Add a new Py_TPFLAGS_DISALLOW_INSTANTIATION type flag to disallow
creating type instances: set tp_new to NULL and don't create the
"__new__" key in the type dictionary.
The flag is set automatically on static types if tp_base is NULL or
&PyBaseObject_Type and tp_new is NULL.
Use the flag on the following types:
* _curses.ncurses_version type
* _curses_panel.panel
* _tkinter.Tcl_Obj
* _tkinter.tkapp
* _tkinter.tktimertoken
* _xxsubinterpretersmodule.ChannelID
* sys.flags type
* sys.getwindowsversion() type
* sys.version_info type
Update MyStr example in the C API documentation to use
Py_TPFLAGS_DISALLOW_INSTANTIATION.
Add _PyStructSequence_InitType() function to create a structseq type
with the Py_TPFLAGS_DISALLOW_INSTANTIATION flag set.
type_new() calls _PyType_CheckConsistency() at exit.
Reorganize pycore_interp_init() to initialize singletons before the
the first PyType_Ready() call. Fix an issue when Python is configured
using --without-doc-strings.
No longer use deprecated aliases to functions:
* Replace PyMem_MALLOC() with PyMem_Malloc()
* Replace PyMem_REALLOC() with PyMem_Realloc()
* Replace PyMem_FREE() with PyMem_Free()
* Replace PyMem_Del() with PyMem_Free()
* Replace PyMem_DEL() with PyMem_Free()
Modify also the PyMem_DEL() macro to use directly PyMem_Free().
These functions are considered not safe because they suppress all internal errors
and can return wrong result. PyDict_GetItemString and _PyDict_GetItemId can
also silence current exception in rare cases.
Remove no longer used _PyDict_GetItemId.
Add _PyDict_ContainsId and rename _PyDict_Contains into
_PyDict_Contains_KnownHash.
Use _PyType_HasFeature() in the _io module and in structseq
implementation. Replace PyType_HasFeature() opaque function call with
_PyType_HasFeature() inlined function.
In ArgumentClinic, value "NULL" should now be used only for unrepresentable default values
(like in the optional third parameter of getattr). "None" should be used if None is accepted
as argument and passing None has the same effect as not passing the argument at all.
No longer limit repr(structseq) to 512 bytes. Use _PyUnicodeWriter
for better performance and to write directly Unicode rather than
encoding repr() value to UTF-8 and then decoding from UTF-8.
Replace strncpy() with memcpy() in structseq_repr() to fix the
following compiler warning:
Objects/structseq.c:187:5: warning: 'strncpy' specified bound depends on the length of the source argument [-Wstringop-overflow=]
strncpy(pbuf, typ->tp_name, len);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Objects/structseq.c:185:11: note: length computed here
len = strlen(typ->tp_name) > TYPE_MAXSIZE ? TYPE_MAXSIZE :
The function writes the terminating NUL byte later.
PyTuple_Pack can fail and return NULL. If this happens, then PyType_FromSpecWithBases will incorrectly create a new type without bases. Also, it will crash on the Py_DECREF that follows. Also free members and type in error conditions.
METH_NOARGS functions need only a single argument but they are cast
into a PyCFunction, which takes two arguments. This triggers an
invalid function cast warning in gcc8 due to the argument mismatch.
Fix this by adding a dummy unused argument.
PyStructSequence_InitType() except that it has a return value (0 on success,
-1 on error).
* PyStructSequence_InitType2() now raises MemoryError on memory allocation failure
* Fix also some calls to PyDict_SetItemString(): handle error
_PyDict_GetItemId() is more efficient: it only builds the Unicode string once.
Identifiers (dictionary keys) are now created at Python initialization, and if
the creation failed, Python does exit with a fatal error.
Before, PyDict_GetItemString() failure was not handled: structseq_new() could
call PyObject_GC_NewVar() with a negative size, and structseq_dealloc() could
also crash.
svn+ssh://pythondev@svn.python.org/python/trunk
TESTED=./python -E -tt ./Lib/test/regrtest.py -uall (both debug and opt modes)
........
r66006 | neal.norwitz | 2008-08-23 22:04:52 -0700 (Sat, 23 Aug 2008) | 25 lines
Fix:
* crashes on memory allocation failure found with failmalloc
* memory leaks found with valgrind
* compiler warnings in opt mode which would lead to invalid memory reads
* problem using wrong name in decimal module reported by pychecker
Update the valgrind suppressions file with new leaks that are small/one-time
leaks we don't care about (ie, they are too hard to fix).
TBR=barry
TESTED=./python -E -tt ./Lib/test/regrtest.py -uall (both debug and opt modes)
in opt mode:
valgrind -q --leak-check=yes --suppressions=Misc/valgrind-python.supp \
./python -E -tt ./Lib/test/regrtest.py -uall,-bsddb,-compiler \
-x test_logging test_ssl test_multiprocessing
valgrind -q --leak-check=yes --suppressions=Misc/valgrind-python.supp \
./python -E -tt ./Lib/test/regrtest.py test_multiprocessing
for i in `seq 1 4000` ; do
LD_PRELOAD=~/local/lib/libfailmalloc.so FAILMALLOC_INTERVAL=$i \
./python -c pass
done
At least some of these fixes should probably be backported to 2.5.
