import urllib2

import requests

import pdb

import time

class PyBingException(Exception):

pass

class PyBingSearch(object):

# QUERY_URL = 'https://api.datamarket.azure.com/Bing/Search/v1/Composite' \

# + '?Sources={}&Query={}&$top={}&$skip={}&$format={}'

QUERY_URL = 'https://api.datamarket.azure.com/Bing/Search/v1/Image' \

+ '?Query={}&$top={}&$skip={}&$format={}'

# QUERY_URL = 'https://api.datamarket.azure.com/Bing/Search/v1/Web' \

# + '?Query={}&$top={}&$skip={}&$format={}'

# proxies = {

# "http": "http://privateproxy.gq"

# }

def __init__(self, api_key, safe=False):

self.api_key = api_key

self.safe = safe

def search(self, query, limit=50, offset=0, format='json'):

return self._search(query, limit, offset, format)

def search_all(self, query, limit=50, format='json'):

results = self._search(query, limit, 0, format)

while results.total > len(results) and len(results) < limit:

max = limit - len(results)

more_results = self._search(query, max, len(results), format)

results += more_results

print len(results)

return results

def _search(self, query, limit, offset, format):

'''

Returns a list of result objects, with the url for the next page bing search url.

'''

url = self.QUERY_URL.format(urllib2.quote("'{}'".format(query)), limit, offset, format)

#r = requests.get(url, auth=("", self.api_key), proxies=self.proxies)

r = requests.get(url, auth=("", self.api_key))

print url

try:

json_results = r.json()

except ValueError as vE:

if not self.safe:

print ("Request returned with code %s, error msg: %s" % (r.status_code, r.text))

#raise PyBingException("Request returned with code %s, error msg: %s" % (r.status_code, r.text))

return [],r.status_code

else:

print "[ERROR] Request returned with code %s, error msg: %s. \nContinuing in 5 seconds." % (r.status_code, r.text)

time.slee(5)

try:

next_link = json_results['d']['__next']

except KeyError as kE:

print "Couldn't extract next_link: KeyError: %s" % kE

next_link = None

return [Result(single_result_json) for single_result_json in json_results['d']['results']], next_link

class Result(object):

'''

The class represents a SINGLE search result.

Each result will come with the following:

#For the actual results#

title: title of the result

url: the url of the result

description: description for the result

id: bing id for the page

#Meta info#:

meta.uri: the search uri for bing

meta.type: for the most part WebResult

'''

class _Meta(object):

'''

Holds the meta info for the result.

'''

def __init__(self, meta):

self.type = meta['type']

self.uri = meta['uri']

def __init__(self, result):

self.source_url = result['SourceUrl']

self.media_url = result['MediaUrl']

self.title = result['Title']

#self.meta = self._Meta(result['__metadata'])

# def __init__(self, result):

# self.url = result['Url']

# self.title = result['Title']

# self.description = result['Description']

# self.id = result['ID']

# self.meta = self._Meta(result['__metadata'])