[PATCH 3/5] lib/string: Add a generic wildcard string matching function

[Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>]

Add strglobmatch() for generic wildcard string matching.
This code is originally from perf-tools. For porting in
the kernel, the limitation of the number of wildcards is
introduced, because of the limitation of the stack.

Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Akinobu Mita <akinobu.mita@gmail.com>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: David Howells <dhowells@redhat.com>
---
 include/linux/string.h |    8 ++++
 lib/string.c           |   91 ++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 99 insertions(+)

diff --git a/include/linux/string.h b/include/linux/string.h
index ac889c5..fcb1506 100644
--- a/include/linux/string.h
+++ b/include/linux/string.h
@@ -122,6 +122,14 @@ extern void argv_free(char **argv);
 extern bool sysfs_streq(const char *s1, const char *s2);
 extern int strtobool(const char *s, bool *res);
 
+/*
+ * Maximum number of wildcards in a glob pattern
+ * This limits the number of wildcards in a glob pattern, because
+ * more wildcards consume more kernel stack.
+ */
+#define MAX_GLOB_WILDCARDS	10
+extern bool strglobmatch(const char *glob, const char *str);
+
 #ifdef CONFIG_BINARY_PRINTF
 int vbin_printf(u32 *bin_buf, size_t size, const char *fmt, va_list args);
 int bstr_printf(char *buf, size_t size, const char *fmt, const u32 *bin_buf);
diff --git a/lib/string.c b/lib/string.c
index e5878de..640fb10 100644
--- a/lib/string.c
+++ b/lib/string.c
@@ -824,3 +824,94 @@ void *memchr_inv(const void *start, int c, size_t bytes)
 	return check_bytes8(start, value, bytes % 8);
 }
 EXPORT_SYMBOL(memchr_inv);
+
+/* Character class matching */
+static bool __match_charclass(const char *pat, char c, const char **npat)
+{
+	bool complement = false, ret = true;
+
+	if (*pat == '!') {
+		complement = true;
+		pat++;
+	}
+	if (*pat++ == c)	/* First character is special */
+		goto end;
+
+	while (*pat && *pat != ']') {	/* Matching */
+		if (*pat == '-' && *(pat + 1) != ']') {	/* Range */
+			if (*(pat - 1) <= c && c <= *(pat + 1))
+				goto end;
+			if (*(pat - 1) > *(pat + 1))
+				goto error;
+			pat += 2;
+		} else if (*pat++ == c)
+			goto end;
+	}
+	if (!*pat)
+		goto error;
+	ret = false;
+
+end:
+	while (*pat && *pat != ']')	/* Searching closing */
+		pat++;
+	if (!*pat)
+		goto error;
+	*npat = pat + 1;
+	return complement ? !ret : ret;
+
+error:
+	return false;
+}
+
+/* Glob pattern(wildcard and char-class) matching */
+static bool __match_glob(const char *pat, const char *str, int nest)
+{
+	if (nest > MAX_GLOB_WILDCARDS)
+		return false;
+
+	while (*str && *pat && *pat != '*') {
+		if (*pat == '?') {	/* Matches any single character */
+			str++;
+			pat++;
+			continue;
+		} else if (*pat == '[')	/* Character classes/Ranges */
+			if (__match_charclass(pat + 1, *str, &pat)) {
+				str++;
+				continue;
+			} else
+				return false;
+		else if (*pat == '\\') /* Escaped char match as normal char */
+			pat++;
+		if (*str++ != *pat++)
+			return false;
+	}
+	/* Check wild card */
+	if (*pat == '*') {
+		while (*pat == '*')
+			pat++;
+		if (!*pat)	/* Tail wild card matches all */
+			return true;
+		while (*str)
+			if (__match_glob(pat, str++, nest + 1))
+				return true;
+	}
+	return !*str && !*pat;
+}
+
+/**
+ * strglobmatch - glob expression pattern matching
+ * @pat: the pattern string to match
+ * @str: the target string to match
+ *
+ * This returns true if the @str matches @pat. @pat can includes wildcards
+ * ('*','?') and character classes ([CHARS], complementation and ranges are
+ * also supported). Also, this supports escape character ('\') to use special
+ * characters as normal character.
+ *
+ * Note: if @pat syntax is broken, this always returns false.
+ */
+bool strglobmatch(const char *pat, const char *str)
+{
+	return __match_glob(pat, str, 0);
+}
+EXPORT_SYMBOL(strglobmatch);