Mailman 3 March 2007 - Python-announce-list

Elixir 0.3.0 released!
by Gaetan de Menten 27 Mar '07

27 Mar '07

We are pleased to announce that version 0.3.0 of Elixir is now available. As always, feedback is very welcome, preferably on Elixir mailing list. Highlights for this release ------------------------------------- - Added support for autoloading/reflecting databases with has_and_belongs_to_many relationships. See the docs for details. - Added support for the "version_id_col" option on entities. This option adds a column to the table which will be used to prevent concurrent modifications on any row of the entity's table (i.e. it will raise an error if it happens). - Added an "ondelete" argument to belongs_to relationships. - The tablename option can now be given a callable so that people can provide their own function to get the table name for an entity. - Made the provided metadata not threadlocal. This could break things for you in some rare case. Please see the (newly created) FAQ file for details about this. - Fixed bug preventing having entities without any statement. - Fixed typo which broke the use_alter argument on belongs_to relationships. The full list of changes can be seen at: http://elixir.ematia.de/svn/elixir/tags/0.3.0/CHANGES What is Elixir? --------------------- Elixir is a declarative layer on top of SQLAlchemy. It is a fairly thin wrapper, which provides the ability to define model objects following the Active Record design pattern, and using a DSL syntax similar to that of the Ruby on Rails ActiveRecord system. Elixir does not intend to replace SQLAlchemy's core features, but instead focuses on providing a simpler syntax for defining model objects when you do not need the full expressiveness of SQLAlchemy's manual mapper definitions. Mailing list ---------------- http://groups.google.com/group/sqlelixir/about -- Gaëtan de Menten http://openhex.org

1 0

Shed Skin Optimizing Python-to-C++ Compiler 0.0.21
by Mark Dufour 27 Mar '07

27 Mar '07

Hi all, Development of Shed Skin is continuing at a steady pace. I have just released 0.0.21, which comes with the following changes: -important type inference fix/cleanup -support for 'bisect', 'collections.deque' and 'string.maketrans' -improved 'copy' support -support for 'try, else' construction -some optimizations ('dict[..] += ..', "''.join(sorted(str))") -several minor bug fixes See the homepage for more information about Shed Skin, and a collection of 27 'largish' programs (in total about 7,000 lines) that compile with the latest release: http://mark.dufour.googlepages.com I am always looking for developers to join the project. So far I have not received much help. An important area where Shed Skin is still lacking is integration with CPython (building extension modules and using non-compiled modules from compiled code). Other areas I could really use help with are systematic testing, bug reporting, building a better website/wiki, etc. Please let me know if you are interested, and/or join the mailing list from the sourceforge site. Thanks, Mark Dufour. -- "One of my most productive days was throwing away 1000 lines of code" - Ken Thompson

