mirrors/cpython
mirrors/cpython
0
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2024-11-24 00:38:00 +01:00
Code Issues Releases Activity

gh-124064: Fix -Wconversion warnings in Parser/string_parser.c (#124204)

Browse Source 
Fix integer overflow check in decode_unicode_with_escapes(): use
PY_SSIZE_T_MAX instead of SIZE_MAX.
This commit is contained in:
Victor Stinner 2024-09-18 19:10:56 +02:00 committed by GitHub
parent 5cd50cb6eb
commit f9fa6ba4f8
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 8 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
Parser/string_parser.c
View File

@ -18,7 +18,7 @@ warn_invalid_escape_sequence(Parser *p, const char *first_invalid_escape, Token
// to avoid showing the warning twice.
return 0;
}
unsigned char c = *first_invalid_escape;
unsigned char c = (unsigned char)*first_invalid_escape;
if ((t->type == FSTRING_MIDDLE || t->type == FSTRING_END) && (c == '{' || c == '}')) {
// in this case the tokenizer has already emitted a warning,
// see Parser/tokenizer/helpers.c:warn_invalid_escape_sequence
@ -90,12 +90,12 @@ decode_unicode_with_escapes(Parser *parser, const char *s, size_t len, Token *t)
const char *end;
/* check for integer overflow */
if (len > SIZE_MAX / 6) {
if (len > (size_t)PY_SSIZE_T_MAX / 6) {
return NULL;
}
/* "ä" (2 bytes) may become "\U000000E4" (10 bytes), or 1:5
"\ä" (3 bytes) may become "\u005c\U000000E4" (16 bytes), or ~1:6 */
u = PyBytes_FromStringAndSize((char *)NULL, len * 6);
u = PyBytes_FromStringAndSize((char *)NULL, (Py_ssize_t)len * 6);
if (u == NULL) {
return NULL;
}
@ -142,11 +142,11 @@ decode_unicode_with_escapes(Parser *parser, const char *s, size_t len, Token *t)
*p++ = *s++;
}
}
len = p - buf;
len = (size_t)(p - buf);
s = buf;
const char *first_invalid_escape;
v = _PyUnicode_DecodeUnicodeEscapeInternal(s, len, NULL, NULL, &first_invalid_escape);
v = _PyUnicode_DecodeUnicodeEscapeInternal(s, (Py_ssize_t)len, NULL, NULL, &first_invalid_escape);
// HACK: later we can simply pass the line no, since we don't preserve the tokens
// when we are decoding the string but we preserve the line numbers.
@ -185,7 +185,7 @@ PyObject *
_PyPegen_decode_string(Parser *p, int raw, const char *s, size_t len, Token *t)
{
if (raw) {
return PyUnicode_DecodeUTF8Stateful(s, len, NULL, NULL);
return PyUnicode_DecodeUTF8Stateful(s, (Py_ssize_t)len, NULL, NULL);
}
return decode_unicode_with_escapes(p, s, len, t);
}
@ -274,9 +274,9 @@ _PyPegen_parse_string(Parser *p, Token *t)
}
}
if (rawmode) {
return PyBytes_FromStringAndSize(s, len);
return PyBytes_FromStringAndSize(s, (Py_ssize_t)len);
}
return decode_bytes_with_escapes(p, s, len, t);
return decode_bytes_with_escapes(p, s, (Py_ssize_t)len, t);
}
return _PyPegen_decode_string(p, rawmode, s, len, t);
}