def refresh(self):
secrets = app_config.get_secrets()
if not self.project.domain:
return
url = 'http://%s%s' % (self.project.domain, self.project.prefix)
response = requests.get('https://free.sharedcount.com/url?apikey=%s&url=%s' % (secrets['SHAREDCOUNT_API_KEY'], url))
if response.status_code != 200:
print 'Failed to refresh social data from SharedCount: %i.' % response.status_code
return
print 'Updating social counts from SharedCount'
data = response.json()
self.facebook_likes = data['Facebook']['like_count'] or 0
self.facebook_shares = data['Facebook']['share_count'] or 0
self.facebook_comments = data['Facebook']['comment_count'] or 0
self.twitter = data['Twitter'] or 0
self.google = data['GooglePlusOne'] or 0
self.pinterest = data['Pinterest'] or 0
self.linkedin = data['LinkedIn'] or 0
self.stumbleupon = data['StumbleUpon'] or 0
self.last_update = timezone.now()
self.save()
def _test_api_init(freq=2627):
secrets = app_config.get_secrets()
url = secrets.get('VERB8TM_SRT_TEST_API', app_config.PROJECT_SLUG)
header = 'Content-Type: application/json'
payload = '{"command":"init","millisecondCaptionDelay":"%s"}' % (freq)
command = 'curl -H "%s" -X POST -d \'%s\' %s' % (header, payload, url)
local(command)
def get_goodreads_id(cls, isbn):
"""
Use GoodReads search API
"""
secrets = app_config.get_secrets()
goodreads_id = None
search_api_tpl = 'https://www.goodreads.com/search/index.xml'
params = {
'key': secrets['GOODREADS_API_KEY'],
'q': isbn.encode('utf-8')
}
query_string = urlencode(params)
search_api_url = '%s?%s' % (search_api_tpl, query_string)
# Get search api results.
r = requests.get(search_api_url, params=params)
if r.status_code == 200:
tree = ElementTree.fromstring(r.content)
best_book = tree.find('.//best_book')
if best_book is not None:
goodreads_id = best_book.find('id').text
else:
logger.warning('could not find a matching book for ISBN %s' % isbn)
else:
logger.warning('did not receive a 200 when using Goodreads search api')
return goodreads_id
def refresh(self):
secrets = app_config.get_secrets()
if not self.project.domain:
return
url = 'http://%s%s' % (self.project.domain, self.project.prefix)
response = requests.get(
'https://free.sharedcount.com/url?apikey=%s&url=%s' %
(secrets['SHAREDCOUNT_API_KEY'], url))
if response.status_code != 200:
print 'Failed to refresh social data from SharedCount: %i.' % response.status_code
return
print 'Updating social counts from SharedCount'
data = response.json()
self.facebook_likes = data['Facebook']['like_count'] or 0
self.facebook_shares = data['Facebook']['share_count'] or 0
self.facebook_comments = data['Facebook']['comment_count'] or 0
self.twitter = data['Twitter'] or 0
self.google = data['GooglePlusOne'] or 0
self.pinterest = data['Pinterest'] or 0
self.linkedin = data['LinkedIn'] or 0
self.stumbleupon = data['StumbleUpon'] or 0
self.last_update = timezone.now()
self.save()
def render_confs():
"""
Renders server configurations.
"""
require('settings', provided_by=[production, staging])
with settings(warn_only=True):
local('mkdir confs/rendered')
context = app_config.get_secrets()
context['PROJECT_SLUG'] = app_config.PROJECT_SLUG
context['CLOUD_SEARCH_DOMAIN'] = app_config.CLOUD_SEARCH_DOMAIN
context['PROJECT_NAME'] = app_config.PROJECT_NAME
context['DEPLOYMENT_TARGET'] = env.settings
context['CONFIG_NAME'] = env.project_slug.replace('-', '').upper()
for service, remote_path, extension in SERVICES:
file_path = 'confs/rendered/%s.%s.%s' % (app_config.PROJECT_SLUG,
service, extension)
with open('confs/%s.%s' % (service, extension), 'r') as read_template:
with open(file_path, 'wb') as write_template:
payload = Template(read_template.read())
write_template.write(payload.render(**context))
def fetch_posts():
"""
Returns a list of all tumblr posts, unsorted.
"""
print "Starting."
# Set constants
secrets = app_config.get_secrets()
base_url = 'http://api.tumblr.com/v2/blog/%s.tumblr.com/posts/photo' % app_config.TUMBLR_BLOG_ID
key_param = '?api_key=%s' % secrets['TUMBLR_APP_KEY']
limit_param = '&limit=20'
limit = 20
new_limit = limit
post_list = []
print base_url + key_param
# Figure out the total number of posts.
r = requests.get(base_url + key_param)
total_count = int(json.loads(r.content)['response']['total_posts'])
print "%s total posts available." % total_count
# Do the pagination math.
pages_count = (total_count / limit)
pages_remainder = (total_count % limit)
if pages_remainder > 0:
pages_count += 1
pages = range(0, pages_count)
print "%s pages required." % len(pages)
# Start requesting pages.
# Note: Maximum of 20 posts per page.
print "Requesting pages."
for page in pages:
# Update all of the pagination shenanigans.
start_number = new_limit - limit
end_number = new_limit
if end_number > total_count:
end_number = total_count
new_limit = new_limit + limit
page_param = '&offset=%s' % start_number
page_url = base_url + key_param + limit_param + page_param
# Actually fetch the page URL.
r = requests.get(page_url)
posts = json.loads(r.content)
for post in posts['response']['posts']:
try:
if 'NSFW' in post['tags']:
pass
elif 'nsfw' in post['tags']:
pass
else:
post_list.append(post)
except KeyError:
pass
return post_list
def fetch_posts():
"""
Returns a list of all tumblr posts, unsorted.
"""
print "Starting."
# Set constants
secrets = app_config.get_secrets()
base_url = 'http://api.tumblr.com/v2/blog/%s.tumblr.com/posts/photo' % app_config.TUMBLR_BLOG_ID
key_param = '?api_key=%s' % secrets['TUMBLR_APP_KEY']
limit_param = '&limit=20'
limit = 20
new_limit = limit
post_list = []
print base_url + key_param
# Figure out the total number of posts.
r = requests.get(base_url + key_param)
total_count = int(json.loads(r.content)['response']['total_posts'])
print "%s total posts available." % total_count
# Do the pagination math.
pages_count = (total_count / limit)
pages_remainder = (total_count % limit)
if pages_remainder > 0:
pages_count += 1
pages = range(0, pages_count)
print "%s pages required." % len(pages)
# Start requesting pages.
# Note: Maximum of 20 posts per page.
print "Requesting pages."
for page in pages:
# Update all of the pagination shenanigans.
start_number = new_limit - limit
end_number = new_limit
if end_number > total_count:
end_number = total_count
new_limit = new_limit + limit
page_param = '&offset=%s' % start_number
page_url = base_url + key_param + limit_param + page_param
# Actually fetch the page URL.
r = requests.get(page_url)
posts = json.loads(r.content)
for post in posts['response']['posts']:
try:
if 'NSFW' in post['tags']:
pass
elif 'nsfw' in post['tags']:
pass
else:
post_list.append(post)
except KeyError:
pass
return post_list
def local_reset_db():
secrets = app_config.get_secrets()
with settings(warn_only=True):
local('dropdb %s' % app_config.PROJECT_SLUG)
local('echo "CREATE USER %s WITH PASSWORD \'%s\';" | psql' % (app_config.PROJECT_SLUG, secrets['POSTGRES_PASSWORD']))
local('createdb -O %s %s' % (app_config.PROJECT_SLUG, app_config.PROJECT_SLUG))
def index():
"""
Example view demonstrating rendering a simple HTML page.
