summaryrefslogtreecommitdiffstats
path: root/Modules
Commit message (Collapse)AuthorAgeFilesLines
* SF 1055820: weakref callback vs gc vs threadsTim Peters2004-10-302-107/+296
| | | | | | | | In cyclic gc, clear weakrefs to unreachable objects before allowing any Python code (weakref callbacks or __del__ methods) to run. This is a critical bugfix, affecting all versions of Python since weakrefs were introduced. I'll backport to 2.3.
* some platforms still need offsetof() from structmember.hFred Drake2004-10-171-0/+1
|
* Fix and test weak referencing of itertools.tee objects.Raymond Hettinger2004-10-171-1/+6
|
* Just remove the #include of signal.h. That it was C++-commented outMichael W. Hudson2004-10-141-1/+0
| | | | was a mistake of mine in updating patch #975056 (I think).
* Use C89 style comment for old compilersNeal Norwitz2004-10-141-1/+1
|
* Synchronize with PyXML 1.79:Martin v. Löwis2004-10-131-1/+10
| | | | | | 73: Restore support for Python 2.1. 74: Factor the previous change differently 79: Conditionalize usage of PyTuple_Pack.
* Patch 977343, Solaris likes sys/loadavg.h. Added support for sys/loadavg.hAnthony Baxter2004-10-131-0/+4
| | | | detection to configure &c.
* Patch #975056 - fixes for restartable signals on *BSD. In addition,Anthony Baxter2004-10-133-11/+5
| | | | a few remaining calls to signal() were converted to PyOS_setsig().
* Simplify delitem() code by calling rotate() directly instead of usingRaymond Hettinger2004-10-091-35/+23
| | | | arguments passed through tuples.
* This is Michiel de Hoon's patch, as attached to the bug report:Michael W. Hudson2004-10-071-7/+13
| | | | | | [ 1030629 ] PyOS_InputHook broken with a couple of utterly inconsequential changes by me.
* SF patch #1035498: -m option to run a module as a scriptRaymond Hettinger2004-10-071-7/+73
| | | | (Contributed by Nick Coghlan.)
* Armin's patch to prevent overflows.Raymond Hettinger2004-10-061-7/+22
|
* Upon insertion, if memory runs out, the deque was left in a corrupted state.Armin Rigo2004-10-021-34/+34
| | | | | | | | | | | | | | | | | | | | | | | | | deque_item(): a performance bug: the linked list of blocks was followed from the left in most cases, because the test (i < (deque->len >> 1)) was after "i %= BLOCKLEN". deque_clear(): replaced a call to deque_len() with deque->len; not sure what this call was here for, nor if all compilers under the sun would inline it. deque_traverse(): I belive that it could be called by the GC when the deque has leftblock==rightblock==NULL, because it is tracked before the first block is allocated (though closely before). Still, a C extension module subclassing deque could provide its own tp_alloc that could trigger a GC collection after the PyObject_GC_Track()... deque_richcompare(): rewrote to cleanly check for end-of-iterations instead of relying on deque.__iter__().next() to succeed exactly len(deque) times -- an assumption which can break if deques are subclassed. Added a test. I wonder if the length should be explicitely bounded to INT_MAX, with OverflowErrors, as in listobject.c. On 64-bit machines, adding more than INT_MAX in the deque will result in trouble. (Note to anyone/me fixing this: carefully check for overflows if len is close to INT_MAX in the following functions: deque_rotate(), deque_item(), deque_ass_item())
* Replace structure member before decreffing.Raymond Hettinger2004-10-021-1/+3
|
* * Bulletproof the method for detecting mutations during iteration.Raymond Hettinger2004-10-021-18/+32
| | | | | | | | | The previous approach was too easily fooled (a rotate() sufficed). * Use it->counter to determine when iteration is complete. The previous approach was too complex. * Strengthen an assertion and add a comment here or there.
* Restore the block length and add a comment.Raymond Hettinger2004-10-011-1/+7
|
* Clarify the relationship between indices.Raymond Hettinger2004-10-011-5/+7
|
* * Elaborate on the invariant comments and make them more precise.Raymond Hettinger2004-10-011-14/+28
| | | | | | * Change the centering by one to make it possible to test the module with BLOCKLEN's as low as two. Testing small blocks makes end-point errors surface more readily.
* deque_traverse(): If the deque had one block, and its rightindex wasTim Peters2004-10-011-10/+13
| | | | | BLOCKLEN-1, this assert-failed in a debug build, or went wild with a NULL pointer in a release build. Reported on c.l.py by Stefan Behnel.
