def run(*fileids):
_redis = redis.client.Redis(settings.REDIS_HOST, settings.REDIS_PORT)
connection = motor.MotorConnection().open_sync()
db = connection.tiler
try:
cursor = db.images.find({'fileid': {'$in': fileids}})
for document in (yield motor.Op(cursor.to_list)):
pprint(document)
yield motor.Op(db.images.update, {'_id': document['_id']},
{'$unset': {
'cdn_domain': 1
}})
metadata_key = 'metadata:%s' % document['fileid']
if _redis.get(metadata_key):
print "Invalidated metadata cache"
_redis.delete(metadata_key)
lock_key = 'uploading:%s' % document['fileid']
# locking it from aws upload for 1 hour
_redis.setex(lock_key, time.time(), 60 * 60)
upload_log = os.path.join(ROOT, 'static',
'upload.%s.txt' % document['fileid'])
if os.path.isfile(upload_log):
os.remove(upload_log)
print "\n"
finally:
IOLoop.instance().stop()
def run(*args):
_redis = redis.client.Redis(settings.REDIS_HOST, settings.REDIS_PORT)
connection = motor.MotorConnection().open_sync()
db = connection.tiler
try:
cursor = (db.images.find())
image = yield motor.Op(cursor.next_object)
while image:
_redis.incr('bytes_downloaded', image['size'])
image = yield motor.Op(cursor.next_object)
finally:
IOLoop.instance().stop()
def run(*fileids):
connection = motor.MotorConnection().open_sync()
db = connection.tiler
cursor = db.images.find({'fileid': {'$in': fileids}})
_ids = []
for document in (yield motor.Op(cursor.to_list)):
print document
image_split = document['fileid'][:1] + '/' + document['fileid'][
1:3] + '/' + document['fileid'][3:]
for each in ('tiles', 'uploads', 'thumbnails'):
d = os.path.join(HERE, '..', 'static', 'tiles', image_split)
d = os.path.normpath(d)
if os.path.isdir(d):
print "DEL", d
shutil.rmtree(d)
yield motor.Op(db.images.remove, {'_id': document['_id']})
IOLoop.instance().stop()
def run(*fileids):
_redis = redis.client.Redis(
settings.REDIS_HOST,
settings.REDIS_PORT
)
connection = motor.MotorConnection().open_sync()
db = connection.tiler
cursor = db.images.find({'fileid': {'$in': fileids}})
_ids = []
for document in (yield motor.Op(cursor.to_list)):
print document
image_split = document['fileid'][:1] + '/' + document['fileid'][1:3] + '/' + document['fileid'][3:]
for each in ('tiles', 'uploads', 'thumbnails'):
d = os.path.join(HERE, '..', 'static', 'tiles', image_split)
d = os.path.normpath(d)
if os.path.isdir(d):
print "DEL", d
shutil.rmtree(d)
metadata_key = 'metadata:%s' % document['fileid']
if _redis.get(metadata_key):
print "Invalidated metadata cache"
_redis.delete(metadata_key)
lock_key = 'uploading:%s' % document['fileid']
_redis.delete(lock_key)
all_fileids_key = 'allfileids'
_redis.delete(all_fileids_key)
all_fileids_key += ':%s' % document['user']
_redis.delete(all_fileids_key)
cache_keys_key = 'thumbnail_grid:keys'
for key in _redis.lrange(cache_keys_key, 0, -1):
_redis.delete(key)
_redis.delete(cache_keys_key)
yield motor.Op(
db.images.remove,
{'_id': document['_id']}
)
IOLoop.instance().stop()
def upload_original(fileid, extension, static_path, bucket_id):
conn = S3Connection(settings.AWS_ACCESS_KEY, settings.AWS_SECRET_KEY)
bucket = conn.lookup(bucket_id) or conn.create_bucket(bucket_id, location=Location.EU)
db_connection = motor.MotorConnection().open_sync()
db = db_connection[settings.DATABASE_NAME]
original = find_original(fileid, static_path, extension)
if original:
relative_path = original.replace(static_path, '')
k = Key(bucket)
k.key = relative_path
print "Uploading original", original
s = os.stat(original)[stat.ST_SIZE]
print "%.1fKb" % (s / 1024.)