"""
context = make_context()
context['INSTAGRAM_CLIENT_ID'] = app_config.get_secrets()['INSTAGRAM_CLIENT_ID']
return render_template('index.html', **context)
def index():
"""
Example view demonstrating rendering a simple HTML page.
"""
context = make_context()
context['INSTAGRAM_CLIENT_ID'] = app_config.get_secrets(
)['INSTAGRAM_CLIENT_ID']
return render_template('index.html', **context)
def authenticate():
secrets = app_config.get_secrets()
auth = tweepy.OAuthHandler(secrets['TWITTER_CONSUMER_KEY'],
secrets['TWITTER_CONSUMER_SECRET'])
auth.set_access_token(secrets['TWITTER_ACCESS_KEY'],
secrets['TWITTER_ACCESS_SECRET'])
api = tweepy.API(auth)
return api
def verb8tm_stop():
"""
Stops the generation of content in Verb8tm test API endpoint
"""
secrets = app_config.get_secrets()
url = secrets.get('VERB8TM_SRT_TEST_API', app_config.PROJECT_SLUG)
header = 'Content-Type: application/json'
payload = '{"command":"stop"}'
command = 'curl -H "%s" -X POST -d \'%s\' %s' % (header, payload, url)
local(command)
def load_images():
"""
Downloads images from Baker and Taylor.
Eschews the API for a magic URL pattern, which is faster.
"""
# Secrets.
secrets = app_config.get_secrets()
# Open the books JSON.
with open("www/static-data/books.json", "rb") as readfile:
books = json.loads(readfile.read())
print "Start load_images(): %i books." % len(books)
# Loop.
for book in books:
# Skip books with no title or ISBN.
if book["title"] == "":
continue
if "isbn" not in book or book["isbn"] == "":
continue
# Construct the URL with secrets and the ISBN.
book_url = "http://imagesa.btol.com/ContentCafe/Jacket.aspx?UserID=%s&Password=%s&Return=T&Type=L&Value=%s" % (
secrets["BAKER_TAYLOR_USERID"],
secrets["BAKER_TAYLOR_PASSWORD"],
book["isbn"],
)
# Request the image.
r = requests.get(book_url)
path = "www/img/cover/%s.jpg" % book["slug"]
# Write the image to www using the slug as the filename.
with open(path, "wb") as writefile:
writefile.write(r.content)
file_size = os.path.getsize(path)
if file_size < 10000:
print "Image not available for ISBN: %s" % book["isbn"]
image = Image.open(path)
width = 250
height = int((float(width) / image.size[0]) * image.size[1])
image.thumbnail([width, height], Image.ANTIALIAS)
image.save(path.replace(".jpg", "-thumb.jpg"), "JPEG")
print "End."
def local_reset_db():
"""
Reset the database locally.
"""
secrets = app_config.get_secrets()
with settings(warn_only=True):
local('dropdb %s' % app_config.PROJECT_SLUG)
local('echo "CREATE USER %s WITH PASSWORD \'%s\';" | psql' % (app_config.PROJECT_SLUG, secrets['POSTGRES_PASSWORD']))
local('createdb -O %s %s' % (app_config.PROJECT_SLUG, app_config.PROJECT_SLUG))
def local_reset_db():
secrets = app_config.get_secrets()
with settings(warn_only=True):
local("dropdb %s" % app_config.PROJECT_SLUG)
local(
"echo \"CREATE USER %s WITH PASSWORD '%s';\" | psql"
% (app_config.PROJECT_SLUG, secrets["POSTGRES_PASSWORD"])
)
local("createdb -O %s %s" % (app_config.PROJECT_SLUG, app_config.PROJECT_SLUG))
def load_images():
"""
Downloads images from Baker and Taylor.
Eschews the API for a magic URL pattern, which is faster.
"""
# Secrets.
secrets = app_config.get_secrets()
# Open the books JSON.
with open('www/static-data/books.json', 'rb') as readfile:
books = json.loads(readfile.read())
print "Start load_images(): %i books." % len(books)
# Loop.
for book in books:
# Skip books with no title or ISBN.
if book['title'] == "":
continue
if 'isbn' not in book or book['isbn'] == "":
continue
# Construct the URL with secrets and the ISBN.
book_url = "http://imagesa.btol.com/ContentCafe/Jacket.aspx?UserID=%s&Password=%s&Return=T&Type=L&Value=%s" % (
secrets['BAKER_TAYLOR_USERID'], secrets['BAKER_TAYLOR_PASSWORD'],
book['isbn'])
# Request the image.
r = requests.get(book_url)
path = 'www/assets/cover/%s.jpg' % book['slug']
# Write the image to www using the slug as the filename.
with open(path, 'wb') as writefile:
writefile.write(r.content)
file_size = os.path.getsize(path)
if file_size < 10000:
print "Image not available for ISBN: %s" % book['isbn']
image = Image.open(path)
width = 250
height = int((float(width) / image.size[0]) * image.size[1])
image.thumbnail([width, height], Image.ANTIALIAS)
image.save(path.replace('.jpg', '-thumb.jpg'), 'JPEG')
print "End."
