Merge remote-tracking branch 'upstream/master' into usermarksusermarks

1 files changed, 356 insertions, 414 deletions

diff --git a/src/nvim/mbyte.c b/src/nvim/mbyte.c
index af9e214d92..8b50ba719a 100644
--- a/src/nvim/mbyte.c
+++ b/src/nvim/mbyte.c
@@ -25,36 +25,51 @@
 /// Vim scripts may contain an ":scriptencoding" command. This has an effect
 /// for some commands, like ":menutrans".
 
-#include <inttypes.h>
+#include <assert.h>
+#include <ctype.h>
+#include <errno.h>
+#include <iconv.h>
 #include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
 #include <string.h>
 #include <wchar.h>
 #include <wctype.h>
 
-#include "nvim/ascii.h"
-#include "nvim/vim.h"
-#ifdef HAVE_LOCALE_H
-# include <locale.h>
-#endif
+#include "auto/config.h"
 #include "nvim/arabic.h"
+#include "nvim/ascii.h"
+#include "nvim/buffer_defs.h"
 #include "nvim/charset.h"
 #include "nvim/cursor.h"
 #include "nvim/drawscreen.h"
-#include "nvim/eval.h"
-#include "nvim/fileio.h"
-#include "nvim/func_attr.h"
+#include "nvim/eval/typval.h"
+#include "nvim/eval/typval_defs.h"
 #include "nvim/getchar.h"
+#include "nvim/gettext.h"
+#include "nvim/globals.h"
+#include "nvim/grid_defs.h"
 #include "nvim/iconv.h"
+#include "nvim/keycodes.h"
+#include "nvim/macros.h"
 #include "nvim/mark.h"
 #include "nvim/mbyte.h"
+#include "nvim/mbyte_defs.h"
 #include "nvim/memline.h"
 #include "nvim/memory.h"
 #include "nvim/message.h"
+#include "nvim/option_defs.h"
 #include "nvim/os/os.h"
-#include "nvim/path.h"
+#include "nvim/os/os_defs.h"
+#include "nvim/pos.h"
 #include "nvim/screen.h"
-#include "nvim/spell.h"
 #include "nvim/strings.h"
+#include "nvim/types.h"
+#include "nvim/vim.h"
+
+#ifdef HAVE_LOCALE_H
+# include <locale.h>
+#endif
 
 typedef struct {
   int rangeStart;
@@ -68,11 +83,12 @@ struct interval {
   long last;
 };
 
+// uncrustify:off
 #ifdef INCLUDE_GENERATED_DECLARATIONS
 # include "mbyte.c.generated.h"
-
 # include "unicode_tables.generated.h"
 #endif
+// uncrustify:on
 
 static char e_list_item_nr_is_not_list[]
   = N_("E1109: List item %d is not a List");
@@ -84,8 +100,8 @@ static char e_list_item_nr_cell_width_invalid[]
   = N_("E1112: List item %d cell width invalid");
 static char e_overlapping_ranges_for_nr[]
   = N_("E1113: Overlapping ranges for 0x%lx");
-static char e_only_values_of_0x100_and_higher_supported[]
-  = N_("E1114: Only values of 0x100 and higher supported");
+static char e_only_values_of_0x80_and_higher_supported[]
+  = N_("E1114: Only values of 0x80 and higher supported");
 
 // To speed up BYTELEN(); keep a lookup table to quickly get the length in
 // bytes of a UTF-8 character from the first byte of a UTF-8 string.  Bytes
@@ -132,14 +148,11 @@ const uint8_t utf8len_tab_zero[] = {
   4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 0, 0,  // F?
 };
 
-/*
- * Canonical encoding names and their properties.
- * "iso-8859-n" is handled by enc_canonize() directly.
- */
+// Canonical encoding names and their properties.
+// "iso-8859-n" is handled by enc_canonize() directly.
 static struct
 {   const char *name;   int prop;              int codepage; }
-enc_canon_table[] =
-{
+enc_canon_table[] = {
 #define IDX_LATIN_1     0
   { "latin1",          ENC_8BIT + ENC_LATIN1,  1252 },
 #define IDX_ISO_2       1
@@ -269,13 +282,10 @@ enc_canon_table[] =
 #define IDX_COUNT       59
 };
 
-/*
- * Aliases for encoding names.
- */
+// Aliases for encoding names.
 static struct
 {   const char *name; int canon; }
-enc_alias_table[] =
-{
+enc_alias_table[] = {
   { "ansi",            IDX_LATIN_1 },
   { "iso-8859-1",      IDX_LATIN_1 },
   { "latin2",          IDX_ISO_2 },
@@ -342,46 +352,40 @@ enc_alias_table[] =
   { NULL,              0 }
 };
 
-/*
- * Find encoding "name" in the list of canonical encoding names.
- * Returns -1 if not found.
- */
-static int enc_canon_search(const char_u *name)
+/// Find encoding "name" in the list of canonical encoding names.
+/// Returns -1 if not found.
+static int enc_canon_search(const char *name)
   FUNC_ATTR_PURE
 {
   for (int i = 0; i < IDX_COUNT; i++) {
-    if (STRCMP(name, enc_canon_table[i].name) == 0) {
+    if (strcmp(name, enc_canon_table[i].name) == 0) {
       return i;
     }
   }
   return -1;
 }
 
