Category: Python

Over the weekend I signed up for PyCon DC 2005, and I’m looking forward to it already. The earlybird deadline has already passed, but PyCon is still probably one of the best values out there as far as geek conferences go. I’m in the DC area so if anyone is going to be in town and would like to get together, let me know and we’ll figure something out. PyCon week also happens to fall during my spring break, which couldn’t have worked out better.

See you there!

Here is one more reason to attend PyCon 2005:

#16. Python for Series 60

Erik Smartt / Nokia

UPDATE 2005-01-04:

A presentation where the Product Manager and one of the engineers do a joint presentation to quickly introduce and demonstrate the port, then cover the technical issues of porting and developing extensions

In addition to the planned presentation, it would be great to get together a Series 60 Python BoF session or some time in the open space for further meets, greets, and hacking. What do you think? Let us know in the wiki.

This gem showed up in my inbox tonight:

You will be happy to know that the PyCon Program Committee, after lengthy deliberations, has now finalized the program for PyCon DC 2005. I can tell you that the decision-making was very difficult, as the standard of submissions was even higher than last year.

The preliminary schedule is online.

It looks like a great collection of stuff this year. I’m going to do my best to make it there and will try to organize a Python for Series 60 BoF or open session. I hope to see everyone there this year!

Steve Holden:

Following my last message, I am pleased to be able to
announce that you can register for PyCon DC 2005 on the
web at

http://www.python.org/pycon/2005/register.html

Rock on! PyCon DC is probably one of the best bang/buck conferences out there, and it’s right in my backyard! I hope to see everyone there this year.

Jim, Ewan, and Erik have already spoken about it, but Python for Series 60 is in the wild!

I’ve been tinkering with the public release and should have some fun stuff to share shortly.

Edd and others have been covering the big news: IronPython, a really fast implementation of Python that runs on .NET and Mono, has finally been released as open source.  You can read more and snag it at IronPython.com.  The author, Jim Hugunin, has also been snagged by Microsoft.

For the past month or two, I have been tinkering with Python for Series 60. I regret that I have not had enough time to pound on it, but what I have played with has been awesome. I can parse HTML, like an adaptation from this page of Dive Into Python. When things like from sgmllib import SGMLParser just work, I’m amazed. Hey, check that out, I’m parsing HTML on my phone. Rock!

Series 60 Python is still a research project at Nokia. There are no guarantees that it will ever see the light of day beyond limited testing. There’s only one problem: sockets. It always comes back to sockets. Don’t get me wrong, sockets are available in Series 60 Python, but you can’t do something like urllib.urlopen(url). That little thing is keeping a lot of people from doing useful things with Series 60 Python. Part of the reason that so many people use and love Python is because it’s sort of runnable pseudocode. It’s intuitive, it makes sense, and the syntax that you think ought to work often does. Once writing a Python app for a platform becomes much harder than that, you start loosing developers quickly. So many killer mobile apps that would be awesome to code in Python involve the following steps:

1. Grab a specific resource (XML, HTML, etc)
2. parse the resource and do something with the data
3. interpret and present the data to the user

If the first step is overly complicated, or doesn’t work as Python programmers expect it to, you’re going to loose a lot of potential developers quickly. There’s a bit of a catch 22 though: Nokia needs to be sure that Python for Series 60 is a worthwhile project, so they’ve seeded it to developers. However the developers are frustraed because they can’t just type urllib.urlopen(url). Which came first, the useful apps taking full advantage of a connected mobile platform, or the infrastructure to do so?

It is obvious that Nokia has put a lot of time and effort in to Series 60 Python. There’s a lot beyond your standard Python distro buried in there. It would be an absolute tragedy to see this die half-ripened on the vine. Hey Nokia: you’ve done a great job so far, keep it up! Hey developers and bloggers: make some noise! Play with it. Have fun. Post your thoughts. Let them know what you think.

Further Reading:

• Russ: Amaretto Thoughts
• Jim: Are All Almonds Bitter?

The Daily Python-URL is chock full of Python linkage today.  The one that has the biggest wow factor to me is a link to Python development with Eclipse and Ant at IBM developerWorks.  The article, which I managed to miss when it came out on June 15, details how to set up Eclipse as a Python development environment.  The story glues together PyDev, a Python Eclipse plugin, Python Ant tasks, and even manages to make mention of Mark’s feed parser.