def execute_setup(script_name=None, doc_id=None, log_id=None):
"""
execute script setup: params script_id, document_id, log_id
"""
require('settings', provided_by=['production', 'staging', 'development'])
# Get secrets
secrets = app_config.get_secrets()
verb8tm_srt_url = secrets.get('VERB8TM_SRT_API',
app_config.PROJECT_SLUG)
verb8tm_timestamp_url = secrets.get('VERB8TM_TIMESTAMP_API',
app_config.PROJECT_SLUG)
# Get the script id from the script name and deployment target
# prioritize passed in parameters
if not script_name:
script_name = '%s_%s' % (app_config.DEPLOYMENT_TARGET,
app_config.SCRIPT_PROJECT_NAME)
script_id = get_gas_project_id(script_name)
URL_PREFIX = 'https://script.googleapis.com/v1/scripts/'
# url
url = '%s%s:run' % (URL_PREFIX, script_id)
# Compose payload we pass documentID and logID to setup script properties
payload = {
'function': 'setup',
'parameters': [verb8tm_srt_url,
verb8tm_timestamp_url,
app_config.LIVEBLOG_GDOC_KEY,
app_config.GAS_LOG_KEY]
}
kwargs = {
'credentials': check_credentials(),
'url': url,
'method': 'POST',
'headers': {'Content-Type': 'application/json'},
'body': json.dumps(payload)
}
resp = send_api_request(kwargs)
if resp.status == 200:
return True
else:
print resp.status
exit()
return False
def _geckoboard_push(widget_key, data):
payload = {
'api_key': app_config.get_secrets()['GECKOBOARD_API_KEY'],
'data': data
}
response = requests.post(
'https://push.geckoboard.com/v1/send/%s' % widget_key,
json.dumps(payload))
if response.status_code != 200:
print 'Failed update update Geckoboard widget %s' % widget_key
print response.content
def _geckoboard_push(widget_key, data):
payload = {
'api_key': app_config.get_secrets()['GECKOBOARD_API_KEY'],
'data': data
}
response = requests.post(
'https://push.geckoboard.com/v1/send/%s' % widget_key,
json.dumps(payload)
)
if response.status_code != 200:
print 'Failed update update Geckoboard widget %s' % widget_key
print response.content
def mount(server=None, path='www', mount_point='mnt'):
'''mount especiales server into local disk'''
local_path = os.path.join(cwd, '..', mount_point)
if os.path.exists(local_path):
if os.listdir(local_path):
logger.error('Mount point is not empty, choose other mount point')
exit(1)
else:
os.makedirs(local_path)
if not server:
secrets = app_config.get_secrets()
server = secrets.get('DEFAULT_SERVER', app_config.PROJECT_SLUG)
command = SMB_MOUNT_TEMPLATE % (server, path, local_path)
local(command)
return local_path
def fetch_tweets(username, days):
"""
Get tweets of a specific user
"""
current_time = datetime.now()
secrets = app_config.get_secrets()
twitter_api = Twitter(
auth=OAuth(
secrets['TWITTER_API_OAUTH_TOKEN'],
secrets['TWITTER_API_OAUTH_SECRET'],
secrets['TWITTER_API_CONSUMER_KEY'],
secrets['TWITTER_API_CONSUMER_SECRET']
)
)
out = []
tweets = twitter_api.statuses.user_timeline(screen_name=username, count=TWITTER_BATCH_SIZE)
i = 0
while True:
if i > (len(tweets)-1):
break
tweet = tweets[i]
created_time = datetime.strptime(tweet['created_at'], '%a %b %d %H:%M:%S +0000 %Y')
time_difference = (current_time - created_time).days
if time_difference > int(days):
break
out.extend(_process_tweet(tweet, username))
i += 1
if i > (TWITTER_BATCH_SIZE-1):
tweets = twitter_api.statuses.user_timeline(screen_name=username, count=TWITTER_BATCH_SIZE, max_id=tweet['id'])
i = 0
out = _dedupe_links(out)
return out
def dump_tumblr_json():
secrets = app_config.get_secrets()
t = Tumblpy(
app_key=secrets['TUMBLR_APP_KEY'],
app_secret=secrets['TUMBLR_APP_SECRET'],
oauth_token=secrets['TUMBLR_OAUTH_TOKEN'],
oauth_token_secret=secrets['TUMBLR_OAUTH_TOKEN_SECRET'])
limit = 10
pages = range(0, 20)
for page in pages:
offset = page * limit
posts = t.get('posts', blog_url=app_config.TUMBLR_URL, params={'limit': limit, 'offset': offset})
with open('data/backups/tumblr_prod_%s.json' % page, 'w') as f:
f.write(json.dumps(posts))
def dump_tumblr_json():
secrets = app_config.get_secrets()
t = Tumblpy(
app_key=secrets['TUMBLR_APP_KEY'],
app_secret=secrets['TUMBLR_APP_SECRET'],
oauth_token=secrets['TUMBLR_OAUTH_TOKEN'],
oauth_token_secret=secrets['TUMBLR_OAUTH_TOKEN_SECRET'])
limit = 10
pages = range(0, 20)
for page in pages:
offset = page * limit
posts = t.get('posts', blog_url=app_config.TUMBLR_URL, params={'limit': limit, 'offset': offset})
with open('data/backups/tumblr_prod_%s.json' % page, 'w') as f:
f.write(json.dumps(posts))
def deploy_to_tumblr():
now = datetime.now()
secrets = app_config.get_secrets()
tumblr_api = pytumblr.TumblrRestClient(secrets['TUMBLR_CONSUMER_KEY'],
secrets['TUMBLR_CONSUMER_SECRET'],
secrets['TUMBLR_TOKEN'],
secrets['TUMBLR_TOKEN_SECRET'])
body = data.make_tumblr_draft_html()
response = tumblr_api.create_text(env.tumblr_blog_name,
state='draft',
format='html',
body=body.encode('utf8'))
print "%s: Created tumblr draft (id: %s)" % (now.isoformat(),
response['id'])
return response
def render_confs():
"""
Renders server configurations.
"""
require('settings', provided_by=['production', 'staging'])
# Copy the app_config so that when we load the secrets they don't
# get exposed to other management commands
context = copy.copy(app_config.__dict__)
context.update(app_config.get_secrets())
for service, remote_path, extension in app_config.SERVER_SERVICES:
template_path = _get_template_conf_path(service, extension)
rendered_path = _get_rendered_conf_path(service, extension)
with open(template_path, 'r') as read_template:
with open(rendered_path, 'wb') as write_template:
payload = Template(read_template.read())
write_template.write(payload.render(**context))
def render_confs():
"""
Renders server configurations.
"""
require('settings', provided_by=['production', 'staging'])
# Copy the app_config so that when we load the secrets they don't
# get exposed to other management commands
context = copy.copy(app_config.__dict__)
context.update(app_config.get_secrets())
for service, remote_path, extension in app_config.SERVER_SERVICES:
template_path = _get_template_conf_path(service, extension)
rendered_path = _get_rendered_conf_path(service, extension)
with open(template_path, 'r') as read_template:
with open(rendered_path, 'wb') as write_template:
payload = Template(read_template.read())
write_template.write(payload.render(**context))
def write_test_posts():
"""
Writes test posts to our test blog as defined by app_config.py
"""
# This is how many posts will be written.
TOTAL_NUMBER = 9
secrets = app_config.get_secrets()
t = Tumblpy(app_key=secrets['TUMBLR_APP_KEY'],
app_secret=secrets['TUMBLR_APP_SECRET'],
oauth_token=secrets['TUMBLR_OAUTH_TOKEN'],
oauth_token_secret=secrets['TUMBLR_OAUTH_TOKEN_SECRET'])
tags = ['featured', '']
images = [
'http://media.npr.org/assets/img/2013/04/24/habitablezones_custom-fa87578c6e6a97788b92a0ecac956b9098607aa6-s40.jpg',
'http://media.npr.org/assets/img/2013/04/24/ocpack-32260770b4090f86ddeb7502175a631d50c3b4a1-s51.jpg',
'http://media.npr.org/assets/img/2013/04/24/dalrymple-c-karoki-lewis-4c9bd790639c870d51c670cbecbca4b802b82b1a-s51.jpg',
'http://media.npr.org/assets/img/2013/04/24/ap111231019469-46289d097a45801ed2ca3464da14b13be40e5adb-s51.jpg'
]
n = 0
while n < TOTAL_NUMBER:
image = choice(images)
tag = choice(tags)
caption = u"<p class='intro'>Introduction,</p><p class='message'>This is a test post.</p><p class='signature-name'>Sincerely,<br/>Test from Test, Test</p>"
tumblr_post = t.post('post',
blog_url=app_config.TUMBLR_URL,
params={
'type': 'photo',
'caption': caption,
'tags': tag,
'source': image
})
print n, tumblr_post['id']
n += 1
def create_tables():
"""
Create the databse tables.
