def command(cls, config_ini, org_names):
common.load_config(config_ini)
common.register_translator()
from ckan.plugins import toolkit
from ckan import model
orgs = [toolkit.get_action('organization_show')(
data_dict={'id': org_name})
for org_name in org_names]
source_org, dest_org = orgs
assert source_org
assert dest_org
search_results = toolkit.get_action('package_search')(
data_dict=dict(fq='publisher:%s' % source_org['name'], rows=1000))
print 'Datasets: %s' % search_results['count']
stats = Stats()
if len(search_results['results']) != search_results['count']:
assert 0, 'need to implement paging'
#context = {
# 'user': get_script_user(__name__)['name'],
# 'ignore_auth': True,
# 'model': model}
rev = model.repo.new_revision()
rev.author = 'script-%s.py' % __file__
for dataset in search_results['results']:
model.Package.get(dataset['id']).owner_org = dest_org['id']
#dataset_ = toolkit.get_action('package_patch')(
# context=context,
# data_dict=dict(id=dataset['id'], owner_org=dest_org['id']))
print stats.add('Changed owner_org', dataset['name'])
print stats.report()
print 'Writing'
model.Session.commit()
def command(cls, config_ini, write):
common.load_config(config_ini)
common.register_translator()
rev = model.repo.new_revision()
rev.author = "script-fix_mandate.py"
for package in model.Session.query(model.Package).filter(model.Package.state == "active"):
if "mandate" in package.extras:
mandate = package.extras.get("mandate")
try:
mandate = json.loads(mandate)
if isinstance(mandate, list):
stats.add("Already list", package.name)
elif isinstance(mandate, basestring):
stats.add("Fixing JSON string", package.name)
package.extras["mandate"] = json.dumps([mandate])
else:
stats.add("Problem JSON", package.name)
except ValueError:
if mandate != "":
stats.add("Fixing string", package.name)
package.extras["mandate"] = json.dumps([mandate])
else:
stats.add("Deleting empty string", package.name)
del package.extras["mandate"]
else:
stats.add("No mandate field", package.name)
print stats.report()
if write:
print "Writing"
model.Session.commit()
def command(cls, config_ini, write):
common.load_config(config_ini)
common.register_translator()
rev = model.repo.new_revision()
rev.author = 'fix_contact_details.py'
for package in model.Session.query(model.Package).filter_by(state='active'):
group = package.get_organization()
if not group:
stats.add('was not in a group', package.name)
continue
if package.extras.get('contact-name') == group.extras.get('contact-name'):
if package_is_effected(package, group):
if write:
package.extras['contact-name'] = ''
package.extras['contact-email'] = ''
package.extras['contact-phone'] = ''
package.extras['foi-name'] = ''
package.extras['foi-email'] = ''
package.extras['foi-web'] = ''
package.extras['foi-phone'] = ''
stats.add('resetting', 'Resetting package %s' % package.name)
print stats.report()
if write:
model.Session.commit()
def command(cls, config_ini, write, options):
common.load_config(config_ini)
common.register_translator()
rev = model.repo.new_revision()
rev.author = 'script-delete_cache_filepath.py'
process_all = True
if options.resource:
cls.process_resource(model.Resource.get(options.resource))
process_all = False
else:
# Get each dataset,
counter = 0
datasets_q = model.Session.query(model.Package) \
.filter_by(state='active')
rounded = int(math.ceil(datasets_q.count() / 100.0)) * 100
for x in xrange(0, rounded, 100):
datasets = datasets_q.offset(x).limit(100)
updated = False
for dataset in datasets.all():
counter += 1
print "Processing dataset %d\r" % counter,
for resource in dataset.resources:
if cls.process_resource(resource):
updated = True
for key in dataset_properties_to_make_null:
if getattr(dataset, key):
stats_dp.add('Making property null: %s' % key, dataset.name)
setattr(dataset, key, None)
updated = True
else:
stats_dp.add('Property has no value: %s' % key, dataset.name)
for key in dataset_extras_to_remove:
if key in dataset.extras:
#stats_de.add('Removing: %s' % key, dataset.name)
del dataset.extras[key]
updated = True
else:
stats_de.add('No field to remove: %s' % key, dataset.name)
# We will be committing 100 at a time
if updated and write:
print "\nCommitting changes"
import time
s = time.time()
model.Session.commit()
print "Committed in ", time.time() - s
print 'Resource Properties:\n', stats_rp.report(show_time_taken=False)
print 'Resource Extras:\n', stats_re.report()
print 'Dataset Properties:\n', stats_dp.report(show_time_taken=False)
print 'Dataset Extras:\n', stats_de.report()
def get_datasets_from_ckan(domain):
common.load_config(config_ini)
common.register_translator()
from pylons import config
apikey = config['dgu.merge_datasets.apikey']
ckan = ckanapi.RemoteCKAN('https://%s' % domain, apikey=apikey)
datasets = ckan.action.package_search(q='organogram', rows=400)
return datasets
def command(cls, config_ini, write):
common.load_config(config_ini)
common.register_translator()
def new_revision():
rev = model.repo.new_revision()
rev.author = 'script_delete_duplicate_datasets.py'
if write:
new_revision()
publisher = model.Group.get(options.publisher)
if publisher is None:
print "Publisher could not be found"
sys.exit(0)
guids = defaultdict(list)
for package in publisher.packages():
guids[package.extras.get('guid')].append(package)
for guid, packages in guids.items():
if guid is None:
for package in packages:
stats.add('Skip package not harvested', package.name)
continue
if len(packages) == 1:
stats.add('Skip guid without duplicates', guid)
continue
best_name = None
for i, package in enumerate(sorted(packages,
key=lambda x: x.metadata_modified,
reverse=options.keep_last)):
if (not best_name or
len(package.name) < len(best_name) or
(len(package.name) == len(best_name) and
package.name < best_name)):
best_name = package.name
if i == 0:
kept_package = package
else:
stats.add('Deleting', package.name)
package.name = package.name + '_'
package.state = 'deleted'