1 0

Pydev 1.3.1 Released
by Fabio Zadrozny 26 Mar '07

26 Mar '07

Hi All, Pydev and Pydev Extensions 1.3.1 have been released Details on Pydev Extensions: http://www.fabioz.com/pydev Details on Pydev: http://pydev.sf.net Details on its development: http://pydev.blogspot.com Release Highlights in Pydev Extensions: ----------------------------------------------------------------- * Interactive Console: jython process can receive vmargs (configured in the preferences page) * Code-Completion: auto-import completions will not be brought for tokens in the current module * Code-Completion: auto-import inserted as last import in imports list (so, __future__ imports will remain in the 1st position) * Code-Analysis: Only adds unused parameters in methods that have some statement that is not a 'pass' or string * Fix: Find Occurrences: will not 'overlap' occurrences anymore * Fix: Code-Analysis: deep attribute access after finding __getattr__ does not signal unresolved import errors anymore Release Highlights in Pydev: ---------------------------------------------- * Mylar integration: the pydev package explorer now supports mylar (packaged as a separate feature: org.python.pydev.mylar.feature) * Code-completion: comment completion is now the same as string completion * Debug: Breakpoints can be set in external files * Debug: Breakpoint annotations now show in external files * Package Explorer: filter for import nodes created * Fix: Package Explorer Actions: Open action does not expand children when opening python file * Fix: Project Explorer (WTP) integration: does not conflict with elements from other plugins anymore (such as java projects) * Fix: halt in new project wizard: when creating a new project from the pydev wizard it was halting in some platforms * Fix: ${string_prompt} in run config: now only evaluated on the actual run * Fix: Code-Completion: jython shell was not handling java.lang.NoClassDefFoundError correctly What is PyDev? --------------------------- PyDev is a plugin that enables users to use Eclipse for Python and Jython development -- making Eclipse a first class Python IDE -- It comes with many goodies such as code completion, syntax highlighting, syntax analysis, refactor, debug and many others. Cheers, -- Fabio Zadrozny ------------------------------------------------------ Software Developer ESSS - Engineering Simulation and Scientific Software http://www.esss.com.br Pydev Extensions http://www.fabioz.com/pydev Pydev - Python Development Enviroment for Eclipse http://pydev.sf.net http://pydev.blogspot.com

1 0

Modular toolkit for Data Processing 2.1 released!
by Tiziano Zito 26 Mar '07

26 Mar '07

MDP version 2.1 and symeig 1.2 have been released! What's new in version 2.1? -------------------------- - Fully compatible with NumpPy 1.0, the first stable release of the descendant of the Numeric python extension module - symeig project resumed and updated - For increased speed, scipy and symeig are automatically used if available - New nodes: Independent Slow Feature Analysis and quadratic forms analysis algorithms - General improvements, several bug fixes, and code cleanups What is it? ----------- Modular toolkit for Data Processing (MDP) is a data processing framework written in Python. From the user's perspective, MDP consists of a collection of trainable supervised and unsupervised algorithms that can be combined into data processing flows. The base of readily available algorithms includes Principal Component Analysis, two flavors of Independent Component Analysis, Slow Feature Analysis, Independent Slow Feature Analysis, and many more. From the developer's perspective, MDP is a framework to make the implementation of new algorithms easier. MDP takes care of tedious tasks like numerical type and dimensionality checking, leaving the developer free to concentrate on the implementation of the training and execution phases. The new elements then seamlessly integrate with the rest of the library. MDP has been written in the context of theoretical research in neuroscience, but it has been designed to be helpful in any context where trainable data processing algorithms are used. Its simplicity on the user side together with the reusability of the implemented nodes make it also a valid educational tool. As its user base is increasing, MDP is becoming a common repository of user-supplied, freely available, Python-implemented data processing algorithms. The optional symeig module contains a Python wrapper for the LAPACK functions to solve the standard and generalized eigenvalue problems for symmetric (hermitian) positive definite matrices. Those specialized algorithms give an important speed-up with respect to the generic LAPACK eigenvalue problem solver used by NumPy. Resources --------- Download: http://sourceforge.net/project/showfiles.php?group_id=116959 Homepage: http://mdp-toolkit.sourceforge.net Mailing list: http://sourceforge.net/mail/?group_id=116959 -- Tiziano Zito Institute for Theoretical Biology Humboldt-Universitaet zu Berlin Invalidenstrasse, 43 10115 Berlin, Germany Pietro Berkes Gatsby Computational Neuroscience Unit Alexandra House, 17 Queen Square London WC1N 3AR, United Kingdom

