#!/Library/Frameworks/Python.framework/Versions/Current/bin/python

from calais import Calais
import feedparser, time, math

API_KEY = 'ENTER_YOUR_KEY_HERE'
calais = Calais(API_KEY, submitter="python-calais")


feeds = ['http://feeds.boingboing.net/boingboing/iBag', \
			'http://www.foxnews.com/xmlfeed/rss/0,4313,0,00.rss', \
			'http://gizmodo.com/tag/top/index.xml', \
			'http://www.nytimes.com/services/xml/rss/nyt/HomePage.xml', \
			'http://rss.people.com/web/people/rss/topheadlines/index.xml', \
			'http://feeds.digg.com/digg/popular.rss', \
			'http://gawker.com/tag/top/index.xml' \
			]

lastPing = time.time()

dataBase = {}

for f in feeds:
	d = feedparser.parse(f)
	if(not d.feed.has_key('title')): d.feed.title = 'untitled'
	
	entitycount = 0
	values = {}
	
	print 'BLOG: ' + d.feed.title + ' has ' + `len(d.entries)` + ' elements'
	
	for e in d.entries[:15]:
		if 'summary' in e: summary=e.summary
		else: summary=e.description
		
		print 'Summary: ' + summary + "\n"
		
		elapsed = time.time()-lastPing
		if(elapsed < 0.25):
			time.sleep(0.25 - elapsed) #maxiumum api rate
		result = calais.analyze(summary.encode('utf-8'))
		lastPing = time.time()
		
		try:
			#print result.entities
		
		 	names = [x['name'] for x in result.entities]
			entities =  [x['_type'] for x in result.entities]
		
			for e in entities:
				if not e in values: values[e] = 0
				values[e] += 1
				entitycount += 1
		except AttributeError:
			print "No results from: " + summary.encode('utf-8')
		
	
	normalizedValues = {}
	for k in values.keys():	normalizedValues[k] = float(values[k]) / entitycount
	"""
	a = normalizedValues.keys()
	def comp(x, y):
		if normalizedValues[x]>normalizedValues[y]:
			return 1
		elif normalizedValues[x]==normalizedValues[y]:
			return 0
		else:
			return -1
	
	a.sort(cmp=comp)
	#for x in a:
	#	print x + ' = ' + `normalizedValues[x]` + "\n"
		
	"""
	
	dataBase[d.feed.title] = normalizedValues

blogList = dataBase.keys()
scoreTable = open('scores.csv', 'w')
scoreTable.write('Blogs')
for blog in blogList:
	scoreTable.write(',"'+blog+'"')
scoreTable.write("\n")
#process each blog results

allKeys = set()
for reference in blogList:
	scoreTable.write('"'+reference+'"')
	for comparison in blogList:
		score = 0.0
		if comparison != reference:
			for e in dataBase[reference].keys():
				allKeys.add(e)
				value = dataBase[reference][e]
				if e in dataBase[comparison]:
					difference = abs( value-dataBase[comparison][e] )
				else:
					difference = value	
				score += difference
			for j in dataBase[comparison].keys():
				if not j in dataBase[reference]:
					score += dataBase[comparison][j]
		print "Score between " + reference + " and " + comparison + ' is ' + `score`		
		scoreTable.write(','+`score`)
	scoreTable.write("\n")

print "all keys: " + `allKeys`

table = open('allblogdata.csv','w')
table.write('Blogs')
for k in allKeys:
	table.write(',"'+k+'"')
table.write("\n")
for b in blogList:
	table.write('"'+b+'"')
	for k in allKeys:
		if k in dataBase[b]:
			score = dataBase[b][k]
		else: score = 0.0
		table.write(',' + `score`)
	table.write("\n")