git.stevedylan.dev

render invalid utf8 sequences as U+FFFD 59936c7d

previously drw_text would do the width calculations as if
invalid utf8 sequences were replaced with U+FFFD but would pass
the invalid utf8 sequence to xft to render where xft would just
cut it off at the first invalid byte.

this change makes invalid utf8 render as U+FFFD and avoids
sending invalid sequences to xft. the following can be used to
check the behavior before and after the patch:

	$ printf "0\xef1234567\ntest" | dmenu

Ref: https://lists.suckless.org/dev/2407/35646.html

NRK · 2024-07-04 21:27 1 file(s) · +15 −4

drw.c +15 −4

	XftResult result;
	int charexists = 0, overflow = 0;
	/* keep track of a couple codepoints for which we have no match. */
	static unsigned int nomatches[128], ellipsis_width;
	static unsigned int nomatches[128], ellipsis_width, invalid_width;
	static const char invalid[] = "�";

	if (!drw || (render && (!drw->scheme || !w)) || !text || !drw->fonts)
		return 0;

	usedfont = drw->fonts;
	if (!ellipsis_width && render)
		ellipsis_width = drw_fontset_getwidth(drw, "...");
	if (!invalid_width) {
		invalid_width = -1; /* stop infinite recursion */
		invalid_width = drw_fontset_getwidth(drw, invalid);
	}
	while (1) {
		ew = ellipsis_len = utf8err = utf8charlen = utf8strlen = 0;
		utf8str = text;

						else
							utf8strlen = ellipsis_len;
					} else if (curfont == usedfont) {
						utf8strlen += utf8charlen;
						text += utf8charlen;
						ew += tmpw;
						utf8strlen += utf8err ? 0 : utf8charlen;
						ew += utf8err ? 0 : tmpw;
					} else {
						nextfont = curfont;
					}

				}
			}

			if (overflow || !charexists || nextfont)
			if (overflow || !charexists || nextfont || utf8err)
				break;
			else
				charexists = 0;

			}
			x += ew;
			w -= ew;
		}
		if (utf8err && (!render || invalid_width < w)) {
			if (render)
				drw_text(drw, x, y, w, h, 0, invalid, invert);
			x += invalid_width;
			w -= invalid_width;
		}
		if (render && overflow)
			drw_text(drw, ellipsis_x, y, ellipsis_w, h, 0, "...", invert);

237	237		XftResult result;
238	238		int charexists = 0, overflow = 0;
239	239		/* keep track of a couple codepoints for which we have no match. */
240		-	static unsigned int nomatches[128], ellipsis_width;
	240	+	static unsigned int nomatches[128], ellipsis_width, invalid_width;
	241	+	static const char invalid[] = "�";
241	242
242	243		if (!drw \|\| (render && (!drw->scheme \|\| !w)) \|\| !text \|\| !drw->fonts)
243	244		return 0;

257	258		usedfont = drw->fonts;
258	259		if (!ellipsis_width && render)
259	260		ellipsis_width = drw_fontset_getwidth(drw, "...");
	261	+	if (!invalid_width) {
	262	+	invalid_width = -1; /* stop infinite recursion */
	263	+	invalid_width = drw_fontset_getwidth(drw, invalid);
	264	+	}
260	265		while (1) {
261	266		ew = ellipsis_len = utf8err = utf8charlen = utf8strlen = 0;
262	267		utf8str = text;

284	289		else
285	290		utf8strlen = ellipsis_len;
286	291		} else if (curfont == usedfont) {
287		-	utf8strlen += utf8charlen;
288	292		text += utf8charlen;
289		-	ew += tmpw;
	293	+	utf8strlen += utf8err ? 0 : utf8charlen;
	294	+	ew += utf8err ? 0 : tmpw;
290	295		} else {
291	296		nextfont = curfont;
292	297		}

294	299		}
295	300		}
296	301
297		-	if (overflow \|\| !charexists \|\| nextfont)
	302	+	if (overflow \|\| !charexists \|\| nextfont \|\| utf8err)
298	303		break;
299	304		else
300	305		charexists = 0;

308	313		}
309	314		x += ew;
310	315		w -= ew;
	316	+	}
	317	+	if (utf8err && (!render \|\| invalid_width < w)) {
	318	+	if (render)
	319	+	drw_text(drw, x, y, w, h, 0, invalid, invert);
	320	+	x += invalid_width;
	321	+	w -= invalid_width;
311	322		}
312	323		if (render && overflow)
313	324		drw_text(drw, ellipsis_x, y, ellipsis_w, h, 0, "...", invert);