1 0

ANN: Urwid 0.9.8 - Console UI Library
by Ian Ward 25 Mar '07

25 Mar '07

Announcing Urwid 0.9.8 ---------------------- Urwid home page: http://excess.org/urwid/ Tarball: http://excess.org/urwid/urwid-0.9.8.tar.gz About this release: =================== This release improves Urwid's performance by 70% to 450% (reducing running time by 41% to 82%)[1] for some benchmarks[2]. New base classes have been introduced for widgets, canvases and list walkers. Some bugs have been fixed. Python 2.2 or later is now required. There were three major changes that contributed to the performance improvement in this release. The Canvas class was rewritten so that canvases can be combined more efficiently. Canvas objects are now cached so that most unchanged widgets will not need to re-render themselves. There is a new C module for some of the heavily used low-level string functions contributed by Rebecca Breu. Urwid's raw_display module may now be used with libraries like Twisted that have their own event loops. Use the new get_input_descriptors() and get_input_nonblocking() methods instead of get_input() when using an external event loop. While I have tried to maintain compatibility with older versions of Urwid, some of the changes in this release may affect existing code. Please report any problems you have to the mailing list or the IRC channel. [1] http://article.gmane.org/gmane.comp.lib.urwid/415 [2] http://excess.org/urwid/browser/contrib/trunk New in this release: ==================== * Rendering is now significantly faster. * New Widget base class for all widgets. It includes automatic caching of rows() and render() methods. It also adds a new __super attribute for accessing methods in superclasses. Widgets must now call self._invalidate() to notify the cache when their content has changed. To disable caching in a widget set the class variable no_cache to a list that includes the string "render". * Canvas classes have been reorganized: Canvas has been renamed to TextCanvas and Canvas is now the base class for all canvases. New canvas classes include BlankCanvas, SolidCanvas and CompositeCanvas. * External event loops may now be used with the raw_display module. The new methods get_input_descriptors() and get_input_nonblocking() should be used instead of get_input() to allow input processing without blocking. * The Columns, Pile and ListBox widgets now choose their first selectable child widget as the focus widget by defaut. * New ListWalker base class for list walker classes. * New Signals class that will be used to improve the existing event callbacks. Currently it is used for ListWalker objects to notify their ListBox when their content has changed. * SimpleListWalker now behaves as a list and supports all list operations. This class now detects when changes are made to the list and notifies the ListBox object. New code should use this class to wrap lists of widgets before passing them to the ListBox constructor. * New PollingListWalker class is now the default list walker that is used when passing a simple list to the ListBox constructor. This class is intended for backwards compatibility only. When this class is used the ListBox object is unable to cache its render() method. * The curses_display module can now draw in the lower-right corner of the screen. * All display modules now have start() and stop() methods that may be used instead of calling run_wrapper(). * The raw_display module now uses an alternate buffer so that the original screen can be restored on exit. The old behaviour is available by seting the alternate_buffer parameter of start() or run_wrapper() to False. * Many internal string processing functions have been rewritten in C to improve their performance. * Compatible with Python >= 2.2. Python 2.1 is no longer supported. About Urwid =========== Urwid is a console UI library for Python. It features fluid interface resizing, UTF-8 support, multiple text layouts, simple attribute markup, powerful scrolling list boxes and flexible interface design. Urwid is released under the GNU LGPL.

1 0

EuroPython 2007: Call for Proposals
by Michael Hudson 25 Mar '07

25 Mar '07

Book Monday 9th July to Wednesday 11th July 2007 in your calendar! EuroPython 2007, the European Python and Zope Conference, will be held in Vilnius, Lithuania. Last year's conference was a great success, featuring a variety of tracks, amazing lightning talks and inspiring keynotes. With your participation, we want to make EuroPython 2007, the sixth EuroPython, even more successful than the previous five. Talks, Papers and Themes ------------------------ This year we have decided to borrow a few good ideas from PyCon, one of which is to move away from the 'track' structure. Instead, speakers are invited to submit presentations about anything they have done that they think would be of interest to the Python community. We will then arrange them into related groups and schedule them in the space available. In the past, EuroPython participants have found the following themes to be of interest: * Science * Python Language and Libraries * Web Related Technologies * Education * Games * Agile Methodologies and Testing * Social Skills In addition to talks, we will also accept full paper submissions about any of the above themes. The Call for Refereed Papers will be posted shortly. The deadline for talk proposals is Friday 18th May at midnight (24:00 CEST, Central European Summer Time, UTC+2). Other ways to participate ------------------------- Apart from giving talks, there are plenty of other ways to participate in the conference. Just attending and talking to people you find here can be satisfying enough, but there are three other kinds of activity you may wish to plan for: Lightning Talks, Open Space and Sprints. Lightning Talks are very short talks that give you just enough time to introduce a topic or project, Open Space is an area reserved for informal discussions, and Sprints are focused gatherings for developers interested in particular projects. For more information please see the following pages: * Lightning Talks: http://www.europython.org/sections/events/lightning_talks * Open Space: http://www.europython.org/sections/events/open_space * Sprints: http://www.europython.org/sections/sprints_and_wiki Your Contribution ----------------- To propose a talk or a paper, go to... * http://www.europython.org/submit For more general information on the conference, please visit... * http://www.europython.org/ Looking forward to seeing what you fine folk have been up to, The EuroPython Team

