[pypy-commit] pypy py3.5: Detect null bytes in unicode or byte strings for the OS functions

[arigo]

Author: Armin Rigo <arigo at tunes.org>
Branch: py3.5
Changeset: r89506:6e3116ffd73c
Date: 2017-01-12 15:17 +0100
http://bitbucket.org/pypy/pypy/changeset/6e3116ffd73c/

Log:	Detect null bytes in unicode or byte strings for the OS functions

diff --git a/pypy/interpreter/test/test_fsencode.py b/pypy/interpreter/test/test_fsencode.py
--- a/pypy/interpreter/test/test_fsencode.py
+++ b/pypy/interpreter/test/test_fsencode.py
@@ -78,3 +78,8 @@
 
             assert space.fsencode_w(w_enc) == space.bytes_w(w_enc)
             assert space.eq_w(space.wrap_fsdecoded(space.bytes_w(w_enc)), w_st2)
+
+    def test_null_byte(self):
+        space = self.space
+        w_u = space.newunicode(u'abc\x00def')
+        space.raises_w(space.w_ValueError, space.fsencode, w_u)
diff --git a/pypy/interpreter/unicodehelper.py b/pypy/interpreter/unicodehelper.py
--- a/pypy/interpreter/unicodehelper.py
+++ b/pypy/interpreter/unicodehelper.py
@@ -1,5 +1,5 @@
 import sys
-from pypy.interpreter.error import OperationError
+from pypy.interpreter.error import OperationError, oefmt
 from rpython.rlib.objectmodel import specialize
 from rpython.rlib import runicode
 from pypy.module._codecs import interp_codecs
@@ -104,6 +104,8 @@
         from pypy.module._codecs.locale import (
             unicode_encode_locale_surrogateescape)
         uni = space.unicode_w(w_uni)
+        if u'\x00' in uni:
+            raise oefmt(space.w_ValueError, "embedded null character")
         bytes = unicode_encode_locale_surrogateescape(
             uni, errorhandler=encode_error_handler(space))
     else:
diff --git a/pypy/module/posix/interp_posix.py b/pypy/module/posix/interp_posix.py
--- a/pypy/module/posix/interp_posix.py
+++ b/pypy/module/posix/interp_posix.py
@@ -148,11 +148,15 @@
     try:
         path_b = space.fsencode_w(w_value)
         return Path(-1, path_b, None, w_value)
-    except OperationError:
+    except OperationError as e:
+        if not e.match(space, space.w_TypeError):
+            raise
         if allow_fd:
             fd = unwrap_fd(space, w_value, "string, bytes or integer")
             return Path(fd, None, None, w_value)
-    raise oefmt(space.w_TypeError, "illegal type for path parameter")
+    raise oefmt(space.w_TypeError,
+                "illegal type for path parameter (expected "
+                "string or bytes, got %T)", w_value)
 
 class _PathOrFd(Unwrapper):
     def unwrap(self, space, w_value):
diff --git a/pypy/module/posix/test/test_posix2.py b/pypy/module/posix/test/test_posix2.py
--- a/pypy/module/posix/test/test_posix2.py
+++ b/pypy/module/posix/test/test_posix2.py
@@ -202,6 +202,8 @@
         excinfo = raises(TypeError, self.posix.stat, 2.)
         assert "should be string, bytes or integer, not float" in str(excinfo.value)
         raises(ValueError, self.posix.stat, -1)
+        raises(ValueError, self.posix.stat, b"abc\x00def")
+        raises(ValueError, self.posix.stat, u"abc\x00def")
 
     if hasattr(__import__(os.name), "statvfs"):
         def test_statvfs(self):