New UTF8 utility functions to parse/generate UTF8 strings.

4 files changed, 159 insertions, 0 deletions

diff --git a/CHANGES b/CHANGES
index f2d1cd8eb3..c246ca9158 100644
--- a/CHANGES
+++ b/CHANGES
@@ -4,6 +4,10 @@
 
  Changes between 0.9.4 and 0.9.5  [xx XXX 1999]
 
+  *) New functions UTF8_getc() and UTF8_putc() that parse and generate
+     UTF8 strings a character at a time.
+     [Steve Henson]
+
   *) Use client_version from client hello to select the protocol
      (s23_srvr.c) and for RSA client key exchange verification
      (s3_srvr.c), as required by the SSL 3.0/TLS 1.0 specifications.
diff --git a/crypto/asn1/a_utf8.c b/crypto/asn1/a_utf8.c
index 4a8a92e9e4..6179bd09a3 100644
--- a/crypto/asn1/a_utf8.c
+++ b/crypto/asn1/a_utf8.c
@@ -81,3 +81,152 @@ ASN1_UTF8STRING *d2i_ASN1_UTF8STRING(ASN1_UTF8STRING **a, unsigned char **pp,
 	return(ret);
 	}
 
+
+/* UTF8 utilities */
+
+/* This parses a UTF8 string one character at a time. It is passed a pointer
+ * to the string and the length of the string. It sets 'value' to the value of
+ * the current character. It returns the number of characters read or a
+ * negative error code:
+ * -1 = string too short
+ * -2 = illegal character
+ * -3 = subsequent characters not of the form 10xxxxxx
+ * -4 = character encoded incorrectly (not minimal length).
+ */
+
+int UTF8_getc(unsigned char *str, int len, unsigned long *val)
+{
+	unsigned char *p;
+	unsigned long value;
+	int ret;
+	if(len <= 0) return 0;
+	p = str;
+
+	/* Check syntax and work out the encoded value (if correct) */
+	if((*p & 0x80) == 0) {
+		value = *p++ & 0x7f;
+		ret = 1;
+	} else if((*p & 0xe0) == 0xc0) {
+		if(len < 2) return -1;
+		if((p[1] & 0xc0) != 0x80) return -3;
+		value = (*p++ & 0x1f) << 6;
+		value |= *p++ & 0x3f;
+		if(value < 0x80) return -4;
+		ret = 2;
+	} else if((*p & 0xf0) == 0xe0) {
+		if(len < 3) return -1;
+		if( ((p[1] & 0xc0) != 0x80)
+		   || ((p[2] & 0xc0) != 0x80) ) return -3;
+		value = (*p++ & 0xf) << 12;
+		value |= (*p++ & 0x3f) << 6;
+		value |= *p++ & 0x3f;
+		if(value < 0x800) return -4;
+		ret = 3;
+	} else if((*p & 0xf8) == 0xf0) {
+		if(len < 4) return -1;
+		if( ((p[1] & 0xc0) != 0x80)
+		   || ((p[2] & 0xc0) != 0x80) 
+		   || ((p[3] & 0xc0) != 0x80) ) return -3;
+		value = (*p++ & 0x7) << 18;
+		value |= (*p++ & 0x3f) << 12;
+		value |= (*p++ & 0x3f) << 6;
+		value |= *p++ & 0x3f;
+		if(value < 0x10000) return -4;
+		ret = 4;
+	} else if((*p & 0xfc) == 0xf8) {
+		if(len < 5) return -1;
+		if( ((p[1] & 0xc0) != 0x80)
+		   || ((p[2] & 0xc0) != 0x80) 
+		   || ((p[3] & 0xc0) != 0x80) 
+		   || ((p[4] & 0xc0) != 0x80) ) return -3;
+		value = (*p++ & 0x3) << 24;
+		value |= (*p++ & 0x3f) << 18;
+		value |= (*p++ & 0x3f) << 12;
+		value |= (*p++ & 0x3f) << 6;
+		value |= *p++ & 0x3f;
+		if(value < 0x200000) return -4;
+		ret = 5;
+	} else if((*p & 0xfe) == 0xfc) {
+		if(len < 6) return -1;
+		if( ((p[1] & 0xc0) != 0x80)
+		   || ((p[2] & 0xc0) != 0x80) 
+		   || ((p[3] & 0xc0) != 0x80) 
+		   || ((p[4] & 0xc0) != 0x80) 
+		   || ((p[5] & 0xc0) != 0x80) ) return -3;
