Extend word commands to use any Unicode whitespace character, GitHub

issue 5562 from Hongyi Zhao.

grid.c

@@ -1,4 +1,4 @@
-/* $OpenBSD: grid.c,v 1.157 2026/08/28 08:11:08 nicm Exp $ */
+/* $OpenBSD: grid.c,v 1.158 2026/09/01 12:49:49 nicm Exp $ */
 
 /*
  * Copyright (c) 2008 Nicholas Marriott <nicholas.marriott@gmail.com>
@@ -1660,21 +1660,29 @@ grid_in_set(struct grid *gd, u_int px, u_int py, const char *set)
 {
 	struct grid_cell	gc, tmp_gc;
 	u_int			pxx;
+	int			has_tab, has_space;
+
+	has_tab = (strchr(set, '\t') != NULL);
+	has_space = (strchr(set, ' ') != NULL);
 
 	grid_get_cell(gd, px, py, &gc);
-	if (strchr(set, '\t')) {
+	if (gc.flags & GRID_FLAG_PADDING) {
-		if (gc.flags & GRID_FLAG_PADDING) {
+		if (!has_tab && !has_space)
-			pxx = px;
+			return (0);
-			do
+		pxx = px;
-				grid_get_cell(gd, --pxx, py, &tmp_gc);
+		do
-			while (pxx > 0 && tmp_gc.flags & GRID_FLAG_PADDING);
+			grid_get_cell(gd, --pxx, py, &tmp_gc);
-			if (tmp_gc.flags & GRID_FLAG_TAB)
+		while (pxx > 0 && tmp_gc.flags & GRID_FLAG_PADDING);
-				return (tmp_gc.data.width - (px - pxx));
+		if (((has_tab || has_space) &&
-		} else if (gc.flags & GRID_FLAG_TAB)
+		    (tmp_gc.flags & GRID_FLAG_TAB)) ||
-			return (gc.data.width);
+		    (has_space && utf8_has_whitespace(&tmp_gc.data)))
-	}
+			return (tmp_gc.data.width - (px - pxx));
-	if (gc.flags & GRID_FLAG_PADDING)
 		return (0);
+	}
+	if ((has_tab || has_space) && (gc.flags & GRID_FLAG_TAB))
+		return (gc.data.width);
+	if (has_space && utf8_has_whitespace(&gc.data))
+		return (gc.data.width == 0 ? 1 : gc.data.width);
 	return (utf8_cstrhas(set, &gc.data));
 }
 

options-table.c

@@ -1,4 +1,4 @@
-/* $OpenBSD: options-table.c,v 1.243 2026/08/25 08:37:08 nicm Exp $ */
+/* $OpenBSD: options-table.c,v 1.244 2026/09/01 12:49:49 nicm Exp $ */
 
 /*
  * Copyright (c) 2011 Nicholas Marriott <nicholas.marriott@gmail.com>
@@ -1251,7 +1251,8 @@ const struct options_table_entry options_table[] = {
 	   * underscore.
 	   */
 	  .default_str = "!\"#$%&'()*+,-./:;<=>?@[\\]^`{|}~",
-	  .text = "Characters considered to separate words."
+	  .text = "Characters considered to separate words; a space matches "
+		  "any character with the Unicode White_Space property."
 	},
 
 	/* Window options. */

tmux.1

@@ -1,4 +1,4 @@
-.\" $OpenBSD: tmux.1,v 1.1162 2026/08/31 19:34:09 nicm Exp $
+.\" $OpenBSD: tmux.1,v 1.1163 2026/09/01 12:49:49 nicm Exp $
 .\"
 .\" Copyright (c) 2007 Nicholas Marriott <nicholas.marriott@gmail.com>
 .\"
@@ -14,7 +14,7 @@
 .\" IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING
 .\" OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 .\"
-.Dd $Mdocdate: August 31 2026 $
+.Dd $Mdocdate: September 1 2026 $
 .Dt TMUX 1
 .Os
 .Sh NAME
@@ -2206,14 +2206,14 @@ Move to the end of the next word.
 .Xc
 Same as
 .Ic next\-word
-but use a space alone as the word separator.
+but treat all non-whitespace characters as part of a word.
 .It Xo
 .Ic next\-space\-end
 (vi: E)
 .Xc
 Same as
 .Ic next\-word\-end
-but use a space alone as the word separator.
+but treat all non-whitespace characters as part of a word.
 .It Xo
 .Ic other\-end
 (vi: o)
@@ -2287,7 +2287,7 @@ Move to the previous word.
 .Xc
 Same as
 .Ic previous\-word
-but use a space alone as the word separator.
+but treat all non-whitespace characters as part of a word.
 .It Xo
 .Ic rectangle\-on
 .Xc
@@ -2572,8 +2572,8 @@ Word separators can be customized with the
 session option.
 Next word moves to the start of the next word, next word end to the end of the
 next word and previous word to the start of the previous word.
-The three next and previous space keys work similarly but use a space alone as
+The three next and previous space keys work similarly but treat all
-the word separator.
+non-whitespace characters as part of a word.
 Setting
 .Em word\-separators
 to the empty string makes next/previous word equivalent to next/previous space.
@@ -5647,6 +5647,11 @@ If set to both, a bell and a message are produced.
 Sets the session's conception of what characters are considered word
 separators, for the purposes of the next and previous word commands in
 copy mode.
+A space in
+.Ar string
+matches any character with the Unicode
+.Em White_Space
+property.
 .El
 .Pp
 Available window options are:

tmux.h

@@ -1,4 +1,4 @@
-/* $OpenBSD: tmux.h,v 1.1432 2026/08/31 19:34:09 nicm Exp $ */
+/* $OpenBSD: tmux.h,v 1.1433 2026/09/01 12:49:49 nicm Exp $ */
 
 /*
  * Copyright (c) 2007 Nicholas Marriott <nicholas.marriott@gmail.com>
@@ -4057,6 +4057,7 @@ void		 session_update_history(struct session *);
 /* utf8.c */
 enum utf8_state	 utf8_towc (const struct utf8_data *, wchar_t *);
 enum utf8_state	 utf8_fromwc(wchar_t wc, struct utf8_data *);
+int		 utf8_has_whitespace(const struct utf8_data *);
 void		 utf8_update_width_cache(void);
 utf8_char	 utf8_build_one(u_char);
 enum utf8_state	 utf8_from_data(const struct utf8_data *, utf8_char *);

utf8.c

@@ -1,4 +1,4 @@
-/* $OpenBSD: utf8.c,v 1.71 2026/05/12 09:37:25 nicm Exp $ */
+/* $OpenBSD: utf8.c,v 1.72 2026/09/01 12:49:49 nicm Exp $ */
 
 /*
  * Copyright (c) 2008 Nicholas Marriott <nicholas.marriott@gmail.com>
@@ -594,6 +594,72 @@ utf8_towc(const struct utf8_data *ud, wchar_t *wc)
 	return (UTF8_DONE);
 }
 
+/* Check for a Unicode whitespace character. */
+int
+utf8_has_whitespace(const struct utf8_data *ud)
+{
+	struct utf8_data tmp;
+	wchar_t		 wc;
+	u_int		 offset = 0, size;
+	u_char		 ch;
+
+	while (offset < ud->size) {
+		ch = ud->data[offset];
+		if (ch < 0x80) {
+			wc = ch;
+			size = 1;
+		} else {
+			if (ch >= 0xc2 && ch <= 0xdf)
+				size = 2;
+			else if (ch >= 0xe0 && ch <= 0xef)
+				size = 3;
+			else if (ch >= 0xf0 && ch <= 0xf4)
+				size = 4;
+			else
+				return (0);
+			if (size > ud->size - offset)
+				return (0);
+
+			memset(&tmp, 0, sizeof tmp);
+			memcpy(tmp.data, ud->data + offset, size);
+			tmp.size = tmp.have = size;
+			if (utf8_towc(&tmp, &wc) != UTF8_DONE)
+				return (0);
+		}
+		offset += size;
+
+		switch (wc) {
+		case 0x0009:
+		case 0x000A:
+		case 0x000B:
+		case 0x000C:
+		case 0x000D:
+		case 0x0020:
+		case 0x0085:
+		case 0x00A0:
+		case 0x1680:
+		case 0x2000:
+		case 0x2001:
+		case 0x2002:
+		case 0x2003:
+		case 0x2004:
+		case 0x2005:
+		case 0x2006:
+		case 0x2007:
+		case 0x2008:
+		case 0x2009:
+		case 0x200A:
+		case 0x2028:
+		case 0x2029:
+		case 0x202F:
+		case 0x205F:
+		case 0x3000:
+			return (1);
+		}
+	}
+	return (0);
+}
+
 /* Convert wide character to UTF-8 character. */
 enum utf8_state
 utf8_fromwc(wchar_t wc, struct utf8_data *ud)