# Write the name changes, so that we can reuse the best_name.
stats.add('Keep', '%s->%s' % (kept_package.name, best_name))
if write:
model.Session.commit()
new_revision()
kept_package.name = best_name
if write:
model.Session.commit()
print stats.report()
def command(cls, config_ini):
common.load_config(config_ini)
common.register_translator()
from ckanext.dgu.model.feedback import Feedback
comment_hashes = []
headers = ["user_id", "package_id", "timestamp", "title", "comment"]
writer = csv.DictWriter(sys.stdout, headers)
for fb in model.Session.query(Feedback)\
.filter(Feedback.visible==True)\
.filter(Feedback.active==True)\
.order_by(Feedback.created):
if not any([fb.economic, fb.social, fb.effective, fb.linked, fb.other]):
stats.add('Missing any content', fb.id )
continue
user = model.User.get(fb.user_id)
pkg = model.Package.get(fb.package_id)
data = {
u"timestamp": fb.created.isoformat(),
u"package": pkg.name,
u"item": fb
}
content = render_template(TEMPLATE, data)
comment = content.replace(u'\r',u'').replace(u'\n',u'').replace(u' ', u'')
# Check for identical comments ... we want users duplicating comments on
# the same package (by mistake most often).
hashkey = u'{}.{}.{}'.format(comment, fb.package_id, fb.user_id).encode('utf8', 'ignore')
comment_hash = hashlib.md5(hashkey).hexdigest()
if comment_hash in comment_hashes:
stats.add('Duplicate post', fb.id )
continue
comment_hashes.append(comment_hash)
row = {
u"user_id": user.name[len("user_d"):],
u"package_id": pkg.name,
u"timestamp": fb.created.isoformat(),
u"title": "Feedback on the value of this dataset ",
u"comment": comment.encode('utf-8', 'ignore')
}
writer.writerow(row)
stats.add('Processed', fb.id )
def test_init_missing_cert():
""" Try to initialize the context with a nonexistant cert. """
config = load_config()
config['name'] = "failboat"
config['sign_messages'] = True
context = FedMsgContext(**config)
context.publish(topic='awesome', msg=dict(foo='bar'))
def main(username, password):
global config
config = common.load_config()
common.mkdirs(config["jive-base"])
os.chdir(config["jive-base"])
l = common.Lock(".lock")
global max_index
try:
with open(".max-index") as f:
max_index = int(f.read())
except IOError:
pass
tls.s = requests.Session()
login(username, password)
threads = int(config["jive-threads"])
for i in range(threads):
t = threading.Thread(target = worker, name = i, args = (tls.s.cookies, ))
t.daemon = True
t.start()
for c in contents():
q.put((iter_content, c))
q.join()
cleanup()
common.write_sync_done()
global index
with open(".max-index", "w") as f:
print(index, file = f)
def install():
config = load_config()
try:
run('ls blink')
except:
run('git clone ' + config.get('blink', 'repository'))
configure()
def __init__(self, configfile):
"""
Following class attributes initialized in __init__:
nova - client for nova service
keystone - client for keystone service
ceilometer - client for ceilometer service
hosts - all hosts for region
servers - all servers in region
"""
self.logger = logging.getLogger(__name__)
self.config = load_config(configfile)
if self.config:
try:
auth = v3.Password(username=self.config['username'],
password=self.config['password'],
project_name=self.config['tenant_name'],
auth_url=self.config['auth_url'],
user_domain_name=self.config['domain'],
project_domain_name=self.config['domain'],)
self.session = session.Session(auth=auth, verify=self.config['cacert'])
self.nova = client_nova.Client(2, session=self.session)
self.keystone = client_keystone.Client(session=self.session)
self.ceilometer = client_ceilometer.Client(2, session=self.session)
self.glance = client_glance.Client(2, session=self.session)
self.hosts = self.nova.hosts.list()
self.servers = self.nova.servers.list(search_opts = { 'all_tenants': 1 })
self.projects = self.getprojects()
self.flavors = self.getflavors()
self.connected = True
except:
self.logger.error('Error for authentication with credentials from ' + configfile)
self.connected = False
def main():
global config
config = common.load_config()
common.mkdirs(config["product-docs-base"])
os.chdir(config["product-docs-base"])
l = common.Lock(".lock")
get_dump()
valid_files = set([".lock", ".sync-done"])
pool = multiprocessing.Pool(processes=int(config["product-docs-threads"]))
for x in iter_dump():
x["product_"] = x["product"].replace("_", " ")
url = "https://access.redhat.com/documentation/%(language)s/" \
"%(product)s/%(version)s/pdf/%(name)s/" \
"%(product)s-%(version)s-%(name)s-%(language)s.pdf" % x
f = "%(product_)s/%(version)s/" \
"%(product)s-%(version)s-%(name)s-%(language)s.pdf" % x
pool.apply_async(download, (url, f))
valid_files.add(f)
pool.close()
pool.join()
remove_invalid_files(valid_files)
common.write_sync_done()
def main():
global config
config = common.load_config()
# Permit write of UTF-8 characters to stderr (required when piping output)
if sys.stderr.encoding == None:
sys.stderr = codecs.getwriter("UTF-8")(sys.stderr)
common.mkdirs(config["pt-base"])
os.chdir(config["pt-base"])
_lock = common.Lock(".lock")
tls.s = requests.Session()
login()
for i in range(int(config["pt-threads"])):
t = threading.Thread(target = worker, name = i, args = [tls.s.cookies])
t.daemon = True
t.start()
read_project_list()
while q.unfinished_tasks:
time.sleep(1)
q.join()
cleanup()
common.write_sync_done()
def xmpp_test():
global conf
if not conf:
conf = load_config()
if not push_in_queue(construct_message(conf['xmpp_recipients'], 'test message for: %s' % conf['xmpp_recipients'])):
return '', 500
return '', 204
