Commit Graph

1444 Commits

Author SHA1 Message Date
Mark Dickinson aa2adc828a Issue #6713: Improve performance of str(n) and repr(n) for integers n
(up to 3.1 times faster in tests), by special-casing base 10 in
_PyLong_Format.  (Backport of r74851 from py3k.)
2009-09-16 22:10:56 +00:00
Thomas Wouters c4dcb3801e Fix issue #1590864, multiple threads and fork() can cause deadlocks, by
acquiring the import lock around fork() calls. This prevents other threads
from having that lock while the fork happens, and is the recommended way of
dealing with such issues. There are two other locks we care about, the GIL
and the Thread Local Storage lock. The GIL is obviously held when calling
Python functions like os.fork(), and the TLS lock is explicitly reallocated
instead, while also deleting now-orphaned TLS data.

This only fixes calls to os.fork(), not extension modules or embedding
programs calling C's fork() directly. Solving that requires a new set of API
functions, and possibly a rewrite of the Python/thread_*.c mess. Add a
warning explaining the problem to the documentation in the mean time.

This also changes behaviour a little on AIX. Before, AIX (but only AIX) was
getting the import lock reallocated, seemingly to avoid this very same
problem. This is not the right approach, because the import lock is a
re-entrant one, and reallocating would do the wrong thing when forking while
holding the import lock.

Will backport to 2.6, minus the tiny AIX behaviour change.
2009-09-16 19:55:54 +00:00
Mark Dickinson 6b3f1ef190 Issue #6848: Fix curses module build failure on OS X 10.6. 2009-09-06 21:21:05 +00:00
Mark Dickinson 2f865b962c Issue #6347: Add inttypes.h to the pyport.h #includes; fixes a build
failure on HP-UX 11.00.
2009-06-30 15:32:30 +00:00
Mark Dickinson 1dcc38456f Remove unused stdint.h includes 2009-06-28 22:37:13 +00:00
Benjamin Peterson 94819cf47c remove some unused symtable constants 2009-06-23 03:01:56 +00:00
Benjamin Peterson d9920c255d remove tmpname support since it's no longer used 2009-06-21 23:03:36 +00:00
Alexandre Vassalotti a6e3474d02 Add const qualifier to PyErr_SetFromErrnoWithFilename and to
PyErr_SetFromErrnoWithUnicodeFilename.
2009-06-12 20:57:12 +00:00
Georg Brandl 944f684ce6 Allow multiple context managers in one with statement, as proposed
in http://codereview.appspot.com/53094 and accepted by Guido.

The construct is transformed into multiple With AST nodes so that
there should be no problems with the semantics.
2009-05-25 21:02:56 +00:00
Benjamin Peterson 1880d8b823 add a SETUP_WITH opcode
It speeds up the with statement and correctly looks up the special
methods involved.
2009-05-25 13:13:44 +00:00
Jeffrey Yasskin 655d835415 Issue #6042:
lnotab-based tracing is very complicated and isn't documented very well.  There
were at least 3 comment blocks purporting to document co_lnotab, and none did a
very good job. This patch unifies them into Objects/lnotab_notes.txt which
tries to completely capture the current state of affairs.

I also discovered that we've attached 2 layers of patches to the basic tracing
scheme. The first layer avoids jumping to instructions that don't start a line,
to avoid problems in if statements and while loops.  The second layer
discovered that jumps backward do need to trace at instructions that don't
start a line, so it added extra lnotab entries for 'while' and 'for' loops, and
added a special case for backward jumps within the same line. I replaced these
patches by just treating forward and backward jumps differently.
2009-05-23 23:23:01 +00:00
Hirokazu Yamamoto b24bb27d22 Issue #3527: Removed Py_WIN_WIDE_FILENAMES which is not used any more. 2009-05-17 02:52:09 +00:00
Jeffrey Yasskin f7f858d141 Issue 5954, PyFrame_GetLineNumber:
Most uses of PyCode_Addr2Line
(http://www.google.com/codesearch?q=PyCode_Addr2Line) are just trying to get
the line number of a specified frame, but there's no way to do that directly.
Forcing people to go through the code object makes them know more about the
guts of the interpreter than they should need.

