must retain the above copyright notice,
-//      this list of conditions and the following disclaimer.
-//
-//   2. Redistributions in binary form must reproduce the above copyright
-//      notice, this list of conditions and the following disclaimer in the
-//      documentation and/or other materials provided with the distribution.
-//
-//   3. The name of the author may be used to endorse or promote products
-//      derived from this software without specific prior written permission.
-//
-// THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
-// WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
-// MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
-// EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
-// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
-// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
-// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
-// WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
-// OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
-// ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-//
-///////////////////////////////////////////////////////////////////////////////
+/* ISO C9x  compliant stdint.h for Microsoft Visual Studio
+ * Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124
+ *
+ *  Copyright (c) 2006-2008 Alexander Chemeris
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ *   1. Redistributions of source code must retain the above copyright notice,
+ *      this list of conditions and the following disclaimer.
+ *
+ *   2. Redistributions in binary form must reproduce the above copyright
+ *      notice, this list of conditions and the following disclaimer in the
+ *      documentation and/or other materials provided with the distribution.
+ *
+ *   3. The name of the author may be used to endorse or promote products
+ *      derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
+ * EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+ * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
+ * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
+ * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
+ * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
+ * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *******************************************************************************/
 
-#ifdef _MSC_VER // [
+#ifdef _MSC_VER /* [ */
 
-#ifndef _MSC_STDINT_H_ // [
+#ifndef _MSC_STDINT_H_ /* [ */
 #define _MSC_STDINT_H_
 
 #if _MSC_VER > 1000
@@ -40,10 +40,11 @@
 
 #include <limits.h>
 
-// For Visual Studio 6 in C++ mode and for many Visual Studio versions when
-// compiling for ARM we should wrap <wchar.h> include with 'extern "C++" {}'
-// or compiler give many errors like this:
-//   error C2733: second C linkage of overloaded function 'wmemchr' not allowed
+/* For Visual Studio 6 in C++ mode and for many Visual Studio versions when
+ * compiling for ARM we should wrap <wchar.h> include with 'extern "C++" {}'
+ * or compiler give many errors like this:
+ *   error C2733: second C linkage of overloaded function 'wmemchr' not allowed
+*/
 #ifdef __cplusplus
 extern "C" {
 #endif
@@ -52,7 +53,7 @@ extern "C" {
 }
 #endif
 
-// Define _W64 macros to mark types changing their size, like intptr_t.
+/* Define _W64 macros to mark types changing their size, like intptr_t. */
 #ifndef _W64
 #  if !defined(__midl) && (defined(_X86_) || defined(_M_IX86)) && _MSC_VER >= 1300
 #     define _W64 __w64
@@ -62,13 +63,14 @@ extern "C" {
 #endif
 
 
-// 7.18.1 Integer types
-
-// 7.18.1.1 Exact-width integer types
-
-// Visual Studio 6 and Embedded Visual C++ 4 doesn't
-// realize that, e.g. char has the same size as __int8
-// so we give up on __intX for them.
+/* 7.18.1 Integer types
+ *
+ * 7.18.1.1 Exact-width integer types
+ *
+ * Visual Studio 6 and Embedded Visual C++ 4 doesn't
+ * realize that, e.g. char has the same size as __int8
+ * so we give up on __intX for them.
+ */
 #if (_MSC_VER < 1300)
    typedef signed char       int8_t;
    typedef signed short      int16_t;
@@ -88,7 +90,7 @@ typedef signed __int64       int64_t;
 typedef unsigned __int64     uint64_t;
 
 
-// 7.18.1.2 Minimum-width integer types
+/* 7.18.1.2 Minimum-width integer types */
 typedef int8_t    int_least8_t;
 typedef int16_t   int_least16_t;
 typedef int32_t   int_least32_t;
@@ -98,7 +100,7 @@ typedef uint16_t  uint_least16_t;
 typedef uint32_t  uint_least32_t;
 typedef uint64_t  uint_least64_t;
 
-// 7.18.1.3 Fastest minimum-width integer types
+/* 7.18.1.3 Fastest minimum-width integer types */
 typedef int8_t    int_fast8_t;
 typedef int16_t   int_fast16_t;
 typedef int32_t   int_fast32_t;
@@ -108,25 +110,25 @@ typedef uint16_t  uint_fast16_t;
 typedef uint32_t  uint_fast32_t;
 typedef uint64_t  uint_fast64_t;
 