"""
import models
secrets = app_config.get_secrets()
print 'Creating database tables'
models.Race.create_table()
models.Candidate.create_table()
models.Slide.create_table()
models.SlideSequence.create_table()
print 'Setting up admin'
admin_app.auth.User.create_table()
admin_user = admin_app.auth.User(username='******', email='', admin=True, active=True)
admin_user.set_password(secrets.get('ADMIN_PASSWORD'))
admin_user.save()
def generate_rdio_playlist():
"""
Generate a list of Rdio track IDs
"""
secrets = app_config.get_secrets()
state = {}
rdio_api = Rdio(secrets['RDIO_CONSUMER_KEY'],
secrets['RDIO_CONSUMER_SECRET'], state)
songs = []
with open('data/songs.csv') as f:
rows = csv.DictReader(f)
for row in rows:
if row['rdio']:
song_object = rdio_api.getObjectFromUrl(url=row['rdio'])
if song_object['type'] == 't':
songs.append(song_object['key'])
print ','.join(songs)
def write_test_posts():
"""
Writes test posts to our test blog as defined by app_config.py
"""
# This is how many posts will be written.
TOTAL_NUMBER = 9
secrets = app_config.get_secrets()
t = Tumblpy(
app_key=secrets['TUMBLR_APP_KEY'],
app_secret=secrets['TUMBLR_APP_SECRET'],
oauth_token=secrets['TUMBLR_OAUTH_TOKEN'],
oauth_token_secret=secrets['TUMBLR_OAUTH_TOKEN_SECRET'])
tags = ['featured', '']
images = [
'http://media.npr.org/assets/img/2013/04/24/habitablezones_custom-fa87578c6e6a97788b92a0ecac956b9098607aa6-s40.jpg',
'http://media.npr.org/assets/img/2013/04/24/ocpack-32260770b4090f86ddeb7502175a631d50c3b4a1-s51.jpg',
'http://media.npr.org/assets/img/2013/04/24/dalrymple-c-karoki-lewis-4c9bd790639c870d51c670cbecbca4b802b82b1a-s51.jpg',
'http://media.npr.org/assets/img/2013/04/24/ap111231019469-46289d097a45801ed2ca3464da14b13be40e5adb-s51.jpg'
]
n = 0
while n < TOTAL_NUMBER:
image = choice(images)
tag = choice(tags)
caption = u"<p class='intro'>Introduction,</p><p class='message'>This is a test post.</p><p class='signature-name'>Sincerely,<br/>Test from Test, Test</p>"
tumblr_post = t.post('post', blog_url=app_config.TUMBLR_URL, params={
'type': 'photo',
'caption': caption,
'tags': tag,
'source': image
})
print n, tumblr_post['id']
n += 1
def render_confs():
"""
Renders server configurations.
"""
require('settings', provided_by=[production, staging])
with settings(warn_only=True):
local('mkdir confs/rendered')
context = app_config.get_secrets()
context['PROJECT_SLUG'] = app_config.PROJECT_SLUG
context['PROJECT_NAME'] = app_config.PROJECT_NAME
context['DEPLOYMENT_TARGET'] = env.settings
for service, remote_path in SERVICES:
file_path = 'confs/rendered/%s.%s.conf' % (app_config.PROJECT_SLUG, service)
with open('confs/%s.conf' % service, 'r') as read_template:
with open(file_path, 'wb') as write_template:
payload = Template(read_template.read())
write_template.write(payload.render(**context))
def generate_spotify_playlist():
"""
Generate a list of Spotify track IDs
"""
sp = spotipy.Spotify()
songs = []
secrets = app_config.get_secrets()
with open('data/songs.csv') as f:
rows = csv.DictReader(f)
for row in rows:
if row['spotify'] and "track" in row['spotify']:
try:
song_object = sp.track(track_id=row['spotify'])
if song_object['track_number']:
songs.append(song_object['uri'])
except:
print 'invalid ID'
print ','.join(songs)
def render_confs():
"""
Renders server configurations.
"""
require('settings', provided_by=[production, staging])
with settings(warn_only=True):
local('mkdir confs/rendered')
context = app_config.get_secrets()
context['PROJECT_SLUG'] = app_config.PROJECT_SLUG
context['PROJECT_NAME'] = app_config.PROJECT_NAME
context['DEPLOYMENT_TARGET'] = env.settings
for service, remote_path in SERVICES:
file_path = 'confs/rendered/%s.%s.conf' % (app_config.PROJECT_SLUG, service)
with open('confs/%s.conf' % service, 'r') as read_template:
with open(file_path, 'wb') as write_template:
payload = Template(read_template.read())
write_template.write(payload.render(**context))
def render_confs():
"""
Renders server configurations.
"""
require("settings", provided_by=[production, staging])
with settings(warn_only=True):
local("mkdir confs/rendered")
context = app_config.get_secrets()
context["PROJECT_SLUG"] = app_config.PROJECT_SLUG
context["PROJECT_NAME"] = app_config.PROJECT_NAME
context["DEPLOYMENT_TARGET"] = env.settings
for service, remote_path in SERVICES:
file_path = "confs/rendered/%s.%s.conf" % (app_config.PROJECT_SLUG, service)
with open("confs/%s.conf" % service, "r") as read_template:
with open(file_path, "wb") as write_template:
payload = Template(read_template.read())
write_template.write(payload.render(**context))
def generate_spotify_playlist():
"""
Generate a list of Spotify track IDs
"""
sp = spotipy.Spotify()
songs = []
secrets = app_config.get_secrets()
with open('data/songs.csv') as f:
rows = csv.DictReader(f)
for row in rows:
if row['spotify'] and "track" in row['spotify']:
try:
song_object = sp.track(track_id=row['spotify'])
if song_object['track_number']:
songs.append(song_object['uri'])
except:
print 'invalid ID'
print ','.join(songs)
def update():
"""
Fetch latests posts from Tumblr API.
"""
secrets = app_config.get_secrets()
offset = 0
total_posts = 1
while offset < total_posts:
print 'Fetching posts %i-%i' % (offset, offset + LIMIT)
response = requests.get('http://api.tumblr.com/v2/blog/%s.tumblr.com/posts' % app_config.TUMBLR_NAME, params={
'api_key':secrets['TUMBLR_CONSUMER_KEY'],
'limit': LIMIT,
'offset': offset
})
data = response.json()
total_posts = data['response']['total_posts']
posts = data['response']['posts']
for post in posts:
if post['type'] in UNSUPPORTED_TYPES:
continue
if datetime.fromtimestamp(post['timestamp']) < app_config.TUMBLR_NOT_BEFORE:
print 'Skipping old post'
continue
elif datetime.fromtimestamp(post['timestamp']) > app_config.TUMBLR_NOT_AFTER:
print 'Skipping newer post'
continue
_create_slide(post)
offset += LIMIT
def render_confs():
"""
Renders server configurations.