# reduced because I'm a cheapskate
k.set_contents_from_filename(original, reduced_redundancy=True)
print "Original uploaded"
else:
print "Original can't be found", repr(original)
def run(*fileids):
_redis = redis.client.Redis(settings.REDIS_HOST, settings.REDIS_PORT)
connection = motor.MotorConnection().open_sync()
db = connection.tiler
try:
cursor = db.images.find({'fileid': {'$in': fileids}})
for document in (yield motor.Op(cursor.to_list)):
pprint(document)
yield motor.Op(db.images.update, {'_id': document['_id']},
{'$set': {
'featured': False
}})
cache_keys_key = 'thumbnail_grid:keys'
for key in _redis.lrange(cache_keys_key, 0, -1):
_redis.delete(key)
_redis.delete(cache_keys_key)
finally:
IOLoop.instance().stop()
def run(*args):
_redis = redis.client.Redis(settings.REDIS_HOST, settings.REDIS_PORT)
connection = motor.MotorConnection().open_sync()
db = connection.tiler
yield motor.Op(db.images.remove, {'width': {'$exists': False}})
try:
cursor = (db.images.find({'featured': {'$exists': False}}))
image = yield motor.Op(cursor.next_object)
_fileids = {}
while image:
yield motor.Op(db.images.update, {'_id': image['_id']},
{'$set': {
'featured': True
}})
print image['fileid']
image = yield motor.Op(cursor.next_object)
finally:
IOLoop.instance().stop()
def run(*args):
_redis = redis.client.Redis(settings.REDIS_HOST, settings.REDIS_PORT)
connection = motor.MotorConnection().open_sync()
db = connection.tiler
try:
cursor = (db.comments.find())
comment = yield motor.Op(cursor.next_object)
_fileids = {}
while comment:
if comment['image'] not in _fileids:
image = yield motor.Op(db.images.find_one,
{'_id': comment['image']})
_fileids[comment['image']] = image['fileid']
fileid = _fileids[comment['image']]
print fileid
_redis.hincrby('comments', fileid, 1)
print _redis.hget('comments', fileid)
#_redis.incr('bytes_downloaded', image['size'])
comment = yield motor.Op(cursor.next_object)
finally:
IOLoop.instance().stop()
except ImportError:
print >> sys.stderr, ("Can't import motor.\n\n"
" Motor is an experimental async driver for"
" MongoDB, get it by cloning\n"
" git://github.com/ajdavis/mongo-python-driver.git"
" and switching to branch 'motor',\n"
" then put the mongo-python-driver directory"
" on your PYTHONPATH\n\n")
raise
if __name__ == "__main__":
opts = options.options()
# TODO: Mongo connection options
db = motor.MotorConnection().open_sync().motorblog
cache.startup(db)
if opts.ensure_indexes:
logging.info('Ensuring indexes...')
indexes.ensure_indexes(db)
logging.info(' done.')
base_url = opts.base_url
class U(tornado.web.URLSpec):
def __init__(self, pattern, *args, **kwargs):
"""Include base_url in pattern"""
super(U, self).__init__(
'/' + base_url.strip('/') + '/' + pattern.lstrip('/'), *args,
**kwargs)
except ImportError:
print >> sys.stderr, ("Can't import motor.\n\n"
" Motor is an experimental async driver for"
" MongoDB, get it by cloning\n"
" git://github.com/ajdavis/mongo-python-driver.git"
" and switching to branch 'motor',\n"
" then put the mongo-python-driver directory"
" on your PYTHONPATH\n\n")
raise
if __name__ == "__main__":
mongodb_uri = os.environ.get('MONGOLAB_URI', 'mongodb://localhost:27017')
db = motor.MotorConnection(mongodb_uri).open_sync().agg
static_path = 'static'
application = tornado.web.Application([
URLSpec(r"/static/(.+)",
StaticFileHandler, {"path": static_path},
name='static'),
URLSpec(r"/answer/(?P<quiz_id>.+)", AnswerHandler, name='answer'),
URLSpec(r"/(?P<quiz_id>[0-9]+)?", MainHandler, name='main'),
URLSpec(r"/example/?", ExampleHandler, name='example'),
],
db=db,
template_path='templates',
debug=True)
http_server = httpserver.HTTPServer(application)
def serve(listenuri, mongodburi):
#TODO add RedisURI
#TODO add Static Path setting
static_path = os.path.join(os.path.dirname(__file__), "static")
static_path_dict = dict(path=static_path)
template_path = os.path.join(os.path.dirname(__file__), "templates")
inbound_path = os.path.join(os.path.dirname(__file__), "../inbound")
motordb = motor.MotorConnection(mongodburi).open_sync().ace
mongodb = pymongo.Connection(mongodburi).ace
tornadoredisdb = tornadoredis.Client()
redisdb = redis.Redis()
application = tornado.web.Application(
handlers = [
tornado.web.URLSpec(r"/static/(.*)", tornado.web.StaticFileHandler, static_path_dict),
