[Python-checkins] [3.6] bpo-31592: Fix an assertion failure in Python parser in case of a bad unicodedata.normalize(). (GH-3767) (#3836)

Sat Sep 30 15:51:44 EDT 2017

https://github.com/python/cpython/commit/a4dfe1c9eaca6faf206f63a178233ffea208c906
commit: a4dfe1c9eaca6faf206f63a178233ffea208c906
branch: 3.6
author: Miss Islington (bot) <31488909+miss-islington at users.noreply.github.com>
committer: Serhiy Storchaka <storchaka at gmail.com>
date: 2017年09月30日T22:51:37+03:00
summary:
[3.6] bpo-31592: Fix an assertion failure in Python parser in case of a bad unicodedata.normalize(). (GH-3767) (#3836)
(cherry picked from commit 7dc46d8cf5854d9f4ce3271b29c21aea4872e8ad)
files:
A Misc/NEWS.d/next/Core and Builtins/2017-09-26-16-05-04.bpo-31592.IFBZj9.rst
M Lib/test/test_ast.py
M Python/ast.c

diff --git a/Lib/test/test_ast.py b/Lib/test/test_ast.py
index 8c62408bd8a..e68d0de2f8b 100644
--- a/Lib/test/test_ast.py
+++ b/Lib/test/test_ast.py
@@ -421,6 +421,16 @@ def test_empty_yield_from(self):
 compile(empty_yield_from, "<test>", "exec")
 self.assertIn("field value is required", str(cm.exception))
 
+ @support.cpython_only
+ def test_issue31592(self):
+ # There shouldn't be an assertion failure in case of a bad
+ # unicodedata.normalize().
+ import unicodedata
+ def bad_normalize(*args):
+ return None
+ with support.swap_attr(unicodedata, 'normalize', bad_normalize):
+ self.assertRaises(TypeError, ast.parse, '\u03D5')
+
 
 class ASTHelpers_Test(unittest.TestCase):
 
diff --git a/Misc/NEWS.d/next/Core and Builtins/2017-09-26-16-05-04.bpo-31592.IFBZj9.rst b/Misc/NEWS.d/next/Core and Builtins/2017-09-26-16-05-04.bpo-31592.IFBZj9.rst
new file mode 100644
index 00000000000..29f3461921e
--- /dev/null
+++ b/Misc/NEWS.d/next/Core and Builtins/2017-09-26-16-05-04.bpo-31592.IFBZj9.rst	
@@ -0,0 +1,2 @@
+Fixed an assertion failure in Python parser in case of a bad `unicodedata.normalize()`.
+Patch by Oren Milman.
diff --git a/Python/ast.c b/Python/ast.c
index aa4acc9b261..2973b9fd866 100644
--- a/Python/ast.c
+++ b/Python/ast.c
@@ -589,7 +589,6 @@ struct compiling {
 PyArena *c_arena; /* Arena for allocating memory. */
 PyObject *c_filename; /* filename */
 PyObject *c_normalize; /* Normalization function from unicodedata. */
- PyObject *c_normalize_args; /* Normalization argument tuple. */
 };
 
 static asdl_seq *seq_for_testlist(struct compiling *, const node *);
@@ -624,12 +623,6 @@ init_normalization(struct compiling *c)
 Py_DECREF(m);
 if (!c->c_normalize)
 return 0;
- c->c_normalize_args = Py_BuildValue("(sN)", "NFKC", Py_None);
- if (!c->c_normalize_args) {
- Py_CLEAR(c->c_normalize);
- return 0;
- }
- PyTuple_SET_ITEM(c->c_normalize_args, 1, NULL);
 return 1;
 }
 
@@ -645,15 +638,29 @@ new_identifier(const char *n, struct compiling *c)
 identifier; if so, normalize to NFKC. */
 if (!PyUnicode_IS_ASCII(id)) {
 PyObject *id2;
+ _Py_IDENTIFIER(NFKC);
 if (!c->c_normalize && !init_normalization(c)) {
 Py_DECREF(id);
 return NULL;
 }
- PyTuple_SET_ITEM(c->c_normalize_args, 1, id);
- id2 = PyObject_Call(c->c_normalize, c->c_normalize_args, NULL);
+ PyObject *form = _PyUnicode_FromId(&PyId_NFKC);
+ if (form == NULL) {
+ Py_DECREF(id);
+ return NULL;
+ }
+ PyObject *args[2] = {form, id};
+ id2 = _PyObject_FastCall(c->c_normalize, args, 2);
 Py_DECREF(id);
 if (!id2)
 return NULL;
+ if (!PyUnicode_Check(id2)) {
+ PyErr_Format(PyExc_TypeError,
+ "unicodedata.normalize() must return a string, not "
+ "%.200s",
+ Py_TYPE(id2)->tp_name);
+ Py_DECREF(id2);
+ return NULL;
+ }
 id = id2;
 }
 PyUnicode_InternInPlace(&id);
@@ -773,7 +780,6 @@ PyAST_FromNodeObject(const node *n, PyCompilerFlags *flags,
 /* borrowed reference */
 c.c_filename = filename;
 c.c_normalize = NULL;
- c.c_normalize_args = NULL;
 
 if (TYPE(n) == encoding_decl)
 n = CHILD(n, 0);
@@ -866,8 +872,6 @@ PyAST_FromNodeObject(const node *n, PyCompilerFlags *flags,
 out:
 if (c.c_normalize) {
 Py_DECREF(c.c_normalize);
- PyTuple_SET_ITEM(c.c_normalize_args, 1, NULL);
- Py_DECREF(c.c_normalize_args);
 }
 return res;
 }