-/*
- * Find canonical encoding "name" in the list and return its properties.
- * Returns 0 if not found.
- */
-int enc_canon_props(const char_u *name)
+// Find canonical encoding "name" in the list and return its properties.
+// Returns 0 if not found.
+int enc_canon_props(const char *name)
   FUNC_ATTR_PURE
 {
-  int i = enc_canon_search(name);
+  int i = enc_canon_search((char *)name);
   if (i >= 0) {
     return enc_canon_table[i].prop;
-  } else if (STRNCMP(name, "2byte-", 6) == 0) {
+  } else if (strncmp(name, "2byte-", 6) == 0) {
     return ENC_DBCS;
-  } else if (STRNCMP(name, "8bit-", 5) == 0 || STRNCMP(name, "iso-8859-", 9) == 0) {
+  } else if (strncmp(name, "8bit-", 5) == 0 || strncmp(name, "iso-8859-", 9) == 0) {
     return ENC_8BIT;
   }
   return 0;
 }
 
-/*
- * Return the size of the BOM for the current buffer:
- * 0 - no BOM
- * 2 - UCS-2 or UTF-16 BOM
- * 4 - UCS-4 BOM
- * 3 - UTF-8 BOM
- */
+// Return the size of the BOM for the current buffer:
+// 0 - no BOM
+// 2 - UCS-2 or UTF-16 BOM
+// 4 - UCS-4 BOM
+// 3 - UTF-8 BOM
 int bomb_size(void)
   FUNC_ATTR_PURE
 {
@@ -389,24 +393,22 @@ int bomb_size(void)
 
   if (curbuf->b_p_bomb && !curbuf->b_p_bin) {
     if (*curbuf->b_p_fenc == NUL
-        || STRCMP(curbuf->b_p_fenc, "utf-8") == 0) {
+        || strcmp(curbuf->b_p_fenc, "utf-8") == 0) {
       n = 3;
-    } else if (STRNCMP(curbuf->b_p_fenc, "ucs-2", 5) == 0
-               || STRNCMP(curbuf->b_p_fenc, "utf-16", 6) == 0) {
+    } else if (strncmp(curbuf->b_p_fenc, "ucs-2", 5) == 0
+               || strncmp(curbuf->b_p_fenc, "utf-16", 6) == 0) {
       n = 2;
-    } else if (STRNCMP(curbuf->b_p_fenc, "ucs-4", 5) == 0) {
+    } else if (strncmp(curbuf->b_p_fenc, "ucs-4", 5) == 0) {
       n = 4;
     }
   }
   return n;
 }
 
-/*
- * Remove all BOM from "s" by moving remaining text.
- */
-void remove_bom(char_u *s)
+// Remove all BOM from "s" by moving remaining text.
+void remove_bom(char *s)
 {
-  char *p = (char *)s;
+  char *p = s;
 
   while ((p = strchr(p, 0xef)) != NULL) {
     if ((uint8_t)p[1] == 0xbb && (uint8_t)p[2] == 0xbf) {
@@ -417,37 +419,33 @@ void remove_bom(char_u *s)
   }
 }
 
-/*
- * Get class of pointer:
- * 0 for blank or NUL
- * 1 for punctuation
- * 2 for an (ASCII) word character
- * >2 for other word characters
- */
-int mb_get_class(const char_u *p)
+// Get class of pointer:
+// 0 for blank or NUL
+// 1 for punctuation
+// 2 for an (ASCII) word character
+// >2 for other word characters
+int mb_get_class(const char *p)
   FUNC_ATTR_PURE
 {
   return mb_get_class_tab(p, curbuf->b_chartab);
 }
 
-int mb_get_class_tab(const char_u *p, const uint64_t *const chartab)
+int mb_get_class_tab(const char *p, const uint64_t *const chartab)
   FUNC_ATTR_PURE
 {
-  if (MB_BYTE2LEN(p[0]) == 1) {
+  if (MB_BYTE2LEN((uint8_t)p[0]) == 1) {
     if (p[0] == NUL || ascii_iswhite(p[0])) {
       return 0;
     }
-    if (vim_iswordc_tab(p[0], chartab)) {
+    if (vim_iswordc_tab((uint8_t)p[0], chartab)) {
       return 2;
     }
     return 1;
   }
-  return utf_class_tab(utf_ptr2char((char *)p), chartab);
+  return utf_class_tab(utf_ptr2char(p), chartab);
 }
 
-/*
- * Return true if "c" is in "table".
- */
+// Return true if "c" is in "table".
 static bool intable(const struct interval *table, size_t n_items, int c)
   FUNC_ATTR_PURE
 {
@@ -484,12 +482,16 @@ static bool intable(const struct interval *table, size_t n_items, int c)
 ///       gen_unicode_tables.lua, which must be manually invoked as needed.
 int utf_char2cells(int c)
 {
-  if (c >= 0x100) {
+  // Use the value from setcellwidths() at 0x80 and higher, unless the
+  // character is not printable.
+  if (c >= 0x80 && vim_isprintc(c)) {
     int n = cw_value(c);
     if (n != 0) {
       return n;
     }
+  }
 