"""
require("settings", provided_by=[production, staging])
with settings(warn_only=True):
local("mkdir confs/rendered")
# Copy the app_config so that when we load the secrets they don't
# get exposed to other management commands
context = copy.copy(app_config.__dict__)
context.update(app_config.get_secrets())
for service, remote_path, extension in app_config.SERVER_SERVICES:
template_path = _get_template_conf_path(service, extension)
rendered_path = _get_rendered_conf_path(service, extension)
with open(template_path, "r") as read_template:
with open(rendered_path, "wb") as write_template:
payload = Template(read_template.read())
write_template.write(payload.render(**context))
def update():
"""
Fetch latests posts from Tumblr API.
"""
secrets = app_config.get_secrets()
offset = 0
total_posts = 1
while offset < total_posts:
print "Fetching posts %i-%i" % (offset, offset + LIMIT)
response = requests.get(
"http://api.tumblr.com/v2/blog/%s.tumblr.com/posts" % app_config.TUMBLR_NAME,
params={"api_key": secrets["TUMBLR_CONSUMER_KEY"], "limit": LIMIT, "offset": offset},
)
data = response.json()
total_posts = data["response"]["total_posts"]
posts = data["response"]["posts"]
for post in posts:
if post["type"] in UNSUPPORTED_TYPES:
continue
if datetime.fromtimestamp(post["timestamp"]) < app_config.TUMBLR_NOT_BEFORE:
print "Skipping old post"
continue
elif datetime.fromtimestamp(post["timestamp"]) > app_config.TUMBLR_NOT_AFTER:
print "Skipping newer post"
continue
_create_slide(post)
offset += LIMIT
def render_confs():
"""
Renders server configurations.
"""
require('settings', provided_by=[production, staging])
with settings(warn_only=True):
local('mkdir confs/rendered')
context = app_config.get_secrets()
context['PROJECT_SLUG'] = app_config.PROJECT_SLUG
context['CLOUD_SEARCH_DOMAIN'] = app_config.CLOUD_SEARCH_DOMAIN
context['PROJECT_NAME'] = app_config.PROJECT_NAME
context['DEPLOYMENT_TARGET'] = env.settings
context['CONFIG_NAME'] = env.project_slug.replace('-', '').upper()
for service, remote_path, extension in SERVICES:
file_path = 'confs/rendered/%s.%s.%s' % (app_config.PROJECT_SLUG, service, extension)
with open('confs/%s.%s' % (service, extension), 'r') as read_template:
with open(file_path, 'wb') as write_template:
payload = Template(read_template.read())
write_template.write(payload.render(**context))
def render_confs():
"""
Renders server configurations.
"""
require("settings", provided_by=[production, staging])
with settings(warn_only=True):
local("mkdir confs/rendered")
context = app_config.get_secrets()
context["PROJECT_SLUG"] = app_config.PROJECT_SLUG
context["CLOUD_SEARCH_DOMAIN"] = app_config.CLOUD_SEARCH_DOMAIN
context["PROJECT_NAME"] = app_config.PROJECT_NAME
context["DEPLOYMENT_TARGET"] = env.settings
context["CONFIG_NAME"] = env.project_slug.replace("-", "").upper()
for service, remote_path, extension in SERVICES:
file_path = "confs/rendered/%s.%s.%s" % (app_config.PROJECT_SLUG, service, extension)
with open("confs/%s.%s" % (service, extension), "r") as read_template:
with open(file_path, "wb") as write_template:
payload = Template(read_template.read())
write_template.write(payload.render(**context))
def generate_rdio_playlist():
"""
Generate a list of Rdio track IDs
"""
secrets = app_config.get_secrets()
state = {}
rdio_api = Rdio(
secrets['RDIO_CONSUMER_KEY'],
secrets['RDIO_CONSUMER_SECRET'],
state
)
songs = []
with open('data/songs.csv') as f:
rows = csv.DictReader(f)
for row in rows:
if row['rdio']:
song_object = rdio_api.getObjectFromUrl(url=row['rdio'])
if song_object['type'] == 't':
songs.append(song_object['key'])
print ','.join(songs)
def create_tables():
"""
Create the databse tables.
"""
import models
secrets = app_config.get_secrets()
print 'Creating database tables'
models.Race.create_table()
models.Candidate.create_table()
models.Slide.create_table()
models.SlideSequence.create_table()
print 'Setting up admin'
admin_app.auth.User.create_table()
admin_user = admin_app.auth.User(username='******',
email='',
admin=True,
active=True)
admin_user.set_password(secrets.get('ADMIN_PASSWORD'))
admin_user.save()
def update_featured_social():
"""
Update featured tweets
"""
COPY = copytext.Copy(app_config.COPY_PATH)
secrets = app_config.get_secrets()
# Twitter
print 'Fetching tweets...'
twitter_api = Twitter(auth=OAuth(secrets['TWITTER_API_OAUTH_TOKEN'],
secrets['TWITTER_API_OAUTH_SECRET'],
secrets['TWITTER_API_CONSUMER_KEY'],
secrets['TWITTER_API_CONSUMER_SECRET']))
tweets = []
for i in range(1, 4):
tweet_url = COPY['share']['featured_tweet%i' % i]
if isinstance(tweet_url,
copytext.Error) or unicode(tweet_url).strip() == '':
continue
tweet_id = unicode(tweet_url).split('/')[-1]
tweet = twitter_api.statuses.show(id=tweet_id)
creation_date = datetime.strptime(tweet['created_at'],
'%a %b %d %H:%M:%S +0000 %Y')
creation_date = '%s %i' % (creation_date.strftime('%b'),
creation_date.day)
tweet_url = 'http://twitter.com/%s/status/%s' % (
tweet['user']['screen_name'], tweet['id'])
photo = None
html = tweet['text']
subs = {}
for media in tweet['entities'].get('media', []):
original = tweet['text'][media['indices'][0]:media['indices'][1]]
replacement = '<a href="%s" target="_blank" onclick="_gaq.push([\'_trackEvent\', \'%s\', \'featured-tweet-action\', \'link\', 0, \'%s\']);">%s</a>' % (
media['url'], app_config.PROJECT_SLUG, tweet_url,
media['display_url'])
subs[original] = replacement
if media['type'] == 'photo' and not photo:
photo = {'url': media['media_url']}
for url in tweet['entities'].get('urls', []):
original = tweet['text'][url['indices'][0]:url['indices'][1]]
replacement = '<a href="%s" target="_blank" onclick="_gaq.push([\'_trackEvent\', \'%s\', \'featured-tweet-action\', \'link\', 0, \'%s\']);">%s</a>' % (
url['url'], app_config.PROJECT_SLUG, tweet_url,
url['display_url'])
subs[original] = replacement
for hashtag in tweet['entities'].get('hashtags', []):
original = tweet['text'][
hashtag['indices'][0]:hashtag['indices'][1]]
replacement = '<a href="https://twitter.com/hashtag/%s" target="_blank" onclick="_gaq.push([\'_trackEvent\', \'%s\', \'featured-tweet-action\', \'hashtag\', 0, \'%s\']);">%s</a>' % (
hashtag['text'], app_config.PROJECT_SLUG, tweet_url,
'#%s' % hashtag['text'])
subs[original] = replacement
for original, replacement in subs.items():
html = html.replace(original, replacement)
# https://dev.twitter.com/docs/api/1.1/get/statuses/show/%3Aid
tweets.append({
'id': tweet['id'],
'url': tweet_url,
'html': html,
'favorite_count': tweet['favorite_count'],
'retweet_count': tweet['retweet_count'],
'user': {
'id': tweet['user']['id'],
'name': tweet['user']['name'],
'screen_name': tweet['user']['screen_name'],
'profile_image_url': tweet['user']['profile_image_url'],
'url': tweet['user']['url'],
},
'creation_date': creation_date,
'photo': photo
})
# Facebook
print 'Fetching Facebook posts...'