def setUp(self):
self.config = load_config()
self.config['name'] = local_name
self.config['mute'] = True
self.config['persistent_store'] = Mock()
self.replay_context = ReplayContext(**self.config)
self.replay_thread = ReplayThread(self.replay_context)
self.context = zmq.Context()
def prometheus_alert():
global conf
if not conf:
conf = load_config()
msg = PrometheusAlert(request.data.decode()).plain()
html = PrometheusAlert(request.data.decode()).html()
push_in_queue(construct_message(conf['xmpp_recipients'], msg, html))
return '', 204
def setUp(self):
self.config = load_config()
self.config['name'] = local_name
self.config['persistent_store'] = Mock()
self.replay_context = ReplayContext(**self.config)
self.request_context = zmq.Context()
self.request_socket = self.request_context.socket(zmq.REQ)
self.request_socket.connect(
self.config['replay_endpoints'][local_name])
def setUp(self):
config = load_config()
self.hub = CentralMokshaHub(config=config)
self.context = FedMsgContext(**config)
# fully qualified
self.fq_topic = "com.test_prefix.dev.unittest.foo"
# short version
self.topic = "foo"
def add_instance(count):
conn = connect()
config = load_config()
ami = config.get('aws', 'ami')
spot_price = config.getfloat('aws', 'spot_price')
key_name = config.get('aws', 'key_name')
instance_type = config.get('aws', 'instance_type')
availability_zone_group = config.get('aws', 'availability_zone_group')
#placement = config.get('aws', 'placement')
security_group = config.get('aws', 'security_group')
create_ondemand_instances(conn, ami, security_group, instance_type, count, key_name)
def test_init_invalid_endpoint():
try:
config = load_config()
config['name'] = local_name
config['persistent_store'] = Mock()
tmp = zmq.Context()
placeholder = tmp.socket(zmq.REP)
placeholder.bind('tcp://*:{0}'.format(
config["replay_endpoints"][local_name].rsplit(':')[-1]
))
context = ReplayContext(**config)
finally:
placeholder.close()
def connect():
config = load_config() #ConfigParser.ConfigParser()
aws_key = config.get('aws', 'aws_key')
aws_secret = config.get('aws', 'aws_secret')
conn = boto.ec2.connect_to_region(
'us-east-1',
aws_access_key_id = aws_key,
aws_secret_access_key = aws_secret,
)
return conn
def init():
global config
config = common.load_config()
if config["thunderbird-base"] is None:
return
if isrunning():
print >>sys.stderr, "thunderbird.py: thunderbird is running, disabling plugin"
config["thunderbird-base"] = None
return
rmpath(config["thunderbird-folder"])
mkpath(config["thunderbird-folder"])
config["thunderbird-base"] = base(spd(config["thunderbird-folder"]))
def main():
# fetch CLI arguments
(
dry_run,
filter_include_list,
filter_exclude_list
) = common.read_arguments()
# load config from file
config_data = common.load_config()
config_auth_token = config_data['AUTH_TOKEN']
# fetch repository list and wiki status of the specified repository type
print('Building repository list:')
all_repository_set = get_repository_name_wiki_status_set(
config_auth_token,
config_data['REPOSITORY_TYPE'],
common.RepositoryFilter(filter_include_list,filter_exclude_list)
)
# get total count, if zero then no work
repository_count = len(all_repository_set)
if (repository_count < 1):
print('\nNo repositories for processing')
return
print('\nTotal repositories: {0}'.format(repository_count))
# determine wiki enabled count
wiki_enabled_repository_set = filter_repository_wiki_enabled(all_repository_set)
wiki_enabled_count = len(wiki_enabled_repository_set)
if (wiki_enabled_count < 1):
# no projects enabled - no work
print('All wikis disabled')
return
print('Wikis enabled: {0}'.format(wiki_enabled_count))
# disable wikis (only simulation if dry run mode)
print('\n\nDisabling wikis{0}:'.format(' [DRY RUN]' if (dry_run) else ''))
for repository_name in wiki_enabled_repository_set:
if (not dry_run):
disable_repository_wiki(config_auth_token,repository_name)
print(repository_name)
def main():
# load config from file
config_data = common.load_config(
config_key_addition_set = { ORGANIZATION_CONFIG_KEY }
)
config_auth_token = config_data['AUTH_TOKEN']
# fetch repository names/sizes of the specified type
print('Building repository list ordered by size:')
repository_list = get_organization_repository_size_sorted_list(
config_auth_token,
config_data[ORGANIZATION_CONFIG_KEY],
config_data['REPOSITORY_TYPE']
)
# output list, repository URI/size - tab separated
for repository_uri,repository_size in repository_list:
print('{0}\t{1}'.format(repository_uri,repository_size))
def main():
conf = load_config()
bot = Bot(conf['xmpp_jid'], conf['xmpp_password'])
if not timeout(bot.start, [conf['xmpp_host'], conf['xmpp_port']]):
terminate()
receiver = IPCReceiver(conf['mq_name'])
try:
while True:
data = json.loads(receiver.receive())
if 'html' in data:
bot.send_message_to(data['message'], data['recipients'], html=data['html'])
else:
bot.send_message_to(data['message'], data['recipients'])
except KeyboardInterrupt:
pass
except ipc.SignalError:
pass
finally:
receiver.cleanup()
terminate()
def main(username, password):
global config
config = common.load_config()
common.mkdirs(config["jive-base"])
os.chdir(config["jive-base"])
l = common.Lock(".lock")
tls.s = requests.Session()
login(username, password)
threads = int(config["jive-threads"])
for i in range(threads):
t = threading.Thread(target = worker, name = i, args = (tls.s.cookies, ))
t.daemon = True
t.start()
for c in contents():
q.put((iter_content, c))
q.join()
cleanup()
common.write_sync_done()
def main():
config = load_config()
