moving: skip combining characters and other zero-width characters

This makes the cursor move smoothly left and right -- instead of
"stuttering" when passing over a zero-width character.

Pressing <Delete> on a normal (spacing) character also deletes
any zero-width characters after it.  But pressing <Backspace>
while the cursor is placed after a zero-width character, just
deletes that zero-width character.  The latter behavior allows
deleting and retyping just the combining diacritic of a character
instead of the whole character.

This addresses https://savannah.gnu.org/bugs/?50773.
Requested-by: Mike Frysinger <vapier@gentoo.org>

src/chars.c

@@ -197,6 +197,12 @@ int mbwidth(const char *c)
 		return 1;
 }
 
+/* Return TRUE when the given character occupies zero cells. */
+bool is_zerowidth(const char *ch)
+{
+	return (use_utf8 && mbwidth(ch) == 0);
+}
+
 /* Convert the given Unicode value to a multibyte character, if possible.
  * If the conversion succeeds, return the (dynamically allocated) multibyte
  * character and its length.  Otherwise, return a length of zero. */

src/cut.c

@@ -121,7 +121,14 @@ void do_delete(void)
 		zap_text();
 	else
 #endif
+	{
 		do_deletion(DEL);
+#ifdef ENABLE_UTF8
+		while (openfile->current->data[openfile->current_x] != '\0' &&
+				is_zerowidth(openfile->current->data + openfile->current_x))
+			do_deletion(DEL);
+#endif
+	}
 }
 
 /* Backspace over one character.  That is, move the cursor left one
@@ -133,7 +140,10 @@ void do_backspace(void)
 		zap_text();
 	else
 #endif
-	if (openfile->current_x > 0 || openfile->current != openfile->filetop) {
+	if (openfile->current_x > 0) {
+		openfile->current_x = step_left(openfile->current->data, openfile->current_x);
+		do_deletion(BACK);
+	} else if (openfile->current != openfile->filetop) {
 		do_left();
 		do_deletion(BACK);
 	}

src/move.c

@@ -291,6 +291,10 @@ void do_prev_word(bool allow_punct)
 			/* If at the head of a line now, this surely is a word start. */
 			if (openfile->current_x == 0)
 				break;
+#ifdef ENABLE_UTF8
+		} else if (is_zerowidth(openfile->current->data + openfile->current_x)) {
+			; /* skip */
+#endif
 		} else if (seen_a_word) {
 			/* This is space now: we've overshot the start of the word. */
 			step_forward = TRUE;
@@ -339,11 +343,20 @@ bool do_next_word(bool after_ends, bool allow_punct)
 			if (is_word_char(openfile->current->data + openfile->current_x,
 								allow_punct))
 				seen_word = TRUE;
+#ifdef ENABLE_UTF8
+			else if (is_zerowidth(openfile->current->data + openfile->current_x))
+				; /* skip */
+#endif
 			else if (seen_word)
 				break;
 		} else
 #endif
 		{
+#ifdef ENABLE_UTF8
+			if (is_zerowidth(openfile->current->data + openfile->current_x))
+				; /* skip */
+			else
+#endif
 			/* If this is not a word character, then it's a separator; else
 			 * if we've already seen a separator, then it's a word start. */
 			if (!is_word_char(openfile->current->data + openfile->current_x,
@@ -582,8 +595,16 @@ void do_left(void)
 	linestruct *was_current = openfile->current;
 
 	if (openfile->current_x > 0)
+	{
 		openfile->current_x = step_left(openfile->current->data,
 												openfile->current_x);
+#ifdef ENABLE_UTF8
+		while (openfile->current_x > 0 &&
+					is_zerowidth(openfile->current->data + openfile->current_x))
+			openfile->current_x = step_left(openfile->current->data,
+												openfile->current_x);
+#endif
+	}
 	else if (openfile->current != openfile->filetop) {
 		openfile->current = openfile->current->prev;
 		openfile->current_x = strlen(openfile->current->data);
@@ -598,8 +619,16 @@ void do_right(void)
 	linestruct *was_current = openfile->current;
 
 	if (openfile->current->data[openfile->current_x] != '\0')
+	{
 		openfile->current_x = step_right(openfile->current->data,
 												openfile->current_x);
+#ifdef ENABLE_UTF8
+		while (openfile->current->data[openfile->current_x] != '\0' &&
+					is_zerowidth(openfile->current->data + openfile->current_x))
+			openfile->current_x = step_right(openfile->current->data,
+												openfile->current_x);
+#endif
+	}
 	else if (openfile->current != openfile->filebot) {
 		openfile->current = openfile->current->next;
 		openfile->current_x = 0;

src/prototypes.h

@@ -204,6 +204,7 @@ bool is_word_char(const char *c, bool allow_punct);
 char control_mbrep(const char *c, bool isdata);
 #ifdef ENABLE_UTF8
 int mbwidth(const char *c);
+bool is_zerowidth(const char *ch);
 char *make_mbchar(long code, int *length);
 #endif
 int char_length(const char *pointer);