[Python-checkins] cpython (merge 3.5 -> default): As per further discussion, re-enable the typeobject.c guard for picklability.
barry.warsaw
python-checkins at python.org
Mon Jan 11 15:53:57 EST 2016
https://hg.python.org/cpython/rev/f1cfac294af2
changeset: 99857:f1cfac294af2
parent: 99849:927fd0e14d49
parent: 99856:54dd5c105334
user: Barry Warsaw <barry at python.org>
date: Mon Jan 11 15:51:27 2016 -0500
summary:
As per further discussion, re-enable the typeobject.c guard for picklability.
upstreams such as Cython will have to adjust.
files:
Misc/NEWS | 3 ---
Objects/typeobject.c | 9 ---------
2 files changed, 0 insertions(+), 12 deletions(-)
diff --git a/Misc/NEWS b/Misc/NEWS
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -10,9 +10,6 @@
Core and Builtins
-----------------
-- Issue #22995: [UPDATE] Comment out the one of the pickleability tests in
- _PyObject_GetState() due to regressions observed in Cython-based projects.
-
- Issue #25961: Disallowed null characters in the type name.
- Issue #25973: Fix segfault when an invalid nonlocal statement binds a name
diff --git a/Objects/typeobject.c b/Objects/typeobject.c
--- a/Objects/typeobject.c
+++ b/Objects/typeobject.c
@@ -3872,14 +3872,6 @@
}
assert(slotnames == Py_None || PyList_Check(slotnames));
-#if 0
- /* 2016-01-11 barry - This clause breaks at least three packages which
- rely on Cython: kivy, pysam, and s3ql. Cython may be doing
- something funny under the hood, but as this is clearly a regression
- and the rationale for this prohibition is suspect, I am commenting
- this out. Perhaps it should just be removed. See issue #22995 for
- details.
- */
if (required) {
Py_ssize_t basicsize = PyBaseObject_Type.tp_basicsize;
if (obj->ob_type->tp_dictoffset)
@@ -3897,7 +3889,6 @@
return NULL;
}
}
-#endif
if (slotnames != Py_None && Py_SIZE(slotnames) > 0) {
PyObject *slots;
--
Repository URL: https://hg.python.org/cpython
More information about the Python-checkins
mailing list