sync drw.{c,h} from dmenu

- drw: minor improvement to the nomatches cache
- overhaul utf8decoding and render invalid utf8 sequences as U+FFFD.

Thanks NRK for these improvements!

ref.
https://git.suckless.org/dwm/commit/8933ebcf50024f4378a78e556b1ac08091197206.html#h0-1

drw.c

@@ -35,12 +35,6 @@ apply_fribidi(const char *str)
 
 #if !BAR_PANGO_PATCH
 #define UTF_INVALID 0xFFFD
-#define UTF_SIZ     4
-
-static const unsigned char utfbyte[UTF_SIZ + 1] = {0x80,    0, 0xC0, 0xE0, 0xF0};
-static const unsigned char utfmask[UTF_SIZ + 1] = {0xC0, 0x80, 0xE0, 0xF0, 0xF8};
-static const long utfmin[UTF_SIZ + 1] = {       0,    0,  0x80,  0x800,  0x10000};
-static const long utfmax[UTF_SIZ + 1] = {0x10FFFF, 0x7F, 0x7FF, 0xFFFF, 0x10FFFF};
 #endif // BAR_PANGO_PATCH
 
 #if BAR_POWERLINE_TAGS_PATCH || BAR_POWERLINE_STATUS_PATCH
@@ -48,47 +42,39 @@ Clr transcheme[3];
 #endif // BAR_POWERLINE_TAGS_PATCH | BAR_POWERLINE_STATUS_PATCH
 
 #if !BAR_PANGO_PATCH
-static long
+static int
-utf8decodebyte(const char c, size_t *i)
+utf8decode(const char *s_in, long *u, int *err)
 {
-	for (*i = 0; *i < (UTF_SIZ + 1); ++(*i))
+	static const unsigned char lens[] = {
-		if (((unsigned char)c & utfmask[*i]) == utfbyte[*i])
+		/* 0XXXX */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-			return (unsigned char)c & ~utfmask[*i];
+		/* 10XXX */ 0, 0, 0, 0, 0, 0, 0, 0,  /* invalid */
-	return 0;
+		/* 110XX */ 2, 2, 2, 2,
-}
+		/* 1110X */ 3, 3,
+		/* 11110 */ 4,
+		/* 11111 */ 0,  /* invalid */
+	};
+	static const unsigned char leading_mask[] = { 0x7F, 0x1F, 0x0F, 0x07 };
+	static const unsigned int overlong[] = { 0x0, 0x80, 0x0800, 0x10000 };
 
-static size_t
+	const unsigned char *s = (const unsigned char *)s_in;
-utf8validate(long *u, size_t i)
+	int len = lens[*s >> 3];
-{
-	if (!BETWEEN(*u, utfmin[i], utfmax[i]) || BETWEEN(*u, 0xD800, 0xDFFF))
 	*u = UTF_INVALID;
-	for (i = 1; *u > utfmax[i]; ++i)
+	*err = 1;
-		;
+	if (len == 0)
-	return i;
-}
-
-static size_t
-utf8decode(const char *c, long *u, size_t clen)
-{
-	size_t i, j, len, type;
-	long udecoded;
-
-	*u = UTF_INVALID;
-	if (!clen)
-		return 0;
-	udecoded = utf8decodebyte(c[0], &len);
-	if (!BETWEEN(len, 1, UTF_SIZ))
 		return 1;
-	for (i = 1, j = 1; i < clen && j < len; ++i, ++j) {
-		udecoded = (udecoded << 6) | utf8decodebyte(c[i], &type);
-		if (type)
-			return j;
-	}
-	if (j < len)
-		return 0;
-	*u = udecoded;
-	utf8validate(u, len);
 
