Commit Graph

322 Commits

Author SHA1 Message Date
Neal Norwitz a84dcd7546 Stop using METH_OLDARGS implicitly 2007-05-22 07:16:44 +00:00
Armin Rigo 9790a27065 Fix for #1303614 and #1174712:
- __dict__ descriptor abuse for subclasses of built-in types
- subclassing from both ModuleType and another built-in types

Thanks zseil for the patch.
2007-05-02 19:23:31 +00:00
Armin Rigo 796fc992ce Fix a usage of the dangerous pattern decref - modify field - incref. 2007-04-19 14:56:48 +00:00
Armin Rigo c0ba52d3fd Revert r53997 as per
http://mail.python.org/pipermail/python-dev/2007-March/071796.html .

I've kept a couple of still-valid extra tests in test_descr, but didn't
bother to sort through the new comments and refactorings added in r53997
to see if some of them could be kept.  If so, they could go in a
follow-up check-in.
2007-04-19 14:44:48 +00:00
Neal Norwitz cbd9ee69ee When __slots__ are set to a unicode string, make it work the same as
setting a plain string, ie don't expand to single letter identifiers.
2007-04-14 05:25:50 +00:00
Guido van Rossum f102e24bd3 Add a type.__init__() method that enforces the same signature as
type.__new__(), and then calls object.__init__(cls), just to be anal.

This allows us to restore the code in string.py's _TemplateMetaclass
that called super(...).__init__(name, bases, dct), which I commented
out yesterday since it broke due to the stricter argument checking
added to object.__init__().
2007-03-23 18:53:03 +00:00
Guido van Rossum 143b564059 - Bug #1683368: The object.__init__() and object.__new__() methods are
now stricter in rejecting excess arguments.  The only time when
  either allows excess arguments is when it is not overridden and the
  other one is.  For backwards compatibility, when both are
  overridden, it is a deprecation warning (for now; maybe a Py3k
  warning later).

When merging this into 3.0, the warnings should become errors.

Note: without the change to string.py, lots of spurious warnings happen.
What's going on there?
2007-03-23 04:58:42 +00:00
Žiga Seilnacht 6f2d09c949 Patch #1623563: allow __class__ assignment for classes with __slots__.
The old and the new class are still required to have the same slot
names, but the order in which they are specified is not relevant.
2007-03-16 11:59:38 +00:00
Žiga Seilnacht 20f43d3018 Patch #1462488: prevent a segfault in object_reduce_ex() by splitting
the implementation for __reduce__ and __reduce_ex__ into two separate
functions. Fixes bug #931877. Will backport.
2007-03-15 11:44:55 +00:00
Žiga Seilnacht 71436f0229 Patch #1680015: Don't modify __slots__ tuple if it contains an unicode
name. Remove a reference leak that happened if the name could not be
converted to string. Will backport.
2007-03-14 12:24:09 +00:00
Žiga Seilnacht 8903208776 Patch #1675981: remove unreachable code from type.__new__() method.
__dict__ and __weakref__ are removed from the slots tuple earlier
in the code, in the loop that mangles slot names. Will backport.
2007-03-11 15:54:54 +00:00
Jeremy Hylton fa955697fa Add checking for a number of metaclass error conditions.
We add some new rules that are required for preserving internal
invariants of types.

1.  If type (or a subclass of type) appears in bases, it must appear
    before any non-type bases.  If a non-type base (like a regular
    new-style class) occurred first, it could trick type into
    allocating the new class an __dict__ which must be impossible.

2. There are several checks that are made of bases when creating a
   type.  Those checks are now repeated when assigning to __bases__.
   We also add the restriction that assignment to __bases__ may not
   change the metaclass of the type.

Add new tests for these cases and for a few other oddball errors that
were no previously tested.  Remove a crasher test that was fixed.

Also some internal refactoring:  Extract the code to find the most
derived metaclass of a type and its bases.  It is now needed in two
places.  Rewrite the TypeError checks in test_descr to use doctest.
The tests now clearly show what exception they expect to see.
2007-02-27 18:29:45 +00:00
Jeremy Hylton 2d1f5c93bb whitespace normalization 2007-02-27 17:24:48 +00:00
Neal Norwitz ee3a1b5244 Variation of patch # 1624059 to speed up checking if an object is a subclass
of some of the common builtin types.

Use a bit in tp_flags for each common builtin type.  Check the bit
to determine if any instance is a subclass of these common types.
The check avoids a function call and O(n) search of the base classes.
The check is done in the various Py*_Check macros rather than calling
PyType_IsSubtype().