+  if (c >= 0x100) {
     if (!utf_printable(c)) {
       return 6;                 // unprintable, displays <xxxx>
     }
@@ -536,13 +538,13 @@ int utf_ptr2cells(const char *p)
 
 /// Like utf_ptr2cells(), but limit string length to "size".
 /// For an empty string or truncated character returns 1.
-int utf_ptr2cells_len(const char_u *p, int size)
+int utf_ptr2cells_len(const char *p, int size)
 {
   int c;
 
   // Need to convert to a wide character.
-  if (size > 0 && *p >= 0x80) {
-    if (utf_ptr2len_len(p, size) < utf8len_tab[*p]) {
+  if (size > 0 && (uint8_t)(*p) >= 0x80) {
+    if (utf_ptr2len_len(p, size) < utf8len_tab[(uint8_t)(*p)]) {
       return 1;        // truncated
     }
     c = utf_ptr2char((char *)p);
@@ -568,8 +570,8 @@ size_t mb_string2cells(const char *str)
 {
   size_t clen = 0;
 
-  for (const char_u *p = (char_u *)str; *p != NUL; p += utfc_ptr2len((char *)p)) {
-    clen += (size_t)utf_ptr2cells((char *)p);
+  for (const char *p = str; *p != NUL; p += utfc_ptr2len(p)) {
+    clen += (size_t)utf_ptr2cells(p);
   }
 
   return clen;
@@ -586,9 +588,9 @@ size_t mb_string2cells_len(const char *str, size_t size)
 {
   size_t clen = 0;
 
-  for (const char_u *p = (char_u *)str; *p != NUL && p < (char_u *)str + size;
-       p += utfc_ptr2len_len(p, (int)size + (int)(p - (char_u *)str))) {
-    clen += (size_t)utf_ptr2cells((char *)p);
+  for (const char *p = str; *p != NUL && p < str + size;
+       p += utfc_ptr2len_len(p, (int)size + (int)(p - str))) {
+    clen += (size_t)utf_ptr2cells(p);
   }
 
   return clen;
@@ -601,7 +603,7 @@ size_t mb_string2cells_len(const char *str, size_t size)
 /// For an overlong sequence this may return zero.
 /// Does not include composing characters for obvious reasons.
 ///
-/// @param[in]  p  String to convert.
+/// @param[in]  p_in  String to convert.
 ///
 /// @return Unicode codepoint or byte value.
 int utf_ptr2char(const char *const p_in)
@@ -646,22 +648,20 @@ int utf_ptr2char(const char *const p_in)
   return p[0];
 }
 
-/*
- * Convert a UTF-8 byte sequence to a wide character.
- * String is assumed to be terminated by NUL or after "n" bytes, whichever
- * comes first.
- * The function is safe in the sense that it never accesses memory beyond the
- * first "n" bytes of "s".
- *
- * On success, returns decoded codepoint, advances "s" to the beginning of
- * next character and decreases "n" accordingly.
- *
- * If end of string was reached, returns 0 and, if "n" > 0, advances "s" past
- * NUL byte.
- *
- * If byte sequence is illegal or incomplete, returns -1 and does not advance
- * "s".
- */
+// Convert a UTF-8 byte sequence to a wide character.
+// String is assumed to be terminated by NUL or after "n" bytes, whichever
+// comes first.
+// The function is safe in the sense that it never accesses memory beyond the
+// first "n" bytes of "s".
+//
+// On success, returns decoded codepoint, advances "s" to the beginning of
+// next character and decreases "n" accordingly.
+//
+// If end of string was reached, returns 0 and, if "n" > 0, advances "s" past
+// NUL byte.
+//
+// If byte sequence is illegal or incomplete, returns -1 and does not advance
+// "s".
 static int utf_safe_read_char_adv(const char_u **s, size_t *n)
 {
   int c;
@@ -701,38 +701,32 @@ static int utf_safe_read_char_adv(const char_u **s, size_t *n)
   return -1;
 }
 
-/*
- * Get character at **pp and advance *pp to the next character.
- * Note: composing characters are skipped!
- */
-int mb_ptr2char_adv(const char_u **const pp)
+// Get character at **pp and advance *pp to the next character.
+// Note: composing characters are skipped!
+int mb_ptr2char_adv(const char **const pp)
 {
   int c;
 
-  c = utf_ptr2char((char *)(*pp));
-  *pp += utfc_ptr2len((char *)(*pp));
+  c = utf_ptr2char(*pp);
+  *pp += utfc_ptr2len(*pp);
   return c;
 }
 
-/*
- * Get character at **pp and advance *pp to the next character.
- * Note: composing characters are returned as separate characters.
- */
-int mb_cptr2char_adv(const char_u **pp)
+// Get character at **pp and advance *pp to the next character.
+// Note: composing characters are returned as separate characters.
+int mb_cptr2char_adv(const char **pp)
 {
   int c;
 
-  c = utf_ptr2char((char *)(*pp));
-  *pp += utf_ptr2len((char *)(*pp));
+  c = utf_ptr2char(*pp);
+  *pp += utf_ptr2len(*pp);
   return c;
 }
 