fb_api = GraphAPI(secrets['FACEBOOK_API_APP_TOKEN'])
facebook_posts = []
for i in range(1, 4):
fb_url = COPY['share']['featured_facebook%i' % i]
if isinstance(fb_url, copytext.Error) or unicode(fb_url).strip() == '':
continue
fb_id = unicode(fb_url).split('/')[-1]
post = fb_api.get_object(fb_id)
user = fb_api.get_object(post['from']['id'])
user_picture = fb_api.get_object('%s/picture' % post['from']['id'])
likes = fb_api.get_object('%s/likes' % fb_id, summary='true')
comments = fb_api.get_object('%s/comments' % fb_id, summary='true')
#shares = fb_api.get_object('%s/sharedposts' % fb_id)
creation_date = datetime.strptime(post['created_time'],
'%Y-%m-%dT%H:%M:%S+0000')
creation_date = '%s %i' % (creation_date.strftime('%b'),
creation_date.day)
# https://developers.facebook.com/docs/graph-api/reference/v2.0/post
facebook_posts.append({
'id': post['id'],
'message': post['message'],
'link': {
'url': post['link'],
'name': post['name'],
'caption': (post['caption'] if 'caption' in post else None),
'description': post['description'],
'picture': post['picture']
},
'from': {
'name': user['name'],
'link': user['link'],
'picture': user_picture['url']
},
'likes': likes['summary']['total_count'],
'comments': comments['summary']['total_count'],
#'shares': shares['summary']['total_count'],
'creation_date': creation_date
})
# Render to JSON
output = {'tweets': tweets, 'facebook_posts': facebook_posts}
with open('data/featured.json', 'w') as f:
json.dump(output, f)
"""
import app_config
import dataset
import ipdb
import logging
import requests
from datetime import datetime
from dateutil import parser
from fabric.api import local, task
from facebook import GraphAPI
from pyquery import PyQuery
from scrapers.utils import get_art_root_url, get_seamus_id_from_url
from scrapers.seamus.models import Story
SECRETS = app_config.get_secrets()
FACEBOOK_USER = '******'
logging.basicConfig()
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)
@task
def local_bootstrap():
local('dropdb --if-exists %s' % app_config.PROJECT_SLUG)
local('createdb %s' % app_config.PROJECT_SLUG)
@task
def drop_tables():
"""
import app_config
import json
import logging
import requests
from datetime import datetime
from fabric.api import local, require, task
logging.basicConfig()
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)
today = datetime.now().strftime('%Y-%m-%d')
WEBHOOK = app_config.get_secrets()['WEBHOOK']
URL = 'https://www.federalregister.gov/api/v1/documents.json'
COLORS_DICT = {
'Rule': '#5cb85c',
'Proposed Rule': '#5bc0de',
'Presidential Document': '#d9534f'
}
@task
def post_message():
"""
Example cron task. Note we use "local" instead of "run"
because this will run on the server.
"""
#!/usr/bin/env python
from datetime import datetime
import logging
import os
import requests
from pyquery import PyQuery
from app_config import get_secrets
from models import ApiEntry, Article
SECRETS = get_secrets()
logging.basicConfig()
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)
class HomepageScraper:
url = 'http://npr.org'
def __init__(self):
self.run_time = datetime.utcnow()
def scrape_homepage(self, **kwargs):
"""
Scrape!
"""
logger.info('Scraping homepage (start time: %s)' % self.run_time)
if not kwargs:
import datetime
import time
from peewee import *
from app_config import get_secrets
secrets = get_secrets()
psql_db = PostgresqlDatabase("breaking", user=secrets["APPS_USER"], password=secrets["APPS_PASS"])
def delete_tables():
try:
Event.drop_table()
except:
pass
try:
Fact.drop_table()
except:
pass
def create_tables():
Event.create_table()
Fact.create_table()
class Event(Model):
"""
def execute_setup(name=None,
doc_id=None,
log_id=None,
cspan=None,
cspan_server=None):
"""
execute script setup
params:
name - script name
document_id - associated google doc
log_id - associated google logs spreadsheet
cspan - False to use Verb8tm, True to use CSPAN
cspan_server - url that will serve the CSPAN captions
"""
require('settings', provided_by=['production', 'staging', 'development'])
# Get secrets
secrets = app_config.get_secrets()
verb8tm_srt_url = secrets.get('VERB8TM_SRT_API', app_config.PROJECT_SLUG)
verb8tm_timestamp_url = secrets.get('VERB8TM_TIMESTAMP_API',
app_config.PROJECT_SLUG)
if cspan:
pass
else:
cspan = str(app_config.CSPAN)
ec2_public_dns_tpl = 'http://ec2-%s.compute-1.amazonaws.com:5000/'
if cspan_server:
pass
else:
if len(app_config.SERVERS):
ip = app_config.SERVERS[0]
ip = ip.replace('.', '-')
cspan_server = ec2_public_dns_tpl % (ip)
else:
logger.error("no servers found, please specify a cspan_server")
exit()
# Get the script id from the script name and deployment target
# prioritize passed in parameters
if not name:
name = '%s_%s' % (app_config.DEPLOYMENT_TARGET,
app_config.SCRIPT_PROJECT_NAME)
script_id = get_gas_project_id(name)
URL_PREFIX = 'https://script.googleapis.com/v1/scripts/'
# url
url = '%s%s:run' % (URL_PREFIX, script_id)
# Compose payload we pass documentID and logID to setup script properties
payload = {
'function':
'setup',
'parameters': [
verb8tm_srt_url, verb8tm_timestamp_url, cspan_server, cspan,
app_config.TRANSCRIPT_GDOC_KEY, app_config.GAS_LOG_KEY
]
}
kwargs = {
'credentials': check_credentials(),
'url': url,
'method': 'POST',
'headers': {
'Content-Type': 'application/json'
},
'body': json.dumps(payload)
}
resp = send_api_request(kwargs)
if resp.status == 200:
return True
else:
print resp.status
exit()
return False
import argparse
import base64
from cStringIO import StringIO
import datetime
import logging
from flask import Flask, abort, redirect, render_template, request, session, url_for
from twython import Twython
import app_config
from render_utils import make_context
app = Flask(__name__)
app.debug = app_config.DEBUG
app.secret_key = app_config.get_secrets()["SESSION_KEY"]
file_handler = logging.FileHandler(app_config.APP_LOG_PATH)
file_handler.setLevel(logging.INFO)
app.logger.addHandler(file_handler)
app.logger.setLevel(logging.INFO)
secrets = app_config.get_secrets()
# Example application views
@app.route("/test/", methods=["GET"])
def _test_app():
"""
Test route for verifying the application is running.
"""
app.logger.info("Test URL requested.")
