Author: Matti Picus <[email protected]>
Branch: unicode-utf8-py3
Changeset: r94950:9c9fca815956
Date: 2018-08-05 10:36 -0700
http://bitbucket.org/pypy/pypy/changeset/9c9fca815956/
Log: use a signature to find where len < 0, fix
diff --git a/pypy/module/_pypyjson/interp_decoder.py
b/pypy/module/_pypyjson/interp_decoder.py
--- a/pypy/module/_pypyjson/interp_decoder.py
+++ b/pypy/module/_pypyjson/interp_decoder.py
@@ -307,6 +307,7 @@
# ascii only, fast path (ascii is a strict subset of
# latin1, and we already checked that all the chars are <
# 128)
+ assert end >= start
return self.space.newutf8(self.getslice(start, end),
end - start)
diff --git a/pypy/module/cpyext/longobject.py b/pypy/module/cpyext/longobject.py
--- a/pypy/module/cpyext/longobject.py
+++ b/pypy/module/cpyext/longobject.py
@@ -200,6 +200,8 @@
string, length gives the number of characters, and base is the radix
for the conversion. The radix must be in the range [2, 36]; if it is
out of range, ValueError will be raised."""
+ if length < 0:
+ length = 0
w_value = space.newutf8(wcharpsize2utf8(space, u, length), length)
return PyLong_FromUnicodeObject(space, w_value, base)
diff --git a/pypy/module/cpyext/unicodeobject.py
b/pypy/module/cpyext/unicodeobject.py
--- a/pypy/module/cpyext/unicodeobject.py
+++ b/pypy/module/cpyext/unicodeobject.py
@@ -472,6 +472,8 @@
object. If the buffer is not NULL, the return value might be a shared
object.
Therefore, modification of the resulting Unicode object is only allowed
when u
is NULL."""
+ if length < 0:
+ length = 0
if wchar_p:
s = wcharpsize2utf8(space, wchar_p, length)
return make_ref(space, space.newutf8(s, length))
@@ -755,6 +757,8 @@
"""Encode the Py_UNICODE buffer of the given size and return a
Python string object. Return NULL if an exception was raised
by the codec."""
+ if size < 0:
+ size = 0
u = wcharpsize2utf8(space, s, size)
w_u = space.newutf8(u, size)
if errors:
diff --git a/pypy/objspace/std/objspace.py b/pypy/objspace/std/objspace.py
--- a/pypy/objspace/std/objspace.py
+++ b/pypy/objspace/std/objspace.py
@@ -13,7 +13,8 @@
from rpython.rlib.rarithmetic import base_int, widen, is_valid_int
from rpython.rlib.objectmodel import import_from_mixin, we_are_translated
from rpython.rlib.objectmodel import not_rpython
-from rpython.rlib import jit, rutf8
+from rpython.rlib import jit, rutf8, types
+from rpython.rlib.signature import signature, finishsigs
# Object imports
from pypy.objspace.std.boolobject import W_BoolObject
@@ -36,7 +37,7 @@
from pypy.objspace.std.typeobject import W_TypeObject, TypeCache
from pypy.objspace.std.unicodeobject import W_UnicodeObject
-
+@finishsigs
class StdObjSpace(ObjSpace):
"""The standard object space, implementing a general-purpose object
library in Restricted Python."""
@@ -381,10 +382,9 @@
@specialize.argtype(1)
def newtext(self, s):
- assert not isinstance(s, unicode)
- #if isinstance(s, unicode):
- #s, lgt = s.encode('utf8'), len(s)
- if isinstance(s, str):
+ if isinstance(s, unicode):
+ s, lgt = s.encode('utf8'), len(s)
+ elif isinstance(s, str):
s, lgt, codepoints = decode_utf8sp(self, s)
elif isinstance(s, tuple):
# result of decode_utf8
@@ -400,8 +400,9 @@
return self.w_None
return self.newtext(s)
+ # XXX find where length is annotated as negative int
+ @signature(types.any(), types.str(), types.int_nonneg(),
returns=types.any())
def newutf8(self, utf8s, length):
- assert length >= 0
assert isinstance(utf8s, str)
return W_UnicodeObject(utf8s, length)
diff --git a/rpython/rlib/rutf8.py b/rpython/rlib/rutf8.py
--- a/rpython/rlib/rutf8.py
+++ b/rpython/rlib/rutf8.py
@@ -28,7 +28,7 @@
# we need a way to accept both r_uint and int(nonneg=True)
-@signature(types.int_nonneg(), types.bool(), returns=types.str())
+#@signature(types.int_nonneg(), types.bool(), returns=types.str())
def unichr_as_utf8(code, allow_surrogates=False):
"""Encode code (numeric value) as utf8 encoded string
"""
diff --git a/rpython/rlib/signature.py b/rpython/rlib/signature.py
--- a/rpython/rlib/signature.py
+++ b/rpython/rlib/signature.py
@@ -9,7 +9,7 @@
def foo(...)
The arguments paramNtype and returntype should be instances
- of the classes in rpython.annotator.types.
+ of the classes in rpython.rlib.types.
"""
returntype = kwargs.pop('returns', None)
if returntype is None:
_______________________________________________
pypy-commit mailing list
[email protected]
https://mail.python.org/mailman/listinfo/pypy-commit