# TODO: break down into more functions?
boto_session = get_boto_session(config)
# TODO: if s3_resources continue to be used only to get bucket, merge get_s3_resource into get_s3_bucket. but wait
# until after any restructuring as class.
s3_resource = get_s3_resource(boto_session)
s3_bucket = get_s3_bucket(s3_resource, config)
# get list of files in relevant dir of bucket
# by converting to list we should trigger just one call to S3 API, unlike iterating over collection
# see https://boto3.amazonaws.com/v1/documentation/api/latest/guide/collections.html#when-collections-make-requests
objs = list(
s3_bucket.objects.filter(
Prefix=config['AWS']['s3_trips_prefix']).all())
# use a pandas.DataFrame for convenient storage of metadata about the trip data csv files
# collect metadata as (vertical) list of (horizontal) lists (rows), then construct DataFrame, for efficiency
# (see "Notes" on
# https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.append.html#pandas-dataframe-append)
# parse filename into year, month, and trip type
# unfortunately, the DataFrame constructor only accepts a singleton dtype argument and the default inference doesn't
# preserve smaller numpy data types in the input. any improvement in the efficiency of the filtration and sorting
# below probably not worth the cost of using astype() to force the series to convert after creation given relatively
# small number of rows.
trip_files = [parse_key(obj.key) for obj in objs]
trip_files = DataFrame(trip_files,
columns=[col[0] for col in TRIPFILE_METADATA_COLS])
# exclude undesired trip files now, using labeled columns, rather than the less convenient file names
# for now, limit to those with usable lat and long columns: green and yellow, through the first half of 2016
undesired_indices = trip_files[
~trip_files['Type'].isin(['green', 'yellow']) |
(trip_files['Year'] > 2016) | ((trip_files['Year'] == 2016) &
(trip_files['Month'] > 6))].index
trip_files.drop(undesired_indices, axis=0, inplace=True)
# assuming pandas isn't optimized enough under the hood that sorting first would speed the lookups involved in the
# drops, more efficient to reduce dataset size before sorting
trip_files.sort_values(['Year', 'Month'],
axis=0,
ascending=True,
inplace=True)
if TESTING and TESTING_CSV_LIMIT is not None:
trip_files = [trip_files.head(min(len(trip_files), TESTING_CSV_LIMIT))]
# separate trip_file entries into separate pandas DataFrames for each time period (year, for now) and store them
# chronologically in time_period_tables list
# note that groupby guarantees the sort order done above will be preserved within each group — see
# https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.groupby.html#pandas-dataframe-groupby
year_groups = trip_files.groupby(['Year'])
time_period_tables = [
year_groups.get_group(year) for year in year_groups.groups
]
# simulate the time periods in sequence
[
simulate_time_period(time_period_table, config)
for time_period_table in time_period_tables
]
#!/usr/bin/env python3
import argparse
from pprint import pprint
from common import load_config
from common import encode
from common import send_msg
def opt_parser():
parser = argparse.ArgumentParser(
description='Network reconfiguration node')
parser.add_argument('--net_config',
default='config/sample_graph3.json',
type=str)
return parser
if __name__ == '__main__':
parser = opt_parser()
opt = parser.parse_args()
config = load_config(opt.net_config)
pprint(config)
start_msg = encode(dict(type="start"))
for node in config.values():
send_msg(start_msg, host=node['host'], port=node['port'])
print(node)
base_dir = os.path.dirname(os.path.abspath(__file__))
script_dir = os.path.join(base_dir, 'Common')
sys.path.insert(0, script_dir)
import common
from polyglotdb.client.client import PGDBClient
token = common.load_token()
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('corpus_name', help='Name of the corpus')
args = parser.parse_args()
corpus_name = args.corpus_name
directories = [
x for x in os.listdir(base_dir) if os.path.isdir(x) and x != 'Common'
]
if args.corpus_name not in directories:
print(
'The corpus {0} does not have a directory (available: {1}). Please make it with a {0}.yaml file inside.'
.format(args.corpus_name, ', '.join(directories)))
sys.exit(1)
corpus_conf = common.load_config(corpus_name)
print('Processing...')
