Google Reader to Lookbook.nu (subscriptions → fans, stars → hypes)

.gitignore

.*.swp
*.pyc
credentials.ini

credentials.ini.dist

; Copy this and rename it to credentials.ini

[google]
username = [email protected]
password = yourpassword

[lookbook]
username = [email protected]
password = yourpassword

credentials.py

import ConfigParser

from lxml.html import document_fromstring
from requests import Session

__all__ = 'google', 'login_lookbook', 'lookbook'


config = ConfigParser.SafeConfigParser()
config.read('credentials.ini')

google = (config.get('google', 'username'),
          config.get('google', 'password'))

lookbook = (config.get('lookbook', 'username'),
            config.get('lookbook', 'password'))


def login_lookbook(username=lookbook[0], password=lookbook[1]):
    session = Session()
    session.headers['Origin'] = 'http://lookbook.nu'
    # authenticity_token
    url = 'http://lookbook.nu/login'
    response = session.get(url)
    html = document_fromstring(response.content)
    form = html.xpath('//form[@action={0!r}][@method="post"]'.format(url))[0]
    authenticity_token, = form.xpath('.//input[@name="authenticity_token"]'
                                     '/@value')
    utf8, = form.xpath('.//input[@name="utf8"]/@value')
    form = {
        'user[email]': username,
        'user[password]': password,
        'authenticity_token': authenticity_token,
        'utf8': utf8.encode('utf-8')
    }
    response = session.post(url, headers={'Referer': url}, data=form)
    assert 'http://lookbook.nu/logout' in response.content, \
           'You seem to fail to login'
    return session

fans.py

"""Migrate Google Reader subscriptions to Lookbook.nu fans."""
from libgreader import ClientAuthMethod, GoogleReader
from lxml.html import document_fromstring

from credentials import google, login_lookbook

auth = ClientAuthMethod(*google)
reader = GoogleReader(auth)

reader.buildSubscriptionList()
feeds = reader.getSubscriptionList()
lookbook_feeds = [feed for feed in feeds if 'lookbook.nu/' in feed.feedUrl]
lookbook_urls = {feed.siteUrl for feed in lookbook_feeds
                              if feed.siteUrl and 'lookbook.nu' in feed.siteUrl}

lookbook = login_lookbook()
for i, lookbook_url in enumerate(lookbook_urls):
    print '[{0}/{1}] {2} ->'.format(
        i + 1,
        len(lookbook_urls),
        lookbook_url
    ),
    response = lookbook.get(lookbook_url)
    lookbook_html = document_fromstring(response.content)
    csrf_token = lookbook_html.xpath('//meta[@name="csrf-token"]/@content')[0]
    fan_url = lookbook_html.xpath('//a[@id="become_a_fan_btn"]/@href')[0]
    print fan_url,
    response = lookbook.post(fan_url, headers={
        'X-Requested-With': 'XMLHttpRequest',
        'X-CSRF-Token': csrf_token
    })
    print '->', response.status_code
    assert response.status_code == 200

hypes.py

"""Migrate Google Reader stars to Lookbook.nu hypes."""
import datetime
import json
import re
import time

from libgreader import ClientAuthMethod, GoogleReader
from lxml.html import document_fromstring

from credentials import google, login_lookbook

auth = ClientAuthMethod(*google)
reader = GoogleReader(auth)

reader.makeSpecialFeeds()
feed = reader.getSpecialFeed('starred')


def iter_feeds(feed):
    feed.loadItems(loadLimit=100)
    loaded = 0
    while 1:
        print '(loaded more)'
        for item in feed.getItems()[loaded:]:
            yield item
        loaded += feed.lastLoadLength
        feed.loadMoreItems(loadLimit=100)


lookbook = login_lookbook()
oldest_limit = datetime.datetime(2008, 1, 1)
hyped_ids_re = re.compile(r'var\s+hypedLookIds\s*=\s*(\[[\d,]*\])\s*;')
look_id_re = re.compile(r'^https?://(?:www\.)?lookbook\.nu/look/(\d+)')
hyped_looks = set()

for item in iter_feeds(feed):
    updated = datetime.datetime.utcfromtimestamp(item.data['updated'])
    if updated < oldest_limit:
        break
    if not item.url or '/lookbook.nu/' not in item.url:
        continue
    print item.url[7:], '->',
    response = lookbook.get(item.url)
    if (response.url == 'http://lookbook.nu/' and
        response.history[-1].status_code == 302):
        print 'GONE'
    look_id = int(look_id_re.match(item.url).group(1))
    if look_id in hyped_looks:
        print 'HYPED'
        continue
    hyped_looks.add(look_id)
    hyped_ids_match = hyped_ids_re.search(response.content)
    if hyped_ids_match:
        if look_id in json.loads(hyped_ids_match.group(1)):
            print 'HYPED'
            continue
    html = document_fromstring(response.content)
    csrf_token = html.xpath('//meta[@name="csrf-token"]/@content')[0]
    hype_url = html.xpath('//*[@class="look_hype"]/@hypeurl')[0]
    print hype_url[7:],
    response = lookbook.get(hype_url, headers={
        'X-Requested-With': 'XMLHttpRequest',
        'X-CSRF-Token': csrf_token
    })
    print '->', response.status_code
    assert response.status_code == 200
    time.sleep(5)

requirements.txt

libgreader >= 0.7.0
lxml >= 3.2.0
requests >= 1.2.0