See http://gmpy.sourceforge.net/ for details.
What is it: a wrapper for the GMP library, to provide multi-precision
arithmetic for Python. Multi-precision floats, and unbounded-precision
rationals, are not present in stock Python; multi-precision integers
('long') are, but gmpy's version of multi-precision integers is faster
for some operations (NOT all -- used to be, but Python 2.3 did serious
enhancements to some operations on longs) and provides lots of nifty
pre-packaged additional functions.
Minor changes and bug-fixes since the latest 0.9 pre-alpha; support for
Python 2.3. The Windows binary release is now for Python 2.3 _only_ (if
you're stuck with Python 2.2 on Windows, you can keep using gmpy 0.9
pre-alpha and not really suffer from that). Known bug on Windows: the
scan0 and scan1 functions appear broken (perhaps related to the lack of
a GMP 4.0 library for Windows -- haven't found one around yet).
New in release 0.2:
# Replace pictures - generate catalogues, worksheets, photo galleries
etc. from an SXW-Template.
# Patch for makeSerialLetters() to make it usable with Asian fonts (no
PDF support for such fonts yet).
Have a look at:
The SpamBayes team is pleased to announce the 1.0 release of SpamBayes.
As is now usual, this is both a release of the source code and of an
installation program for all Microsoft Windows users.
The Windows installation program will install either the Outlook add-in (for
Microsoft Outlook users), or the SpamBayes server program (for all other
POP3 mail client users, including Microsoft Outlook Express). All Windows
users (including existing users of the Outlook add-in) are encouraged to use
the installation program.
If you wish to use the source-code version, you will also need to install
Python - see README.txt in the source tree for more information.
This release includes no changes from the successful (but now rather dated)
1.0rc2 release. However, we still highly recommend that existing users
upgrade to the final version. Work has already begun towards the first 1.1
release, and we expect to release a (bug fix only) 1.0.1 release around the
same time as 1.1a1.
September 2004 is Spambayes' 2nd birthday, and (as many users know) we have
gone through a very long release process, including 8 alpha releases, a
beta, and two release candidates, all tested by a large number of users. As
such, we are very confident that this 1.0 release is stable and suitable for
regular use. We do welcome any and all contributions for improvements, of
Get it via the 'Download' page at
Enjoy the new release and your spam-free mailbox :-)
Thanks to everyone involved in this release, particularly Richie Hindle and
(on behalf of the SpamBayes team)
--- What is SpamBayes? ---
The SpamBayes project is working on developing a Bayesian (of sorts)
anti-spam filter (in Python), initially based on the work of Paul Graham.
The major difference between this and other, similar, projects is the
emphasis on testing newer approaches to scoring messages.
The project includes a number of different applications, all using the same
core code, ranging from a plug-in for Microsoft Outlook, to a POP3 proxy, to
various command-line tools.
* Middleware early registration deadline has been extended to *
* Thursday, September 30th (3 days from now) *
* Wednesday, October 6th: online registration closes *
ACM/IFIP/USENIX Middleware 2004
Call for Participation
Renaissance Toronto Hotel at SkyDome
Toronto, Ontario, Canada
October 18th - 22nd, 2004
Early Registration Rate Ends: 30 September 2004 (extended)
Last Day to Pre-Register: 06 October 2004
Hotel Cut-Off Date: 25 September 2004
Hotel room reservations (Canada/US): 1-800-237-1512
For details and exact dates see: http://www.eecg.utoronto.ca/middleware2004/
How Wrong Can You Be? Getting Lost on the Road to Massive Scalability
Werner Vogels, Director of Systems Research, Amazon.com
Experiences Building a 24x7 Real-time ASP Service at Citrix Online
Thorsten von Eicken, Chief Architect, Citrix Online
Aspect-Oriented Programming - The promise and the controversy
Gregor Kiczales, University of British Columbia
Main conference program
Single-track conference with 26 technical paper presentations
and 6 work-in-progress presentations.
For details see
Tutorials (Monday and Tuesday):
Building Distributed .NET Applications - Technologies,
Architecture, Comparison with J2EE. By Michael Stal, Siemens AG
Germany (Monday, full day.)
Getting Started with Aspect-Oriented Programming for Middleware.
By Yvonne Coady, University of Victoria, Canada (Monday,
Middleware and Data Management for Sensor Networks. By Nalini
Venkatasubramanian, University of California, Irvine, USA and
Sharad Mehrotra,University of California, Irvine, USA (Tuesday,
Enterprise Integration Patterns: Asynchronous Messaging
Architectures in Practice. By Gregor Hohpe, ThoughtWorks Inc.,
USA (Tuesday, morning.)
Data Grid Management Systems. By Arun Swaran Jagatheesan,
University of California, San Diego, USA (Tuesday, afternoon.)
DDS: The Real-Time Publish-Subscribe Standard for Data
Distribution By Gerardo Pardo-Castellote, Real-Time Innovations
Inc., USA (Tuesday, afternoon.)
