Simple cli for pages jaunes

PagesJaunes-CLI.py

import requests
import urllib
import bs4
import json
import base64

def parse_card(elem):
	result = {
		"name":"",
		"phone":"",
		"street":"",
		"town":"",
		"postal-code":""
	}
	
	person = elem.findAll("span",{"class","bi-pos-links pj-lb pj-link"})[0]
	data =  person.attrs["data-pjlb"]
	info = json.loads(data)
	url = base64.b64decode(info["url"])

	if ":80" in url:
		#We do not care about companies
		return result
	
	r = requests.get(url)

	if r.status_code != 200:
		print "Error while requesting data"
		return result
	
	document = bs4.BeautifulSoup(r.text,"lxml")

	temp = {}

	temp["name"] =  document.find(itemprop="name")
	temp["phone"] = document.find("span",{"class","coord-numero"})
	temp["street"] = document.find(itemprop="streetAddress")
	temp["town"] = document.find(itemprop="addressLocality")
	temp["postal-code"] = document.find(itemprop="postalCode")

	for key in temp:
		#We need to make sure we are not trying to call null objects (no data on some fields)
		if temp[key] != None:
			result[key] = temp[key].text.strip().encode("UTF-8")

	return result

def search(name,town):
	endpoint = "http://www.pagesjaunes.fr/pagesblanches/recherche?"
	form_data = {
		"quoiqui": name,
		"ou": town
	}

	results = []

	r = requests.post(endpoint+urllib.urlencode(form_data),data=form_data)
	if r.status_code != 200:
		print "Error"
		return

	document = bs4.BeautifulSoup(r.text,"lxml")
	
	for elem in document.find_all("header"):
		if elem.attrs["class"][0] == "v-card":
			card = parse_card(elem)

			if card["name"] != "":
				results.append(card)
	print "\n----- Found {} result(s) -----".format(len(results))
	for record in results:
		for key in record:
			print "[{}] --> {}".format(key,record[key])
		print "----------\n"
	




name = raw_input("Enter name : ")
town = raw_input("Enter town : ")

search(name,town)