'ERROR', '500', e, app_config.SERVERS[0], svg_path, name))
# These bits build a nicer error page that has the real stack trace on it.
context = {}
context['title'] = 'CairoSVG is unhappy.'
context['message'] = e.output
return render_template('500.html', **context)
context = {
'name': name,
'location': location,
}
caption = render_template('caption.html', **context)
secrets = app_config.get_secrets()
t = Tumblpy(
app_key=secrets['TUMBLR_APP_KEY'],
app_secret=secrets['TUMBLR_APP_SECRET'],
oauth_token=secrets['TUMBLR_OAUTH_TOKEN'],
oauth_token_secret=secrets['TUMBLR_OAUTH_TOKEN_SECRET'])
params = {
"type": "photo",
"caption": caption,
"tags": app_config.TUMBLR_TAGS,
"source": "http://%s%s" % (app_config.SERVERS[0], png_path)
}
try:
def update_featured_social():
"""
Update featured tweets
"""
COPY = copytext.Copy(app_config.COPY_PATH)
secrets = app_config.get_secrets()
# Twitter
print 'Fetching tweets...'
twitter_api = Twitter(
auth=OAuth(
secrets['TWITTER_API_OAUTH_TOKEN'],
secrets['TWITTER_API_OAUTH_SECRET'],
secrets['TWITTER_API_CONSUMER_KEY'],
secrets['TWITTER_API_CONSUMER_SECRET']
)
)
tweets = []
for i in range(1, 4):
tweet_url = COPY['share']['featured_tweet%i' % i]
if isinstance(tweet_url, copytext.Error) or unicode(tweet_url).strip() == '':
continue
tweet_id = unicode(tweet_url).split('/')[-1]
tweet = twitter_api.statuses.show(id=tweet_id)
creation_date = datetime.strptime(tweet['created_at'],'%a %b %d %H:%M:%S +0000 %Y')
creation_date = '%s %i' % (creation_date.strftime('%b'), creation_date.day)
tweet_url = 'http://twitter.com/%s/status/%s' % (tweet['user']['screen_name'], tweet['id'])
photo = None
html = tweet['text']
subs = {}
for media in tweet['entities'].get('media', []):
original = tweet['text'][media['indices'][0]:media['indices'][1]]
replacement = '<a href="%s" target="_blank" onclick="_gaq.push([\'_trackEvent\', \'%s\', \'featured-tweet-action\', \'link\', 0, \'%s\']);">%s</a>' % (media['url'], app_config.PROJECT_SLUG, tweet_url, media['display_url'])
subs[original] = replacement
if media['type'] == 'photo' and not photo:
photo = {
'url': media['media_url']
}
for url in tweet['entities'].get('urls', []):
original = tweet['text'][url['indices'][0]:url['indices'][1]]
replacement = '<a href="%s" target="_blank" onclick="_gaq.push([\'_trackEvent\', \'%s\', \'featured-tweet-action\', \'link\', 0, \'%s\']);">%s</a>' % (url['url'], app_config.PROJECT_SLUG, tweet_url, url['display_url'])
subs[original] = replacement
for hashtag in tweet['entities'].get('hashtags', []):
original = tweet['text'][hashtag['indices'][0]:hashtag['indices'][1]]
replacement = '<a href="https://twitter.com/hashtag/%s" target="_blank" onclick="_gaq.push([\'_trackEvent\', \'%s\', \'featured-tweet-action\', \'hashtag\', 0, \'%s\']);">%s</a>' % (hashtag['text'], app_config.PROJECT_SLUG, tweet_url, '#%s' % hashtag['text'])
subs[original] = replacement
for original, replacement in subs.items():
html = html.replace(original, replacement)
# https://dev.twitter.com/docs/api/1.1/get/statuses/show/%3Aid
tweets.append({
'id': tweet['id'],
'url': tweet_url,
'html': html,
'favorite_count': tweet['favorite_count'],
'retweet_count': tweet['retweet_count'],
'user': {
'id': tweet['user']['id'],
'name': tweet['user']['name'],
'screen_name': tweet['user']['screen_name'],
'profile_image_url': tweet['user']['profile_image_url'],
'url': tweet['user']['url'],
},
'creation_date': creation_date,
'photo': photo
})
# Facebook
print 'Fetching Facebook posts...'
fb_api = GraphAPI(secrets['FACEBOOK_API_APP_TOKEN'])
facebook_posts = []
for i in range(1, 4):
fb_url = COPY['share']['featured_facebook%i' % i]
if isinstance(fb_url, copytext.Error) or unicode(fb_url).strip() == '':
continue
fb_id = unicode(fb_url).split('/')[-1]
post = fb_api.get_object(fb_id)
user = fb_api.get_object(post['from']['id'])
user_picture = fb_api.get_object('%s/picture' % post['from']['id'])
likes = fb_api.get_object('%s/likes' % fb_id, summary='true')
comments = fb_api.get_object('%s/comments' % fb_id, summary='true')
#shares = fb_api.get_object('%s/sharedposts' % fb_id)
creation_date = datetime.strptime(post['created_time'],'%Y-%m-%dT%H:%M:%S+0000')
creation_date = '%s %i' % (creation_date.strftime('%b'), creation_date.day)
# https://developers.facebook.com/docs/graph-api/reference/v2.0/post
facebook_posts.append({
'id': post['id'],
'message': post['message'],
'link': {
'url': post['link'],
'name': post['name'],
'caption': (post['caption'] if 'caption' in post else None),
'description': post['description'],
'picture': post['picture']
},
'from': {
'name': user['name'],
'link': user['link'],
'picture': user_picture['url']
},
'likes': likes['summary']['total_count'],
'comments': comments['summary']['total_count'],
#'shares': shares['summary']['total_count'],
'creation_date': creation_date
})
# Render to JSON
output = {
'tweets': tweets,
'facebook_posts': facebook_posts
}
with open('data/featured.json', 'w') as f:
json.dump(output, f)
"""
import app_config
import json
import logging
import os
import requests
from fabric.api import local, require, task
from pyquery import PyQuery as pq
logging.basicConfig()
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)
secrets = app_config.get_secrets()
WEBHOOK = secrets['WEBHOOK']
@task
def post_message():
message = check_page()
if message:
r = requests.post(WEBHOOK, data=json.dumps(message))
logger.info(r.text)
def check_page():
attachments = []
last_title = get_last_title()
#!/usr/bin/env python
from datetime import datetime
import logging
import os
import requests
from pyquery import PyQuery
from app_config import get_secrets
from models import ApiEntry, Article
SECRETS = get_secrets()
logging.basicConfig()
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)
class HomepageScraper:
url = 'http://npr.org'
def __init__(self):
self.run_time = datetime.utcnow()
def scrape_homepage(self, **kwargs):
"""
Scrape!
"""
logger.info('Scraping homepage (start time: %s)' % self.run_time)
def boundaries():
context = make_context()
context['name'] = 'Boundaries'
context['id'] = context['name']
context['secrets'] = app_config.get_secrets()
return make_response(render_template('boundaries.html', **context))
#!/usr/bin/env python
import app_config
import json
import os
import time
from datetime import datetime
from elections import AP
from fabric.api import task
from time import sleep
SECRETS = app_config.get_secrets()
# If main FTP server goes down, uncomment the next line
#AP.FTP_HOSTNAME = 'electionsonline2.ap.org'
def _generate_candidate_id(candidate, race):
"""
Makes an unique compound ID
"""
return '%s-%s' % (candidate.ap_polra_number, race.state_postal)
@task
def bootstrap():
init()
update()
@task
def init(output_dir='data'):
"""
def load_images():
"""
Downloads images from Baker and Taylor.