+		value = (*p++ & 0x1) << 30;
+		value |= (*p++ & 0x3f) << 24;
+		value |= (*p++ & 0x3f) << 18;
+		value |= (*p++ & 0x3f) << 12;
+		value |= (*p++ & 0x3f) << 6;
+		value |= *p++ & 0x3f;
+		if(value < 0x4000000) return -4;
+		ret = 6;
+	} else return -2;
+	*val = value;
+	return ret;
+}
+
+/* This takes a character 'value' and writes the UTF8 encoded value in
+ * 'str' where 'str' is a buffer containing 'len' characters. Returns
+ * the number of characters written or -1 if 'len' is too small. 'str' can
+ * be set to NULL in which case it just returns the number of characters.
+ * It will need at most 6 characters.
+ */
+
+int UTF8_putc(unsigned char *str, int len, unsigned long value)
+{
+	if(!str) len = 6;	/* Maximum we will need */
+	else if(len <= 0) return -1;
+	if(value < 0x80) {
+		if(str) *str = value;
+		return 1;
+	}
+	if(value < 0x800) {
+		if(len < 2) return -1;
+		if(str) {
+			*str++ = ((value >> 6) & 0x1f) | 0xc0;
+			*str = (value & 0x3f) | 0x80;
+		}
+		return 2;
+	}
+	if(value < 0x10000) {
+		if(len < 3) return -1;
+		if(str) {
+			*str++ = ((value >> 12) & 0xf) | 0xe0;
+			*str++ = ((value >> 6) & 0x3f) | 0x80;
+			*str = (value & 0x3f) | 0x80;
+		}
+		return 3;
+	}
+	if(value < 0x200000) {
+		if(len < 4) return -1;
+		if(str) {
+			*str++ = ((value >> 18) & 0x7) | 0xf0;
+			*str++ = ((value >> 12) & 0x3f) | 0x80;
+			*str++ = ((value >> 6) & 0x3f) | 0x80;
+			*str = (value & 0x3f) | 0x80;
+		}
+		return 4;
+	}
+	if(value < 0x4000000) {
+		if(len < 5) return -1;
+		if(str) {
+			*str++ = ((value >> 24) & 0x3) | 0xf8;
+			*str++ = ((value >> 18) & 0x3f) | 0x80;
+			*str++ = ((value >> 12) & 0x3f) | 0x80;
+			*str++ = ((value >> 6) & 0x3f) | 0x80;
+			*str = (value & 0x3f) | 0x80;
+		}
+		return 5;
+	}
+	if(len < 6) return -1;
+	if(str) {
+		*str++ = ((value >> 30) & 0x1) | 0xfc;
+		*str++ = ((value >> 24) & 0x3f) | 0x80;
+		*str++ = ((value >> 18) & 0x3f) | 0x80;
+		*str++ = ((value >> 12) & 0x3f) | 0x80;
+		*str++ = ((value >> 6) & 0x3f) | 0x80;
+		*str = (value & 0x3f) | 0x80;
+	}
+	return 6;
+}
diff --git a/crypto/asn1/asn1.h b/crypto/asn1/asn1.h
index 5c2d8999bc..0ef18ac41c 100644
--- a/crypto/asn1/asn1.h
+++ b/crypto/asn1/asn1.h
@@ -553,6 +553,10 @@ int i2d_ASN1_BMPSTRING(ASN1_BMPSTRING *a, unsigned char **pp);
 ASN1_BMPSTRING *d2i_ASN1_BMPSTRING(ASN1_BMPSTRING **a, unsigned char **pp,
 	long length);
 
+
+int UTF8_getc(unsigned char *str, int len, unsigned long *val);
+int UTF8_putc(unsigned char *str, int len, unsigned long value);
+
 int i2d_ASN1_PRINTABLE(ASN1_STRING *a,unsigned char **pp);
 ASN1_STRING *d2i_ASN1_PRINTABLE(ASN1_STRING **a,
 	unsigned char **pp, long l);
diff --git a/util/libeay.num b/util/libeay.num
index 8e7e63d5c1..d28fa26550 100755
--- a/util/libeay.num
+++ b/util/libeay.num
@@ -1874,3 +1874,5 @@ NETSCAPE_SPKI_set_pubkey                1898
 NETSCAPE_SPKI_b64_encode                1899
 NETSCAPE_SPKI_get_pubkey                1900
 NETSCAPE_SPKI_b64_decode                1901
+UTF8_putc                               1902
+UTF8_getc                               1903