From 91879ab8ea3ccb611a2b46b81a9984f451cb0669 Mon Sep 17 00:00:00 2001 From: Tim Peters Date: Fri, 27 Aug 2004 22:35:44 +0000 Subject: 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. --- Objects/unicodeobject.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c index 668d239..f78788e 100644 --- a/Objects/unicodeobject.c +++ b/Objects/unicodeobject.c @@ -4000,6 +4000,13 @@ PyUnicode_Join(PyObject *separator, PyObject *seq) 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 @@ PyUnicode_Join(PyObject *separator, PyObject *seq) 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 @@ PyUnicode_Join(PyObject *separator, PyObject *seq) 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; -- cgit v0.12