* Typos in new comments.Tim Peters2004-10-011-2/+2
|
* Document some reverse-engineered invariants and pragmatic hints.Tim Peters2004-10-011-2/+17
|
* Definition consistency.Tim Peters2004-10-011-1/+2
|
* Trimmed trailing whitespace.Tim Peters2004-10-011-26/+26
|
* * Increase test coverage.Raymond Hettinger2004-09-281-5/+9
| | | | * Have groupby() be careful about decreffing structure members.
* Silence a compiler warning by supplying the correct argument type toRaymond Hettinger2004-09-281-2/+2
| | | | the htons() function.
* Plug a leak and beef-up test coverage.Raymond Hettinger2004-09-281-4/+12
|
* Patch #1009075, bug #952953: allow execve with empty 2nd argumentArmin Rigo2004-09-271-12/+0
|
* Patch #1011240: SystemError generated by struct.pack('P', 'foo').Armin Rigo2004-09-271-7/+9
|
* Add docstrings for regular expression objects and methods.Raymond Hettinger2004-09-241-8/+51
|
* SF bug #1028306: date-datetime comparisonTim Peters2004-09-161-1/+11
| | | | Treat comparing a date to a datetime like a mixed-type comparison.
* Patch for compilation on IRIX from rwgk on http://python.org/sf/728330Trent Mick2004-09-131-0/+10
|
* SF patch #998993: The UTF-8 and the UTF-16 stateful decoders now supportWalter Dörwald2004-09-071-24/+57
| | | | | | | | | | | decoding incomplete input (when the input stream is temporarily exhausted). codecs.StreamReader now implements buffering, which enables proper readline support for the UTF-16 decoders. codecs.StreamReader.read() has a new argument chars which specifies the number of characters to return. codecs.StreamReader.readline() and codecs.StreamReader.readlines() have a new argument keepends. Trailing "\n"s will be stripped from the lines if keepends is false. Added C APIs PyUnicode_DecodeUTF8Stateful and PyUnicode_DecodeUTF16Stateful.
* Apply patch from http://python.org/sf/728330 to fix socket module ↵Trent Mick2004-09-071-1/+13
| | | | compilation on Solaris 2.6, HP-UX 11, AIX 5.1 and (possibly) some IRIX versions.
* Remove redunandant assertions from last checkin.Raymond Hettinger2004-09-061-9/+0
|
* SF #1022953: binascii.a2b_hqx("") raises SystemErrorRaymond Hettinger2004-09-061-5/+16
| | | | | | | | | | | | | Several functions adopted the strategy of altering a full lengthed string copy and resizing afterwards. That would fail if the initial string was short enough (0 or 1) to be interned. Interning precluded the subsequent resizing operation. The solution was to make sure the initial string was at least two characters long. Added tests to verify that all binascii functions do not crater when given an empty string argument.
* Fix erroneous docstring comment.Raymond Hettinger2004-09-061-2/+3
|
* SF bug 1017405: the keys() values() and items() DB methods wereGregory P. Smith2004-09-041-1/+1
| | | | ignoring their transaction (txn) argument.
* Fixing bug #817234, which made SRE get into an infinite loop onGustavo Niemeyer2004-09-031-5/+3
| | | | | empty final matches with finditer(). New test cases included for this bug and for #581080.
* SF patch #1020188: Use Py_CLEAR where necessary to avoid crashesRaymond Hettinger2004-09-011-2/+1
| | | | (Contributed by Dima Dorfman)
* Remove mpz, rotor, xreadlines modulesAndrew M. Kuchling2004-08-311-22/+1
|
* Remove mpz, rotor, xreadlines modulesAndrew M. Kuchling2004-08-313-2488/+0
|
* SF patch #1007189, multi-line imports, for instance:Anthony Baxter2004-08-311-36/+88
| | | | | "from blah import (foo, bar baz, bongo)"
* win32_urandom(): There's no need to copy the generated byte string, soTim Peters2004-08-301-17/+11
| | | | don't.
* win32_urandom(): pass the function name to PyArg_ParseTuple, for betterTim Peters2004-08-301-1/+1
| | | | error msgs.
* win32_urandom(): Raise ValueError if the argument is negative.Tim Peters2004-08-301-0/+3
|
* win32_urandom(): Rewrite to Python C standards (hard tabs, function nameTim Peters2004-08-301-52/+59
| | | | in first column, no parens around return value).
* Patch #934711: Expose platform-specific entropy.Martin v. Löwis2004-08-291-1/+72
|
* SF feature request #992967: array.array objects should support sequences.Raymond Hettinger2004-08-291-6/+20
| | | | Made the constructor accept general iterables.
* Fix the spelling of Fredrik Lundh's last name (unless there really is aBrett Cannon2004-08-271-1/+1
| | | | "Fredrik Lund" who contributed the code in question).
* Patch #1015012. Improve markup and punctuation in libsocket.texDave Cole2004-08-261-2/+2
|
generated by cgit v0.12 at 2025-01-31 06:41:03 (GMT)