-/*
- * Check if the character pointed to by "p2" is a composing character when it
- * comes after "p1".  For Arabic sometimes "ab" is replaced with "c", which
- * behaves like a composing character.
- */
-bool utf_composinglike(const char_u *p1, const char_u *p2)
+/// Check if the character pointed to by "p2" is a composing character when it
+/// comes after "p1".  For Arabic sometimes "ab" is replaced with "c", which
+/// behaves like a composing character.
+bool utf_composinglike(const char *p1, const char *p2)
 {
   int c2;
 
@@ -754,28 +748,25 @@ bool utf_composinglike(const char_u *p1, const char_u *p2)
 ///                   space at least for #MAX_MCO + 1 elements.
 ///
 /// @return leading character.
-int utfc_ptr2char(const char_u *p, int *pcc)
+int utfc_ptr2char(const char *p, int *pcc)
 {
-  int len;
-  int c;
-  int cc;
   int i = 0;
 
-  c = utf_ptr2char((char *)p);
-  len = utf_ptr2len((char *)p);
+  int c = utf_ptr2char(p);
+  int len = utf_ptr2len(p);
 
   // Only accept a composing char when the first char isn't illegal.
-  if ((len > 1 || *p < 0x80)
-      && p[len] >= 0x80
+  if ((len > 1 || (uint8_t)(*p) < 0x80)
+      && (uint8_t)p[len] >= 0x80
       && utf_composinglike(p, p + len)) {
-    cc = utf_ptr2char((char *)p + len);
+    int cc = utf_ptr2char(p + len);
     for (;;) {
       pcc[i++] = cc;
       if (i == MAX_MCO) {
         break;
       }
-      len += utf_ptr2len((char *)p + len);
-      if (p[len] < 0x80 || !utf_iscomposing(cc = utf_ptr2char((char *)p + len))) {
+      len += utf_ptr2len(p + len);
+      if ((uint8_t)p[len] < 0x80 || !utf_iscomposing(cc = utf_ptr2char(p + len))) {
         break;
       }
     }
@@ -788,13 +779,11 @@ int utfc_ptr2char(const char_u *p, int *pcc)
   return c;
 }
 
-/*
- * Convert a UTF-8 byte string to a wide character.  Also get up to MAX_MCO
- * composing characters.  Use no more than p[maxlen].
- *
- * @param [out] pcc: composing chars, last one is 0
- */
-int utfc_ptr2char_len(const char_u *p, int *pcc, int maxlen)
+// Convert a UTF-8 byte string to a wide character.  Also get up to MAX_MCO
+// composing characters.  Use no more than p[maxlen].
+//
+// @param [out] pcc: composing chars, last one is 0
+int utfc_ptr2char_len(const char *p, int *pcc, int maxlen)
 {
   assert(maxlen > 0);
 
@@ -803,14 +792,14 @@ int utfc_ptr2char_len(const char_u *p, int *pcc, int maxlen)
   int len = utf_ptr2len_len(p, maxlen);
   // Is it safe to use utf_ptr2char()?
   bool safe = len > 1 && len <= maxlen;
-  int c = safe ? utf_ptr2char((char *)p) : *p;
+  int c = safe ? utf_ptr2char(p) : (uint8_t)(*p);
 
   // Only accept a composing char when the first char isn't illegal.
-  if ((safe || c < 0x80) && len < maxlen && p[len] >= 0x80) {
+  if ((safe || c < 0x80) && len < maxlen && (uint8_t)p[len] >= 0x80) {
     for (; i < MAX_MCO; i++) {
       int len_cc = utf_ptr2len_len(p + len, maxlen - len);
       safe = len_cc > 1 && len_cc <= maxlen - len;
-      if (!safe || (pcc[i] = utf_ptr2char((char *)p + len)) < 0x80
+      if (!safe || (pcc[i] = utf_ptr2char(p + len)) < 0x80
           || !(i == 0 ? utf_composinglike(p, p + len) : utf_iscomposing(pcc[i]))) {
         break;
       }
@@ -849,31 +838,27 @@ int utf_ptr2len(const char *const p_in)
   return len;
 }
 
-/*
- * Return length of UTF-8 character, obtained from the first byte.
- * "b" must be between 0 and 255!
- * Returns 1 for an invalid first byte value.
- */
+// Return length of UTF-8 character, obtained from the first byte.
+// "b" must be between 0 and 255!
+// Returns 1 for an invalid first byte value.
 int utf_byte2len(int b)
 {
   return utf8len_tab[b];
 }
 
-/*
- * Get the length of UTF-8 byte sequence "p[size]".  Does not include any
- * following composing characters.
- * Returns 1 for "".
- * Returns 1 for an illegal byte sequence (also in incomplete byte seq.).
- * Returns number > "size" for an incomplete byte sequence.
- * Never returns zero.
- */
-int utf_ptr2len_len(const char_u *p, int size)
+// Get the length of UTF-8 byte sequence "p[size]".  Does not include any
+// following composing characters.
+// Returns 1 for "".
+// Returns 1 for an illegal byte sequence (also in incomplete byte seq.).
+// Returns number > "size" for an incomplete byte sequence.
+// Never returns zero.
+int utf_ptr2len_len(const char *p, int size)
 {
   int len;
   int i;
   int m;
 
