Mailman 3 January 2009 - Python-Dev

Why is there still a PRINT_EXPR opcode in Python 3?
by skip＠pobox.com Jan. 6, 2009

Jan. 6, 2009

Since print is now a builtin function why is there still a PRINT_EXPR opcode? Skip

6 10

Re: [Python-Dev] [Python-checkins] r68182 - in python/trunk: Lib/decimal.py Misc/NEWS
by Jim Jewett Jan. 6, 2009

Jan. 6, 2009

Our of curiousity, why are these constants for internal use only? Is there concern that people might start using "is", or is it just to keep the (beyond the spec) API small, or ...? -jJ On Fri, Jan 2, 2009 at 6:07 PM, mark. dickinson <python-checkins(a)python.org> wrote: > Author: mark.dickinson > Date: Sat Jan 3 00:07:08 2009 > New Revision: 68182 > > Log: > Issue #4812: add missing underscore prefix to some internal-use-only > constants in the decimal module. (… [View More]Dec_0 becomes _Dec_0, etc.) > > > > Modified: > python/trunk/Lib/decimal.py > python/trunk/Misc/NEWS > > Modified: python/trunk/Lib/decimal.py > ============================================================================== > --- python/trunk/Lib/decimal.py (original) > +++ python/trunk/Lib/decimal.py Sat Jan 3 00:07:08 2009 > @@ -216,7 +216,7 @@ > if args: > ans = _dec_from_triple(args[0]._sign, args[0]._int, 'n', True) > return ans._fix_nan(context) > - return NaN > + return _NaN > > class ConversionSyntax(InvalidOperation): > """Trying to convert badly formed string. > @@ -226,7 +226,7 @@ > syntax. The result is [0,qNaN]. > """ > def handle(self, context, *args): > - return NaN > + return _NaN > > class DivisionByZero(DecimalException, ZeroDivisionError): > """Division by 0. > @@ -242,7 +242,7 @@ > """ > > def handle(self, context, sign, *args): > - return Infsign[sign] > + return _Infsign[sign] > > class DivisionImpossible(InvalidOperation): > """Cannot perform the division adequately. > @@ -253,7 +253,7 @@ > """ > > def handle(self, context, *args): > - return NaN > + return _NaN > > class DivisionUndefined(InvalidOperation, ZeroDivisionError): > """Undefined result of division. > @@ -264,7 +264,7 @@ > """ > > def handle(self, context, *args): > - return NaN > + return _NaN > > class Inexact(DecimalException): > """Had to round, losing information. > @@ -290,7 +290,7 @@ > """ > > def handle(self, context, *args): > - return NaN > + return _NaN > > class Rounded(DecimalException): > """Number got rounded (not necessarily changed during rounding). > @@ -340,15 +340,15 @@ > def handle(self, context, sign, *args): > if context.rounding in (ROUND_HALF_UP, ROUND_HALF_EVEN, > ROUND_HALF_DOWN, ROUND_UP): > - return Infsign[sign] > + return _Infsign[sign] > if sign == 0: > if context.rounding == ROUND_CEILING: > - return Infsign[sign] > + return _Infsign[sign] > return _dec_from_triple(sign, '9'*context.prec, > context.Emax-context.prec+1) > if sign == 1: > if context.rounding == ROUND_FLOOR: > - return Infsign[sign] > + return _Infsign[sign] > return _dec_from_triple(sign, '9'*context.prec, > context.Emax-context.prec+1) > > @@ -1171,12 +1171,12 @@ > if self._isinfinity(): > if not other: > return context._raise_error(InvalidOperation, '(+-)INF * 0') > - return Infsign[resultsign] > + return _Infsign[resultsign] > > if other._isinfinity(): > if not self: > return context._raise_error(InvalidOperation, '0 * (+-)INF') > - return Infsign[resultsign] > + return _Infsign[resultsign] > > resultexp = self._exp + other._exp > > @@ -1226,7 +1226,7 @@ > return context._raise_error(InvalidOperation, '(+-)INF/(+-)INF') > > if self._isinfinity(): > - return Infsign[sign] > + return _Infsign[sign] > > if other._isinfinity(): > context._raise_error(Clamped, 'Division by infinity') > @@ -1329,7 +1329,7 @@ > ans = context._raise_error(InvalidOperation, 'divmod(INF, INF)') > return ans, ans > else: > - return (Infsign[sign], > + return (_Infsign[sign], > context._raise_error(InvalidOperation, 'INF % x')) > > if not other: > @@ -1477,7 +1477,7 @@ > if other._isinfinity(): > return context._raise_error(InvalidOperation, 'INF // INF') > else: > - return Infsign[self._sign ^ other._sign] > + return _Infsign[self._sign ^ other._sign] > > if not other: > if self: > @@ -1732,12 +1732,12 @@ > if not other: > return context._raise_error(InvalidOperation, > 'INF * 0 in fma') > - product = Infsign[self._sign ^ other._sign] > + product = _Infsign[self._sign ^ other._sign] > elif other._exp == 'F': > if not self: > return context._raise_error(InvalidOperation, > '0 * INF in fma') > - product = Infsign[self._sign ^ other._sign] > + product = _Infsign[self._sign ^ other._sign] > else: > product = _dec_from_triple(self._sign ^ other._sign, > str(int(self._int) * int(other._int)), > @@ -2087,7 +2087,7 @@ > if not self: > return context._raise_error(InvalidOperation, '0 ** 0') > else: > - return Dec_p1 > + return _Dec_p1 > > # result has sign 1 iff self._sign is 1 and other is an odd integer > result_sign = 0 > @@ -2109,19 +2109,19 @@ > if other._sign == 0: > return _dec_from_triple(result_sign, '0', 0) > else: > - return Infsign[result_sign] > + return _Infsign[result_sign] > > # Inf**(+ve or Inf) = Inf; Inf**(-ve or -Inf) = 0 > if self._isinfinity(): > if other._sign == 0: > - return Infsign[result_sign] > + return _Infsign[result_sign] > else: > return _dec_from_triple(result_sign, '0', 0) > > # 1**other = 1, but the choice of exponent and the flags > # depend on the exponent of self, and on whether other is a > # positive integer, a negative integer, or neither > - if self == Dec_p1: > + if self == _Dec_p1: > if other._isinteger(): > # exp = max(self._exp*max(int(other), 0), > # 1-context.prec) but evaluating int(other) directly > @@ -2154,7 +2154,7 @@ > if (other._sign == 0) == (self_adj < 0): > return _dec_from_triple(result_sign, '0', 0) > else: > - return Infsign[result_sign] > + return _Infsign[result_sign] > > # from here on, the result always goes through the call > # to _fix at the end of this function. > @@ -2674,9 +2674,9 @@ > """ > # if one is negative and the other is positive, it's easy > if self._sign and not other._sign: > - return Dec_n1 > + return _Dec_n1 > if not self._sign and other._sign: > - return Dec_p1 > + return _Dec_p1 > sign = self._sign > > # let's handle both NaN types > @@ -2686,51 +2686,51 @@ > if self_nan == other_nan: > if self._int < other._int: > if sign: > - return Dec_p1 > + return _Dec_p1 > else: > - return Dec_n1 > + return _Dec_n1 > if self._int > other._int: > if sign: > - return Dec_n1 > + return _Dec_n1 > else: > - return Dec_p1 > - return Dec_0 > + return _Dec_p1 > + return _Dec_0 > > if sign: > if self_nan == 1: > - return Dec_n1 > + return _Dec_n1 > if other_nan == 1: > - return Dec_p1 > + return _Dec_p1 > if self_nan == 2: > - return Dec_n1 > + return _Dec_n1 > if other_nan == 2: > - return Dec_p1 > + return _Dec_p1 > else: > if