client = PGDBClient('http://localhost:{}'.format(8080), token=token)
client.delete_database(corpus_name)
import os
import re
import glob
from openpyxl import Workbook
from formula.parser import FormulaParser
from settings import BASE_DIR
from common import load_config
CONFIG = load_config()
WORK_ORDERS_PATH = os.path.join(BASE_DIR, CONFIG.get('default', 'formula_dir'))
def extract_viscosity():
"""提取粘度数据"""
all_formula_files = glob.glob(f"{WORK_ORDERS_PATH}/**/*.xlsx",
recursive=True)
result = []
for filepath in all_formula_files:
parser = FormulaParser(filepath)
formulas = parser.parse()
for formula in formulas:
after_adding_requirement = formula['metas'][
'after_adding_requirement']
if isinstance(after_adding_requirement, list):
for requirement in after_adding_requirement:
if requirement.find("粘度要求") >= 0:
result.append(
dict(name=formula['name'], viscosity=requirement))
elif after_adding_requirement and isinstance(
self.signal_hup = 0
ts_msg = TsMessage(_ts, _msg)
ts_msg.print_raw()
self.msgs_curr_total = self.msgs_curr_total + 1
def handler_atexit():
log("Terminated")
if __name__ == "__main__":
atexit.register(handler_atexit)
log("Running: Pid {:5d}".format(os.getpid()))
params = {}
load_config(params, "config/config.txt")
host = params['addr']
port = params['port']
type = params['type']
print(params['vers'], params['name'])
client = MyClient(_host=host, _port=port, _datatype=type)
client.run()
sys.exit(0)
def command(cls, config_ini, dataset_names, options):
common.load_config(config_ini)
common.register_translator()
from pylons import config
apikey = config['dgu.merge_datasets.apikey']
ckan = ckanapi.RemoteCKAN('https://data.gov.uk', apikey=apikey)
#ckan = ckanapi.LocalCKAN()
if options.publisher:
org_name = common.name_stripped_of_url(options.publisher)
if options.search:
results = ckan.action.package_search(q=options.search,
fq='publisher:%s' %
org_name,
rows=100,
escape_q=False)
dataset_names.extend(
[dataset['name'] for dataset in results['results']])
else:
org = ckan.action.organization_show(id=org_name,
include_datasets=True)
dataset_names.extend([d['name'] for d in org['packages']])
datasets = []
datasets_by_name = {}
def get_extra(dataset, key):
for extra in dataset['extras']:
if extra['key'] == key:
return extra['value']
for dataset_name in dataset_names:
print 'Dataset: %s' % dataset_name
for dataset_name in dataset_names:
# strip off the url part of the dataset name, if there is one
dataset_name = common.name_stripped_of_url(dataset_name)
dataset = ckan.action.package_show(id=dataset_name)
harvest_source_ref = get_extra(dataset, 'harvest_source_reference')
if harvest_source_ref:
print '** Discarding dataset %s due to harvest source: %s **' \
% (dataset_name, harvest_source_ref)
continue
datasets.append(dataset)
datasets_by_name[dataset['name']] = dataset
datasets.sort(key=lambda x: x['metadata_modified'])
# aggregate resources
def resource_identity(res_dict, dataset_name):
return (res_dict.get('date'), res_dict['url'],
res_dict.get('title') or res_dict['description'],
res_dict.get('format'), dataset_name)
combined_resources = {} # identity
res_stats = Stats()
for dataset in datasets:
for resource in dataset['resources']:
identity = resource_identity(resource, dataset['name'])
resource['dataset_name'] = dataset['name']
if identity in combined_resources:
print res_stats.add(
'Discarding duplicate', '\n%s duplicate of \n%s' %
(resource, combined_resources[identity]))
else:
combined_resources[identity] = resource
resources = combined_resources.values()
# find dates for resources
if options.frequency:
url_munge_re = re.compile('(%20|-|_|\.)')
def fields_to_hunt_for_date(res):
date = res.get('date')
if date:
yield 'date', date
title = res.get('title')
if title:
yield 'title', title
yield 'description', res['description']
yield 'url', url_munge_re.sub(' ', res['url'])
if not options.update:
dataset = datasets_by_name[res['dataset_name']]
yield 'dataset-title', dataset['title']
yield 'dataset-notes', dataset['notes']
ensure_regexes_are_initialized()
global regexes
for resource in resources:
for field_name, field_value in fields_to_hunt_for_date(
resource):
if options.frequency in ('monthly', 'quarterly',
'twice annually'):
month, year = hunt_for_month_and_year(field_value)
if year and month:
resource['date'] = '%02d/%s' % (month, year)
res_stats.add(
'Found date in %s' % field_name,
'%s %r' % (resource['date'], resource))
if resource.get(
'resource_type') == 'documentation':
resource['resource_type'] = 'file'
res_stats.add('Converted additional resource',
resource)
break
elif options.frequency == 'annually':
year = regexes['year'].search(field_value)
if year:
resource['date'] = year.groups()[0]
res_stats.add(
'Found date in %s' % field_name,
'%s %r' % (resource['date'], resource))
if resource.get(
'resource_type') == 'documentation':
resource['resource_type'] = 'file'
res_stats.add('Converted additional resource',
resource)
break
else:
if resource.get('resource_type') == 'documentation':
print res_stats.add(
'Could not find date but it\'s Additional Resource',
resource)
continue
print res_stats.add('Could not find date', resource)
continue
print 'Resources: \n', res_stats
resources_without_date = [
res for res in resources if not res.get('date')
and res.get('resource_type') != 'documentation'
]
for i, res in enumerate(resources_without_date):
print 'Resources without dates %s/%s' % (
i + 1, len(resources_without_date))
for field_name, field_value in fields_to_hunt_for_date(res):
print ' %s: %s' % (
field_name, field_value.encode('latin-1', 'ignore'))
print 'https://data.gov.uk/dataset/%s/resource/%s' % (
res['dataset_name'], res['id'])
date_format = {
'annually': 'YYYY',
'monthly': 'MM/YYYY',
'twice annually': 'MM/YYYY',
'quarterly': 'MM/YYYY'
}
input_ = raw_input(
'Date (%s) or DOCS to make it an Additional Resource: ' %
date_format[options.frequency])
if input_.strip().lower() == 'docs':
res['date'] = ''
res['resource_type'] = 'documentation'
else:
res['date'] = input_
resources.sort(key=lambda x: x.get('date', '').split('/')[::-1])
# Ensure there is not a mixture of resources with and without a date
have_dates = None
for res in resources:
if res.get('resource_type') == 'documentation':
continue
if have_dates is None:
have_dates = bool(res.get('date'))
else:
has_date = bool(res.get('date'))
if has_date != have_dates:
print[res.get('date') for res in resources]
print 'Cannot mix resources with dates and others without!'
