about summary refs log tree commit diff
diff options
context:
space:
mode:
authorRich Felker <dalias@aerifal.cx>2013-06-04 16:09:36 -0400
committerRich Felker <dalias@aerifal.cx>2013-06-04 16:09:36 -0400
commitf18846dd3a048598676e10b2a7b9f931bb3d1d6a (patch)
tree55582d95c77451bfdb0a63e936be144109fca256
parenta6d272127bb6a2eb09dc182cc39c49e77310ade4 (diff)
downloadmusl-f18846dd3a048598676e10b2a7b9f931bb3d1d6a.tar.gz
musl-f18846dd3a048598676e10b2a7b9f931bb3d1d6a.tar.xz
musl-f18846dd3a048598676e10b2a7b9f931bb3d1d6a.zip
refactor scanf core to use common code path for all string formats
the concept here is that %s and %c are essentially special-cases of
%[, with some minimal additional special-casing.

aside from simplifying the code and reducing the number of complex
code-paths that would need changing to make optimizations later, the
main purpose of this change is to simplify addition of the 'm'
modifier which causes scanf to allocate storage for the string being
read.
-rw-r--r--src/stdio/vfscanf.c137
1 files changed, 52 insertions, 85 deletions
diff --git a/src/stdio/vfscanf.c b/src/stdio/vfscanf.c
index fe071e95..62bf47f6 100644
--- a/src/stdio/vfscanf.c
+++ b/src/stdio/vfscanf.c
@@ -88,6 +88,7 @@ int vfscanf(FILE *restrict f, const char *restrict fmt, va_list ap)
 	unsigned long long x;
 	long double y;
 	off_t pos = 0;
+	unsigned char scanset[257];
 
 	FLOCK(f);
 
@@ -170,104 +171,92 @@ int vfscanf(FILE *restrict f, const char *restrict fmt, va_list ap)
 
 		switch (t) {
 		case 'C':
+			if (width < 1) width = 1;
+		case 'S':
+			t |= 32;
+			size = SIZE_l;
+			break;
 		case 'c':
 			if (width < 1) width = 1;
-		case 's':
-			if (size == SIZE_l) t &= ~0x20;
 		case 'd': case 'i': case 'o': case 'u': case 'x':
 		case 'a': case 'e': case 'f': case 'g':
 		case 'A': case 'E': case 'F': case 'G': case 'X':
-		case '[': case 'S':
+		case '[': case 's':
 		case 'p': case 'n':
-			if (width < 1) width = 0;
 			break;
 		default:
 			goto fmt_fail;
 		}
 
-		shlim(f, width);
+		if (t == 'n') {
+			store_int(dest, size, pos);
+			/* do not increment match count, etc! */
+			continue;
+		}
 
-		if (t != 'n') {
-			if (shgetc(f) < 0) goto input_fail;
+		if (t != '[' && (t|32) != 'c') {
+			shlim(f, 0);
+			while (isspace(shgetc(f)));
 			shunget(f);
+			pos += shcnt(f);
 		}
 
