diff options
| author | John Mark Bell <jmb@netsurf-browser.org> | 2009-04-02 00:29:07 +0000 |
|---|---|---|
| committer | John Mark Bell <jmb@netsurf-browser.org> | 2009-04-02 00:29:07 +0000 |
| commit | 1a8ae5bb267dc3fcd75856dd62b40c252f444c7c (patch) | |
| tree | 8a610ce7e97a06d294f30fdc2091ee44ddf44115 | |
| parent | 9b8208fca65e6ff52634bb5055d2566e1d2c82c1 (diff) | |
| download | libcss-1a8ae5bb267dc3fcd75856dd62b40c252f444c7c.tar.gz libcss-1a8ae5bb267dc3fcd75856dd62b40c252f444c7c.tar.bz2 | |
Reject any non-ASCII charset detected using the ASCII-compatible detector
svn path=/trunk/libcss/; revision=7027
| -rw-r--r-- | src/charset/detect.c | 18 |
1 files changed, 18 insertions, 0 deletions
diff --git a/src/charset/detect.c b/src/charset/detect.c index 78b4663..2612fa8 100644 --- a/src/charset/detect.c +++ b/src/charset/detect.c @@ -414,6 +414,24 @@ parserutils_error try_ascii_compatible_charset(const uint8_t *data, size_t len, /* Convert to MIB enum */ charset = parserutils_charset_mibenum_from_name( (const char *) start, end - start); + + /* Any non-ASCII compatible charset must be ignored, as + * we've just used an ASCII parser to read it. */ + if (charset == parserutils_charset_mibenum_from_name( + "UTF-32", SLEN("UTF-32")) || + charset == parserutils_charset_mibenum_from_name( + "UTF-32LE", SLEN("UTF-32LE")) || + charset == parserutils_charset_mibenum_from_name( + "UTF-32BE", SLEN("UTF-32BE")) || + charset == parserutils_charset_mibenum_from_name( + "UTF-16", SLEN("UTF-16")) || + charset == parserutils_charset_mibenum_from_name( + "UTF-16LE", SLEN("UTF-16LE")) || + charset == parserutils_charset_mibenum_from_name( + "UTF-16BE", SLEN("UTF-16BE"))) { + + charset = 0; + } } #undef CHARSET |