1 0

[ANN] argparse 0.7 - Command-line parsing library
by Steven Bethard 25 Mar '07

25 Mar '07

Announcing argparse 0.7 ----------------------- The argparse module is an optparse-inspired command line parser that improves on optparse by supporting: * positional arguments * sub-commands * required options * options with a variable number of args * better usage messages * a much simpler extension mechanism and a number of other improvements on the optparse API. Download argparse ================= argparse home: http://argparse.python-hosting.com/ argparse single module download: http://argparse.python-hosting.com/file/trunk/argparse.py?format=raw argparse bundled downloads at PyPI: http://www.python.org/pypi/argparse/ About this release ================== This release adds support for single-dash long options and combining single-dash options in a single option string. Note that the 'outfile' type is still deprecated and will likely be removed in the next release. Please update your code to use the new FileType factory. New in this release =================== * Single-dash long options (e.g. ``-foo`` or ``-foo-bar``) * Combining single-dash options (e.g. ``-xy`` is now the same as ``-x -y`` when ``-x`` does not take an argument) Example argparse code ===================== Here's a simple program that sums its command-line arguments and writes them to a file:: parser = argparse.ArgumentParser() parser.add_argument('integers', nargs='+', type=int) parser.add_argument('--log', default=sys.stdout, type=argparse.FileType('w')) args = parser.parse_args() args.log.write('%s\n' % sum(args.integers)) args.log.close()