self_nan == 1: > - return Dec_p1 > + return _Dec_p1 > if other_nan == 1: > - return Dec_n1 > + return _Dec_n1 > if self_nan == 2: > - return Dec_p1 > + return _Dec_p1 > if other_nan == 2: > - return Dec_n1 > + return _Dec_n1 > > if self < other: > - return Dec_n1 > + return _Dec_n1 > if self > other: > - return Dec_p1 > + return _Dec_p1 > > if self._exp < other._exp: > if sign: > - return Dec_p1 > + return _Dec_p1 > else: > - return Dec_n1 > + return _Dec_n1 > if self._exp > other._exp: > if sign: > - return Dec_n1 > + return _Dec_n1 > else: > - return Dec_p1 > - return Dec_0 > + return _Dec_p1 > + return _Dec_0 > > > def compare_total_mag(self, other): > @@ -2771,11 +2771,11 @@ > > # exp(-Infinity) = 0 > if self._isinfinity() == -1: > - return Dec_0 > + return _Dec_0 > > # exp(0) = 1 > if not self: > - return Dec_p1 > + return _Dec_p1 > > # exp(Infinity) = Infinity > if self._isinfinity() == 1: > @@ -2927,15 +2927,15 @@ > > # ln(0.0) == -Infinity > if not self: > - return negInf > + return _negInf > > # ln(Infinity) = Infinity > if self._isinfinity() == 1: > - return Inf > + return _Inf > > # ln(1.0) == 0.0 > - if self == Dec_p1: > - return Dec_0 > + if self == _Dec_p1: > + return _Dec_0 > > # ln(negative) raises InvalidOperation > if self._sign == 1: > @@ -3007,11 +3007,11 @@ > > # log10(0.0) == -Infinity > if not self: > - return negInf > + return _negInf > > # log10(Infinity) = Infinity > if self._isinfinity() == 1: > - return Inf > + return _Inf > > # log10(negative or -Infinity) raises InvalidOperation > if self._sign == 1: > @@ -3063,7 +3063,7 @@ > > # logb(+/-Inf) = +Inf > if self._isinfinity(): > - return Inf > + return _Inf > > # logb(0) = -Inf, DivisionByZero > if not self: > @@ -3220,7 +3220,7 @@ > return ans > > if self._isinfinity() == -1: > - return negInf > + return _negInf > if self._isinfinity() == 1: > return _dec_from_triple(0, '9'*context.prec, context.Etop()) > > @@ -3243,7 +3243,7 @@ > return ans > > if self._isinfinity() == 1: > - return Inf > + return _Inf > if self._isinfinity() == -1: > return _dec_from_triple(1, '9'*context.prec, context.Etop()) > > @@ -5490,15 +5490,15 @@ > ##### Useful Constants (internal use only) ################################ > > # Reusable defaults > -Inf = Decimal('Inf') > -negInf = Decimal('-Inf') > -NaN = Decimal('NaN') > -Dec_0 = Decimal(0) > -Dec_p1 = Decimal(1) > -Dec_n1 = Decimal(-1) > +_Inf = Decimal('Inf') > +_negInf = Decimal('-Inf') > +_NaN = Decimal('NaN') > +_Dec_0 = Decimal(0) > +_Dec_p1 = Decimal(1) > +_Dec_n1 = Decimal(-1) > > -# Infsign[sign] is infinity w/ that sign > -Infsign = (Inf, negInf) > +# _Infsign[sign] is infinity w/ that sign > +_Infsign = (_Inf, _negInf) > > > > > Modified: python/trunk/Misc/NEWS > ============================================================================== > --- python/trunk/Misc/NEWS (original) > +++ python/trunk/Misc/NEWS Sat Jan 3 00:07:08 2009 > @@ -108,6 +108,9 @@ > Library > ------- > > +- Issue #4812: add missing underscore prefix to some internal-use-only > + constants in the decimal module. (Dec_0 becomes _Dec_0, etc.) > + > - Issue #4795: inspect.isgeneratorfunction() returns False instead of None when > the function is not a generator. > > _______________________________________________ > Python-checkins mailing list > Python-checkins(a)python.org > http://mail.python.org/mailman/listinfo/python-checkins > [View Less]