The remaining uses of PyCode_Addr2Line seem to be getting the line from a
traceback (for example,
http://www.google.com/codesearch/p?hl=en#u_9_nDrchrw/pygame-1.7.1release/src/base.c&q=PyCode_Addr2Line),
which is replaced by the tb_lineno field.  So we may be able to deprecate
PyCode_Addr2Line entirely for external use.
2009-05-08 22:23:21 +00:00
Jeffrey Yasskin 1aa4700234 PyCode_NewEmpty:
Most uses of PyCode_New found by http://www.google.com/codesearch?q=PyCode_New
are trying to build an empty code object, usually to put it in a dummy frame
object. This patch adds a PyCode_NewEmpty wrapper which lets the user specify
just the filename, function name, and first line number, instead of also
requiring lots of code internals.
2009-05-08 21:51:06 +00:00
Benjamin Peterson 399e4c4f8f add _PyObject_LookupSpecial to handle fetching special method lookup 2009-05-08 03:06:00 +00:00
Eric Smith a985a3aee4 Issue #5920: Changed format.__float__ and complex.__float__ to use a precision of 12 when using the empty presentation type. This more closely matches str()'s behavior and reduces surprises when adding alignment flags to an empty format string. Patch by Mark Dickinson. 2009-05-05 18:26:08 +00:00
Antoine Pitrou 653dece278 Issue #4426: The UTF-7 decoder was too strict and didn't accept some legal sequences.
Patch by Nick Barnes and Victor Stinner.
2009-05-04 18:32:32 +00:00
Eric Smith 9139cc6a3b Issue #1588: Add complex.__format__. 2009-04-30 00:58:58 +00:00
Eric Smith cac7af6863 Issue #5793: rationalize isdigit / isalpha / tolower, etc. Will port to py3k. Should fix Windows buildbot errors. 2009-04-27 19:04:37 +00:00
Eric Smith 068f06568b Issue #5835, deprecate PyOS_ascii_formatd.
If anyone wants to clean up the documentation, feel free. It's my first documentation foray, and it's not that great.

Will port to py3k with a different strategy.
2009-04-25 21:40:15 +00:00
Eric Smith aca19e6a74 Backport of some of the work in r71665 to trunk. This reworks much of
int, long, and float __format__(), and it keeps their implementation
in sync with py3k.

Also added PyOS_double_to_string. This is the "fallback" version
that's also available in trunk, and should be kept in sync with that
code. I'll add an issue to document PyOS_double_to_string in the C
API.

There are many internal cleanups. Externally visible changes include:

- Implement PEP 378, Format Specifier for Thousands Separator, for
  floats, ints, and longs.

- Issue #5515: 'n' formatting for ints, longs, and floats handles
  leading zero formatting poorly.

- Issue #5772: For float.__format__, don't add a trailing ".0" if
  we're using no type code and we have an exponent.
2009-04-22 13:29:05 +00:00
Ronald Oussoren 6c0b872f85 Fix for issue5657. 2009-04-19 10:38:20 +00:00
Benjamin Peterson 01c6e6fb35 many more types to initialize (I had to expose some of them) 2009-04-18 22:15:26 +00:00
Mark Dickinson 8e5446f902 Backport r71704 (add configure check for C99 round function) to trunk. 2009-04-18 14:41:37 +00:00
Matthias Klose 91a3b9e4f0 Issue #1113244: Py_XINCREF, Py_DECREF, Py_XDECREF: Add `do { ... } while (0)'
to avoid compiler warnings.
2009-04-05 21:19:13 +00:00
Matthias Klose 0610e0808b - Py_DECREF: Add `do { ... } while (0)' to avoid compiler warnings.
(avoiding brown paper typo this time)
2009-04-05 12:43:08 +00:00
Benjamin Peterson ea0d7444c8 revert r71159 since it broke the build 2009-04-04 16:05:51 +00:00
Matthias Klose 8212a82371 - Py_DECREF: Add `do { ... } while (0)' to avoid compiler warnings. 2009-04-04 15:51:23 +00:00
Antoine Pitrou 789be0c0a0 Issue #2396: backport the memoryview object. 2009-04-02 21:18:34 +00:00
Brett Cannon 238cedcd63 Fix two issues introduced by issue #71031 by changing the signature of
PyImport_AppendInittab() to take a const char *.
2009-04-02 03:34:53 +00:00
Antoine Pitrou f8387af262 Issue #4688: Add a heuristic so that tuples and dicts containing only
untrackable objects are not tracked by the garbage collector. This can
reduce the size of collections and therefore the garbage collection overhead
on long-running programs, depending on their particular use of datatypes.

