Peter Hoffmann Code Blog

This Week in Google with Chris Messina on OAuth and OpenID http://peter-hoffmann.com/2010/twig-chris-messina-oauth-openid.html

PubSubHubbub for Json http://peter-hoffmann.com/2010/pubsubhubbub-for-json.html

iterate over list by pairs

The following code will iterate over an list by pairs of elements:

from itertools import chain

def by_pairs(iterable):
    sequence = iter(iterable)
    previous = sequence.next()
    while True:
        current = sequence.next()
        yield previous, current
        previous = current

store json data with sqlobject in mysql database

Especially while developing the final database schema I find pleasant to store JSON data structures direct in the database. SQLObject has a PickleCol, but I has the downside, that serialized objects are binary based and thus not very well to debug.

The following codes need sqlobject as well as formencode:

from sqlobject import *
sqlhub.processConnection = connectionForURI("...")

import cjson
from formencode import validators


class JSONValidator(validators.Validator):

    def to_python(self, value, state):
        if value is None:
            return None
        try:
            return cjson.decode(value)
        except:
            raise validators.Invalid("expected a decodable JSON object in the JSONCol '%s', got %s %r instead" % (self.name, type(value), value), value, state)

    def from_python(self, value, state):
        if value is None:
            return None
        try:
            return cjson.encode(value)
        except:
            raise validators.Invalid("expected an encodable JSON object in the JSONCol '%s', got %s %r instead" % (self.name, type(value), value), value, state)


class SOJSONCol(SOStringCol):

    def createValidators(self):
        return [JSONValidator()] + super(SOJSONCol, self).createValidators()


class JSONCol(Col):
    baseClass = SOJSONCol


class Person(SQLObject):
    name = StringCol()
    data = JSONCol()

The code was adopted from http://openxcap.org/browser/xcap/interfaces/backend/sipthor.py

installing cloudkit on ubuntu 8.10

Installing cloudkit on ubuntu 8.10:

sudo aptitude install rubygems ruby1.8-dev libsqlite3-ruby libhttp-access2-ruby
sudo gem install hoe
sudo gem install sequel -v 2.6.0
sudo gem install thin
sudo gem install cloudkit

Create a configuration file config.ru

require 'cloudkit'
expose :notes, :projects

Start thin with:

/var/lib/gems/1.8/bin/thin -R config.ru start

For more Information about the api see http://getcloudkit.com/rest-api.html

vim toogle spell

Toggle spell language and on/off in vim

function! ToggleSpell()
    if &spell
        if &spelllang == "de"
            set spelllang=de,en
            echo "toggle spell" &spelllang
        elseif &spelllang == "de,en"
            set spelllang=en
            echo "toggle spell" &spelllang
        else
            set spell!
            echo "toggle spell off"
        endif
    else
        set spelllang=de
        set spell!
        echo "toogle spell" &spelllang
    endif
endfunction
map <F2> :call ToggleSpell()<CR>

blinkenlights project stereoscope in toronto

The latest blinkenlights project is to transform the Toronto City Hall into a big interactive computer matrix. The installation is part of the nuit blache.

The project also provides an iphone simulator (app store) which displays a live stream from Toronto.

More images can be found on flickr

http pubsub

I’m looking for some kind of http pubsub messaging server. Just found some blog posts but nothing I really like:

• Webhooks
• Joshua Schachter about beyond rest
• xmlrouter

working with uuids on python

Python has a module to generate different UUIDs like defined in RFC 4122. uuids include creation time. This recipe extracts the timestamp information:

def get_posixtime(uuid1):
     """Convert the uuid1 timestamp to a standard posix timestamp
    """
    assert uuid1.version == 1, ValueError('only applies to type 1')
    t = uuid1.time
    t = t - 0x01b21dd213814000
    t = t / 1e7
    return t

Another task is to sort uuids by creation time. The default cmp function seems not to do this:

...  
    int = ((time_low << 96L) | (time_mid << 80L) |
               (time_hi_version << 64L) | (clock_seq << 48L) | node)
def __cmp__(self, other):
    if isinstance(other, UUID):
        return cmp(self.int, other.int)
    return NotImplemented

But since uuids have time attribute, a list with them can be sorted with

from operator import attrgetter
lst = [..]
lst.sort(key=attrgetter("time"))

middleware to monitor python wsgi applications

I’m working on some RESTful python wsgi application. For debugging and monitoring I needed a way to see the request and response HTTP header without modifying the actual code.

So I posted the question to stackoverflow where Florian Boenisch came up with a nice solution.

from wsgiref.util import request_uri
import sys

def logging_middleware(application, stream=sys.stdout):
    def _logger(environ, start_response):
        stream.write('REQUEST\n')
        stream.write('%s %s\n' %(
            environ['REQUEST_METHOD'],
            request_uri(environ),
        ))

        for name, value in sorted(environ.items()):
            if name.startswith('HTTP_'):
                stream.write('    %s: %s\n' %(
                    name[5:].title().replace('_', '-'),
                    value,
                ))
        stream.flush()
        def _start_response(code, headers):
            stream.write('RESPONSE\n')
            stream.write('%s\n' % code)
            for data in sorted(headers):
                stream.write('    %s: %s\n' % data)
            stream.flush()
            start_response(code, headers)
        return application(environ, _start_response)
    return _logger

def application(environ, start_response):
    start_response('200 OK', [
        ('Content-Type', 'text/html')
    ])
    return ['Hello World']

if __name__ == '__main__':
    logger = logging_middleware(application)
    from wsgiref.simple_server import make_server
    httpd = make_server('', 1234, logger)
    httpd.serve_forever()

This middelware is is a example of accessing the response from the wrapped middleware.

The output will be something like this:

REQUEST
PUT http://localhost:5000/testdb/1
    Accept-Encoding: identity
    Host: localhost:5000
    User-Agent: Python-httplib2/$Rev: 196 $
RESPONSE
201 CREATED
    content-type: application/json
    ['{"rev": "1222169702", "ok": true, "id": "1"}']

REQUEST
GET http://localhost:5000/_all_dbs
    Accept-Encoding: compress, gzip
    Host: localhost:5000
    User-Agent: Python-httplib2/$Rev: 196 $
RESPONSE
200 OK
    Content-Type: text/plain; charset=utf-8
   ['["testdb"]']

REQUEST
DELETE http://localhost:5000/testdb
    Accept-Encoding: identity
    Host: localhost:5000
    User-Agent: Python-httplib2/$Rev: 196 $
RESPONSE
200 OK
    Content-Type: text/plain; charset=utf-8
    ['{"ok":true}']