Mailman 3 April 2001 - Python-Dev

PEP 245 Prototype
by Michel Pelletier 25 Apr '01

25 Apr '01

I have created a prototype of PEP 245 based on the Mobius Python library. This prototype requires no changes to Python 2.x. I have also updated PEP 245 to reflect using the prototype. Refer to the 'doc' directory for the revised PEP. The prototype can be downloaded from: http://www.zope.org/Members/michel/InterfacesPEP/Interface.tgz It contains four directories and a README that should get you started. Please follow up any comments specific to the PEP onto the type-sig(a)python.org list. Thanks, -Michel

1 0

Attn: Christian Tismer (apologies to others)
by Gordon McMillan 25 Apr '01

25 Apr '01

Sorry to blast everybody with this. Christian, your mail server is getting a new HD. You can reach Jean-Claude at jcwippler(a)hotmail.com. - Gordon

1 0

WWW.090000900.COM GAGNEZ 1 AN DE COMMUNICATIONS GSM ! WIN 1 JAAR GRATIS GSM-GEBRUIK !
by thelink 24 Apr '01

24 Apr '01

--{ Liste hébergée par PoPList }------{ http://www.poplist.fr/ }-- --{ Voir en bas de ce mail les options de désabonnement }-- ______________________________________________________________________ GAGNEZ 1 AN DE COMMUNICATIONS GSM GRATUITES ! WIN 1 JAAR GRATIS GSM-GEBRUIK ! TELECHARGEZ PAR SMS 1500 LOGOS et SONNERIES AU TARIF LE PLUS BAS ( 30 bef / unite ) ! DOWNLOAD PER SMS 1500 LOGOS en RINGTONES AAN HET LAAGSTE TARIEF ( 30 bef / stuk ) ! http://www.090000900.com $ Pour vous désabonner, rendez-vous simplement sur cette page : -> <http://cgi.poplist.fr/@/u/electro/python-dev@python.org>

1 0

tickling version numbers during release
by Skip Montanaro 24 Apr '01

24 Apr '01

It seems like there is always a flurry of checkins associated with bumping version numbers whenever a release is impending. Wouldn't it make sense to stuff the version number into a file somewhere then add a make target to the makefile to update the relevant files and check them into cvs? Skip

4 3

ineffective optimization: method tables
by Martin v. Loewis 23 Apr '01

23 Apr '01

> Unfortunately, all my benchmarks show this patch to be ineffective > in terms of speeding up the interpreter. Anyone know why? I guess because you took PyObject_IsTrue. After profiling some application, I found that this is a frequent function, so I thought it should be optimized. I then found that most of the time, it gets Py_True, Py_False, or Py_None as an argument, so I checked for identity with these objects. Indeed, that covered the majority of the calls - but with no significant speed gain when special-cased. So I think I agree with Guido: even as these functions are frequently called, this is not where the time is consumed. Regards, Martin

1 0

PyChecker 0.4 - new release
by Neal Norwitz 23 Apr '01

23 Apr '01

I've released a new version of PyChecker, the source code checking tool. The most notable change is support for .pycheckrc file to specify your own defaults. There were also a few new warnings added and some bug fixes. Here's the CHANGELOG: * Add .pycheckrc file processing to specify options (like on command line) * Add new warning if module.Attribute doesn't exist * Add new warning: Module (%s) re-imported locally * Add glob'ing support for windows * Handle apply(BaseClass.__init__(self, args)) * Fix command line handling so you can pass module, package, or filename * Fix **kwArgs warning if named parameter is not first * Don't exit from checker when import checker from interpreter PyChecker is available on Source Forge: Web page: http://pychecker.sourceforge.net/ Project page: http://sourceforge.net/projects/pychecker/ Neal

1 0

RE: Python-Dev digest, Vol 1 #1324 - 16 msgs
by Greg Wilson 23 Apr '01

23 Apr '01

> > Thomas Heller: > > PS: I find it strange that everyone so far seems to be against it. > Tim Peters: > I didn't get that sense yet. I did get the sense they're not > actively *for* it yet, and the questions asked so far explain why: > What does it buy us? Greg Wilson: I'd really like class methods so that my classes can carry their factory methods around with them, and so that these factories can be selectively overridden in derived classes. I have machinery to do all of this using freestanding functions, but it's clumsy and error-prone. Of course, so is a lot of my code... :-)

6 7

'iter' as a function name
by Martin v. Loewis 23 Apr '01

23 Apr '01

I should probably be silent on the issue of picking names for things, but I feel that 'iter' is an unfortunte choice for a function name: it is an abbreviated word. Now, you could argue that there is plenty of precedent for that in Python, but some of these are also confusing. Just today, I asked colleague what he thought 'repr' might do, and he had no clue. Anyway, I've been browsing dictionaries somewhat, and here are a few proposals, taking a well-known dictionary as an argument: harp(sys.modules) # or harp_on(sys.modules)? traverse(sys.modules) # not shorter than iterate, though... narrate(sys.modules) Of course, spelling-out "iterate" would be just as fine. Just my 0.02EUR, Martin

1 0

Class Methods
by Thomas Heller 23 Apr '01

23 Apr '01

There are some signs :-) that Python's object model is going to be revised even _before_ Python 3000. Is someone willing to join me fighting for class methods (I mean 'real' class-methods in the Smalltalk style here, _not_ static methods like in Java or C++). Thomas

12 17

Re: [Python-Dev] ineffective optimization: method tables
by Samuele Pedroni 23 Apr '01

23 Apr '01

Hi. [Neil Schemenauer] > I did have 2-D tables for binary operations but since they were > quite sparse I took them out in favor of arrays and case > statements. Unfortunately, all my benchmarks show this patch to > be ineffective in terms of speeding up the interpreter. Anyone > know why? I just noticed that your changes add a 1 more call price also were there was no such price (int + int, etc), do not know if that matters ... regards.

1 0