-  len = utf8len_tab[*p];
+  len = utf8len_tab[(uint8_t)(*p)];
   if (len == 1) {
     return 1;           // NUL, ascii or illegal lead byte
   }
@@ -882,7 +867,7 @@ int utf_ptr2len_len(const char_u *p, int size)
   } else {
     m = len;
   }
-  for (i = 1; i < m; ++i) {
+  for (i = 1; i < m; i++) {
     if ((p[i] & 0xc0) != 0x80) {
       return 1;
     }
@@ -893,21 +878,20 @@ int utf_ptr2len_len(const char_u *p, int size)
 /// Return the number of bytes occupied by a UTF-8 character in a string.
 /// This includes following composing characters.
 /// Returns zero for NUL.
-int utfc_ptr2len(const char *const p_in)
+int utfc_ptr2len(const char *const p)
   FUNC_ATTR_PURE FUNC_ATTR_WARN_UNUSED_RESULT FUNC_ATTR_NONNULL_ALL
 {
-  uint8_t *p = (uint8_t *)p_in;
-  uint8_t b0 = *p;
+  uint8_t b0 = (uint8_t)(*p);
 
   if (b0 == NUL) {
     return 0;
   }
-  if (b0 < 0x80 && p[1] < 0x80) {  // be quick for ASCII
+  if (b0 < 0x80 && (uint8_t)p[1] < 0x80) {  // be quick for ASCII
     return 1;
   }
 
   // Skip over first UTF-8 char, stopping at a NUL byte.
-  int len = utf_ptr2len((char *)p);
+  int len = utf_ptr2len(p);
 
   // Check for illegal byte.
   if (len == 1 && b0 >= 0x80) {
@@ -918,23 +902,21 @@ int utfc_ptr2len(const char *const p_in)
   // skip all of them (otherwise the cursor would get stuck).
   int prevlen = 0;
   for (;;) {
-    if (p[len] < 0x80 || !utf_composinglike(p + prevlen, p + len)) {
+    if ((uint8_t)p[len] < 0x80 || !utf_composinglike(p + prevlen, p + len)) {
       return len;
     }
 
     // Skip over composing char.
     prevlen = len;
-    len += utf_ptr2len((char *)p + len);
+    len += utf_ptr2len(p + len);
   }
 }
 
-/*
- * Return the number of bytes the UTF-8 encoding of the character at "p[size]"
- * takes.  This includes following composing characters.
- * Returns 0 for an empty string.
- * Returns 1 for an illegal char or an incomplete byte sequence.
- */
-int utfc_ptr2len_len(const char_u *p, int size)
+/// Return the number of bytes the UTF-8 encoding of the character at "p[size]"
+/// takes.  This includes following composing characters.
+/// Returns 0 for an empty string.
+/// Returns 1 for an illegal char or an incomplete byte sequence.
+int utfc_ptr2len_len(const char *p, int size)
 {
   int len;
   int prevlen;
@@ -942,7 +924,7 @@ int utfc_ptr2len_len(const char_u *p, int size)
   if (size < 1 || *p == NUL) {
     return 0;
   }
-  if (p[0] < 0x80 && (size == 1 || p[1] < 0x80)) {  // be quick for ASCII
+  if ((uint8_t)p[0] < 0x80 && (size == 1 || (uint8_t)p[1] < 0x80)) {  // be quick for ASCII
     return 1;
   }
 
@@ -950,26 +932,22 @@ int utfc_ptr2len_len(const char_u *p, int size)
   len = utf_ptr2len_len(p, size);
 
   // Check for illegal byte and incomplete byte sequence.
-  if ((len == 1 && p[0] >= 0x80) || len > size) {
+  if ((len == 1 && (uint8_t)p[0] >= 0x80) || len > size) {
     return 1;
   }
 
-  /*
-   * Check for composing characters.  We can handle only the first six, but
-   * skip all of them (otherwise the cursor would get stuck).
-   */
+  // Check for composing characters.  We can handle only the first six, but
+  // skip all of them (otherwise the cursor would get stuck).
   prevlen = 0;
   while (len < size) {
     int len_next_char;
 
-    if (p[len] < 0x80) {
+    if ((uint8_t)p[len] < 0x80) {
       break;
     }
 
-    /*
-     * Next character length should not go beyond size to ensure that
-     * utf_composinglike(...) does not read beyond size.
-     */
+    // Next character length should not go beyond size to ensure that
+    // utf_composinglike(...) does not read beyond size.
     len_next_char = utf_ptr2len_len(p + len, size - len);
     if (len_next_char > size - len) {
       break;
@@ -1048,26 +1026,21 @@ int utf_char2bytes(const int c, char *const buf)
   }
 }
 
-/*
- * Return true if "c" is a composing UTF-8 character.  This means it will be
- * drawn on top of the preceding character.
- * Based on code from Markus Kuhn.
- */
+// Return true if "c" is a composing UTF-8 character.  This means it will be
+// drawn on top of the preceding character.
+// Based on code from Markus Kuhn.
 bool utf_iscomposing(int c)
 {
   return intable(combining, ARRAY_SIZE(combining), c);
 }
 
