regluit/core/goodreads.py

import re
import json
import logging
from itertools import islice
from urllib import urlencode
from urlparse import urlparse, urlunparse, urljoin

import httplib
import oauth2 as oauth
from requests import request
from xml.etree import ElementTree as ET
import django.utils.encoding

import regluit.core
from regluit.core import bookloader
from regluit.core import models

# import parse_qsl from cgi if it doesn't exist in urlparse
try:
  from urlparse import parse_qsl
except:
  from cgi import parse_qsl

from django.conf import settings

logger = logging.getLogger(__name__)

# QUESTION: should the request_token, access_token be part of the state of the client?
# for simplicity for now, I will make them part of the state of GoodReadsClient

class GoodreadsException(Exception):
    pass

class GoodreadsAuthorizationRequired(GoodreadsException):
    pass

def filter_none(d):
    d2 = {}
    for (k,v) in d.iteritems():
        if v is not None:
            d2[k] = v
    return d2

def safe_strip(a_string): 
    if a_string: 
        return a_string.strip()
    else: 
        return '' 

class GoodreadsClient(object):
    
    url = 'http://www.goodreads.com'
    request_token_url = urljoin(url,'oauth/request_token')
    authorize_url = urljoin(url, '/oauth/authorize')
    access_token_url = urljoin(url,'/oauth/access_token')  
  
    def __init__(self,key,secret,user=None, access_token=None):
        self.key = key
        self.secret = secret
        self.consumer = oauth.Consumer(key=self.key,
                              secret=self.secret)
        
        self.client = oauth.Client(self.consumer)
        #self.unauth_client = None
        
        if access_token is not None:
          self.__load_access_token(access_token)
        else:
          self.access_token = None
          
        if user is not None:
          self.load_user_access_token(user)
          
    @property     
    def is_authorized(self):
        return (self.access_token is not None)

    def begin_authorization (self, callback_url=None):
        # get request token
        response, content = self.client.request(GoodreadsClient.request_token_url, 'GET')
        
        if int(response['status']) != httplib.OK:
          raise Exception('Invalid response: %s' % response['status'])
        
        request_token = dict(parse_qsl(content))
        
        q = {'oauth_token':request_token['oauth_token']}
        if callback_url is not None:
          q['oauth_callback'] = callback_url
        
        authorize_link = GoodreadsClient.authorize_url + '?' + urlencode(q)
        return (authorize_link, request_token)
        
    def complete_authorization(self, request_token):
      token = oauth.Token(request_token['oauth_token'],
                          request_token['oauth_token_secret'])
      
      self.client = oauth.Client(self.consumer, token)
      response, content = self.client.request(GoodreadsClient.access_token_url, 'POST')
      if int(response['status']) != httplib.OK:
          raise Exception('Invalid response: %s' % response['status'])
      
      access_token_raw = dict(parse_qsl(content))
      self.__load_access_token(access_token_raw)
      return access_token_raw
    
    def load_user_access_token(self,user):
      access_token = {'oauth_token':user.profile.goodreads_auth_token,
                      'oauth_token_secret':user.profile.goodreads_auth_secret}
      self.__load_access_token(access_token)
      
    def __load_access_token(self, access_token):
        token = oauth.Token(access_token['oauth_token'],
                            access_token['oauth_token_secret'])
        self.access_token = token
        self.client = oauth.Client(self.consumer, self.access_token)        
  
    def __clear_access_token(self):
        self.access_token = None
        self.consumer = oauth.Consumer(key=self.key,
                              secret=self.secret)

    def auth_user(self):
        if self.is_authorized:
            response, content = self.client.request('%s/api/auth_user' % GoodreadsClient.url,
                                         'GET')
            if int(response['status']) != httplib.OK:
                raise GoodreadsException('Error authenticating Goodreads user ' )
            else:
                doc = ET.fromstring(content)
                user = doc.find('user')
                userid = user.get('id')
                name = user.find('name').text
                link = user.find('link').text
                return({'userid':userid, 'name':name, 'link':link})
        else:
            raise GoodreadsAuthorizationRequired('Attempt to access auth_user without authorization.')
      