Workshops (Monday and Tuesday):
2nd Workshop on Middleware for Grid Computing. Organizers: Bruno
Schulze - National Scientific Computing Laboratory (Brazil) and
Radha Nandkumar - NCSA/UIUC (USA) (Monday, October 18th.)
2nd Workshop on Middleware for Pervasive and Ad-Hoc Computing
Organizer: Paddy Nixon - Strathclyde University (UK) (Monday,
3rd Workshop on Adaptive and Reflective Middleware. Organizer:
Fabio Costa, Federal University of Goi�s - (Brazil) and Nanbor
Wang - Tech-X Corporation (USA) (Tuesday, October 19th.)
1st Middleware Doctoral Symposium. Organizers: Edward Curry -
National University of Ireland (Ireland) and Doug Lea - SUNY
Oswego (USA) (Tuesday, October 19th.)
GET it here: http://www.nevow.com/releases/0.3.0/nevow-0.3.tar.gz
Nevow (pronounced nou-veau) is a web application construction kit written
in Python. It is designed to allow the programmer to express as much of
the view logic as desired in Python, and includes a pure Python XML
expression syntax named stan to facilitate this. However it also provides
rich support for designer-edited templates, using a very small XML
attribute language to provide bi-directional template manipulation
Nevow also includes formless, a declarative syntax for specifying the
types of method parameters and exposing these methods to the web. Forms
can be rendered automatically, and form posts will be validated and input
coerced, rendering error pages if appropriate. Once a form post has
validated successfully, the method will be called with the coerced values.
a browser and Python on the server. For 0.3, LivePage has been updated to
operate on Mozilla, Firefox, Windows Internet Explorer 6, and Safari on
implementation details are hidden from the programmer, with Nevow taking
care of routing data to and from the server using XmlHttpRequest.
* XHTML templates: contain no programming logic, only nodes tagged
with nevow attributes
* data/render methods: simplify the task of separating data from
presentation and writing view logic
* stan: An s-expression-like syntax for expressing xml in pure python
* formless: For describing the types of objects which may be passed to
methods of your classes, validating and coercing string input from
either web or command-line sources, and calling your methods
automatically once validation passes
* webform: For rendering web forms based on formless type
descriptions, accepting form posts and passing them to formless
validators, and rendering error forms in the event validation fails
events to the server and server side events to the client after the
page has loaded, without causing the entire page to refresh
Bug Tracker: http://divmod.org/users/roundup.twistd/nevow/
What's New in Nevow 0.3:
2004-09-26 Donovan Preston <dp(a)divmod.org>
* Added nevow.inevow.IQ interface, an interface for querying the
stan DOM. Eventually, this interface will contain APIs for doing
traditional DOM introspection, such as iterating children,
examining tag names, and examining attributes. For now, it contains
only the patternGenerator, onePattern, and allPatterns APIs. These
APIs have been deprecated from Context.
The main benefit of this is the ability to do:
which would be nice for creating "pattern library" files containing
common skin idioms which can then be copied and used throughout
2004-09-25 Donovan Preston <dp(a)divmod.org>
* Chatola received a major facelift, bringing it from cool demo
up to almost a full fledged web-based chat server. The helper API
LiveEvil.call(...) was added, which deprecates
2004-09-23 Tommi Virtanen <tv(a)twistedmatrix.com>
* Make guard.SessionWrapper store its URL location in all requests it
passes to its children. This allows the children know where to post
the __login__ and __logout__ forms, even deep inside the resource
tree (fixes issue59).
* Guard now works as a non-root resource, with help from the above
change. Semantics of __login__ clarified in unit tests; if your guard
is at /foo, posting to /foo/__login__ redirects to /foo, and posting
to /foo/__login__/ redirects to /foo/. The two unit tests that failed
earlier now pass (with that change in their __login__ URLs).
* If URL-based sessions are used, login no longer loses session
information due to redirect to / (fixes issue56).
2004-09-20 Matt Goodall <matt(a)pollenation.net>
* Added URL.secure() method to make switching between http and
2004-09-08 Donovan Preston <dp(a)divmod.org>
* Nevow now includes a very simple proof-of-concept WSGI Application
implementation, and can be used with no Twisted dependency. Nevow can
also be used to write CGIs, either using a simple CGI WSGI gateway
(which supports URL traversal), or by using Page.renderString (which does not).
* Two new context interfaces, ICurrentSegments and IRemainingSegments,
replace the need to examine the Request prepath and postpath attributes
* ISerializable is deprecated, and has been replaced with a simple
Flattener registry. nevow.flat.registerFlattener and
nevow.flat.getFlattener have been added to support this.