Eschews the API for a magic URL pattern, which is faster.
"""
# Secrets.
secrets = app_config.get_secrets()
# Open the books JSON.
with open('www/static-data/books.json', 'rb') as readfile:
books = json.loads(readfile.read())
print "Start load_images(): %i books." % len(books)
# Loop.
for book in books:
# Skip books with no title or ISBN.
if book['title'] == "":
continue
if 'isbn' not in book or book['isbn'] == "":
continue
# Construct the URL with secrets and the ISBN.
book_url = "http://imagesa.btol.com/ContentCafe/Jacket.aspx?UserID=%s&Password=%s&Return=T&Type=L&Value=%s" % (
secrets['BAKER_TAYLOR_USERID'],
secrets['BAKER_TAYLOR_PASSWORD'],
book['isbn'])
# Request the image.
r = requests.get(book_url)
path = 'www/assets/cover/%s.jpg' % book['slug']
# Write the image to www using the slug as the filename.
with open(path, 'wb') as writefile:
writefile.write(r.content)
file_size = os.path.getsize(path)
if file_size < 10000:
print u'LOG (%s): Image not available from Baker and Taylor, using NPR book page' % book['title']
url = 'http://www.npr.org/%s' % book['book_seamus_id']
npr_r = requests.get(url)
soup = BeautifulSoup(npr_r.content)
try:
if book['title'] == 'The Three-Body Problem':
alt_img_url = 'http://media.npr.org/assets/bakertaylor/covers/t/the-three-body-problem/9780765377067_custom-d83e0e334f348e6c52fe5da588ec3448921af64f-s600-c85.jpg'
else:
alt_img_url = soup.select('.bookedition .image img')[0].attrs.get('src').replace('s99', 's400')
print 'LOG (%s): Getting alternate image from %s' % (book['title'], alt_img_url)
alt_img_resp = requests.get(alt_img_url)
with open(path, 'wb') as writefile:
writefile.write(alt_img_resp.content)
except IndexError:
print u'ERROR (%s): Image not available on NPR book page either (%s)' % (book['title'], url)
image = Image.open(path)
image.save(path, optimize=True, quality=75)
print "End."
def _post_to_tumblr():
"""
Handles the POST to Tumblr.
"""
def strip_html(value):
"""
Strips HTML from a string.
"""
return re.compile(r'</?\S([^=]*=(\s*"[^"]*"|\s*\'[^\']*\'|\S*)|[^>])*?>', re.IGNORECASE).sub('', value)
def strip_breaks(value):
"""
Converts newlines, returns and other breaks to <br/>.
"""
value = re.sub(r'\r\n|\r|\n', '\n', value)
return value.replace('\n', do_mark_safe('<br/>'))
# Request is a global. Import it down here where we need it.
from flask import request
# These should match the form fields.
message = strip_html(request.form.get('message', None))
message = escape(message)
message = strip_breaks(message)
name = strip_html(request.form.get('signed_name', None))
email = strip_html(request.form.get('email', None))
context = {
'message': message,
'name': name,
'email': email,
'app_config': app_config
}
caption = render_template('caption.html', **context)
secrets = app_config.get_secrets()
t = Tumblpy(
app_key=secrets['TUMBLR_APP_KEY'],
app_secret=secrets['TUMBLR_APP_SECRET'],
oauth_token=secrets['TUMBLR_OAUTH_TOKEN'],
oauth_token_secret=secrets['TUMBLR_OAUTH_TOKEN_SECRET'])
file_path = '/uploads/%s/%s_%s' % (
app_config.PROJECT_SLUG,
str(time.mktime(datetime.datetime.now().timetuple())).replace('.', ''),
secure_filename(request.files['image'].filename.replace(' ', '-'))
)
with open('/var/www%s' % file_path, 'w') as f:
f.write(request.files['image'].read())
params = {
"type": "photo",
"caption": caption,
"tags": app_config.TUMBLR_TAGS,
"source": "http://%s%s" % (app_config.SERVERS[0], file_path)
}
try:
tumblr_post = t.post('post', blog_url=app_config.TUMBLR_URL, params=params)
tumblr_url = u"http://%s/%s" % (app_config.TUMBLR_URL, tumblr_post['id'])
logger.info('200 %s reader(%s %s) (times in EST)' % (tumblr_url, name, email))
return redirect(tumblr_url, code=301)
except TumblpyError, e:
logger.error('%s %s http://%s%s reader(%s %s) (times in EST)' % (
e.error_code, e.msg, app_config.SERVERS[0], file_path, name, email))
return 'TUMBLR ERROR'
def load_images():
"""
Downloads images from Baker and Taylor.
Eschews the API for a magic URL pattern, which is faster.
"""
# Secrets.
secrets = app_config.get_secrets()
# Open the books JSON.
with open('www/static-data/books.json', 'rb') as readfile:
books = json.loads(readfile.read())
print "Start load_images(): %i books." % len(books)
always_use_npr_cover = set(app_config.ALWAYS_USE_NPR_COVER)
# Loop.
for book in books:
# Skip books with no title or ISBN.
if book['title'] == "":
logger.warning('found book with no title')
continue
if 'isbn' not in book or book['isbn'] == "":
logger.warning('This book has no isbn: %s' % book['title'])
continue
# Construct the URL with secrets and the ISBN.
book_url = "http://imagesa.btol.com/ContentCafe/Jacket.aspx"
params = {
'UserID': secrets['BAKER_TAYLOR_USERID'],
'Password': secrets['BAKER_TAYLOR_PASSWORD'],
'Value': book['isbn'],
'Return': 'T',
'Type': 'L'
}
# Request the image.
r = requests.get(book_url, params=params)
path = 'www/assets/cover'
if not os.path.exists(path):
os.makedirs(path)
imagepath = '%s/%s.jpg' % (path, book['slug'])
if os.path.exists(imagepath):
logger.debug('image already downloaded for: %s' % book['slug'])
# Write the image to www using the slug as the filename.
with open(imagepath, 'wb') as writefile:
writefile.write(r.content)
file_size = os.path.getsize(imagepath)
use_npr_book_page = (
file_size < 10000 or
book['isbn'] in always_use_npr_cover
)
if use_npr_book_page:
msg = ('(%s): Image not available from Baker and Taylor, '
'using NPR book page') % book['title']
logger.info(msg)
try:
alt_img_url = _get_npr_cover_img_url(book)
msg = 'LOG (%s): Getting alternate image from %s' % (
book['title'], alt_img_url)
logger.info(msg)
alt_img_resp = requests.get(alt_img_url)
with open(imagepath, 'wb') as writefile:
writefile.write(alt_img_resp.content)
except ValueError:
msg = (
'ERROR (%s): Image not available on NPR book page either'
) % (book['title'])
logger.info(msg)
image = Image.open(imagepath)
image.save(imagepath, optimize=True, quality=75)
logger.info("Load Images End.")
