X-Git-Url: https://mattmccutchen.net/rsync/rsync.git/blobdiff_plain/c9a59880f0ddf58c700834fea80c08c39c27ccb3..ec497df1a0f9af53b07fbe23667ff6dd831c3cfd:/lib/wildmatch.c

diff --git a/lib/wildmatch.c b/lib/wildmatch.c
index 9276954c..2f0023d8 100644
--- a/lib/wildmatch.c
+++ b/lib/wildmatch.c
@@ -12,87 +12,194 @@
 #include "rsync.h"
 
 /* What character marks an inverted character class? */
-#define NEGATE_CLASS '!'
+#define NEGATE_CLASS	'!'
+#define NEGATE_CLASS2	'^'
 
 #define FALSE 0
 #define TRUE 1
 #define ABORT_ALL -1
 #define ABORT_TO_STARSTAR -2
 
+#define CC_EQ(class, len, litmatch) ((len) == sizeof (litmatch)-1 \
+				    && *(class) == *(litmatch) \
+				    && strncmp((char*)class, litmatch, len) == 0)
+
+#if defined STDC_HEADERS || !defined isascii
+# define ISASCII(c) 1
+#else
+# define ISASCII(c) isascii(c)
+#endif
+
+#ifdef isblank
+# define ISBLANK(c) (ISASCII(c) && isblank(c))
+#else
+# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
+#endif
+
+#ifdef isgraph
+# define ISGRAPH(c) (ISASCII(c) && isgraph(c))
+#else
+# define ISGRAPH(c) (ISASCII(c) && isprint(c) && !isspace(c))
+#endif
+
+#define ISPRINT(c) (ISASCII(c) && isprint(c))
+#define ISDIGIT(c) (ISASCII(c) && isdigit(c))
+#define ISALNUM(c) (ISASCII(c) && isalnum(c))
+#define ISALPHA(c) (ISASCII(c) && isalpha(c))
+#define ISCNTRL(c) (ISASCII(c) && iscntrl(c))
+#define ISLOWER(c) (ISASCII(c) && islower(c))
+#define ISPUNCT(c) (ISASCII(c) && ispunct(c))
+#define ISSPACE(c) (ISASCII(c) && isspace(c))
+#define ISUPPER(c) (ISASCII(c) && isupper(c))
+#define ISXDIGIT(c) (ISASCII(c) && isxdigit(c))
+
 #ifdef WILD_TEST_ITERATIONS
 int wildmatch_iteration_count;
 #endif
 
-static int domatch(const char *p, const char *text)
+static int force_lower_case = 0;
+
+/* Match pattern "p" against string "text". */
+static int dowild(const uchar *p, const uchar *text)
 {
-    int matched, special;
-    char ch, prev;
+    uchar p_ch;
 
