| Mon, 29 Oct 2001 14:33:44 +0000 |
Guido van Rossum |
When overriding __str__ or __repr__, set the tp_print slot to NULL.
legacy-trunk
|
| Sat, 27 Oct 2001 19:37:48 +0000 |
Tim Peters |
SF bug #475327: type() produces incorrect error msg
legacy-trunk
|
| Fri, 26 Oct 2001 04:26:12 +0000 |
Guido van Rossum |
Allow assignment to newinstance.__dict__.
legacy-trunk
|
| Mon, 22 Oct 2001 00:43:43 +0000 |
Guido van Rossum |
Methods of built-in types now properly check for keyword arguments
legacy-trunk
|
| Sun, 21 Oct 2001 22:28:58 +0000 |
Neil Schemenauer |
Add missing "static" declarations (found by "make smelly").
legacy-trunk
|
| Sun, 21 Oct 2001 00:44:31 +0000 |
Guido van Rossum |
Big internal change that should have no external effects: unify the
legacy-trunk
|
| Thu, 18 Oct 2001 15:49:21 +0000 |
Guido van Rossum |
Fix SF bug #472234: type(obj) calls type->tp_init (Roeland Rengelink)
legacy-trunk
|
| Wed, 17 Oct 2001 13:59:09 +0000 |
Guido van Rossum |
Remove a bunch of stuff that's no longer needed now that update_slot()
legacy-trunk
|
| Wed, 17 Oct 2001 07:15:43 +0000 |
Guido van Rossum |
slot_sq_item(): ensure that self is an instance of the wrapper's
legacy-trunk
|
| Tue, 16 Oct 2001 17:00:48 +0000 |
Guido van Rossum |
Refactored the update_slot() code a bit to be hopefully slightly more
legacy-trunk
|
| Mon, 15 Oct 2001 22:03:32 +0000 |
Guido van Rossum |
Get rid of __defined__ and tp_defined -- there's no need to
legacy-trunk
|
| Mon, 15 Oct 2001 21:05:10 +0000 |
Guido van Rossum |
Completely get rid of __dynamic__ and the corresponding
legacy-trunk
|
| Mon, 15 Oct 2001 19:44:24 +0000 |
Guido van Rossum |
Add (void *) casts to solve some problems on HP-UX 11.0, as discussed
legacy-trunk
|
| Sat, 13 Oct 2001 20:02:41 +0000 |
Guido van Rossum |
Redid the slot computation. The initial slot assignments are now done
legacy-trunk
|
| Fri, 12 Oct 2001 14:13:21 +0000 |
Guido van Rossum |
Now that COPYBUF is a new local macro, add #undef COPYBUF.
legacy-trunk
|
| Fri, 12 Oct 2001 02:38:24 +0000 |
Tim Peters |
SF bug [#470040] ParseTuple t# vs subclasses.
legacy-trunk
|
| Thu, 11 Oct 2001 18:33:53 +0000 |
Guido van Rossum |
Another step in the right direction: when a new class's attribute
legacy-trunk
|
| Tue, 09 Oct 2001 20:17:57 +0000 |
Guido van Rossum |
The slot definition table entry for mp_getitem had a bogus wrapper
legacy-trunk
|
| Tue, 09 Oct 2001 19:39:46 +0000 |
Guido van Rossum |
Halfway checkin. This is still messy, but it's beginning to address
legacy-trunk
|
| Tue, 09 Oct 2001 11:07:24 +0000 |
Guido van Rossum |
It turned out not so difficult to support old-style numbers (those
legacy-trunk
|
| Mon, 08 Oct 2001 16:49:26 +0000 |
Tim Peters |
type_subclasses(): debug build was broken due to typo in new assert().
legacy-trunk
|
| Mon, 08 Oct 2001 15:18:27 +0000 |
Guido van Rossum |
Keep track of a type's subclasses (subtypes), in tp_subclasses, which
legacy-trunk
|
| Sun, 07 Oct 2001 03:54:51 +0000 |
Tim Peters |
Guido suggests, and I agree, to insist that SIZEOF_VOID_P be a power of 2.
legacy-trunk
|
| Sat, 06 Oct 2001 21:27:34 +0000 |
Tim Peters |
_PyObject_VAR_SIZE: always round up to a multiple-of-pointer-size value.
legacy-trunk
|
| Sat, 06 Oct 2001 19:04:01 +0000 |
Tim Peters |
Repaired the debug Windows deaths in test_descr, by allocating enough
legacy-trunk
|
| Fri, 05 Oct 2001 20:51:39 +0000 |
Guido van Rossum |
Enable GC for new-style instances. This touches lots of files, since
legacy-trunk
|
| Thu, 04 Oct 2001 19:46:06 +0000 |
Guido van Rossum |
Make new classes dynamic by default.
legacy-trunk
|
| Thu, 04 Oct 2001 05:43:02 +0000 |
Tim Peters |
type_new(): cast PyObject_MALLOC's result to char*, for clarity.
legacy-trunk
|
| Thu, 04 Oct 2001 05:27:00 +0000 |
Tim Peters |
SF bug [#467331] ClassType.__doc__ always None.