    def add_book(self, book_id=871441, shelf_name='to-read'):
      # the book is: "Moby-Dick: A Pop-Up Book" 871441
      body = urlencode({'name': 'to-read', 'book_id': book_id})
      headers = {'content-type': 'application/x-www-form-urlencoded'}
      response, content = self.client.request('%s/shelf/add_to_shelf.xml' % GoodreadsClient.url,
                                         'POST', body, headers)
      # check that the new resource has been created
      if int(response['status']) != httplib.CREATED:
          raise GoodreadsException('Cannot create resource: %s' % response['status'])
          logger.info('response,content: %s | %s ' % (response,content))
      else:
          return True
        
    def review_list_unauth(self, user_id, shelf='all',page=1,sort=None,per_page=20,order='a',search=None,v=2):
        path="/review/list.xml"
        method = "GET"
        params = filter_none({'id':user_id,'shelf':shelf,'page':page,'sort':sort,'per_page':per_page,'order':order,
                'search':search, 'v':2})
        params["key"] = self.key
        
        request_url = urljoin(GoodreadsClient.url, path)
        logger.info("request_url:{0}, params: {1}".format(request_url, params))
        
        more_pages = True
        
        while (more_pages):
        
          r = request(method,request_url,params=params)
          # print request_url, params
          if r.status_code != httplib.OK:
              raise GoodreadsException('Error in review_list_unauth, http status_code: {0}'.format(r.status_code))
          else:
              doc = ET.fromstring(r.content.encode('utf-8'))
              # for the moment convert to a iterable of book data presented as dict -- one the way to paging through all results
              reviews = doc.findall('reviews/review')
              for review in reviews:
                  yield ({'id':review.find('id').text,
                          'book': {'id': safe_strip(review.find('book/id').text),
                                   'isbn10': review.find('book/isbn').text,
                                   'isbn13': review.find('book/isbn13').text,
                                   'title': safe_strip(review.find('book/title').text),
                                   'text_reviews_count': safe_strip(review.find('book/text_reviews_count').text),
                                   'link': safe_strip(review.find('book/link').text),
                                   'small_image_url': safe_strip(review.find('book/small_image_url').text),
                                   'ratings_count': safe_strip(review.find('book/ratings_count').text),
                                   'description': safe_strip(review.find('book/description').text)}
                          })
              if len(reviews) == 0:
                  more_pages = False
              else:
                  params["page"] += 1       
    
            
    def review_list(self, user_id, shelf='all',page=1,sort=None,per_page=20,order='a',search=None,v=2):
        """have to account for situation in which we might need authorized access
        for now:  assume no need for auth
        sort: available_for_swap, position, num_pages, votes, recommender, rating, shelves, format,
        avg_rating, date_pub, isbn, comments, author, title, notes, cover, isbn13, review, date_pub_edition,
        condition, asin, date_started, owned, random, date_read, year_pub, read_count, date_added,
        date_purchased, num_ratings, purchase_location, date_updated (optional)
        """
        
        path="/review/list.xml"
        method = "GET"
        params = filter_none({'id':user_id,'shelf':shelf,'page':page,'sort':sort,'per_page':per_page,'order':order,
                'search':search, 'v':2})
        
        request_url = urljoin(GoodreadsClient.url, path)
        
        more_pages = True
        
        while (more_pages):
        
          response, content = self.client.request('%s?%s' % (request_url, urlencode(params)),
                            method)
          if int(response['status']) != httplib.OK:
              raise GoodreadsException('Error in review_list:  ' )
          else:
              #logger.info(' %s' % (content))
              doc = ET.fromstring(content)
              # for the moment convert to a iterable of book data presented as dict -- one the way to paging through all results
              reviews = doc.findall('reviews/review')
              for review in reviews:
                  yield ({'id':review.find('id').text,
                          'book': {'id': review.find('book/id').text.strip(),
                                   'isbn10':review.find('book/isbn').text,
                                   'isbn13':review.find('book/isbn13').text,
                                   'title':review.find('book/title').text.strip(),
                                   'text_reviews_count':review.find('book/text_reviews_count').text.strip(),
                                   'link':review.find('book/link').text.strip(),
                                   'small_image_url':review.find('book/small_image_url').text.strip(),
                                   'ratings_count':review.find('book/ratings_count').text.strip(),
                                   'description':review.find('book/description').text.strip()}
                          })
              if len(reviews) == 0:
                  more_pages = False
              else:
                  params["page"] += 1 
        