The eclipse solution sounds like it would be much more elegant and coherent than my current uses of Pythonwin and IDLE on Windows and the text editor of the moment, the command line, and the interactive interpreter on Linux (usually over SSH).

The Universal Feed Parser turned 3.0 yesterday.  It has a new home, online docs, and oodles of tests.  It also makes julian fries.  Seriously though, Mark’s feed parser can snag data from pretty much any feed you throw at it, no matter how whacked out it may be.

PyPI listed this little gem this evening: paramiko.  What is paramiko?  It is an SSH2 implementation in Python.  Here’s some more information from PyPI:

This is a library for making SSH2 connections (client or server). Emphasis is on using SSH2 as an alternative to SSL for making secure connections between python scripts. All major ciphers and hash methods are supported.

SFTP client mode is now supported too.

paramiko relies on PyCrypt, which is like a Pythonic Bouncy Castle.

EWeek:

For about six hours yesterday, the staff of eWEEK.com was treated to a lesson in emergency improvisation. But thanks to 123 lines of Python code, running on a $7-a-month personal Web site on an Apache Web server somewhere in California, you probably didn’t notice.

This is an excellent story of improvisation at its finest.

Rawdog has been segfaulting on me for awhile now.  It was definitely not Adam Sampson’s fault, it was totally mine.  I aparently copied and pasted an HTML link to my feed list instead of an RSS feed.  Oops!

Anyway, I’ve not been able to get my RSS fix for some time, and I finally got around to writing a few lines of Python to diagnose the problem.  In about 9 lines of code I’m able to read my config file line by line, check to see if the line represents a feed listing, grab the url from the line, print the feed url, and parse it using Mark’s feed parser (the same version that Rawdog is using).

The feed section of a Rawdog config file (usually in ~/.rawdog/config) can look like either of the following:

feed 60 http://postneo.com/rss.xml
feed 1h http://postneo.com/rss.xml

The top line is what the older version of Rawdog uses, the lower line is what newer versions of Rawdog uses.  Luckily each line is “feed” + <time> + <url>, so the string can just be split and I can grab the URL with foo[2].

Here is the result of my two minute hack to figure out where the segfaults are coming from:

import feedparser

f=open('config', 'r')
for line in f.readlines():
  if line.startswith('feed'):
    foo = line.split()
    print 'parsing ' + foo[2]
    data = feedparser.parse(foo[2])
f.close()

This will skip over any comments (lines that start with #) and other directives in the Rawdog config file.  I ended up making a backup, running the checker until I came across an error, commented out the offending feed in the backup file, and then removed all feeds in the file I was checking up to and including the offending feed.  This way I wasn’t hammering feeds in the beginning of the list.  Rinse, later, repeat.  It wasn’t until down at the bottom that I found the link that was causing the segfault: a link to one of my posts.  Sheesh.

But hey, with a few lines of python and a few minutes, I’m back up and running.

PyCon DC 2004 is coming.  It happens March 24-26.  I should be there for sure on Wednesday, but I don’t know if I’ll be able to make it Thursday or Friday.  The schedule of talks makes me think that this is going to be the Python event of 2004.

The official Call for Participation at O’Reilly’s Open Source Convention 2004 went out today.  The general theme is “Opening the Future: Discover, Develop, Deliver,” which you can interpret to suit your needs.  Proposals for OSCON are due by Febrary 9.

Proposal submissions for PyCon DC 2004 are also online.  The deadline for proposals for PyCon DC 2004 is January 15.  More information is at the call for proposals page.

Jeremy Hylton:

The Python conference will be held March 20-26, 2004 at George Washington University in Washington, D.C. PyCon registration is open, and the final call for papers will be posted next week. Mitch Kapor is the keynote speaker.

I punked out last year because of funds, but there’s really no excuse not to attend this $175 conference.  I can’t wait!

According to the Daily Python-URL, Python 2.3.3c1 is out.  This release features major bugsquashing, tweaks to xml.processors.expat, and a digest auth fix in urllib2.

Via PyPI, OpenPGP in Python is just that.  The author is also learning about Python as he goes:

This is an attempt at implementing OpenPGP in Python. This is also my first attempt at doing anything real in Python so [insert disclaimers and cries for help here]. I’m classifying this as “Pre-Alpha” since major changes are still pretty frequent. For testing in the meantime, an API and fairly usable command line application are included.

PyOpenPGP requires PyCrypto.  Don’t let the humble author fool you, check out the demo.