🌐 Mailman 3 January 2002 - Python-list - python.org

RE: "Private" attributes - a possible idea (maybe PEP-worthy).
by Delaney, Timothy Jan. 31, 2002

Jan. 31, 2002

> From: Skip Montanaro [mailto:skip@pobox.com] > Tim> a.Klass -> _1__attr1 and _1__attr2 > Tim> b.Klass -> _2__attr1 and _2__attr2 > > This would be unworkable. Private attributes couldn't be pickled. Damn - knew there had to be *something* big against it. However, I can't help but feel that there must be *some* way to enforce that a "private" variable won't be accidentally used by a subclass. Hmm - would it be possible to save a class:id mapping in the pickle, and have the private attributes re-written with the "real" class ID when unpickling? No - don't think that would be possible - and would be undesirable - we need a solution that doesn't require a change to the pickle code (and hence everything that emulates the pickle code). Any other ideas of how to get around this problem? Unfortunately, whilst it's not a problem that is likely to crop up commonly, it's one that when it *does* crop up is likely to cause real debugging headaches. It is also (IMO) a wart in the language because it can cause such obscure problems. Tim Delaney

2 1

Status of Python for .NET
by gerson.kurz＠t-online.de Jan. 31, 2002

Jan. 31, 2002

Two questions: - The current version of VisualPython .NET seems not to generate CLR code, but just provide a fancy IDE for "Python as we know and love it". Will Python support the CLR, and specifically the language-interoperability thing? (I'm asking because I want to be spared the ... bad language-lookalike that is C#). - With the leave of Mark Hammond from Activestate, does anybody have a guess about the future of Visual Python .NET?

6 5

RE: Java vs Python Benchmarks: Java is faster
by brueckd＠tbye.com Jan. 31, 2002

Jan. 31, 2002

Bryan wrote: > For what it is worth to people, I have taken the liberty of > benchmarking the new Java SDK ( 1.4.0 ) against Python 2.1 and 2.2. > The tests originate from: [snip] Thanks for running those tests, Bryan. Pretty interesting stuff. I wasn't sure from reading your post, but I'm assuming that the Java stuff had the JIT enabled? I'd be curious to know what Java numbers would be without it - I was really expecting Java + JIT to be a lot faster than those tests showed... all that extra technology and meager (except ObjectTest!) speed improvements. -Dave

1 0

ANN: Spectix Multi-language GUI Builder
by support＠internetdiscovery.com Jan. 31, 2002

Jan. 31, 2002

SpecTix provides a development environment to build applications with graphical user interfaces that run on multiple platforms with multiple languages. It lets you design GUIs interactively and graphically. It creates applications on one platform that can run on any of the other platforms with no changes to the application. It also uses a geometry smart enough to keep elements aligned across all platforms, and enables fast development of applications that require many more lines of code in traditional procedural languages. It generates code for Tk/Tcl, perl/Tk, Python/Tkinter, and Ruby, SpecTix is based on the SpecTcl GUI builder from Sun Microsystems Laboratories, with an improved UI using the Tix extension widgets (http://tix.sourceforge.net) See: http://starship.python.net/crew/mike/

1 0

Searching for pattern matches with k mismatches
by Mark Robinson Jan. 31, 2002

Jan. 31, 2002

Hi guys, I was wondering if anyone could help with ideas for the following problem. I want to search a string for all matches to a substring (easy enough with re) but allowing the matches to contain k number of mismatches. I can do it the brute force way, by checking every single pattern individually and comparing character by character but that is as slow as......well me doing it. Is there a fast, efficient way to do it. I tried to think of a way to achieve it with regular expressions but just gave myself a headache. tia Blobby -- Mark Robinson Research Programmer Biocomputation Group Science and Technology Centre University of Hertfordshire Herts, UK AL10 9AB http://strc.herts.ac.uk/bio/markr/

2 1

IMAP over SSL
by Silvano Maffeis Jan. 31, 2002

Jan. 31, 2002

Hi, Is it possible to establish SSL connections from within a Python application? In particaluar I would like to use the Python IMAP module over SSL. Thanks. Silvano

3 2

Re: Problems using SocketServer.ThreadingTCPServer
by Joseph A Knapka Jan. 31, 2002

Jan. 31, 2002

Joao Prado Maia wrote: > > Well, correct me if I'm wrong but if OE is dropping the connection, isn't > SocketServer.ThreadingTCPServer' job to call the correct method on my class > ? I'm not completely sure, but I think it's your job to detect end-of-connection in your handle() method and return. The TCPServer, I believe, just does: rqh = YourRequestHandlerClass(...) rqh.setup() rqh.handle() rqh.finish() Cheers, -- Joe "I should like to close this book by sticking out any part of my neck which is not yet exposed, and making a few predictions about how the problem of quantum gravity will in the end be solved." --- Physicist Lee Smolin, "Three Roads to Quantum Gravity"