-/*
- * Return true for characters that can be displayed in a normal way.
- * Only for characters of 0x100 and above!
- */
+// Return true for characters that can be displayed in a normal way.
+// Only for characters of 0x100 and above!
 bool utf_printable(int c)
 {
   // Sorted list of non-overlapping intervals.
   // 0xd800-0xdfff is reserved for UTF-16, actually illegal.
-  static struct interval nonprint[] =
-  {
+  static struct interval nonprint[] = {
     { 0x070f, 0x070f }, { 0x180b, 0x180e }, { 0x200b, 0x200f }, { 0x202a, 0x202e },
     { 0x2060, 0x206f }, { 0xd800, 0xdfff }, { 0xfeff, 0xfeff }, { 0xfff9, 0xfffb },
     { 0xfffe, 0xffff }
@@ -1076,12 +1049,10 @@ bool utf_printable(int c)
   return !intable(nonprint, ARRAY_SIZE(nonprint), c);
 }
 
-/*
- * Get class of a Unicode character.
- * 0: white space
- * 1: punctuation
- * 2 or bigger: some class of word character.
- */
+// Get class of a Unicode character.
+// 0: white space
+// 1: punctuation
+// 2 or bigger: some class of word character.
 int utf_class(const int c)
 {
   return utf_class_tab(c, curbuf->b_chartab);
@@ -1210,11 +1181,9 @@ bool utf_ambiguous_width(int c)
                        || intable(emoji_all, ARRAY_SIZE(emoji_all), c));
 }
 
-/*
- * Generic conversion function for case operations.
- * Return the converted equivalent of "a", which is a UCS-4 character.  Use
- * the given conversion "table".  Uses binary search on "table".
- */
+// Generic conversion function for case operations.
+// Return the converted equivalent of "a", which is a UCS-4 character.  Use
+// the given conversion "table".  Uses binary search on "table".
 static int utf_convert(int a, const convertStruct *const table, size_t n_items)
 {
   size_t start, mid, end;   // indices into table
@@ -1235,15 +1204,12 @@ static int utf_convert(int a, const convertStruct *const table, size_t n_items)
       && a <= table[start].rangeEnd
       && (a - table[start].rangeStart) % table[start].step == 0) {
     return a + table[start].offset;
-  } else {
-    return a;
   }
+  return a;
 }
 
-/*
- * Return the folded-case equivalent of "a", which is a UCS-4 character.  Uses
- * simple case folding.
- */
+// Return the folded-case equivalent of "a", which is a UCS-4 character.  Uses
+// simple case folding.
 int utf_fold(int a)
 {
   if (a < 0x80) {
@@ -1267,12 +1233,9 @@ int mb_toupper(int a)
     return TOUPPER_ASC(a);
   }
 
-#if defined(__STDC_ISO_10646__)
-  // If towupper() is available and handles Unicode, use it.
   if (!(cmp_flags & CMP_INTERNAL)) {
     return (int)towupper((wint_t)a);
   }
-#endif
 
   // For characters below 128 use locale sensitive toupper().
   if (a < 128) {
@@ -1298,12 +1261,9 @@ int mb_tolower(int a)
     return TOLOWER_ASC(a);
   }
 
-#if defined(__STDC_ISO_10646__)
-  // If towlower() is available and handles Unicode, use it.
   if (!(cmp_flags & CMP_INTERNAL)) {
     return (int)towlower((wint_t)a);
   }
-#endif
 
   // For characters below 128 use locale sensitive tolower().
   if (a < 128) {
@@ -1398,7 +1358,7 @@ static int utf_strnicmp(const char_u *s1, const char_u *s2, size_t n1, size_t n2
   return n1 == 0 ? -1 : 1;
 }
 
-#ifdef WIN32
+#ifdef MSWIN
 # ifndef CP_UTF8
 #  define CP_UTF8 65001  // magic number from winnls.h
 # endif
@@ -1500,16 +1460,16 @@ int utf16_to_utf8(const wchar_t *utf16, int utf16len, char **utf8)
 /// @param len maximum length (an earlier NUL terminates)
 /// @param[out] codepoints incremented with UTF-32 code point size
 /// @param[out] codeunits incremented with UTF-16 code unit size
-void mb_utflen(const char_u *s, size_t len, size_t *codepoints, size_t *codeunits)
+void mb_utflen(const char *s, size_t len, size_t *codepoints, size_t *codeunits)
   FUNC_ATTR_NONNULL_ALL
 {
   size_t count = 0, extra = 0;
   size_t clen;
-  for (size_t i = 0; i < len && s[i] != NUL; i += clen) {
+  for (size_t i = 0; i < len; i += clen) {
     clen = (size_t)utf_ptr2len_len(s + i, (int)(len - i));
     // NB: gets the byte value of invalid sequence bytes.
     // we only care whether the char fits in the BMP or not
-    int c = (clen > 1) ? utf_ptr2char((char *)s + i) : s[i];
+    int c = (clen > 1) ? utf_ptr2char(s + i) : (uint8_t)s[i];
     count++;
     if (c > 0xFFFF) {
       extra++;
@@ -1519,7 +1479,7 @@ void mb_utflen(const char_u *s, size_t len, size_t *codepoints, size_t *codeunit
   *codeunits += count + extra;
 }
 
