extras: use libudev code

4 files changed, 179 insertions, 2 deletions

diff --git a/udev/lib/libudev-private.h b/udev/lib/libudev-private.h
index 32837e368d..1f4abc07ba 100644
--- a/udev/lib/libudev-private.h
+++ b/udev/lib/libudev-private.h
@@ -123,4 +123,5 @@ extern size_t util_path_decode(char *s);
 extern void util_remove_trailing_chars(char *path, char c);
 extern size_t util_strlcpy(char *dst, const char *src, size_t size);
 extern size_t util_strlcat(char *dst, const char *src, size_t size);
+extern int util_replace_chars(char *str, const char *white);
 #endif
diff --git a/udev/lib/libudev-util.c b/udev/lib/libudev-util.c
index ae0adf4b88..7aea8c67fb 100644
--- a/udev/lib/libudev-util.c
+++ b/udev/lib/libudev-util.c
@@ -26,6 +26,7 @@
 #include <errno.h>
 #include <string.h>
 #include <dirent.h>
+#include <ctype.h>
 #include <sys/stat.h>
 
 #include "libudev.h"
@@ -255,3 +256,179 @@ size_t util_strlcat(char *dst, const char *src, size_t size)
 	*q = '\0';
 	return bytes;
 }
+
+/* count of characters used to encode one unicode char */
+static int utf8_encoded_expected_len(const char *str)
+{
+	unsigned char c = (unsigned char)str[0];
+
+	if (c < 0x80)
+		return 1;
+	if ((c & 0xe0) == 0xc0)
+		return 2;
+	if ((c & 0xf0) == 0xe0)
+		return 3;
+	if ((c & 0xf8) == 0xf0)
+		return 4;
+	if ((c & 0xfc) == 0xf8)
+		return 5;
+	if ((c & 0xfe) == 0xfc)
+		return 6;
+	return 0;
+}
+
+/* decode one unicode char */
+static int utf8_encoded_to_unichar(const char *str)
+{
+	int unichar;
+	int len;
+	int i;
+
+	len = utf8_encoded_expected_len(str);
+	switch (len) {
+	case 1:
+		return (int)str[0];
+	case 2:
+		unichar = str[0] & 0x1f;
+		break;
+	case 3:
+		unichar = (int)str[0] & 0x0f;
+		break;
+	case 4:
+		unichar = (int)str[0] & 0x07;
+		break;
+	case 5:
+		unichar = (int)str[0] & 0x03;
+		break;
+	case 6:
+		unichar = (int)str[0] & 0x01;
+		break;
+	default:
+		return -1;
+	}
+
+	for (i = 1; i < len; i++) {
+		if (((int)str[i] & 0xc0) != 0x80)
+			return -1;
+		unichar <<= 6;
+		unichar |= (int)str[i] & 0x3f;
+	}
+
+	return unichar;
+}
+
+/* expected size used to encode one unicode char */
+static int utf8_unichar_to_encoded_len(int unichar)
+{
+	if (unichar < 0x80)
+		return 1;
+	if (unichar < 0x800)
+		return 2;
+	if (unichar < 0x10000)
+		return 3;
+	if (unichar < 0x200000)
+		return 4;
+	if (unichar < 0x4000000)
+		return 5;
+	return 6;
+}
+
+/* check if unicode char has a valid numeric range */
+static int utf8_unichar_valid_range(int unichar)
+{
+	if (unichar > 0x10ffff)
+		return 0;
+	if ((unichar & 0xfffff800) == 0xd800)
+		return 0;
+	if ((unichar > 0xfdcf) && (unichar < 0xfdf0))
+		return 0;
+	if ((unichar & 0xffff) == 0xffff)
+		return 0;
+	return 1;
+}
+
+/* validate one encoded unicode char and return its length */
+static int utf8_encoded_valid_unichar(const char *str)
+{
+	int len;
+	int unichar;
+	int i;
+
+	len = utf8_encoded_expected_len(str);
+	if (len == 0)
+		return -1;
+
+	/* ascii is valid */
+	if (len == 1)
+		return 1;
+
+	/* check if expected encoded chars are available */
+	for (i = 0; i < len; i++)
+		if ((str[i] & 0x80) != 0x80)
+			return -1;
+
+	unichar = utf8_encoded_to_unichar(str);
+
+	/* check if encoded length matches encoded value */
+	if (utf8_unichar_to_encoded_len(unichar) != len)
+		return -1;
+
+	/* check if value has valid range */
+	if (!utf8_unichar_valid_range(unichar))
+		return -1;
+
+	return len;
+}
+
+/* allow chars in whitelist, plain ascii, hex-escaping and valid utf8 */
+int util_replace_chars(char *str, const char *white)
+{
+	size_t i = 0;
+	int replaced = 0;
+
+	while (str[i] != '\0') {
+		int len;
+
+		/* accept whitelist */
+		if (white != NULL && strchr(white, str[i]) != NULL) {
+			i++;
+			continue;
+		}
+
+		/* accept plain ascii char */
+		if ((str[i] >= '0' && str[i] <= '9') ||
+		    (str[i] >= 'A' && str[i] <= 'Z') ||
+		    (str[i] >= 'a' && str[i] <= 'z')) {
+			i++;
+			continue;
+		}
+
+		/* accept hex encoding */
+		if (str[i] == '\\' && str[i+1] == 'x') {
+			i += 2;
+			continue;
+		}
+
+		/* accept valid utf8 */
+		len = utf8_encoded_valid_unichar(&str[i]);
+		if (len > 1) {
+			i += len;
+			continue;
+		}
+
+		/* if space is allowed, replace whitespace with ordinary space */
+		if (isspace(str[i]) && strchr(white, ' ') != NULL) {
+			str[i] = ' ';
+			i++;
+			replaced++;
+			continue;
+		}
+
+		/* everything else is replaced with '_' */
+		str[i] = '_';
+		i++;
+		replaced++;
+	}
+
+	return replaced;
+}
diff --git a/udev/udev.h b/udev/udev.h
index 24646ece59..7f5bd10a7c 100644
--- a/udev/udev.h
+++ b/udev/udev.h
@@ -176,7 +176,6 @@ extern int string_is_true(const char *str);
 extern void remove_trailing_chars(char *path, char c);
 extern size_t path_encode(char *s, size_t len);
 extern size_t path_decode(char *s);
-extern int utf8_encoded_valid_unichar(const char *str);
 extern int replace_chars(char *str, const char *white);
 
 /* udev_utils_file.c */
diff --git a/udev/udev_utils_string.c b/udev/udev_utils_string.c
index 0dea6686db..540fb5caaa 100644
--- a/udev/udev_utils_string.c
+++ b/udev/udev_utils_string.c
@@ -185,7 +185,7 @@ static int utf8_unichar_valid_range(int unichar)
 }
 
 /* validate one encoded unicode char and return its length */
-int utf8_encoded_valid_unichar(const char *str)
+static int utf8_encoded_valid_unichar(const char *str)
 {
 	int len;
 	int unichar;