-// 7.18.1.4 Integer types capable of holding object pointers
-#ifdef _WIN64 // [
+/* 7.18.1.4 Integer types capable of holding object pointers */
+#ifdef _WIN64 /* [ */
    typedef signed __int64    intptr_t;
    typedef unsigned __int64  uintptr_t;
-#else // _WIN64 ][
+#else /* _WIN64 ][ */
    typedef _W64 signed int   intptr_t;
    typedef _W64 unsigned int uintptr_t;
-#endif // _WIN64 ]
+#endif /* _WIN64 ] */
 
-// 7.18.1.5 Greatest-width integer types
+/* 7.18.1.5 Greatest-width integer types */
 typedef int64_t   intmax_t;
 typedef uint64_t  uintmax_t;
 
 
-// 7.18.2 Limits of specified-width integer types
+/* 7.18.2 Limits of specified-width integer types */
 
-#if !defined(__cplusplus) || defined(__STDC_LIMIT_MACROS) // [   See footnote 220 at page 257 and footnote 221 at page 259
+#if !defined(__cplusplus) || defined(__STDC_LIMIT_MACROS) /* [   See footnote 220 at page 257 and footnote 221 at page 259 */
 
-// 7.18.2.1 Limits of exact-width integer types
+/* 7.18.2.1 Limits of exact-width integer types */
 #define INT8_MIN     ((int8_t)_I8_MIN)
 #define INT8_MAX     _I8_MAX
 #define INT16_MIN    ((int16_t)_I16_MIN)
@@ -140,7 +142,7 @@ typedef uint64_t  uintmax_t;
 #define UINT32_MAX   _UI32_MAX
 #define UINT64_MAX   _UI64_MAX
 
-// 7.18.2.2 Limits of minimum-width integer types
+/* 7.18.2.2 Limits of minimum-width integer types */
 #define INT_LEAST8_MIN    INT8_MIN
 #define INT_LEAST8_MAX    INT8_MAX
 #define INT_LEAST16_MIN   INT16_MIN
@@ -154,7 +156,7 @@ typedef uint64_t  uintmax_t;
 #define UINT_LEAST32_MAX  UINT32_MAX
 #define UINT_LEAST64_MAX  UINT64_MAX
 
-// 7.18.2.3 Limits of fastest minimum-width integer types
+/* 7.18.2.3 Limits of fastest minimum-width integer types */
 #define INT_FAST8_MIN    INT8_MIN
 #define INT_FAST8_MAX    INT8_MAX
 #define INT_FAST16_MIN   INT16_MIN
@@ -168,62 +170,62 @@ typedef uint64_t  uintmax_t;
 #define UINT_FAST32_MAX  UINT32_MAX
 #define UINT_FAST64_MAX  UINT64_MAX
 
-// 7.18.2.4 Limits of integer types capable of holding object pointers
-#ifdef _WIN64 // [
+/* 7.18.2.4 Limits of integer types capable of holding object pointers */
+#ifdef _WIN64 /* [ */
 #  define INTPTR_MIN   INT64_MIN
 #  define INTPTR_MAX   INT64_MAX
 #  define UINTPTR_MAX  UINT64_MAX
-#else // _WIN64 ][
+#else /* _WIN64 ][ */
 #  define INTPTR_MIN   INT32_MIN
 #  define INTPTR_MAX   INT32_MAX
 #  define UINTPTR_MAX  UINT32_MAX
-#endif // _WIN64 ]
+#endif /* _WIN64 ] */
 
-// 7.18.2.5 Limits of greatest-width integer types
+/* 7.18.2.5 Limits of greatest-width integer types */
 #define INTMAX_MIN   INT64_MIN
 #define INTMAX_MAX   INT64_MAX
 #define UINTMAX_MAX  UINT64_MAX
 
-// 7.18.3 Limits of other integer types
+/* 7.18.3 Limits of other integer types */
 
-#ifdef _WIN64 // [
+#ifdef _WIN64 /* [ */
 #  define PTRDIFF_MIN  _I64_MIN
 #  define PTRDIFF_MAX  _I64_MAX
-#else  // _WIN64 ][
+#else  /* _WIN64 ][ */
 #  define PTRDIFF_MIN  _I32_MIN
 #  define PTRDIFF_MAX  _I32_MAX
-#endif  // _WIN64 ]
+#endif  /* _WIN64 ] */
 