+	long cp = s[0] & leading_mask[len - 1];
+	for (int i = 1; i < len; ++i) {
+		if (s[i] == '\0' || (s[i] & 0xC0) != 0x80)
+			return i;
+		cp = (cp << 6) | (s[i] & 0x3F);
+	}
+	/* out of range, surrogate, overlong encoding */
+	if (cp > 0x10FFFF || (cp >> 11) == 0x1B || cp < overlong[len - 1])
+		return len;
+
+	*err = 0;
+	*u = cp;
 	return len;
 }
 #endif // BAR_PANGO_PATCH
@@ -500,26 +486,27 @@ drw_text(Drw *drw, int x, int y, unsigned int w, unsigned int h, unsigned int lp
 
 	return x + (render ? w : 0);
 #else
-	char buf[1024];
+	int ty, ellipsis_x = 0;
-	int ty;
+	unsigned int tmpw, ew, ellipsis_w = 0, ellipsis_len, hash, h0, h1;
-	unsigned int ew;
 	XftDraw *d = NULL;
 	Fnt *usedfont, *curfont, *nextfont;
-	size_t i, len;
+	int utf8strlen, utf8charlen, utf8err, render = x || y || w || h;
-	int utf8strlen, utf8charlen, render = x || y || w || h;
 	long utf8codepoint = 0;
 	const char *utf8str;
 	FcCharSet *fccharset;
 	FcPattern *fcpattern;
 	FcPattern *match;
 	XftResult result;
-	int charexists = 0;
+	int charexists = 0, overflow = 0;
+	/* keep track of a couple codepoints for which we have no match. */
+	static unsigned int nomatches[128], ellipsis_width, invalid_width;
+	static const char invalid[] = "<EFBFBD>";
 
-	if (!drw || (render && !drw->scheme) || !text || !drw->fonts)
+	if (!drw || (render && (!drw->scheme || !w)) || !text || !drw->fonts)
 		return 0;
 
 	if (!render) {
-		w = ~w;
+		w = invert ? invert : ~invert;
 	} else {
 		XSetForeground(drw->dpy, drw->gc, drw->scheme[invert ? ColFg : ColBg].pixel);
 		XFillRectangle(drw->dpy, drw->drawable, drw->gc, x, y, w, h);
@@ -535,18 +522,40 @@ drw_text(Drw *drw, int x, int y, unsigned int w, unsigned int h, unsigned int lp
 	}
 
 	usedfont = drw->fonts;
+	if (!ellipsis_width && render)
+		ellipsis_width = drw_fontset_getwidth(drw, "...", markup);
+	if (!invalid_width && render)
+		invalid_width = drw_fontset_getwidth(drw, invalid, markup);
 	while (1) {
-		utf8strlen = 0;
+		ew = ellipsis_len = utf8err = utf8charlen = utf8strlen = 0;
 		utf8str = text;
 		nextfont = NULL;
 		while (*text) {
-			utf8charlen = utf8decode(text, &utf8codepoint, UTF_SIZ);
+			utf8charlen = utf8decode(text, &utf8codepoint, &utf8err);
 			for (curfont = drw->fonts; curfont; curfont = curfont->next) {
 				charexists = charexists || XftCharExists(drw->dpy, curfont->xfont, utf8codepoint);
 				if (charexists) {
-					if (curfont == usedfont) {
+					drw_font_getexts(curfont, text, utf8charlen, &tmpw, NULL);
-						utf8strlen += utf8charlen;
+					if (ew + ellipsis_width <= w) {
+						/* keep track where the ellipsis still fits */
+						ellipsis_x = x + ew;
+						ellipsis_w = w - ew;
+						ellipsis_len = utf8strlen;
+					}
+
+					if (ew + tmpw > w) {
+						overflow = 1;
+						/* called from drw_fontset_getwidth_clamp():
+						 * it wants the width AFTER the overflow
+						 */
+						if (!render)
+							x += tmpw;
+						else
+							utf8strlen = ellipsis_len;
+					} else if (curfont == usedfont) {
 						text += utf8charlen;
+						utf8strlen += utf8err ? 0 : utf8charlen;
+						ew += utf8err ? 0 : tmpw;
 					} else {
 						nextfont = curfont;
 					}
@@ -554,36 +563,31 @@ drw_text(Drw *drw, int x, int y, unsigned int w, unsigned int h, unsigned int lp
 				}
 			}
 