 #ifdef WILD_TEST_ITERATIONS
     wildmatch_iteration_count++;
 #endif
 
-    for ( ; (ch = *p) != '\0'; text++, p++) {
-	if (*text == '\0' && ch != '*')
-	    return FALSE;
-	switch (ch) {
+    for ( ; (p_ch = *p) != '\0'; text++, p++) {
+	int matched, special;
+	uchar t_ch, prev_ch;
+	if ((t_ch = *text) == '\0' && p_ch != '*')
+	    return ABORT_ALL;
+	if (force_lower_case && ISUPPER(t_ch))
+	    t_ch = tolower(t_ch);
+	switch (p_ch) {
 	  case '\\':
 	    /* Literal match with following character.  Note that the test
 	     * in "default" handles the p[1] == '\0' failure case. */
-	    ch = *++p;
+	    p_ch = *++p;
 	    /* FALLTHROUGH */
 	  default:
-	    if (*text != ch)
+	    if (t_ch != p_ch)
 		return FALSE;
 	    continue;
 	  case '?':
 	    /* Match anything but '/'. */
-	    if (*text == '/')
+	    if (t_ch == '/')
 		return FALSE;
 	    continue;
 	  case '*':
 	    if (*++p == '*') {
 		while (*++p == '*') {}
 		special = TRUE;
-	    }
-	    else
+	    } else
 		special = FALSE;
 	    if (*p == '\0') {
 		/* Trailing "**" matches everything.  Trailing "*" matches
 		 * only if there are no more slash characters. */
-		return special? TRUE : strchr(text, '/') == NULL;
+		return special? TRUE : strchr((char*)text, '/') == NULL;
 	    }
-	    for ( ; *text; text++) {
-		if ((matched = domatch(p, text)) != FALSE) {
+	    for ( ; t_ch; (t_ch = *++text)) {
+		if ((matched = dowild(p, text)) != FALSE) {
 		    if (!special || matched != ABORT_TO_STARSTAR)
 			return matched;
-		}
-		else if (!special && *text == '/')
+		} else if (!special && t_ch == '/')
 		    return ABORT_TO_STARSTAR;
 	    }
 	    return ABORT_ALL;
 	  case '[':
-	    ch = *++p;
+	    p_ch = *++p;
+#ifdef NEGATE_CLASS2
+	    if (p_ch == NEGATE_CLASS2)
+		p_ch = NEGATE_CLASS;
+#endif
 	    /* Assign literal TRUE/FALSE because of "matched" comparison. */
-	    special = ch == NEGATE_CLASS? TRUE : FALSE;
+	    special = p_ch == NEGATE_CLASS? TRUE : FALSE;
 	    if (special) {
 		/* Inverted character class. */
-		ch = *++p;
+		p_ch = *++p;
 	    }
-	    prev = 0;
+	    prev_ch = 0;
 	    matched = FALSE;
 	    do {
-		if (!ch)
-		    return FALSE;
-		if (ch == '-' && prev && p[1] && p[1] != ']') {
-		    if (*text <= *++p && *text >= prev)
+		if (!p_ch)
+		    return ABORT_ALL;
+		if (p_ch == '\\') {
+		    p_ch = *++p;
+		    if (!p_ch)
+			return ABORT_ALL;
+		    if (t_ch == p_ch)
+			matched = TRUE;
+		} else if (p_ch == '-' && prev_ch && p[1] && p[1] != ']') {
+		    p_ch = *++p;
+		    if (p_ch == '\\') {
+			p_ch = *++p;
+			if (!p_ch)
+			    return ABORT_ALL;
+		    }
+		    if (t_ch <= p_ch && t_ch >= prev_ch)
 			matched = TRUE;
-		    ch = 0; /* This makes "prev" get set to 0. */
-		}
-		else if (*text == ch)
+		    p_ch = 0; /* This makes "prev_ch" get set to 0. */
+		} else if (p_ch == '[' && p[1] == ':') {
+		    const uchar *s;
+		    int i;
+		    for (s = p += 2; (p_ch = *p) && p_ch != ']'; p++) {}
+		    if (!p_ch)
+			return ABORT_ALL;
+		    i = p - s - 1;
+		    if (i < 0 || p[-1] != ':') {
+			/* Didn't find ":]", so treat like a normal set. */
+			p = s - 2;
+			p_ch = '[';
+			if (t_ch == p_ch)
+			    matched = TRUE;
+			continue;
+		    }
+		    if (CC_EQ(s,i, "alnum")) {
+			if (ISALNUM(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "alpha")) {
+			if (ISALPHA(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "blank")) {
+			if (ISBLANK(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "cntrl")) {
+			if (ISCNTRL(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "digit")) {
+			if (ISDIGIT(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "graph")) {
+			if (ISGRAPH(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "lower")) {
+			if (ISLOWER(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "print")) {
+			if (ISPRINT(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "punct")) {
+			if (ISPUNCT(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "space")) {
+			if (ISSPACE(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "upper")) {
+			if (ISUPPER(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "xdigit")) {
+			if (ISXDIGIT(t_ch))
+			    matched = TRUE;
+		    } else /* malformed [:class:] string */
+			return ABORT_ALL;
+		    p_ch = 0; /* This makes "prev_ch" get set to 0. */
+		} else if (t_ch == p_ch)
 		    matched = TRUE;
-	    } while (prev = ch, (ch = *++p) != ']');
-	    if (matched == special)
+	    } while (prev_ch = p_ch, (p_ch = *++p) != ']');
+	    if (matched == special || t_ch == '/')
 		return FALSE;
 	    continue;
 	}
@@ -101,10 +208,24 @@ static int domatch(const char *p, const char *text)
     return *text == '\0';
 }
 
-int wildmatch(const char *p, const char *text)
+/* Match the "pattern" against the "text" string. */
+int wildmatch(const char *pattern, const char *text)
+{
+#ifdef WILD_TEST_ITERATIONS
+    wildmatch_iteration_count = 0;
+#endif
+    return dowild((const uchar*)pattern, (const uchar*)text) == TRUE;
+}
+
+/* Match the "pattern" against the forced-to-lower-case "text" string. */
+int iwildmatch(const char *pattern, const char *text)
 {
+    int ret;
 #ifdef WILD_TEST_ITERATIONS
     wildmatch_iteration_count = 0;
 #endif
-    return domatch(p, text) == TRUE;
+    force_lower_case = 1;
+    ret = dowild((const uchar*)pattern, (const uchar*)text) == TRUE;
+    force_lower_case = 0;
+    return ret;
 }