 #define SIG_ATOMIC_MIN  INT_MIN
 #define SIG_ATOMIC_MAX  INT_MAX
 
-#ifndef SIZE_MAX // [
-#  ifdef _WIN64 // [
+#ifndef SIZE_MAX /* [ */
+#  ifdef _WIN64 /* [ */
 #     define SIZE_MAX  _UI64_MAX
-#  else // _WIN64 ][
+#  else /* _WIN64 ][  */
 #     define SIZE_MAX  _UI32_MAX
-#  endif // _WIN64 ]
-#endif // SIZE_MAX ]
+#  endif /* _WIN64 ] */
+#endif /* SIZE_MAX ] */
 
-// WCHAR_MIN and WCHAR_MAX are also defined in <wchar.h>
-#ifndef WCHAR_MIN // [
+/* WCHAR_MIN and WCHAR_MAX are also defined in <wchar.h> */
+#ifndef WCHAR_MIN /* [ */
 #  define WCHAR_MIN  0
-#endif  // WCHAR_MIN ]
-#ifndef WCHAR_MAX // [
+#endif  /* WCHAR_MIN ] */
+#ifndef WCHAR_MAX /* [ */
 #  define WCHAR_MAX  _UI16_MAX
-#endif  // WCHAR_MAX ]
+#endif  /* WCHAR_MAX ] */
 
 #define WINT_MIN  0
 #define WINT_MAX  _UI16_MAX
 
-#endif // __STDC_LIMIT_MACROS ]
+#endif /* __STDC_LIMIT_MACROS ] */
 
 
-// 7.18.4 Limits of other integer types
+/* 7.18.4 Limits of other integer types
 
-#if !defined(__cplusplus) || defined(__STDC_CONSTANT_MACROS) // [   See footnote 224 at page 260
+#if !defined(__cplusplus) || defined(__STDC_CONSTANT_MACROS) /* [    See footnote 224 at page 260 */
 
-// 7.18.4.1 Macros for minimum-width integer constants
+/* 7.18.4.1 Macros for minimum-width integer constants */
 
 #define INT8_C(val)  val##i8
 #define INT16_C(val) val##i16
@@ -235,13 +237,13 @@ typedef uint64_t  uintmax_t;
 #define UINT32_C(val) val##ui32
 #define UINT64_C(val) val##ui64
 
-// 7.18.4.2 Macros for greatest-width integer constants
+/* 7.18.4.2 Macros for greatest-width integer constants */
 #define INTMAX_C   INT64_C
 #define UINTMAX_C  UINT64_C
 
-#endif // __STDC_CONSTANT_MACROS ]
+#endif /* __STDC_CONSTANT_MACROS ] */
 
 
-#endif // _MSC_STDINT_H_ ]
+#endif /* _MSC_STDINT_H_ ] */
 
-#endif // _MSC_VER ]
\ No newline at end of file
+#endif /* _MSC_VER ] */
diff --git a/src/cli/CMakeLists.txt b/src/cli/CMakeLists.txt
index 97797e33b..8069d1560 100644
--- a/src/cli/CMakeLists.txt
+++ b/src/cli/CMakeLists.txt
@@ -41,6 +41,7 @@ add_executable(git2_cli ${CLI_SRC_C} ${CLI_SRC_OS} ${CLI_OBJECTS}
 target_link_libraries(git2_cli ${CLI_LIBGIT2_LIBRARY} ${LIBGIT2_SYSTEM_LIBS})
 
 set_target_properties(git2_cli PROPERTIES C_STANDARD 90)
+set_target_properties(git2_cli PROPERTIES C_EXTENSIONS OFF)
 set_target_properties(git2_cli PROPERTIES RUNTIME_OUTPUT_DIRECTORY ${libgit2_BINARY_DIR})
 set_target_properties(git2_cli PROPERTIES OUTPUT_NAME ${LIBGIT2_FILENAME})
 
diff --git a/src/libgit2/CMakeLists.txt b/src/libgit2/CMakeLists.txt
index bc7cb5b35..7c7cc4ea1 100644
--- a/src/libgit2/CMakeLists.txt
+++ b/src/libgit2/CMakeLists.txt
@@ -61,6 +61,7 @@ target_link_libraries(libgit2package ${LIBGIT2_SYSTEM_LIBS})
 target_include_directories(libgit2package SYSTEM PRIVATE ${LIBGIT2_INCLUDES})
 
