edg-l/ddnet
1
0
Fork 0
mirror of https://github.com/ddnet/ddnet.git synced 2024-11-10 10:08:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

Error out on overlong UTF-8 sequences

Browse source 
This not only makes `str_utf8_decode` a compliant parser, but also fixes an
issue where overlong encodings led to undetected line breaks.

Implemented as per https://encoding.spec.whatwg.org/#utf-8-decoder.

Audited the following files, checking whether the new semantics work with the
existing code:
- src/base/system.c
- src/engine/client/text.cpp
- src/game/client/components/chat.cpp
- src/game/server/gamecontext.cpp

They all check whether the string is at its end separately.
This commit is contained in:
heinrich5991 2015-06-30 21:09:43 +02:00
parent ce465b584c
commit 8ccfd0188d
2 changed files with 69 additions and 43 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

103
src/base/system.c
View file

@ -2274,54 +2274,79 @@ int str_utf8_encode(char *ptr, int chr)
return 0;
}
static unsigned char str_byte_next(const char **ptr)
{
unsigned char byte = **ptr;
(*ptr)++;
return byte;
}
static void str_byte_rewind(const char **ptr)
{
(*ptr)--;
}
int str_utf8_decode(const char **ptr)
{
const char *buf = *ptr;
int ch = 0;
do
// As per https://encoding.spec.whatwg.org/#utf-8-decoder.
unsigned char utf8_lower_boundary = 0x80;
unsigned char utf8_upper_boundary = 0xBF;
int utf8_code_point = 0;
int utf8_bytes_seen = 0;
int utf8_bytes_needed = 0;
while(1)
{
if((*buf&0x80) == 0x0) /* 0xxxxxxx */
unsigned char byte = str_byte_next(ptr);
if(utf8_bytes_needed == 0)
{
ch = *buf;
buf++;
if(0x00 <= byte && byte <= 0x7F)
{
return byte;
}
else if(0x80 <= byte && byte <= 0xDF)
{
utf8_bytes_needed = 1;
utf8_code_point = byte - 0xC0;
}
else if(0xE0 <= byte && byte <= 0xEF)
{
if(byte == 0xE0) utf8_lower_boundary = 0xA0;
if(byte == 0xED) utf8_upper_boundary = 0x9F;
utf8_bytes_needed = 2;
utf8_code_point = byte - 0xE0;
}
else if(0xF0 <= byte && byte <= 0xF4)
{
if(byte == 0xF0) utf8_lower_boundary = 0x90;
if(byte == 0xF4) utf8_upper_boundary = 0x8F;
utf8_bytes_needed = 3;
utf8_code_point = byte - 0xF0;
}
else
{
return -1; // Error.
}
utf8_code_point = utf8_code_point << (6 * utf8_bytes_needed);
continue;
}
else if((*buf&0xE0) == 0xC0) /* 110xxxxx */
if(!(utf8_lower_boundary <= byte && byte <= utf8_upper_boundary))
{
ch = (*buf++ & 0x3F) << 6; if(!(*buf)) break;
ch += (*buf++ & 0x3F);
if(ch == 0) ch = -1;
// Resetting variables not necessary, will be done when
// the function is called again.
str_byte_rewind(ptr);
return -1;
}
else if((*buf & 0xF0) == 0xE0) /* 1110xxxx */
utf8_lower_boundary = 0x80;
utf8_upper_boundary = 0xBF;
utf8_bytes_seen += 1;
utf8_code_point = utf8_code_point + ((byte - 0x80) << (6 * (utf8_bytes_needed - utf8_bytes_seen)));
if(utf8_bytes_seen != utf8_bytes_needed)
{
ch = (*buf++ & 0x1F) << 12; if(!(*buf)) break;
ch += (*buf++ & 0x3F) << 6; if(!(*buf)) break;
ch += (*buf++ & 0x3F);
if(ch == 0) ch = -1;
continue;
}
else if((*buf & 0xF8) == 0xF0) /* 11110xxx */
{
ch = (*buf++ & 0x0F) << 18; if(!(*buf)) break;
ch += (*buf++ & 0x3F) << 12; if(!(*buf)) break;
ch += (*buf++ & 0x3F) << 6; if(!(*buf)) break;
ch += (*buf++ & 0x3F);
if(ch == 0) ch = -1;
}
else
{
/* invalid */
buf++;
break;
}
*ptr = buf;
return ch;
} while(0);
/* out of bounds */
*ptr = buf;
return -1;
// Resetting variables not necessary, see above.
return utf8_code_point;
}
}
int str_utf8_check(const char *str)

9
src/base/system.h
View file

@ -1281,16 +1281,17 @@ int str_utf8_forward(const char *str, int cursor);
/*
Function: str_utf8_decode
Decodes an utf8 character
Decodes a utf8 codepoint
Parameters:
ptr - pointer to an utf8 string. this pointer will be moved forward
ptr - Pointer to a utf8 string. This pointer will be moved forward.
Returns:
Unicode value for the character. -1 for invalid characters and 0 for end of string.
The Unicode codepoint. -1 for invalid input and 0 for end of string.
Remarks:
- This function will also move the pointer forward.
- You may call this function again after an error occured.
*/
int str_utf8_decode(const char **ptr);
@ -1299,7 +1300,7 @@ int str_utf8_decode(const char **ptr);
Encode an utf8 character
Parameters:
ptr - Pointer to a buffer that should recive the data. Should be able to hold at least 4 bytes.
ptr - Pointer to a buffer that should receive the data. Should be able to hold at least 4 bytes.
Returns:
Number of bytes put into the buffer.