import pdb
pdb.set_trace()
# Remove 'dataset_name' and others fields from resources
ignore_res_fields = set(('dataset_name', 'created', 'position',
'revision_id', 'id', 'tracking_summary'))
for res in resources:
for field in ignore_res_fields & set(res.keys()):
del res[field]
# Merge dataset fields
def get_all_fields_and_values(datasets):
ignore_fields = set((
'id',
'resources',
'last_major_modification',
'data_dict',
'revision_timestamp',
'num_tags',
'metadata_created',
'metadata_modified',
'odi_certificate',
'extras', # they are at top level already
'timeseries_resources',
'individual_resources',
'additional_resources',
'revision_id',
'organization',
'tracking_summary',
'num_resources',
'license_title',
'author',
'author_email',
'maintainer',
'maintainer_email',
'temporal_granularity',
'geographic_granularity',
'state',
'isopen',
'url',
'date_update_future',
'date_updated',
'date_released',
'precision',
'taxonomy_url',
'temporal_coverage-from',
'temporal_coverage-to',
'published_via',
'creator_user_id',
))
first_fields = [
'title', 'name', 'notes', 'theme-primary', 'theme-secondary'
]
all_field_values = defaultdict(list)
for dataset in datasets:
for field in dataset:
if field not in ignore_fields and dataset[field]:
all_field_values[field].append(dataset[field])
for field in first_fields:
yield field, all_field_values.get(field, [])
for field in all_field_values:
if field not in first_fields:
yield field, all_field_values[field]
spend_data_defaults = {
'geographic_coverage': None,
'theme-primary': 'Government Spending',
'theme-secondary': None,
'update_frequency': 'monthly',
}
combined_dataset = {'resources': resources}
all_fields_and_values = get_all_fields_and_values(datasets)
for field, values in all_fields_and_values:
if field == 'notes':
values = [value.strip() for value in values]
if field == 'tags':
# just merge them up-front and
# dont offer user any choice
tags_by_name = {}
for dataset_tags in values:
for tag in dataset_tags:
if tag['name'] not in tags_by_name:
tags_by_name[tag['name']] = tag
values = [tags_by_name.values()]
if field in ('codelist', 'schema'):
# just merge them up-front
# And convert the dict into just an id string
ids = set()
for dataset_values in values:
for value_dict in dataset_values:
ids.add(value_dict['id'])
values = [list(ids)]
print '\n%s:' % field
pprint(list(enumerate(values)))
if options.spend and field in spend_data_defaults:
value = spend_data_defaults[field]
print 'Spend data defaults to: %s' % value
values = [value] if value is not None else None
# dont be case-sensitive for boolean fields
if field == 'core-dataset':
values = [v.lower() for v in values]
try:
values_identicle = len(set(values)) == 1
except TypeError:
if values and len(values):
val1 = values[0]
for val in values[1:]:
if val != val1:
values_identicle = False
break
else:
values_identicle = True
if (not values) or (not len(values)):
pass
elif values_identicle:
value = values[0]
elif field == 'name':
while True:
from ckan.lib.munge import munge_title_to_name
munged_title = munge_title_to_name(
combined_dataset['title'])
print munge_title_to_name(
datasets[0]['organization']['title'])
value = raw_input('Type new value (%s): ' % (munged_title))
if not value:
value = munged_title
if len(value) < 3:
print 'Too short'
continue
if value in values:
print 'That name is taken'
continue
existing = ckan.action.package_autocomplete(q=value)
if value in existing:
print 'That name is taken on CKAN'
continue
break
else:
while True:
response = raw_input(
'%s: value (number) or type new one: ' % field)
try:
value_index = int(response)
value = values[value_index]
print value
except ValueError:
# fix pound signs if the user pasted from the repr'd version
response = re.sub(r'\\xa3', u'\xa3', response)
value = response
if not value and field in ('title', 'owner_org', 'notes',
'license_id'):
print 'You must have a value for this field!'
continue
break
if value:
combined_dataset[field] = value
# Store
print '\nMerged dataset:\n'
pprint(combined_dataset)
response = raw_input(
'Press enter to write or pdb to edit in pdb first: ')
if response == 'pdb':
import pdb
pdb.set_trace()
try:
if options.update:
ckan.action.dataset_update(**combined_dataset)
else:
ckan.action.dataset_create(**combined_dataset)
except Exception, e:
print e
import pdb
pdb.set_trace()
def test_init_missing_endpoint():
""" Try to initialize the context with a nonexistant service name. """
config = load_config()
config['name'] = "failboat"
config['sign_messages'] = True
context = FedMsgContext(**config)
ap.add_argument("list", nargs="?")