-ssize_t mb_utf_index_to_bytes(const char_u *s, size_t len, size_t index, bool use_utf16_units)
+ssize_t mb_utf_index_to_bytes(const char *s, size_t len, size_t index, bool use_utf16_units)
   FUNC_ATTR_NONNULL_ALL
 {
   size_t count = 0;
@@ -1527,11 +1487,11 @@ ssize_t mb_utf_index_to_bytes(const char_u *s, size_t len, size_t index, bool us
   if (index == 0) {
     return 0;
   }
-  for (i = 0; i < len && s[i] != NUL; i += clen) {
+  for (i = 0; i < len; i += clen) {
     clen = (size_t)utf_ptr2len_len(s + i, (int)(len - i));
     // NB: gets the byte value of invalid sequence bytes.
     // we only care whether the char fits in the BMP or not
-    int c = (clen > 1) ? utf_ptr2char((char *)s + i) : s[i];
+    int c = (clen > 1) ? utf_ptr2char(s + i) : (uint8_t)s[i];
     count++;
     if (use_utf16_units && c > 0xFFFF) {
       count++;
@@ -1543,17 +1503,16 @@ ssize_t mb_utf_index_to_bytes(const char_u *s, size_t len, size_t index, bool us
   return -1;
 }
 
-/*
- * Version of strnicmp() that handles multi-byte characters.
- * Needed for Big5, Shift-JIS and UTF-8 encoding.  Other DBCS encodings can
- * probably use strnicmp(), because there are no ASCII characters in the
- * second byte.
- * Returns zero if s1 and s2 are equal (ignoring case), the difference between
- * two characters otherwise.
- */
-int mb_strnicmp(const char_u *s1, const char_u *s2, const size_t nn)
+/// Version of strnicmp() that handles multi-byte characters.
+/// Needed for Big5, Shift-JIS and UTF-8 encoding.  Other DBCS encodings can
+/// probably use strnicmp(), because there are no ASCII characters in the
+/// second byte.
+///
+/// @return  zero if s1 and s2 are equal (ignoring case), the difference between
+///          two characters otherwise.
+int mb_strnicmp(const char *s1, const char *s2, const size_t nn)
 {
-  return utf_strnicmp(s1, s2, nn, nn);
+  return utf_strnicmp((char_u *)s1, (char_u *)s2, nn, nn);
 }
 
 /// Compare strings case-insensitively
@@ -1570,72 +1529,73 @@ int mb_strnicmp(const char_u *s1, const char_u *s2, const size_t nn)
 /// @return 0 if strings are equal, <0 if s1 < s2, >0 if s1 > s2.
 int mb_stricmp(const char *s1, const char *s2)
 {
-  return mb_strnicmp((const char_u *)s1, (const char_u *)s2, MAXCOL);
+  return mb_strnicmp(s1, s2, MAXCOL);
 }
 
-/*
- * "g8": show bytes of the UTF-8 char under the cursor.  Doesn't matter what
- * 'encoding' has been set to.
- */
+// "g8": show bytes of the UTF-8 char under the cursor.  Doesn't matter what
+// 'encoding' has been set to.
 void show_utf8(void)
 {
   int len;
   int rlen = 0;
-  char_u *line;
+  char *line;
   int clen;
   int i;
 
   // Get the byte length of the char under the cursor, including composing
   // characters.
   line = get_cursor_pos_ptr();
-  len = utfc_ptr2len((char *)line);
+  len = utfc_ptr2len(line);
   if (len == 0) {
     msg("NUL");
     return;
   }
 
   clen = 0;
-  for (i = 0; i < len; ++i) {
+  for (i = 0; i < len; i++) {
     if (clen == 0) {
       // start of (composing) character, get its length
       if (i > 0) {
         STRCPY(IObuff + rlen, "+ ");
         rlen += 2;
       }
-      clen = utf_ptr2len((char *)line + i);
+      clen = utf_ptr2len(line + i);
     }
-    sprintf((char *)IObuff + rlen, "%02x ",
-            (line[i] == NL) ? NUL : line[i]);          // NUL is stored as NL
+    sprintf(IObuff + rlen, "%02x ",  // NOLINT(runtime/printf)
+            (line[i] == NL) ? NUL : (uint8_t)line[i]);          // NUL is stored as NL
     clen--;
-    rlen += (int)STRLEN(IObuff + rlen);
+    rlen += (int)strlen(IObuff + rlen);
     if (rlen > IOSIZE - 20) {
       break;
     }
   }
 
-  msg((char *)IObuff);
+  msg(IObuff);
 }
 
 /// Return offset from "p" to the start of a character, including composing characters.
 /// "base" must be the start of the string, which must be NUL terminated.
 /// If "p" points to the NUL at the end of the string return 0.
 /// Returns 0 when already at the first byte of a character.
-int utf_head_off(const char_u *base, const char_u *p)
+int utf_head_off(const char *base_in, const char *p_in)
 {
   int c;
   int len;
 
-  if (*p < 0x80) {              // be quick for ASCII
+  if ((uint8_t)(*p_in) < 0x80) {              // be quick for ASCII
     return 0;
   }
 
+  const uint8_t *base = (uint8_t *)base_in;
+  const uint8_t *p = (uint8_t *)p_in;
+
   // Skip backwards over trailing bytes: 10xx.xxxx
   // Skip backwards again if on a composing char.
-  const char_u *q;
-  for (q = p;; --q) {
+  const uint8_t *q;
+  for (q = p;; q--) {
     // Move s to the last byte of this char.
-    const char_u *s;
-    for (s = q; (s[1] & 0xc0) == 0x80; ++s) {}
+    const uint8_t *s;
+    for (s = q; (s[1] & 0xc0) == 0x80; s++) {}
 