    def shelves_list(self,user_id,page=1):
        """BUG to fix:  should go through all the pages, not just page 1
        """
        path = "/shelf/list.xml"
        params = {'user_id':user_id, 'page':page}
        params["key"] = self.key
        method = "GET"
        request_url = urljoin(GoodreadsClient.url, path)
        
        r = request(method,request_url,params=params)
        
        if r.status_code != httplib.OK:
            raise GoodreadsException('Error in shelves_list: %s ' % (r.headers))
        else:
            logger.info('headers: %s' % (r.headers))
            doc = ET.fromstring(r.content.encode('utf-8'))
            shelves = doc.find('shelves')
            # do a simple parsing to a dictionary
            
            d = dict(  [ (k,int(shelves.attrib[k])) for k in shelves.attrib ]  )
            d["user_shelves"] = [{'name':shelf.find('name').text,
                                  'book_count':int(shelf.find('book_count').text),
                                  'description':shelf.find('description').text if shelf.find('description').attrib['nil'] != 'true' else None,
                                  'exclusive_flag':shelf.find('exclusive_flag').text} \
                for shelf in shelves.findall('user_shelf')]
  
            d["total_book_count"] = sum([shelf['book_count'] if shelf['exclusive_flag'] == 'true' else 0 for shelf in d["user_shelves"]])
            return d
        
        
def load_goodreads_shelf_into_wishlist(user, shelf_name='all', goodreads_user_id=None, max_books=None, expected_number_of_books=None):
    """
    Load a specified Goodreads shelf (by default:  all the books from the Goodreads account associated with user)
    """
    
    logger.info('Entering load_goodreads_shelf_into_wishlist.  user: %s, shelf_name: %s, goodreads_user_id: %s, max_books: %s, expected_number_of_books: %s',
                user, shelf_name, goodreads_user_id, max_books, expected_number_of_books)
    gc = GoodreadsClient(key=settings.GOODREADS_API_KEY, secret=settings.GOODREADS_API_SECRET, user=user)
        
    if goodreads_user_id is None:
      if user.profile.goodreads_user_id is not None:
        goodreads_user_id = user.profile.goodreads_user_id
      else:
        raise Exception("No Goodreads user_id is associated with user.")
        
    logger.info('computed goodreads_user_id: %s ', goodreads_user_id)
    
    for (i, review) in enumerate(islice(gc.review_list(goodreads_user_id,shelf=shelf_name),max_books)):
        isbn = review["book"]["isbn10"] if review["book"]["isbn10"] is not None else review["book"]["isbn13"]
        logger.info("%d %s %s %s ", i, review["book"]["title"], isbn, review["book"]["small_image_url"])
        try:
            edition = bookloader.add_by_isbn(isbn)
            if not edition:
                continue
            # save the goodreads id since we know it at this point
            # we need to extract it from the link since review['id']
            # is the id for a users review, not the book
            link = review['book']['link']
            match = re.search('/show/(\d+)', link)
            if match:
                identifier= models.Identifier.get_or_add(type = 'gdrd', value = match.group(1), edition = edition, work = edition.work)
                user.wishlist.add_work(edition.work, 'goodreads')
                logger.info("Work with isbn %s added to wishlist.", isbn)
            else:
                logger.error("unable to extract goodreads id from %s", link)
            if edition.new:
                regluit.core.tasks.populate_edition.delay(edition.isbn_13)

        except Exception, e:
            logger.info ("Exception adding ISBN %s: %s", isbn, e) 

    logger.info('Leaving load_goodreads_shelf_into_wishlist.  Length of wishlist for user %s is %s', user, len(user.wishlist.works.all()))

    return user.wishlist