return vars(ap.parse_args())
def isgzip(f):
bytes = f.read(2)
f.seek(0)
return bytes == b"\x1F\x8B"
if __name__ == "__main__":
warnings = 0
global config
config = common.load_config()
args = parse_args()
if args["quiet"]:
common.progress = lambda x, y: None
common.progress_finish = lambda: None
if not config["lists-sync"]:
print("Please configure lists in $HOME/.satools before running %s." %
sys.argv[0],
file=sys.stderr)
sys.exit(1)
common.mkdirs(config["lists-base"])
os.chdir(config["lists-base"])
def main():
parser = argparse.ArgumentParser()
parser.add_argument('--no-download', action='store_true')
parser.add_argument('--no-remux', action='store_true')
parser.add_argument('--no-upload', action='store_true')
parser.add_argument('--no-notify', action='store_true')
parser.add_argument('--no-delete', action='store_true')
parser.add_argument('--force', action='store_true')
parser.add_argument('--force-log-to-file', action='store_true')
parser.add_argument('--override-channel-name')
parser.add_argument('--override-video-name')
parser.add_argument('video_id')
args = parser.parse_args()
config = load_config()
log_filename = LOGDIR / f'{args.video_id}-{os.getpid()}.log' if not sys.stdout.isatty(
) or args.force_log_to_file else None
# For subprocess
log_file = log_filename.open('a') if log_filename else None
if log_file:
sys.stderr = sys.stdout = log_file
setup_logging(filename=log_filename)
log.info(f'Starting download for {args.video_id}')
pid_exists, active_downloaders = check_pid(args.video_id)
if pid_exists and not args.force:
raise ValueError('Another downloader is still alive, exiting')
else:
active_downloaders[args.video_id] = os.getpid()
with open_state() as state:
state['active_downloaders'] = active_downloaders
if args.override_channel_name and args.override_video_name:
log.info(
'Using overridden channel and video name, setting is_upcoming to false'
)
channel_name = args.override_channel_name
video_name = args.override_video_name
# There's no reason to use these overrides for an upcoming video
is_upcoming = False
else:
player_response = get_video_info(args.video_id)
if 'videoDetails' not in player_response:
log.error(f'{args.video_id} has no details, cannot proceed '
'(playability: {}, {})'.format(
player_response["playabilityStatus"]["status"],
player_response["playabilityStatus"]["reason"],
))
sys.exit(1)
else:
channel_name = player_response['videoDetails']['author']
video_name = player_response['videoDetails']['title']
is_upcoming = player_response['videoDetails'].get(
'isUpcoming', False)
log.info(f'Channel: {channel_name}')
log.info(f'Title: {video_name}')
log.info(f'Upcoming: {is_upcoming}')
if is_upcoming:
wait(player_response, config)
filename_base = sanitize_filename(video_name)
log.info(f'Filename base: {filename_base}')
# Copy youtube-dl's naming scheme
filepath_streamlink = WORKDIR / f'{filename_base}-{args.video_id}.ts'
# TODO: If file already exists, rename it and concatenate it later?
# XXX: youtube-dl used to be less reliable than streamlink for downloading
# streams - that may no longer be the case.
# XXX: Invoke this in a less hacky manner
# The reason for doing this is that I wanted to use streamlink
# inside the venv but in a separate process,
# without hardcoding the path of the venv.
streamlink_args = [
'--force', # Overwrite any existing file
'--hls-timeout',
'60',
# XXX: This doesn't work right now!
# See https://github.com/streamlink/streamlink/issues/2936
'--hls-live-restart',
'--retry-streams',
'10',
'--retry-max',
'10',
'-o',
str(filepath_streamlink),
f'https://www.youtube.com/watch?v={args.video_id}',
'best',
]
if not args.no_download:
log.info(f'Starting streamlink with args: {streamlink_args}')
fork_return = os.fork()
if fork_return == 0:
sys.argv = streamlink_args
streamlink_main()
else:
os.wait()
else:
log.info('Skipping download')
filename_output = f'{filename_base}-{args.video_id}.mp4'
filepath_output = WORKDIR / filename_output
ffmpeg_args = (
'ffmpeg',
'-y',
'-i',
filepath_streamlink,
'-c',
'copy',
'-movflags',
'faststart',
'-metadata',
f'title={video_name}',
'-metadata',
f'artist={channel_name}',
'-metadata',
f'comment=https://www.youtube.com/watch?v={args.video_id}',
filepath_output,
)
if not args.no_remux:
log.info('Remuxing to mp4')
subprocess.run(ffmpeg_args, stdout=log_file)
else:
log.info('Skipping remux')
# Upload
if not args.no_upload:
link_url, thumbnail = upload(
sanitize_filename(channel_name),
# This argument duplication is kind of silly...
filename_output,
filepath_output,
)
# We won't have link and thumb if not uploading without
# going through a bunch more effort.
if not args.no_notify:
notify(
channel_name,
video_name,
link_url,
thumbnail,
)
else:
log.info('Skipping notify')
else:
log.info('Skipping upload')
if not args.no_delete:
log.info('Deleting work files')
filepath_streamlink.unlink()
filepath_output.unlink()
log.info('Cleaning up state')
with open_state() as state:
active_downloaders = state.get('active_downloaders', {})
active_downloaders.pop(args.video_id, None)
else:
log.info('Skipping cleanup')
log.info('All done!')
import os
from os import environ
from common import load_config
load_config(environ.get('PYTHONPATH', ''))
class Config:
SECRET_KEY = os.urandom(24)
SESSION_COOKIE_NAME = environ.get('SESSION_COOKIE_NAME')
SESSION_COOKIE_SECURE = True
SQLALCHEMY_DATABASE_URI = environ.get('SQLALCHEMY_DATABASE_URI')
SQLALCHEMY_ECHO = False
SQLALCHEMY_TRACK_MODIFICATIONS = False
class ProdConfig(Config):
FLASK_ENV = 'production'
DEBUG = False
TESTING = False
class DevConfig(Config):
FLASK_ENV = 'development'
DEBUG = True
TESTING = True
def fixup_versions(config: dict = None):
if config is None:
config = common.load_config()
import subprocess
import glob
import mmap
import sys
root = common.root_dir()
script_dir = pathlib.Path(__file__).absolute().parent
res_hacker = glob.glob(f"{root}/packages/**/ResourceHacker.exe",
recursive=True)[0]
data_dir = common.unity_data_dir(config["unityBuildDir"])
if data_dir is None:
print("No unity build dir found", file=sys.stderr)
return
plugins_dir = data_dir / "Managed"
with open(script_dir / "version.rc.in", "r", newline="") as file:
res = file.read()
version_rc = script_dir / "version.rc"
version_res = version_rc.with_suffix(".res")
version = "1.0.0.0"
for plugin_dll in config["invalidFileVersion"]:
plugin = plugins_dir / plugin_dll
with open(version_rc, "w") as file:
file.write(res.format(name=plugin_dll, version=version))
subprocess.check_call([
res_hacker,
"-open",
version_rc,
"-save",
version_res,
"-action",
"compile",
"-log",
"CON",
])
# damn ResourceHacker puts invalid length bytes as far .NET is concerned,
# doesn't seem to affect any other tools though
fileversion_bytes = bytearray.fromhex(
"46 00 69 00 6c 00 65 00 56 00 65 00 72 00 73 00 69 00 6F 00 6E 00"
)
with open(version_res, "r+b") as f:
with mmap.mmap(f.fileno(), 0) as mm:
index = -1
while True:
index = mm.find(fileversion_bytes, index + 1)
if index == -1:
break
# +1 for the null character I guess
len_str = str(hex(len(version) + 1))[2:]
if len(len_str) == 1:
len_str = "0" + len_str
assert len(len_str) == 2
mm[index - 4] = bytearray.fromhex(len_str)[0]
subprocess.check_call([
res_hacker,
"-open",
plugin,
"-save",
plugin,
"-action",
"addskip",
"-res",
version_res,
"-log",
"CON",
])
print(f"{plugin}: fixed version resource")
def main():
global conf
conf = load_config()
service.run(host=conf['flask_host'], port=conf['flask_port'])
def __init__(self):
self.params = {}
load_config(self.params, "config/config.txt")
return
def get_cache_status(archival):
if not archival.cache_filepath:
return 'Not cached'
if os.path.exists(archival.cache_filepath):
return 'Cached'
return 'Cache missing on disk!'
