procyberian/emacs
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Reject invalid 5-byte sequences when detecting UTF-8 encoding

Browse source 
* src/coding.c (detect_coding_utf_8): Reject multibyte sequences
whose leading byte is greater than MAX_MULTIBYTE_LEADING_CODE.
(Bug#31829)
* src/character.h (MAX_MULTIBYTE_LEADING_CODE): Add commentary
about the connection between the value of this macro and MAX_CHAR.
This commit is contained in:
Eli Zaretskii 2018-06-15 17:39:34 +03:00
parent 0d3c35807d
commit 22aa665c9b
2 changed files with 6 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
src/character.h
View file

@ -57,7 +57,8 @@ INLINE_HEADER_BEGIN
/* Minimum leading code of multibyte characters. */
#define MIN_MULTIBYTE_LEADING_CODE 0xC0
/* Maximum leading code of multibyte characters. */
/* Maximum leading code of multibyte characters. Note: this must be
updated if we ever increase MAX_CHAR above. */
#define MAX_MULTIBYTE_LEADING_CODE 0xF8
/* Unicode character values. */

5
src/coding.c
View file

@ -1225,7 +1225,10 @@ detect_coding_utf_8 (struct coding_system *coding,
ONE_MORE_BYTE (c4);
if (c4 < 0 || ! UTF_8_EXTRA_OCTET_P (c4))
break;
if (UTF_8_5_OCTET_LEADING_P (c))
if (UTF_8_5_OCTET_LEADING_P (c)
/* If we ever need to increase MAX_CHAR, the below may need
to be reviewed. */
&& c < MAX_MULTIBYTE_LEADING_CODE)
{
nchars++;
continue;