bogsio

python manage.py syncdb

bogsio

pip install Django South
django-admin.py startproject tptutorial

bogsio

mkdir tastypie-tutorial
cd tastypie-tutorial/
virtualenv tptut
pip install Django South
source tptut/bin/activate

bogsio

from django.core.servers.basehttp import FileWrapper
from django.http import HttpResponse
import tempfile
import urllib

# ...

def xml_view(request):
    # ...
    filename, _str = generate_xml()

bogsio

def generate_xml():
    return 'items.xml', '<items><item>Item 1</item></items>'

bogsio

print convert("death", WN_NOUN, WN_VERB)
print convert("story", WN_NOUN, WN_VERB)
print convert("boring", WN_ADJECTIVE, WN_NOUN)
print convert("trouble", WN_NOUN, WN_ADJECTIVE)
print convert("solve", WN_VERB, WN_ADJECTIVE_SATELLITE)
print convert("think", WN_VERB, WN_ADJECTIVE)

[('die', 0.75), ('end', 0.2), ('decease', 0.05)]
[('report', 0.2222222222222222), ('tell', 0.2222222222222222), ('narrate', 0.2222222222222222),...
[('tedium', 0.3333333333333333), ('dullness', 0.16666666666666666),...

bogsio

from nltk.corpus import wordnet as wn

# Just to make it a bit more readable
WN_NOUN = 'n'
WN_VERB = 'v'
WN_ADJECTIVE = 'a'
WN_ADJECTIVE_SATELLITE = 's'
WN_ADVERB = 'r'

def convert(word, from_pos, to_pos):    

bogsio

print index.lookup('loves')
['Tunnel of Love', 'Whole Lotta Love', "Since I've Been Loving You"]
print index.lookup('loved')
['Tunnel of Love', 'Whole Lotta Love', "Since I've Been Loving You"]
print index.lookup('daze')
['Dazed and Confused']
print index.lookup('confusion')
['Dazed and Confused']

bogsio

# TOP10 Dire straits
index.add('Industrial Disease')
index.add('Private Investigations')
index.add('So Far Away')
index.add('Twisting by the Pool')
index.add('Skateaway')
index.add('Walk of Life')
index.add('Romeo and Juliet')
index.add('Tunnel of Love')
index.add('Money for Nothing')

bogsio

import nltk
from collections import defaultdict
from nltk.stem.snowball import EnglishStemmer

class Index:
	""" Inverted index datastructure """
	
	def __init__(self, tokenizer, stemmer=None, stopwords=None):
		"""
		tokenizer	-- NLTK compatible tokenizer function