tornado.web.URLSpec(r"/(stylesheets/.*)", tornado.web.StaticFileHandler, static_path_dict),
tornado.web.URLSpec(r"/(images/.*)", tornado.web.StaticFileHandler, static_path_dict),
tornado.web.URLSpec(r"/(javascripts/.*)", tornado.web.StaticFileHandler, static_path_dict),
tornado.web.URLSpec(r"/(favicon.ico)$", tornado.web.StaticFileHandler, static_path_dict),
tornado.web.URLSpec(r"/(robots.txt)$", tornado.web.StaticFileHandler, static_path_dict),
#tornado.web.URLSpec(r'/login$', AuthLoginHandler, name='login'),
#tornado.web.URLSpec(r'/logout$', AuthLogoutHandler, name='logout'),
#tornado.web.URLSpec(r'/dashboard/(?P<search_oid_str_or_request>\w{24})$', DashboardHandler, name='dashboard+search'),
#tornado.web.URLSpec(r'/dashboard/(?P<search_oid_str_or_request>\w+)$', DashboardHandler, name='dashboard+newsearch'),
#tornado.web.URLSpec(r'/dashboard$', DashboardHandler, name='dashboard'),
#tornado.web.URLSpec(r'/search/(?P<search_oid_str>\w{24})/(?P<collection_oid_str>\w{24})$', SearchCollectionHandler, name='search+collection'),
#tornado.web.URLSpec(r'/search/(?P<search_oid_str>\w{24})$', SearchAllHandler, name='search'),
#tornado.web.URLSpec(r'/export/csv/(?P<search_oid_str>\w{24})/(?P<collection_oid_str>\w{24})$', ExportCSVSearchCollectionHandler, name='export+search+collection+csv'),
#tornado.web.URLSpec(r'/api/document/batch$', APIDocumentBatchHandler),
#tornado.web.URLSpec(r'/api/document/insert$', APIDocumentInsertHandler),
#tornado.web.URLSpec(r'/api/document/link$', APIDocumentLinkHandler),
#tornado.web.URLSpec(r"/$", tornado.web.RedirectHandler, kwargs=dict(url='/dashboard')), #Temporary pending main advert/news page
tornado.web.URLSpec(r"/jsontest$", JSONTestHandler),
tornado.web.URLSpec(r"/region/(.*)", RegionHandler),
tornado.web.URLSpec(r"/purchase$", PurchaseHandler),
tornado.web.URLSpec(r"/login$", LoginHandler, name='login'),
tornado.web.URLSpec(r"/logout$", LogoutHandler, name='logout'),
tornado.web.URLSpec(r"/$", MainHandler, name='index'),
tornado.web.URLSpec(r"/(.*)", PageErrorHandler, kwargs=dict(error=404)),
],
**{
'template_path': template_path,
'static_path': static_path,
'inbound_path': inbound_path,
'cookie_secret': 'cookiemonster',
'xsrf_cookies': True,
'debug': True, #FIXME
'login_url': '/login',
'mongodb': mongodb,
'motordb': motordb,
'tornadoredisdb': tornadoredisdb,
'redisdb': redisdb,
'siteuser': bson.ObjectId('50bb047f17a78f9c422b45da'),
'cache': {
'region': {
'key': 'aceregion',
'stamp': None,
'array': [],
'sectioned': {},
'map': {
'_id': {},
'slug': {},
},
},
},
'pycket': {
'engine': 'redis',
'storage': {
'host': 'localhost',
'port': 6379,
'db_sessions': 10,
'db_notifications': 11,
},
'cookies': {
'expires_days': 120,
},
},
}
)
server = tornado.httpserver.HTTPServer(application, xheaders=True)
server.listen(8080)
tornado.ioloop.IOLoop.instance().start()
class MessagesHandler(tornado.web.RequestHandler):
@tornado.web.asynchronous
def get(self):
"""Display all messages
"""
self.write('<a href="/compose">Compose a message</a><br>')
self.write('<ul>')
db = self.settings['db']
db.messages.find().sort([('_id', -1)]).each(self._got_message)
def _got_message(self, message, error):
if error:
raise tornado.web.HTTPError(500, error)
elif message:
self.write('<li>%s</li>' % message['msg'])
else:
# Iteration complete
self.write('</ul>')
self.finish()
db = motor.MotorConnection().open_sync().test
application = tornado.web.Application([(r'/compose', NewMessageHandler),
(r'/', MessagesHandler)],
db=db)
print 'Listening on http://localhost:8888'
application.listen(8888)
tornado.ioloop.IOLoop.instance().start()
def db(self):
if not self._db_connection:
self._db_connection = motor.MotorConnection().open_sync()
return self._db_connection[settings.DATABASE_NAME]
def upload_all_tiles(fileid,
static_path,
bucket_id,
max_count=0,
only_if_no_cdn_domain=False,
replace=True):
#print "upload_all_tiles", fileid, static_path, bucket_id, max_count, only_if_no_cdn_domain, replace
log_file = os.path.join(static_path, 'upload.%s.txt' % fileid)
conn = connect_to_s3()
bucket = conn.lookup(bucket_id) or conn.create_bucket(bucket_id,
location=Location.EU)
#bucket.set_acl('public-read')
db_connection = motor.MotorConnection().open_sync()