1 0

ANN: wxPython 2.8.3.0
by Robin Dunn 23 Mar '07

23 Mar '07

Announcing ---------- The 2.8.3.0 release of wxPython is now available for download at http://wxpython.org/download.php. This release includes a number of bug fixes and also some new enhancements, including updates to the XRCed tool and the new InspectionTool. Source code is available, as well as binaries for both Python 2.4 and 2.5, for Windows and Mac, as well some pacakges for various Linux distributions. A summary of changes is listed below and also at http://wxpython.org/recentchanges.php. What is wxPython? ----------------- wxPython is a GUI toolkit for the Python programming language. It allows Python programmers to create programs with a robust, highly functional graphical user interface, simply and easily. It is implemented as a Python extension module that wraps the GUI components of the popular wxWidgets cross platform library, which is written in C++. wxPython is a cross-platform toolkit. This means that the same program will usually run on multiple platforms without modifications. Currently supported platforms are 32-bit Microsoft Windows, most Linux or other Unix-like systems using GTK2, and Mac OS X 10.3+, in most cases the native widgets are used on each platform to provide a 100% native look and feel for the application. Changes in 2.8.3.0 ------------------ Added wx.ToolBar.SetToolNormalBitmap and SetToolDisabledBitmap methods. (Keep in mind however that the disabled bitmap is currently generated on the fly by most native toolbar widgets, so this SetToolDisabledBitmap method won't have any affect on them...) Refactored the inspection tool such that it can be used as a wx.App mix-in class as it was used before (with the wx.lib.mixins.inspect module) and also as a non mix-in tool (using wx.lib.inspect.InspectionTool). Add wx.lib.mixins.treemixin from Frank Niessink. Added the wx.SizerFlags class, and also added AddF, InsertF and PrependF methods to wx.Sizer. The wxSizerFlags class provides a convienient and easier to read way to add items to a sizer. It was added as a new set of methods of the wx.Sizer class so as to not disturb existing code. For example, instead of writing:: sizer.Add(ctrl, 0, wx.EXPAND | wx.ALL, 10) you can now write:: sizer.AddF(ctrl, wx.SizerFlags().Expand().Border(wx.ALL,10)) Will Sadkin provided a patch for the wx.lib.masked package that fixes its support for using the navigation keys on the numeric keypad. wx.lib.plot: patch #1663937 to allow user to turn off scientific notation on plot. wxGTK: Most of the remaining TODOs for the wx.GraphicsContext on wxGTK have been done. This includes implementations for GetTextExtent, Clip, DrawBitmap, fixing the drawing position of text to be at the upper left corner instead of the baseline, etc. wx.lib.customtreectrl patches from Andrea: 1. ExpandAll has been renamed as ExpandAllChildren, and the new ExpandAll now takes no input arguments (consistent with wx.TreeCtrl) 2. ctstyle keyword is now defaulted to 0: every style related to CustomTreeCtrl and the underlying wx.PyScrolledWindow should be declared using the keyword "style" only. For backward compatibility, ctstyle continues to work as I merged ctstyle and style in the __init__ method. 3. GetClassDefaultAttributes is now a classmethod. 4. UnselectAll bug fixed. Renamed the wx.lib.inspect and wx.lib.mixins.inspect modules to inspection, in order to avoid conflicts with the inspect module in the standard Python library. Lots of changes to XRCed from Roman Rolinsky: * Preferences for default "sizeritem" parameters for new panels and controls can be configured ("File">"Preferences..."). * Implemented comment object for including simple one-line comments and comment directives as tree nodes. No validation is performed for a valid XML string so comments must not contain "-->". Comment directive is a special comment starting with '%' character, followed by a line of python code. It is executed using 'exec' when the resource file is opened. This is useful to import plugin modules containing custom handlers which are specific to the resource file, hovewer this is of course a security hole if you use foreign XRC files. A warning is displayed if the preference option 'ask' is selected (by default). * Added support for custom controls and plugin modules. Refer to this wxPythonWiki for the details: http://wiki.wxpython.org/index.cgi/XRCed#custom * Tool panel sections can be collapsed/expanded by clicking on the label of a tool group. * Some undo/redo and other fixes. * Fixes for wxMSW (notebook highlighting, control sizes, tree Unselect). * Notebook page highlighting fix. Highlight resizes when the window is resized. ParamUnit spin button detects event handler re-entry (wxGTK probably has a bug in wxSpinButton with repeated events). * Fix for dealing with empty 'growable' property, using MiniFrame for properties panel, the panel is restored together with the main window. -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython!