........
PyUnicode_AsStringAndSize -> _PyUnicode_AsStringAndSize to mark
them for interpreter internal use only.
We'll have to rework these APIs or create new ones for the
purpose of accessing the UTF-8 representation of Unicode objects
for 3.1.
svn+ssh://pythondev@svn.python.org/python/trunk
........
r59935 | raymond.hettinger | 2008-01-13 07:15:15 +0100 (Sun, 13 Jan 2008) | 1 line
Named tuple is a concept, not a specific type.
........
r59936 | raymond.hettinger | 2008-01-13 07:18:07 +0100 (Sun, 13 Jan 2008) | 1 line
Fix spelling.
........
r59937 | georg.brandl | 2008-01-13 10:36:18 +0100 (Sun, 13 Jan 2008) | 2 lines
Clarify the effect of text mode.
........
r59938 | thomas.heller | 2008-01-13 12:19:43 +0100 (Sun, 13 Jan 2008) | 1 line
Make Modules/socketobject.c compile for Windows again.
........
r59939 | ka-ping.yee | 2008-01-13 12:25:13 +0100 (Sun, 13 Jan 2008) | 9 lines
Check in the patch proposed by Ben Hayden (benjhayden) for issue
#1550: help('modules') broken by several 3rd party libraries.
Tested with Python build: trunk:54235:59936M -- the reported error
occurs with Django installed (or with any __init__.py present on
the path that raises an exception), and such errors indeed go away
when this change is applied.
........
r59940 | georg.brandl | 2008-01-13 16:04:05 +0100 (Sun, 13 Jan 2008) | 2 lines
Back out r59931 - test_ctypes fails with it.
........
r59943 | amaury.forgeotdarc | 2008-01-14 01:22:44 +0100 (Mon, 14 Jan 2008) | 6 lines
As discussed in issue 1700288:
ctypes takes some liberties when creating python types: it modifies the types'
__dict__ directly, bypassing all the machinery of type objects which deal with
special methods. And this broke recent optimisations of method lookup.
Now we try to modify the type with more "official" functions.
........
r59944 | amaury.forgeotdarc | 2008-01-14 01:29:41 +0100 (Mon, 14 Jan 2008) | 5 lines
Re-apply patch #1700288 (first applied in r59931, rolled back in r59940)
now that ctypes uses a more supported method to create types:
Method cache optimization, by Armin Rigo, ported to 2.6 by Kevin Jacobs.
........
r59946 | amaury.forgeotdarc | 2008-01-14 02:07:27 +0100 (Mon, 14 Jan 2008) | 4 lines
?Why did my tests not notice this before?
Slots inheritance is very different from OO inheritance.
This code lead to infinite recursion on classes derived from StructType.
........
r59947 | christian.heimes | 2008-01-14 04:33:52 +0100 (Mon, 14 Jan 2008) | 1 line
Added new an better structseq representation. E.g. repr(time.gmtime(0)) now returns 'time.struct_time(tm_year=1970, tm_mon=1, tm_mday=1, tm_hour=0, tm_min=0, tm_sec=0, tm_wday=3, tm_yday=1, tm_isdst=0)' instead of '(1970, 1, 1, 0, 0, 0, 3, 1, 0)'. The feature is part of #1816: sys.flags
........
r59948 | christian.heimes | 2008-01-14 04:35:38 +0100 (Mon, 14 Jan 2008) | 1 line
I missed the most important file
........
r59949 | christian.heimes | 2008-01-14 04:42:48 +0100 (Mon, 14 Jan 2008) | 1 line
Applied patch #1816: sys.flags patch
........
r59950 | christian.heimes | 2008-01-14 05:13:37 +0100 (Mon, 14 Jan 2008) | 2 lines
Now that I've learnt about structseq objects I felt like converting sys.float_info to a structseq. It's
readonly and help(sys.float_info) explains the attributes nicely.
........
r59951 | christian.heimes | 2008-01-14 07:06:19 +0100 (Mon, 14 Jan 2008) | 1 line
Added more comments to the new structseq repr code and implemented several of Neal's suggestions.
........
Pablo Galindo
Victor Stinner
Jozef Grajciar
Serhiy Storchaka
Raymond Hettinger
Eddie Elizondo
Zackery Spytz
Victor Stinner
Sergey Fedoseev
Siddhesh Poyarekar
Victor Stinner
Victor Stinner
Stefan Krah
Antoine Pitrou
Martin v. Löwis
Georg Brandl
Benjamin Peterson
Mark Dickinson
Neal Norwitz
Marc-André Lemburg
Christian Heimes