     // Move q to the first byte of this char.
     while (q > base && (*q & 0xc0) == 0x80) {
@@ -1659,7 +1619,7 @@ int utf_head_off(const char_u *base, const char_u *p)
 
     if (arabic_maycombine(c)) {
       // Advance to get a sneak-peak at the next char
-      const char_u *j = q;
+      const uint8_t *j = q;
       j--;
       // Move j to the first byte of this char.
       while (j > base && (*j & 0xc0) == 0x80) {
@@ -1828,11 +1788,12 @@ void mb_copy_char(const char **const fp, char **const tp)
   *fp += l;
 }
 
-/// Return the offset from "p" to the first byte of a character.  When "p" is
+/// Return the offset from "p_in" to the first byte of a character.  When "p_in" is
 /// at the start of a character 0 is returned, otherwise the offset to the next
 /// character.  Can start anywhere in a stream of bytes.
-int mb_off_next(const char_u *base, const char_u *p)
+int mb_off_next(const char *base, const char *p_in)
 {
+  const uint8_t *p = (uint8_t *)p_in;
   int i;
   int j;
 
@@ -1844,7 +1805,7 @@ int mb_off_next(const char_u *base, const char_u *p)
   for (i = 0; (p[i] & 0xc0) == 0x80; i++) {}
   if (i > 0) {
     // Check for illegal sequence.
-    for (j = 0; p - j > base; j++) {
+    for (j = 0; p - j > (uint8_t *)base; j++) {
       if ((p[-j] & 0xc0) != 0x80) {
         break;
       }
@@ -1920,16 +1881,14 @@ int utf_cp_head_off(const char_u *base, const char_u *p)
   return i;
 }
 
-/*
- * Find the next illegal byte sequence.
- */
+// Find the next illegal byte sequence.
 void utf_find_illegal(void)
 {
   pos_T pos = curwin->w_cursor;
-  char_u *p;
+  char *p;
   int len;
   vimconv_T vimconv;
-  char_u *tofree = NULL;
+  char *tofree = NULL;
 
   vimconv.vc_type = CONV_NONE;
   if (enc_canon_props(curbuf->b_p_fenc) & ENC_8BIT) {
@@ -1954,9 +1913,8 @@ void utf_find_illegal(void)
     while (*p != NUL) {
       // Illegal means that there are not enough trail bytes (checked by
       // utf_ptr2len()) or too many of them (overlong sequence).
-      len = utf_ptr2len((char *)p);
-      if (*p >= 0x80 && (len == 1
-                         || utf_char2len(utf_ptr2char((char *)p)) != len)) {
+      len = utf_ptr2len(p);
+      if ((uint8_t)(*p) >= 0x80 && (len == 1 || utf_char2len(utf_ptr2char(p)) != len)) {
         if (vimconv.vc_type == CONV_NONE) {
           curwin->w_cursor.col += (colnr_T)(p - get_cursor_pos_ptr());
         } else {
@@ -1964,7 +1922,7 @@ void utf_find_illegal(void)
 
           len = (int)(p - tofree);
           for (p = get_cursor_pos_ptr(); *p != NUL && len-- > 0; p += l) {
-            l = utf_ptr2len((char *)p);
+            l = utf_ptr2len(p);
             curwin->w_cursor.col += l;
           }
         }
@@ -1975,7 +1933,7 @@ void utf_find_illegal(void)
     if (curwin->w_cursor.lnum == curbuf->b_ml.ml_line_count) {
       break;
     }
-    ++curwin->w_cursor.lnum;
+    curwin->w_cursor.lnum++;
     curwin->w_cursor.col = 0;
   }
 
@@ -1989,8 +1947,7 @@ theend:
 }
 
 /// @return  true if string "s" is a valid utf-8 string.
-/// When "end" is NULL stop at the first NUL.
-/// When "end" is positive stop there.
+/// When "end" is NULL stop at the first NUL.  Otherwise stop at "end".
 bool utf_valid_string(const char_u *s, const char_u *end)
 {
   const char_u *p = s;
@@ -2013,10 +1970,8 @@ bool utf_valid_string(const char_u *s, const char_u *end)
   return true;
 }
 
-/*
- * If the cursor moves on an trail byte, set the cursor on the lead byte.
- * Thus it moves left if necessary.
- */
+// If the cursor moves on an trail byte, set the cursor on the lead byte.
+// Thus it moves left if necessary.
 void mb_adjust_cursor(void)
 {
   mark_mb_adjustpos(curbuf, &curwin->w_cursor);
@@ -2033,8 +1988,8 @@ void mb_check_adjust_col(void *win_)
 
   // Column 0 is always valid.
   if (oldcol != 0) {
-    char *p = (char *)ml_get_buf(win->w_buffer, win->w_cursor.lnum, false);
-    colnr_T len = (colnr_T)STRLEN(p);
+    char *p = ml_get_buf(win->w_buffer, win->w_cursor.lnum, false);
+    co