1 0

Jinja 1.0 Released
by Armin Ronacher 23 Mar '07

23 Mar '07

Jinja 1.0 Released ================== Jinja 1.0 is out now. Jinja is a sandboxed template engine written in pure Python licensed under the BSD license. It provides a Django-like non-XML syntax and compiles templates into executable python code. It's basically a combination of Django templates and python code. Differences to Django --------------------- While the Jinja default syntax looks pretty much like Django the rest of the template engine works completely different. Jinja was designed to be used without Django and to support Python expressions. Features that Jinja has in common with Django: * filter functions * block based template inheritance * the {% cycle %} and {% filter %} tag * being sandboxed * similar default syntax Differences: * advanced lexer that allows escaping of tags * generation of bytecode for faster template execution * definition of macros (often recurring tasks like rendering dialog boxes, form elements etc.) * raw blocks * uncoupled internationalization support * expression support in any tag * abbility of modifying the context from the template by using the {% set %} tag. * support for function calling with arguments * filters can be called with a variable argument count * full unicode support * deferred objects (objects that are resolved on first access, not on context instanciation) * Missing support for {{ block.super }} * Support for recursing by using recursing for loops or self calling macros. * If blocks have support for multiple elif statements * for loops have an optional else block executed if there was no iteration. * test functions like is even etc. * block delimiters are exchangeable. * no implicit function calling * {% endblock %} doesn't support the optional name argument * {% ifchanged %} is not implemented because of performance reasons. If there's enough interest an implementation might be added * {% templatetag %} is not implemented because Jinja supports escaping via {% raw %} or the variable syntax * {% comment %} is not implemented, rather use for commenting out stuff. * {% now %} is not implemented because Jinja doesn't handle l10n. Datetime specific stuff should be provided by the application. * {% load %}, loading components for Jinja works from the application side. Template designers don't have to import stuff on their own. * {% debug %} is not implemented. You can use the global function debug(). * custom tags are not supported. Most of the stuff that is possible with django template tags is possible with either functions, filters, tests or a combination of them. * using keywords like endfor as identifer raises an exception. Links ----- * Webpage - http://jinja.pocoo.org/ * Downloads - http://jinja.pocoo.org/download.html * Documentation - http://jinja.pocoo.org/documentation/

1 0

ffnet 0.6 released
by Marek Wojciechowski 22 Mar '07

22 Mar '07

ffnet version 0.6 is now released. Source packages, Gentoo ebuilds and Windows binaries are now available for download at: http://ffnet.sourceforge.net The last public release was 0.5. If you are unfamiliar with this package, see the end of this message for a description. NEW FEATURES - trained network can be now exported to fortran source code and compiled - added new architecture generator (imlgraph) - added rprop training algorithm - added draft network plotting facility (based on networkx and matplotlib) CHANGES & BUG FIXES - fixed bug preventing ffnet form working with networkx-0.33 - training data can be now numpy arrays - ffnet became a package now, but API should be compatibile with previous version DOCUMENTATION - docstrings of all objects have been improved - docs (automatically generated with epydoc) are avilable online and have been included to source distribution What is ffnet? -------------- ffnet is a fast and easy-to-use feed-forward neural network training solution for python. Unique features --------------- 1. Any network connectivity without cycles is allowed. 2. Training can be performed with use of several optimization schemes including: standard backpropagation with momentum, rprop, conjugate gradient, bfgs, tnc, genetic alorithm based optimization. 3. There is access to exact partial derivatives of network outputs vs. its inputs. 4. Automatic normalization of data. Basic assumptions and limitations: ---------------------------------- 1. Network has feed-forward architecture. 2. Input units have identity activation function, all other units have sigmoid activation function. 3. Provided data are automatically normalized, both input and output, with a linear mapping to the range (0.15, 0.85). Each input and output is treated separately (i.e. linear map is unique for each input and output). 4. Function minimized during training is a sum of squared errors of each output for each training pattern. Performance ----------- Excellent computational performance is achieved implementing core functions in fortran 77 and wrapping them with f2py. ffnet outstands in performance pure python training packages and is competitive to 'compiled language' software. Moreover, a trained network can be exported to fortran sources, compiled and called from many programming languages. Usage ----- Basic usage of the package is outlined below: from ffnet import ffnet, mlgraph, savenet, loadnet, exportnet conec = mlgraph( (2,2,1) ) net = ffnet(conec) input = [ [0.,0.], [0.,1.], [1.,0.], [1.,1.] ] target = [ [1.], [0.], [0.], [1.] ] net.train_tnc(input, target, maxfun = 1000) net.test(input, target, iprint = 2) savenet(net, "xor.net") exportnet(net, "xor.f") net = loadnet("xor.net") answer = net( [ 0., 0. ] ) partial_derivatives = net.derivative( [ 0., 0. ] ) Usage examples with full description can be found in examples directory of the source distribution or browsed at http://ffnet.sourceforge.net. -- Marek Wojciechowski

1 0