(trivia: this makes the "binary_trees" benchmark from the Computer Language
Shootout 40% faster)
2009-03-23 18:41:45 +00:00
Mark Dickinson f35f8044e4 Rewrite Py_ARITHMETIC_RIGHT_SHIFT so that it's valid for all signed
integer types T, not just those for which "unsigned T" is legal.
2009-03-20 23:16:14 +00:00
Mark Dickinson efc82f7e8e Issue #4258: Use 30-bit digits for Python longs, on 64-bit platforms.
Backport of r70459.
2009-03-20 15:51:55 +00:00
Jeffrey Yasskin 68d6852006 Backport r69961 to trunk, replacing JUMP_IF_{TRUE,FALSE} with
POP_JUMP_IF_{TRUE,FALSE} and JUMP_IF_{TRUE,FALSE}_OR_POP. This avoids executing
a POP_TOP on each conditional and sometimes allows the peephole optimizer to
skip a JUMP_ABSOLUTE entirely. It speeds up list comprehensions significantly.
2009-02-28 19:03:21 +00:00
Benjamin Peterson 5781f32b4f remove some PyBytes_* aliases that are not in 3.x 2009-02-16 21:09:09 +00:00
Mark Dickinson bcf6b18eb7 A few more minor fixes in longobject.c 2009-02-15 15:48:39 +00:00
Mark Dickinson 2ffb26fb83 Issue #5260: Various portability and standards compliance fixes, optimizations
and cleanups in Objects/longobject.c.  The most significant change is that
longs now use less memory:  average savings are 2 bytes per long on 32-bit
systems and 6 bytes per long on 64-bit systems.  (This memory saving already
exists in py3k.)
2009-02-15 10:13:41 +00:00
Hirokazu Yamamoto bcd3ea86a3 Issue #5204: Define _PyVerify_fd on VC6 to make
test_fdopen (test_os.py) pass.
2009-02-11 04:13:06 +00:00
Kristján Valur Jónsson 6a743d3694 Issue 4804. Add a function to test the validity of file descriptors on Windows, and stop using global runtime settings to silence the warnings / assertions. 2009-02-10 13:32:24 +00:00
Mark Dickinson c97c9096ed Issue #4575: fix Py_IS_INFINITY macro to work correctly on x87 FPUs.
It now forces its argument to double before testing for infinity.
2009-02-09 14:18:43 +00:00
Raymond Hettinger b516370bcb Issue 1242657: list(obj) can swallow KeyboardInterrupt. 2009-02-02 21:50:13 +00:00
Amaury Forgeot d'Arc a40d573664 #3720: Interpreter crashes when an evil iterator removes its own next function.
Now the slot is filled with a function that always raises.

Will not backport: extensions compiled with 2.6.x would not run on 2.6.0.
2009-01-12 23:36:55 +00:00
Jeffrey Yasskin d89f5b20b2 Fix issue 4884, preventing a crash in the socket code when python is compiled
with llvm-gcc and run with a glibc <2.10.
2009-01-09 16:47:07 +00:00
Marc-André Lemburg 3741bfdd3c Fix #4846 (Py_UNICODE_ISSPACE causes linker error) by moving the declaration
into the extern "C" section.

Add a few more comments and apply some minor edits to make the file contents
fit the original structure again.
2009-01-05 19:43:35 +00:00
Mark Dickinson 1bf6401ab2 Use C99 'isfinite' macro in preference to BSD-derived 'finite' function. 2009-01-04 19:53:00 +00:00
Mark Dickinson 0ebfedf38e Fix HAVE_DECL_ISINF/ISNAN test (again). 2009-01-04 17:02:05 +00:00
Mark Dickinson 41fbae1ac9 Oops. Need to check not only that HAVE_DECL_ISINF is defined, but also
that it's equal to 1.  (If isinf isn't defined, HAVE_DECL_ISINF is
defined to be 0, rather than being undefined.)
2009-01-04 16:06:40 +00:00
Mark Dickinson d181e3a246 isinf and isnan are macros, not functions; fix configure script
to use AC_CHECK_DECLS instead of AC_CHECK_FUNCS for these.
(See discussion in issue #4506)
2009-01-04 13:57:26 +00:00
Benjamin Peterson fd9633ed36 fix WORD_BIGEDIAN declaration in Universal builds; fixes #4060 and #4728 2008-12-28 15:37:31 +00:00
Alexandre Vassalotti a925bed208 Sort UCS-2/UCS-4 name mangling list. 2008-12-28 02:10:35 +00:00