1 0

zope's acquisition inheritance
by Wostenberg Jan. 31, 2002

Jan. 31, 2002

This Smalltalker has been studying Zope, the Opensourced python product for internet deployment (see http://www.zope.org), to learn. It has an unusual model of interitance called acquisition, which blends class heiarchy with containment. Read http://www.zope.org/Members/Amos/WhatIsAcquisition for introduction. Basically "objects automatically gather. services from their containers.". Everything is potentially a collector-class, and if I drop objects into a collection, they inherit behavior from the container. Weird, but somehow natural in this Web environment which organises content around a tree-structured containment plan. Anybody heard of acquisition outside Zope, or is it a new contribution to OO modelling? -Alan

7 6

ANNOUNCE: SCons 0.04 now available
by Steven Knight Jan. 31, 2002

Jan. 31, 2002

Version 0.04 of SCons has been released and is available for download from the SCons web site: http://www.scons.org/ Or through the download link at the SCons project page at SourceForge: http://sourceforge.net/projects/scons/ RPM and Debian packages and a Win32 installer are all available, in addition to the traditional .tar.gz files. WHAT'S NEW IN THIS RELEASE? Version 0.04 of SCons contains the following changes: - targets are automatically removed before being built unless the new Precious() method suppresses their removal - a new Ignore() method allows dependencies to be ignored - changes to -I or -L options (include or lib directories) don't cause a rebuild unless an underlying file changed - a --debug=pdb option re-invokes SCons under the Python debugger - you can specify var=value arguments on the command line - a directory can now be a Default() target - function actions now return None upon success, and raise exceptions properly - Action() and Scanner() can be called from subsidiary SConscripts - binary file signatures on Win32 systems are now calculated correctly - LIBS and LIBPATH construction variables can now be strings - an action is now executed just once for a list of targets, not once for each target in the list - SCons will now find the library directory properly if you specify your own --prefix= value at installation - more performance improvements, including calculating a build signature just once per node - documentation improvements, including corrected mistakes and additional examples For a complete list of changes in version 0.05, see the CHANGES.txt file in the release itself. WHAT IS SCONS? SCons is a software construction tool (build tool, make tool) written in Python. Its design is based on the design which won the Software Carpentry build tool competition in August 2000 (in turn derived from the Perl-based Cons build tool). Distinctive features of SCons include: - configuration files are Python scripts, allowing the full use of a real scripting language to solve build problems - a modular architecture allows the SCons Build Engine to be embedded in other Python software - a global view of all dependencies; no multiple passes to get everything built - the ability to scan files for implicit dependencies (#include files); - improved parallel build (-j) support - use of MD5 signatures to decide if a file has changed - easily extensible through user-defined Builder and Scanner objects An scons-users mailing list has been created for those interested in getting started using SCons. You can subscribe at: http://lists.sourceforge.net/lists/listinfo/scons-users Alternatively, we invite you to subscribe to the low-volume scons-announce mailing list to receive notification when new versions of SCons become available: http://lists.sourceforge.net/lists/listinfo/scons-announce ACKNOWLEDGEMENTS Special thanks to Charles Crain, Steven Leblanc, and Anthony Roach for their contributions to this release. On behalf of the SCons team, --SK

1 0

Announce: dataflow toolkit v0.5
by Oren Tirosh Jan. 31, 2002

Jan. 31, 2002

dataflow.py is a toolkit for dataflow-oriented programming. It requires Python 2.2 since it makes extensive use of iterators and generator functions. http://www.tothink.com/python/dataflow The module defines the functions xmap, xzip and xfilter which are equivalent to map, zip, filter but return an iterable object rather than a list and are lazy in reading from their upstream sources. Other lazy functions and objects for contructing data flow processes are also provided. The flow type describes a recipe for a dataflow. A complete flow includes a source, optional transformations and a destination. * Sources Sources may be Python containers or any other object with an __iter__ method. * Transformations A transformation is an object with a __tran__ method. This method takes an upstream iterator and returns an iterator. It is usually implemented as a generator function. A function or any other callable object may also be used as a stateless transformation. * Destinations Destinations have a __sink__ method returning an object conforming to the sink protocol, analogous to the iterator protocol. The sink() function also supports the Python built-in container and file types that do not have a __sink__ method. Partial flows may be used for the construction of complete flows or used by themselves. A flow consisting of a source and one or more transformations may be iterated using a for statement, for example. One final feature is the ability to construct dataflow using shell-like | and > operators. This makes Python quite usable for one-liners. The resulting syntax looks familiar and alien at the same time. I'm still trying to figure out whether that's a good thing or not :-) Oren

2 1