All the bits are set in tp_flags when the type is declared
in the Objects/*object.c files because PyType_Ready() is not called
for all the types.  Should PyType_Ready() be called for all types?
If so and the change is made, the changes to the Objects/*object.c files
can be reverted (remove setting the tp_flags).  Objects/typeobject.c
would also have to be modified to add conditions
for Py*_CheckExact() in addition to each the PyType_IsSubtype check.
2007-02-25 19:44:48 +00:00
Martin v. Löwis fd963265e2 Bug #1653736: Properly discard third argument to slot_nb_inplace_power.
Will backport.
2007-02-09 12:19:32 +00:00
Brett Cannon f5bee30e30 Fix crasher for when an object's __del__ creates a new weakref to itself.
Patch only fixes new-style classes; classic classes still buggy.

Closes bug #1377858.  Already backported.
2007-01-23 23:21:22 +00:00
Andrew M. Kuchling b3f37556bc Fix wording in comment 2006-10-09 18:05:19 +00:00
Armin Rigo 7ccbca93a2 Forward-port of r52136,52138: a review of overflow-detecting code.
* unified the way intobject, longobject and mystrtoul handle
  values around -sys.maxint-1.

* in general, trying to entierely avoid overflows in any computation
  involving signed ints or longs is extremely involved.  Fixed a few
  simple cases where a compiler might be too clever (but that's all
  guesswork).

* more overflow checks against bad data in marshal.c.

* 2.5 specific: fixed a number of places that were still confusing int
  and Py_ssize_t.  Some of them could potentially have caused
  "real-world" breakage.

* list.pop(x): fixing overflow issues on x was messy.  I just reverted
  to PyArg_ParseTuple("n"), which does the right thing.  (An obscure
  test was trying to give a Decimal to list.pop()... doesn't make
  sense any more IMHO)

* trying to write a few tests...
2006-10-04 12:17:45 +00:00
Georg Brandl 5d59c09834 Patch #1567691: super() and new.instancemethod() now don't accept
keyword arguments any more (previously they accepted them, but didn't
use them).
2006-09-30 08:43:30 +00:00
Neal Norwitz 1872b1c01f Fix a couple of bugs exposed by the new __index__ code. The 64-bit buildbots
were failing due to inappropriate clipping of numbers larger than 2**31
with new-style classes. (typeobject.c)  In reviewing the code for classic
classes, there were 2 problems.  Any negative value return could be returned.
Always return -1 if there was an error.  Also make the checks similar
with the new-style classes.  I believe this is correct for 32 and 64 bit
boxes, including Windows64.

Add a test of classic classes too.
2006-08-12 18:44:06 +00:00
Neal Norwitz 8a87f5d37e Patch #1538606, Patch to fix __index__() clipping.
I modified this patch some by fixing style, some error checking, and adding
XXX comments.  This patch requires review and some changes are to be expected.
I'm checking in now to get the greatest possible review and establish a
baseline for moving forward.  I don't want this to hold up release if possible.
2006-08-12 17:03:09 +00:00
Armin Rigo 51fc8c456e Fix and test for an infinite C recursion. 2006-08-09 14:55:26 +00:00
Martin v. Löwis ab2f8f7bd5 __hash__ may now return long int; the final hash
value is obtained by invoking hash on the long int.
Fixes #1536021.
2006-08-09 07:57:39 +00:00
Fred Drake 7a36f5f344 SF patch #1534048 (bug #1531003): fix typo in error message 2006-08-04 05:17:21 +00:00
Neal Norwitz e1fdb32ff2 Handle allocation failures gracefully. Found with failmalloc.
Many (all?) of these could be backported.
2006-07-21 05:32:28 +00:00
Neal Norwitz b114984225 Fix refleak 2006-06-23 03:32:44 +00:00
Armin Rigo 53c1692f6a Fix for an obscure bug introduced by revs 46806 and 46808, with a test.
The problem of checking too eagerly for recursive calls is the
following: if a RuntimeError is caused by recursion, and if code needs
to normalize it immediately (as in the 2nd test), then
PyErr_NormalizeException() needs a call to the RuntimeError class to
instantiate it, and this hits the recursion limit again...  causing
PyErr_NormalizeException() to never finish.

Moved this particular recursion check to slot_tp_call(), which is not
involved in instantiating built-in exceptions.