 set_target_properties(libgit2package PROPERTIES C_STANDARD 90)
+set_target_properties(libgit2package PROPERTIES C_EXTENSIONS OFF)
 set_target_properties(libgit2package PROPERTIES LIBRARY_OUTPUT_DIRECTORY ${PROJECT_BINARY_DIR})
 set_target_properties(libgit2package PROPERTIES RUNTIME_OUTPUT_DIRECTORY ${PROJECT_BINARY_DIR})
 set_target_properties(libgit2package PROPERTIES ARCHIVE_OUTPUT_DIRECTORY ${PROJECT_BINARY_DIR})
diff --git a/tests/libgit2/CMakeLists.txt b/tests/libgit2/CMakeLists.txt
index af70f55a7..0202b8c79 100644
--- a/tests/libgit2/CMakeLists.txt
+++ b/tests/libgit2/CMakeLists.txt
@@ -41,8 +41,8 @@ set_source_files_properties(
 add_executable(libgit2_tests ${SRC_CLAR} ${SRC_TEST} ${LIBGIT2_OBJECTS})
 
 set_target_properties(libgit2_tests PROPERTIES C_STANDARD 90)
+set_target_properties(libgit2_tests PROPERTIES C_EXTENSIONS OFF)
 set_target_properties(libgit2_tests PROPERTIES RUNTIME_OUTPUT_DIRECTORY ${PROJECT_BINARY_DIR})
-
 target_include_directories(libgit2_tests PRIVATE ${TEST_INCLUDES} ${LIBGIT2_INCLUDES} ${LIBGIT2_DEPENDENCY_INCLUDES})
 target_include_directories(libgit2_tests SYSTEM PRIVATE ${LIBGIT2_SYSTEM_INCLUDES})
 target_link_libraries(libgit2_tests ${LIBGIT2_SYSTEM_LIBS})
diff --git a/tests/libgit2/grafts/shallow.c b/tests/libgit2/grafts/shallow.c
index 289d1b191..349e9e9b2 100644
--- a/tests/libgit2/grafts/shallow.c
+++ b/tests/libgit2/grafts/shallow.c
@@ -110,8 +110,8 @@ void test_grafts_shallow__revwalk_behavior(void)
 	cl_git_pass(git_revwalk_new(&w, g_repo));
 	cl_git_pass(git_revwalk_push_head(w));
 
-	cl_git_pass(git_revwalk_next(&oid_1, w)); // a65fedf39aefe402d3bb6e24df4d4f5fe4547750
-	cl_git_pass(git_revwalk_next(&oid_2, w)); // be3563ae3f795b2b4353bcce3a527ad0a4f7f644
+	cl_git_pass(git_revwalk_next(&oid_1, w)); /* a65fedf39aefe402d3bb6e24df4d4f5fe4547750 */
+	cl_git_pass(git_revwalk_next(&oid_2, w)); /* be3563ae3f795b2b4353bcce3a527ad0a4f7f644 */
 	cl_git_fail_with(GIT_ITEROVER, git_revwalk_next(&oid_3, w));
 
 	cl_assert_equal_s(git_oid_tostr_s(&oid_1), "a65fedf39aefe402d3bb6e24df4d4f5fe4547750");
diff --git a/tests/libgit2/index/tests256.c b/tests/libgit2/index/tests256.c
index fed8bfb93..97246ce4d 100644
--- a/tests/libgit2/index/tests256.c
+++ b/tests/libgit2/index/tests256.c
@@ -678,7 +678,7 @@ void test_index_tests256__write_tree_invalid_unowned_index(void)
 
 	cl_git_pass(git_index__new(&idx, GIT_OID_SHA256));
 
-	// TODO: this one is failing
+	/* TODO: this one is failing */
 	cl_git_pass(git_oid__fromstr(&entry.id, "a8c2e0a89a9cbab77c732b6bc39b51a783e3a318a847f46cba7614cac9814291", GIT_OID_SHA256));
 	entry.path = "foo";
 	entry.mode = GIT_FILEMODE_BLOB;
diff --git a/tests/util/CMakeLists.txt b/tests/util/CMakeLists.txt
index 232590ffd..ad57493b5 100644
--- a/tests/util/CMakeLists.txt
+++ b/tests/util/CMakeLists.txt
@@ -40,6 +40,7 @@ set_source_files_properties(
 add_executable(util_tests ${SRC_CLAR} ${SRC_TEST} ${LIBGIT2_OBJECTS})
 
 set_target_properties(util_tests PROPERTIES C_STANDARD 90)
+set_target_properties(util_tests PROPERTIES C_EXTENSIONS OFF)
 set_target_properties(util_tests PROPERTIES RUNTIME_OUTPUT_DIRECTORY ${libgit2_BINARY_DIR})
 
 target_include_directories(util_tests PRIVATE ${TEST_INCLUDES} ${LIBGIT2_INCLUDES} ${LIBGIT2_DEPENDENCY_INCLUDES})