pyva/pyva/src/main/python/internets/api/google.py

   1 import json
   2 import re
   3 from urllib import urlencode
   4 from feed import HtmlFeed, XmlFeed, get_json
   5 from utils import unescape
   6
   7 class Google(object):
   8         def __init__(self):
   9                 pass
  10
  11         # Google shutdown their iGoogle "API",
  12         #def calc(self, expr):
  13         #       url = 'http://www.google.com/ig/calculator?'
  14         #       url += urlencode({'q': expr})
  15         #       page = HtmlFeed(url) #, fake_ua=True)
  16         #       text = page.html()
  17         #       # Convert to valid JSON: {foo: "1"} -> {"foo" : "1"},
  18         #       #                        {"foo": "\x25"} -> {"foo": "\u0025"}
  19         #       result = re.sub("([a-z]+):", '"\\1" :', text)
  20         #       # XXX: HACK; using two substitutes instead of one, but I can't into regex
  21         #       result = re.sub('\\\\x([0-9a-f]{2})', '\\u00\\1', result)
  22         #       result = re.sub('\\\\x([0-9a-f]{1})', '\\u000\\1', result)
  23         #
  24         #       result = unicode(result, 'unicode_escape')
  25         #       result = unescape(result)
  26         #       # Special case: fractions; those are <sup></sup> unicode-slash <sub></sub>
  27         #       result = re.sub(u'<sup>([^<]*)</sup>\u2044<sub>([^<]*)</sub>', ' \\1/\\2', result)
  28         #       # Also un-html-ify <sup>/</sup> because google doesn't send proper plaintext
  29         #       result = re.sub('<sup>([^<]*)</sup>', '^\\1', result)
  30         #       # Same with <sub>.
  31         #       result = re.sub('<sub>([^<]*)</sub>', '_\\1', result)
  32         #       result = json.loads(result.encode('utf-8'))
  33         #       return result
  34
  35         def search(self, query, userip=None):
  36                 url = 'http://ajax.googleapis.com/ajax/services/search/web?v=1.0&'
  37                 parameters = {'q': query}
  38                 if userip:
  39                         parameters['userip'] = userip
  40                 url += urlencode(parameters)
  41                 json = get_json(url)
  42                 return json
  43
  44         def yt_search(self, query, num=1):
  45                 url = 'https://gdata.youtube.com/feeds/api/videos?v=2&max-results=5&'
  46                 url += urlencode({'q': query})
  47                 xml = XmlFeed(url)
  48                 entry = xml.elements('/feed/entry[%d]' % num)
  49                 if not entry:
  50                         return None
  51
  52                 v = entry[0]
  53                 return {
  54                         'title': v.text('title'),
  55                         'duration': v.int('media:group/yt:duration/@seconds'),
  56                         'uploaded': v.text('media:group/yt:uploaded'),
  57                         'id': v.text('media:group/yt:videoid'),
  58                         'rating': v.decimal('gd:rating/@average'),
  59                         'rate_count': v.int('gd:rating/@numRaters'),
  60                         'favorite_count': v.int('yt:statistics/@favoriteCount'),
  61                         'view_count': v.int('yt:statistics/@viewCount'),
  62                         'liked': v.int('yt:rating/@numLikes'),
  63                         'disliked': v.int('yt:rating/@numDislikes')
  64                         }