Backport candidate.
2006-06-21 21:58:50 +00:00
Georg Brandl ccff785258 Patch #1507676: improve exception messages in abstract.c, object.c and typeobject.c. 2006-06-18 22:17:29 +00:00
Georg Brandl 684fd0c8ec Replace PyObject_CallFunction calls with only object args
with PyObject_CallFunctionObjArgs, which is 30% faster.
2006-05-25 19:15:31 +00:00
Thomas Wouters 4e908107b0 Fix variable/format-char discrepancy in new-style class __getitem__,
__delitem__, __setslice__ and __delslice__ hooks. This caused test_weakref
and test_userlist to fail in the p3yk branch (where UserList, like all
classes, is new-style) on amd64 systems, with open-ended slices: the
sys.maxint value for empty-endpoint was transformed into -1.
2006-04-21 11:26:56 +00:00
Skip Montanaro 429433b30b C++ compiler cleanup: bunch-o-casts, plus use of unsigned loop index var in a couple places 2006-04-18 00:35:43 +00:00
Tim Peters ffe2395777 Remove now-unused variables from tp_traverse and tp_clear methods. 2006-04-15 22:51:26 +00:00
Thomas Wouters c6e55068ca Use Py_VISIT in all tp_traverse methods, instead of traversing manually or
using a custom, nearly-identical macro. This probably changes how some of
these functions are compiled, which may result in fractionally slower (or
faster) execution. Considering the nature of traversal, visiting much of the
address space in unpredictable patterns, I'd argue the code readability and
maintainability is well worth it ;P
2006-04-15 21:47:09 +00:00
Thomas Wouters edf17d8798 Use Py_CLEAR instead of in-place DECREF/XDECREF or custom macros, for
tp_clear methods.
2006-04-15 17:28:34 +00:00
Martin v. Löwis ee36d650bb Correct casts to char*. 2006-04-11 09:08:02 +00:00
Anthony Baxter a62862120d More low-hanging fruit. Still need to re-arrange some code (or find a better
solution) in the same way as listobject.c got changed. Hoping for a better
solution.
2006-04-11 07:42:36 +00:00
Armin Rigo 314861c568 Minor bugs in the __index__ code (PEP 357), with tests. 2006-03-30 14:04:02 +00:00
Georg Brandl 347b30042b Remove unnecessary casts in type object initializers. 2006-03-30 11:57:00 +00:00
Georg Brandl 5c170fd4a9 Fix some missing checks after PyTuple_New, PyList_New, PyDict_New 2006-03-17 19:03:25 +00:00
Guido van Rossum 692cdbc5d6 Fix three nits found by Coverity, adding null checks and comments. 2006-03-10 02:04:28 +00:00
Guido van Rossum 38fff8c4e4 Checking in the code for PEP 357.
This was mostly written by Travis Oliphant.
I've inspected it all; Neal Norwitz and MvL have also looked at it
(in an earlier incarnation).
2006-03-07 18:50:55 +00:00
Martin v. Löwis 725507b52e Change int to Py_ssize_t in several places.
Add (int) casts to silence compiler warnings.
Raise Python exceptions for overflows.
2006-03-07 12:08:51 +00:00
Martin v. Löwis 15e62742fa Revert backwards-incompatible const changes. 2006-02-27 16:46:16 +00:00
Georg Brandl c255c7bef7 Bug #1086854: Rename PyHeapType members adding ht_ prefix. 2006-02-20 22:27:28 +00:00
Martin v. Löwis eb079f1c25 Use Py_ssize_t for counts and sizes.
Convert Py_ssize_t using PyInt_FromSsize_t
2006-02-16 14:32:27 +00:00
Martin v. Löwis 2c95cc6d72 Support %zd in PyErr_Format and PyString_FromFormat. 2006-02-16 06:54:25 +00:00
Martin v. Löwis 18e165558b Merge ssize_t branch. 2006-02-15 17:27:45 +00:00
Neal Norwitz 50bf51a3a9 Fix ref/memory leak introduced in rev 41845. 2006-01-02 02:46:54 +00:00
Armin Rigo 037d1e0ff3 SF bug #1153075: "PyXxx_Check(x) trusts x->ob_type->tp_mro".
A patch by mwh to check that user-defined mro's are reasonable
enough.
2005-12-29 17:07:39 +00:00
Armin Rigo fd163f92ce SF patch #1390657:
* set sq_repeat and sq_concat to NULL for user-defined new-style
  classes, as a way to fix a number of related problems.  See
  test_descr.notimplemented()).  One of these problems was fixed
  in r25556 and r25557 but many more existed; this is a general
  fix and thus reverts r25556-r25557.

* to avoid having PySequence_Repeat()/PySequence_Concat() failing
  on user-defined classes, they now fall back to nb_add/nb_mul if
  sq_concat/sq_repeat are not defined and the arguments appear to
  be sequences.

* added tests.

Backport candidate.
2005-12-29 15:59:19 +00:00