+		shlim(f, width);
+		if (shgetc(f) < 0) goto input_fail;
+		shunget(f);
+
 		switch (t) {
-		case 'n':
-			store_int(dest, size, pos);
-			/* do not increment match count, etc! */
-			continue;
-		case 'C':
-			wcs = dest;
-			st = (mbstate_t){ 0 };
-			while ((c=shgetc(f)) >= 0) {
-				if (readwc(c, &wcs, &st) < 0)
-					goto input_fail;
-			}
-			if (!mbsinit(&st)) goto input_fail;
-			if (shcnt(f) != width) goto match_fail;
-			break;
+		case 's':
 		case 'c':
-			if (dest) {
-				s = dest;
-				while ((c=shgetc(f)) >= 0) *s++ = c;
-			} else {
-				while (shgetc(f)>=0);
-			}
-			if (shcnt(f) < width) goto match_fail;
-			break;
 		case '[':
-			s = dest;
-			wcs = dest;
-
-			if (*++p == '^') p++, invert = 1;
-			else invert = 0;
-
-			unsigned char scanset[257];
-			memset(scanset, invert, sizeof scanset);
-
-			scanset[0] = 0;
-			if (*p == '-') p++, scanset[1+'-'] = 1-invert;
-			else if (*p == ']') p++, scanset[1+']'] = 1-invert;
-			for (; *p != ']'; p++) {
-				if (!*p) goto fmt_fail;
-				if (*p=='-' && p[1] && p[1] != ']')
-					for (c=p++[-1]; c<*p; c++)
-						scanset[1+c] = 1-invert;
-				scanset[1+*p] = 1-invert;
+			if (t == 'c' || t == 's') {
+				memset(scanset, -1, sizeof scanset);
+				scanset[0] = 0;
+				if (t == 's') {
+					scanset[1+'\t'] = 0;
+					scanset[1+'\n'] = 0;
+					scanset[1+'\v'] = 0;
+					scanset[1+'\f'] = 0;
+					scanset[1+'\r'] = 0;
+					scanset[1+' '] = 0;
+				}
+			} else {
+				if (*++p == '^') p++, invert = 1;
+				else invert = 0;
+				memset(scanset, invert, sizeof scanset);
+				scanset[0] = 0;
+				if (*p == '-') p++, scanset[1+'-'] = 1-invert;
+				else if (*p == ']') p++, scanset[1+']'] = 1-invert;
+				for (; *p != ']'; p++) {
+					if (!*p) goto fmt_fail;
+					if (*p=='-' && p[1] && p[1] != ']')
+						for (c=p++[-1]; c<*p; c++)
+							scanset[1+c] = 1-invert;
+					scanset[1+*p] = 1-invert;
+				}
 			}
-
+			wcs = 0;
+			s = 0;
 			if (size == SIZE_l) {
+				wcs = dest;
 				st = (mbstate_t){0};
 				while (scanset[(c=shgetc(f))+1]) {
 					if (readwc(c, &wcs, &st) < 0)
 						goto input_fail;
 				}
 				if (!mbsinit(&st)) goto input_fail;
-				s = 0;
-			} else if (s) {
+			} else if ((s = dest)) {
 				while (scanset[(c=shgetc(f))+1])
 					*s++ = c;
-				wcs = 0;
 			} else {
 				while (scanset[(c=shgetc(f))+1]);
 			}
 			shunget(f);
 			if (!shcnt(f)) goto match_fail;
-			if (s) *s = 0;
+			if (t == 'c' && shcnt(f) != width) goto match_fail;
 			if (wcs) *wcs = 0;
+			if (s) *s = 0;
 			break;
-		default:
-			shlim(f, 0);
-			while (isspace(shgetc(f)));
-			shunget(f);
-			pos += shcnt(f);
-			shlim(f, width);
-			if (shgetc(f) < 0) goto input_fail;
-			shunget(f);
-		}
-
-		switch (t) {
 		case 'p':
 		case 'X':
 		case 'x':
@@ -306,28 +295,6 @@ int vfscanf(FILE *restrict f, const char *restrict fmt, va_list ap)
 				break;
 			}
 			break;
-		case 'S':
-			wcs = dest;
-			st = (mbstate_t){ 0 };
-			while (!isspace(c=shgetc(f)) && c!=EOF) {
-				if (readwc(c, &wcs, &st) < 0)
-					goto input_fail;
-			}
-			shunget(f);
-			if (!mbsinit(&st)) goto input_fail;
-			if (dest) *wcs++ = 0;
-			break;
-		case 's':
-			if (dest) {
-				s = dest;
-				while (!isspace(c=shgetc(f)) && c!=EOF)
-					*s++ = c;
-				*s = 0;
-			} else {
-				while (!isspace(c=shgetc(f)) && c!=EOF);
-			}
-			shunget(f);
-			break;
 		}
 
 		pos += shcnt(f);