db = db_connection[settings.DATABASE_NAME]
document = yield motor.Op(db.images.find_one, {'fileid': fileid})
if not document:
logging.warning("Image %r does not exist" % fileid)
IOLoop.instance().stop()
return
if document.get('cdn_domain'):
if only_if_no_cdn_domain:
IOLoop.instance().stop()
return
else:
warnings.warn("%s already has a cdn_domain (%s)" %
(fileid, document['cdn_domain']))
try:
count = 0
all_done = True
all_tiles = list(find_all_tiles(fileid, static_path))
random.shuffle(all_tiles)
#if len(all_tiles) > max_count:
# total = max_count
#else:
# total = len(all_tiles)
total = len(all_tiles)
aggressive_headers = get_aggressive_headers()
for each in all_tiles:
# load which ones we've done every time to prevent
# parallel workers uploading the same file more than once
try:
done = [x.strip() for x in open(log_file) if x.strip()]
except IOError:
done = []
if each not in done:
done.append(each)
relative_path = each.replace(static_path, '')
k = Key(bucket)
k.key = relative_path
# docs:
# http://boto.cloudhackers.com/en/latest/ref/s3.html#boto.s3.\
# key.Key.set_contents_from_filename
print "uploading", relative_path,
try:
count_done = set(x.strip() for x in open(log_file))
except IOError:
count_done = []
print "(%d of %d)" % (len(count_done), total)
k.set_contents_from_filename(
each,
replace=replace,
reduced_redundancy=True,
headers=aggressive_headers,
)
k.make_public()
open(log_file, 'a').write(each + '\n')
count += 1
if max_count > 0 and count >= max_count:
print "STOPPING @", count
all_done = False
break
if all_done:
data = {'cdn_domain': settings.DEFAULT_CDN_TILER_DOMAIN}
print "Updating document finally"
yield motor.Op(db.images.update, {'_id': document['_id']},
{'$set': data})
# invalidate some redis keys
_redis = redis.client.Redis(settings.REDIS_HOST,
settings.REDIS_PORT)
lock_key = 'uploading:%s' % fileid
_redis.delete(lock_key)
metadata_key = 'metadata:%s' % fileid
# make it expire in a minute
data = _redis.get(metadata_key)
if data:
# this gives all workers a chance to finish
# any leftover jobs such as optimizations
_redis.setex(metadata_key, data, 60)
finally:
print "# done", count
IOLoop.instance().stop()
def main(args):
start = time.time()
opts = options.options()
destination_url = '/' + opts.base_url.lstrip('/')
parts = urlparse(args.source_url)
source_base_url = urljoin('%s://%s' % (parts[0], parts[1]),
parts[2].split('/xmlrpc.php')[0])
print 'Base URL', source_base_url
db = pymongo.Connection(safe=True).motorblog
motordb = motor.MotorConnection().open_sync().motorblog
if args.wipe:
print 'Wiping motorblog database'
db.connection.drop_database('motorblog')
print 'Creating capped collection "events"'
create_events_collection(motordb)
print 'Recreating indexes'
ensure_indexes(db)
source = Blog(args.source_url,
args.source_username,
args.source_password,
use_cache=not args.refresh,
verbose=args.verbose)
print 'Getting media library'
media_library = set([m['link'] for m in source.get_media_library()])
print ' %s assets\n' % len(media_library)
print 'Getting posts and pages'
post_structs = source.get_recent_posts(args.nposts)
print ' %s posts' % len(post_structs)
page_structs = source.get_pages()
print ' %s pages' % len(page_structs)
print
for structs, type in [
(post_structs, 'post'),
(page_structs, 'page'),
]:
print '%sS' % type.upper()
for struct in structs:
categories = struct.pop('categories', [])
struct['description'] = wordpress_to_markdown(
struct, media_library, db, destination_url, source_base_url)
post = Post.from_metaweblog(struct, type)
print '%-34s %s' % (post.title, post.status.upper())
for category_name in categories:
doc = db.categories.find_one({'name': category_name})
if doc:
category = Category(**doc)
else:
category = Category(name=category_name,
slug=slugify(category_name))
category.id = db.categories.insert(category.to_python())
print ' %-30s %s' % (category_name,
' NEW' if not doc else '')
post.categories.append(category)
db.posts.insert(post.to_python())
print '\nFinished %s %ss' % (len(structs), type)
print 'Posting "categories_changed" event'
db.events.insert(
{
'ts': datetime.datetime.utcnow(),
'name': 'categories_changed'
},
manipulate=False) # No need to add _id
print '\nFinished in %.2f seconds' % (time.time() - start)