2 1

[PATCH] Allow Python to build on MIPS Targets
by Mark Miller Jan. 6, 2009

Jan. 6, 2009

When the merging of the libffi3 branch took place in March, it broke the logic in configure and fficonfig.py.in to deal with MIPS architecture, specifically differentiating in which files to include for MIPS_IRIX versus MIPS_LINUX. I've re-added that logic based on the older code, and adjusted a few things to deal with the new format. I just tested this on my QEMU instance of a MIPSEL Linux install, and works successfully. Before, it died with the error noted here: http://bugs.python.… [View More]

1 0

Roundup version numbers
by Facundo Batista Jan. 6, 2009

Jan. 6, 2009

Hi! To craete this issue compilation [0] I use roundup through its web interface. For example, to know which version names corresponds to each number, I consulted them through: http://bugs.python.org/version But since two weeks ago, this list was trimmed down. I think that it was to not be able to submit bugs for older Python versions, which is great, but there're some bugs assigned to older versions (for example, [1]). So, question: Should I use another way to query the version number-… [View More]

2 2

Flushing email queue
by Guido van Rossum Jan. 6, 2009

Jan. 6, 2009

If there's anything (be it a python-dev issue, or something for python-committers, or a bug) that needs my attention, please resend. In order to start getting work done today, I am archiving all python-related email from the last 2.5 weeks unread that doesn't have me explicitly in the To: or CC: header. -- --Guido van Rossum (home page: http://www.python.org/~guido/)

1 0

ParseTuple question
by Ulrich Eckhardt Jan. 5, 2009

Jan. 5, 2009

Hi! I'm looking at NullImporter_init in import.c and especially at the call to PyArg_ParseTuple there. What I'm wondering is what that call will do when I call the function with a Unicode object. Will it convert the Unicode to a char string first, will it return the Unicode object in a certain (default) encoding, will it fail? I'm working on the MS Windows CE port, and I don't have stat() there. Also, I don't have GetFileAttributesA(char const*) there, so I need a wchar_t (UTF-16) … [View More]

3 2

issue 3582
by Kristján Valur Jónsson Jan. 5, 2009

Jan. 5, 2009

http://bugs.python.org/issue3582 I submitted a patch last august, but have had no comments. Any thoughts? Here is a suggested update to thread_nt.c. It has two significant changes: 1) it uses the new and safer _beginthreadex API to start a thread 2) it implements native TLS functions on NT, which are certain to be as fast as possible. Kristján

1 0

How to configure with icc on Mac?
by skip＠pobox.com Jan. 5, 2009

Jan. 5, 2009

I downloaded an evaluation copy of the Intel compiler for Mac and tried (so far unsuccessfully) to configure with it. I have tried: CC=icc ./configure --prefix=$HOME/tmp/icc-python That failed computing the size of size_t because it tries to incorrectly link with -lgcc_s. Then I tried forcing it to not use gcc: CC=icc ./configure --without-gcc --prefix=$HOME/tmp/icc-python That failed because of a bug in configure.in: case $withval in no) CC=cc … [View More]

6 11

Wrong buildbot page for 3.x-stable
by Antoine Pitrou Jan. 5, 2009

Jan. 5, 2009

Hello, it seems that both the following pages are identical, and are for release30-maint: http://www.python.org/dev/buildbot/3.0.stable/ http://www.python.org/dev/buildbot/3.x.stable/ Isn't the latter supposed to point to the py3k branch buildbots? Regards Antoine.

2 1

py3k, bad sys.path and abort()
by Mark Hammond Jan. 5, 2009

Jan. 5, 2009

I've recently noticed that in py3k, the lack of a suitable sys.path will cause Py_FatalError() to be called, which immediately terminates the entire application. On Windows, it is fairly easy for this to happen for developers or anyone who hasn't run the official Python installation; just have Python used in a 'service', web server, COM object or anything else which loads python3x.dll from the system32 etc directory, and neglect to have added the PYTHONPATH entry in the registry or … [View More]

2 1