2004-09-06 Donovan Preston <dp(a)divmod.org>
* BACKWARDS INCOMPATIBLE CHANGE. Page.configurable_ *always* returns
self, and a new Page.configurable_original *always* returns
self.original. If you were relying on Page.configurable_'s
introspection behavior and are now getting errors about adapting
to IConfigurable, change your renderForms() calls to:
This causes Page.configurable_original to be invoked and
introspected for form rendering.
2004-08-23 Donovan Preston <dp(a)divmod.org>
* LivePage uses a simpler implementation strategy which requires
the browser to make one request per output event. As a result,
LivePage now works on Mozilla, Safari, and Internet Explorer Windows.
2004-08-05 Donovan Preston <dp(a)divmod.org>
* Implemented support for IFoo(ctx) synonym syntax. It does the
same thing as ctx.locate(IFoo)
* Removed Resource Generators, a feature of NevowSite that nobody
used and wasn't really useful.
* Changed all inevow.IResource apis to take a Context object
where they used to take the request. Remembering objects in
PageContexts is now much easier, and fewer hacks are required to
build the context chain. The context chain now looks like:
2004-7-28 Donovan Preston <dp(a)divmod.org>
* Parameterize data_* methods in the same way as render_* methods.
Patch by k3mper.
For example, <div nevow:data="foo bar,baz" /> will cause
data_foo(self, bar, baz) to be called with the strings "bar" and
"baz". This data method should return a callable taking ctx, data.
The return value of this callable will be remembered as IData at
this point in the context stack.
* Added list-slicing support to ListContainer. You may now use
list slicing syntax in a data directive in addition to a simple
index. For example:
def data_aList(self, ctx, data):
return ["Buckle", "My", "Shoe"]
<span nevow:data="1:-1" nevow:render="string" />
Will render as <div><span>My</span></div>
2004-7-20 Matt Goodall <matt(a)pollenation.net>
* Modified sax loader to retain doctypes, comments and xmlns attributes.
It's now possible to build XHTML valid pages :).
xmlns attributes are always kept but there are options to ignore the doctype
and comment (at the request of foom, not sure why yet). Right now, the default
is to retain doctypes and comments but you can use the ignoreDocType and
ignoreComment args to the xml loaders and flatsax parse functions. This bit
* Add a URL -> IResource adapter that performs a HTTP redirect. URLs can then
be returned from locateChild().
2004-06-07 Donovan Preston <dp(a)divmod.org>
* Added nevow.canvas, an experimental module similar to LivePage
which provides a Python server-side API for drawing arbitrary lines,
curves, and text in the browser. The implementation is socket-based
and asynchronous, so the server can issue drawing commands to the
client at any time.
The idea is to provide a server-side API to the Python programmer
and shield them from implementation details, but the current
implementation uses a pre-compiled Flash movie (which never changes;
we are not generating Flash bytecodes). An implementation using SVG
or Safari's Canvas (nevow.canvas was written before the Safari Canvas
announcement) would be possible.
2004-05-26 Donovan Preston <dp(a)divmod.org>
* Add URLOverlay.keep, an API which lets you specify which query args
will be carried on from the current page render into the new URL.
2004-05-24 Matt Goodall <matt(a)pollenation.net>
* Extracted Fragment from Page. Hopefully, it will make it more obvious
that embedding an object with data_ and render_ methods in a stan tree is
possible without using something as "heavy" as Page which is really meant
to be a web resource.
2004-05-23 Donovan Preston <dp(a)divmod.org>
* Added some useful APIs to LiveEvil for manipulating the client-side
- flt(stan): Flatten some stan, quoting apostrophes as
- set(what, to): Set the contents of the client-side node
with the id 'what' to the stan 'to'.
- append(where, what): Append the stan 'what' to the client-
side node with the id 'where'
- alert(what): Show an alert to the user with the text "what"
Valentino Volonghi aka Dialtone
Linux User #310274, Proud Gentoo User
Home Page: http://xoomer.virgilio.it/dialtone/
The Python Imaging Library (PIL) adds image processing capabilities
to your Python interpreter. This library supports many file formats,
and provides powerful image processing and graphics capabilities,
including display support for Windows and Tkinter.
PIL 1.1.5 alpha 5 is now available from effbot.org:
(look for Imaging-1.1.5a5.tar.gz)
Changes in this release include:
+ Build improvements: Fixed building under AIX, improved detection of
FreeType2 and Mac OS X framework libraries, and more. Many thanks
to everyone who helped test the new "setup.py" script!
+ Added "getcolors()" method. This is similar to the existing histo-
gram method, but looks at color values instead of individual layers,
and returns an unsorted list of (count, color) tuples.
By default, the method returns None if finds more than 256 colors.
If you need to look for more colors, you can pass in a limit (this
is used to allocate internal tables, so you probably don't want to
pass in too large values).
+ Fixed BILINEAR/BICUBIC/ANTIALIAS filtering for mode "LA".
For a list of other changes, see this page:
for the pil team at secret labs ab
"Secret Labs AB -- makers of fine pythonware since 1997"