legacy-trunk
|
| Wed, 03 Oct 2001 13:58:35 +0000 |
Guido van Rossum |
typeobject.c, slot_tp_gettattr_hook(): fix the speedup hack -- the
legacy-trunk
|
| Wed, 03 Oct 2001 12:09:30 +0000 |
Guido van Rossum |
*EXPERIMENTAL* speedup of slot_sq_item. This sped up the following
legacy-trunk
|
| Wed, 03 Oct 2001 00:50:18 +0000 |
Guido van Rossum |
call_method(), call_maybe(): fix a performance bug: the argument
legacy-trunk
|
| Tue, 02 Oct 2001 21:24:57 +0000 |
Guido van Rossum |
Add Garbage Collection support to new-style classes (not yet to their
legacy-trunk
|
| Mon, 01 Oct 2001 17:18:22 +0000 |
Guido van Rossum |
Miscellaneous code fiddling:
legacy-trunk
|
| Mon, 01 Oct 2001 16:42:49 +0000 |
Guido van Rossum |
slot_sq_length(): squash a leak.
legacy-trunk
|
| Mon, 01 Oct 2001 15:55:28 +0000 |
Guido van Rossum |
slot_tp_new(): newargs was leaking.
legacy-trunk
|
| Mon, 01 Oct 2001 13:17:24 +0000 |
Guido van Rossum |
Fix typo found by doerwalter.
legacy-trunk
|
| Sun, 30 Sep 2001 05:58:42 +0000 |
Tim Peters |
SF bug [#466173] unpack TypeError unclear
legacy-trunk
|
| Fri, 28 Sep 2001 22:58:52 +0000 |
Guido van Rossum |
Ouch. The wrapper for __rpow__ was the same as for __pow__, resulting
legacy-trunk
|
| Thu, 27 Sep 2001 20:30:07 +0000 |
Guido van Rossum |
Merge branch changes (coercion, rich comparisons) into trunk.
legacy-trunk
|
| Tue, 25 Sep 2001 21:16:33 +0000 |
Guido van Rossum |
add_operators(): the __floordiv__ and __truediv__ descriptors (and
legacy-trunk
|
| Tue, 25 Sep 2001 16:25:58 +0000 |
Guido van Rossum |
- Provisional support for pickling new-style objects. (*)
legacy-trunk
|
| Tue, 25 Sep 2001 03:56:29 +0000 |
Guido van Rossum |
Change repr() of a new-style class to say <class 'ClassName'> rather
legacy-trunk
|
| Tue, 25 Sep 2001 03:43:42 +0000 |
Guido van Rossum |
Make __class__ assignment possible, when the object structures are the
legacy-trunk
|
| Mon, 24 Sep 2001 18:47:40 +0000 |
Guido van Rossum |
Another comparison patch-up: comparing a type with a dynamic metatype
legacy-trunk
|
| Mon, 24 Sep 2001 16:03:59 +0000 |
Guido van Rossum |
Fix the baffler that Tim reported: sometimes the repr() of an object
legacy-trunk
|
| Fri, 21 Sep 2001 21:24:49 +0000 |
Guido van Rossum |
Add the __getattr__ hook back. The rules are now:
legacy-trunk
|
| Fri, 21 Sep 2001 19:29:08 +0000 |
Guido van Rossum |
Change the name of the __getattr__ special method for new-style
legacy-trunk
|
| Thu, 20 Sep 2001 21:45:26 +0000 |
Guido van Rossum |
Add optional docstrings to getset descriptors. Fortunately, there's
legacy-trunk
|
| Thu, 20 Sep 2001 20:46:19 +0000 |
Guido van Rossum |
Add optional docstrings to member descriptors. For backwards
legacy-trunk
|
| Tue, 18 Sep 2001 20:38:53 +0000 |
Guido van Rossum |
Hopefully fix 3-way comparisons. This unfortunately adds yet another
legacy-trunk
|
| Tue, 18 Sep 2001 20:03:57 +0000 |
Guido van Rossum |
wrap_cmpfunc(): added a safety check for the __cmp__ wrapper. We can
legacy-trunk
|
| Tue, 18 Sep 2001 00:23:33 +0000 |
Tim Peters |
type_new(): Didn't compile anymore, due to change in struct memberlist
legacy-trunk
|
| Fri, 14 Sep 2001 19:43:36 +0000 |
Guido van Rossum |
tp_new_wrapper(): A subtle change in the check for safe use.
legacy-trunk
|
| Fri, 14 Sep 2001 17:51:50 +0000 |
Guido van Rossum |
Add call_maybe(): a variant of call_method() that returns
legacy-trunk
|
| Fri, 14 Sep 2001 16:58:08 +0000 |
Guido van Rossum |
call_method():
legacy-trunk
|
| Thu, 13 Sep 2001 19:18:27 +0000 |
Tim Peters |
type_call(): Change in policy. The keyword args (if any) are now passed
legacy-trunk
|
| Sat, 08 Sep 2001 04:00:12 +0000 |
Tim Peters |
Generalize operator.indexOf (PySequence_Index) to work with any
legacy-trunk
|
| Fri, 07 Sep 2001 18:52:13 +0000 |
Guido van Rossum |
PyType_IsSubtype(): test tp_flags for HAVE_CLASS bit before accessing
legacy-trunk
|
| Thu, 30 Aug 2001 23:13:11 +0000 |
Guido van Rossum |
Give 'super' a decent repr(), and readonly attributes to access the
legacy-trunk
|