-			if (!charexists || nextfont)
+			if (overflow || !charexists || nextfont || utf8err)
 				break;
 			else
 				charexists = 0;
 		}
 
 		if (utf8strlen) {
-			drw_font_getexts(usedfont, utf8str, utf8strlen, &ew, NULL);
-			/* shorten text if necessary */
-			for (len = MIN(utf8strlen, sizeof(buf) - 1); len && ew > w; drw_font_getexts(usedfont, utf8str, len, &ew, NULL))
-				len--;
-
-			if (len) {
-				memcpy(buf, utf8str, len);
-				buf[len] = '\0';
-				if (len < utf8strlen)
-					for (i = len; i && i > len - 3; buf[--i] = '.')
-						; /* NOP */
-
 			if (render) {
 				ty = y + (h - usedfont->h) / 2 + usedfont->xfont->ascent;
 				XftDrawStringUtf8(d, &drw->scheme[invert ? ColBg : ColFg],
-					                  usedfont->xfont, x, ty, (XftChar8 *)buf, len);
+				                  usedfont->xfont, x, ty, (XftChar8 *)utf8str, utf8strlen);
 			}
 			x += ew;
 			w -= ew;
 		}
+		if (utf8err && (!render || invalid_width < w)) {
+			if (render)
+				drw_text(drw, x, y, w, h, 0, invalid, invert, markup);
+			x += invalid_width;
+			w -= invalid_width;
 		}
+		if (render && overflow)
+			drw_text(drw, ellipsis_x, y, ellipsis_w, h, 0, "...", invert, markup);
 
-		if (!*text) {
+		if (!*text || overflow) {
 			break;
 		} else if (nextfont) {
 			charexists = 0;
@@ -593,6 +597,15 @@ drw_text(Drw *drw, int x, int y, unsigned int w, unsigned int h, unsigned int lp
 			 * character must be drawn. */
 			charexists = 1;
 
+			hash = (unsigned int)utf8codepoint;
+			hash = ((hash >> 16) ^ hash) * 0x21F0AAAD;
+			hash = ((hash >> 15) ^ hash) * 0xD35A2D97;
+			h0 = ((hash >> 15) ^ hash) % LENGTH(nomatches);
+			h1 = (hash >> 17) % LENGTH(nomatches);
+			/* avoid expensive XftFontMatch call when we know we won't find a match */
+			if (nomatches[h0] == utf8codepoint || nomatches[h1] == utf8codepoint)
+				goto no_match;
+
 			fccharset = FcCharSetCreate();
 			FcCharSetAddChar(fccharset, utf8codepoint);
 
@@ -623,6 +636,8 @@ drw_text(Drw *drw, int x, int y, unsigned int w, unsigned int h, unsigned int lp
 					curfont->next = usedfont;
 				} else {
 					xfont_free(usedfont);
+					nomatches[nomatches[h0] ? h1 : h0] = utf8codepoint;
+no_match:
 					usedfont = drw->fonts;
 				}
 			}
@@ -754,4 +769,3 @@ drw_cur_free(Drw *drw, Cur *cursor)
 	XFreeCursor(drw->dpy, cursor->cursor);
 	free(cursor);
 }
-

dwm.c

@@ -101,7 +101,6 @@
 #else
 #define ISVISIBLE(C)            ((C->tags & C->mon->tagset[C->mon->seltags]))
 #endif // ATTACHASIDE_PATCH
-#define LENGTH(X)               (sizeof X / sizeof X[0])
 #define MOUSEMASK               (BUTTONMASK|PointerMotionMask)
 #define WIDTH(X)                ((X)->w + 2 * (X)->bw)
 #define HEIGHT(X)               ((X)->h + 2 * (X)->bw)

util.h

@@ -7,6 +7,7 @@
 #define MIN(A, B)               ((A) < (B) ? (A) : (B))
 #endif
 #define BETWEEN(X, A, B)        ((A) <= (X) && (X) <= (B))
+#define LENGTH(X)               (sizeof (X) / sizeof (X)[0])
 
 #ifdef _DEBUG
 #define DEBUG(...) fprintf(stderr, __VA_ARGS__)