Mercurial > cpython-lru_cache-weakref
diff Objects/unicodeobject.c @ 33311:59bd121eeeb7 legacy-trunk
PyUnicode_Join(): Bozo Alert. While this is chugging along, it may
need to convert str objects from the iterable to unicode. So, if
someone set the system default encoding to something nasty enough,
the conversion process could mutate the input iterable as a side
effect, and PySequence_Fast doesn't hide that from us if the input was
a list. IOW, can't assume the size of PySequence_Fast's result is
invariant across PyUnicode_FromObject() calls.
| author | Tim Peters <tim.peters@gmail.com> |
|---|---|
| date | Fri, 27 Aug 2004 22:35:44 +0000 |
| parents | 8579859f198c |
| children | 57478a13fbf6 |
line wrap: on
line diff
--- a/Objects/unicodeobject.c Fri Aug 27 21:32:02 2004 +0000 +++ b/Objects/unicodeobject.c Fri Aug 27 22:35:44 2004 +0000 @@ -4000,6 +4000,13 @@ return NULL; } + /* Grrrr. A codec may be invoked to convert str objects to + * Unicode, and so it's possible to call back into Python code + * during PyUnicode_FromObject(), and so it's possible for a sick + * codec to change the size of fseq (if seq is a list). Therefore + * we have to keep refetching the size -- can't assume seqlen + * is invariant. + */ seqlen = PySequence_Fast_GET_SIZE(fseq); /* If empty sequence, return u"". */ if (seqlen == 0) { @@ -4029,6 +4036,8 @@ goto onError; sep = PyUnicode_AS_UNICODE(internal_separator); seplen = PyUnicode_GET_SIZE(internal_separator); + /* In case PyUnicode_FromObject() mutated seq. */ + seqlen = PySequence_Fast_GET_SIZE(fseq); } } @@ -4057,6 +4066,9 @@ goto onError; /* We own a reference to item from here on. */ + /* In case PyUnicode_FromObject() mutated seq. */ + seqlen = PySequence_Fast_GET_SIZE(fseq); + /* Make sure we have enough space for the separator and the item. */ itemlen = PyUnicode_GET_SIZE(item); new_res_used = res_used + itemlen;