if __name__ == '__main__':
usage = __doc__ + """
usage:
%prog [-w] <ckan.ini>
"""
parser = OptionParser(usage=usage)
parser.add_option("-w", "--write",
action="store_true", dest="write",
help="write the theme to the datasets")
parser.add_option('-d', '--dataset', dest='dataset')
parser.add_option('-r', '--resource', dest='resource')
parser.add_option('-o', '--organization', dest='organization')
(options, args) = parser.parse_args()
if len(args) != 1:
parser.error('Wrong number of arguments (%i)' % len(args))
config_filepath = args[0]
print 'Loading CKAN config...'
common.load_config(config_filepath)
common.register_translator()
print 'Done'
fix_links(options)
if __name__ == '__main__':
USAGE = '''Daily script for government
Usage: python %s <config.ini> [task]
Where:
[task] - task to run (optional), picked from:
%s
or run multiple by separating by a comma.
''' % (sys.argv[0], ','.join(TASKS_TO_RUN))
if set(sys.argv) & set(('--help', '-h')):
print USAGE
sys.exit(1)
if len(sys.argv) < 2:
err = 'Error: Please specify config file.'
print USAGE, err
logging.error('%s' % err)
sys.exit(1)
config_file = sys.argv[1]
config_ini_filepath = os.path.abspath(config_file)
if len(sys.argv) == 3:
TASKS_TO_RUN = sys.argv[2].split(',')
load_config(config_ini_filepath)
register_translator()
logging.config.fileConfig(config_ini_filepath)
command(config_file)
if admin_count:
print " -> ",
print ', '.join(
u.name
for u in group.members_of_type(model.User, 'admin').all())
editor_count = group.members_of_type(model.User, 'editor').count()
print "Editors: {uc}".format(uc=editor_count)
if editor_count:
print " -> ",
print ', '.join(
u.name
for u in group.members_of_type(model.User, 'editor').all())
print "Dataset count: {dc}".format(
dc=group.members_of_type(model.Package).count())
if __name__ == '__main__':
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('config', help='CKAN config .ini filepath')
parser.add_argument('old_name', metavar='old_name')
parser.add_argument('new_name', metavar='new-name')
parser.add_argument('-t', '--title', help='Title to set')
args = parser.parse_args()
common.load_config(args.config)
common.register_translator()
PublisherRenamer().rename(args.old_name, args.new_name, args.title)
def start(demo=False):
if demo:
# load demo options. it will escape config file.
cookie_secret = common.hmacstr(common.randomstr(), common.randomstr())
args = [
sys.argv[0], "--debug", "--host=0.0.0.0", "--port=8080",
"--base_url=/ssweb", "--service_name=shadowsocks",
"--cookie_secret=" + cookie_secret, "--logging=debug"
]
options.parse_command_line(args)
else:
# pre-parse the command line options. it will be over write by 'load
# options from config file'. by then, it yet loaded.
options.parse_command_line()
if options.config is not None:
# load options from specified config file
if not os.path.isfile(options.config):
err_("Can't find config file '%s'." % options.config)
exit(1)
else:
config = common.load_config(options.config)
if config is not None:
info_("Load config from file '%s'." % options.config)
args = [sys.argv[0]]
for item in config:
args += ["--%s=%s" % (item, config[item])]
try:
options.parse_command_line(args)
except tornado.options.Error:
err_("Error on config file option.")
sys.exit(1)
else:
# load options from config file, if the file exists.
config_file = common.find_config_file()
if config_file is not None:
config = common.load_config(config_file)
if config is not None:
info_("Load config from file '%s'." % config_file)
args = [sys.argv[0]]
for item in config:
args += ["--%s=%s" % (item, config[item])]
try:
options.parse_command_line(args)
except tornado.options.Error:
err_("Error on config file option.")
sys.exit(1)
# load options from command line
try:
options.parse_command_line()
except tornado.options.Error:
err_("Error on command line option.")
sys.exit(1)
debug_("options: %s" % json.dumps(options.as_dict(), sort_keys=True))
logging.debug("options: %s" %
json.dumps(options.as_dict(), sort_keys=True))
# load shadowsocks configuration
ss_config_filename = common.find_shadowsocks_config_file()
if ss_config_filename is None:
err_("Can't find any shadowsocks config file. Are you sure there "
"installed shadowsocks already?")
exit(1)
config = common.load_shadowsocks_config(ss_config_filename)
info_("Loading shadowsocks config from file '%s'." % ss_config_filename)
start_tornado(config, ss_config_filename)
global_log.warn(msg, *params)
def usage():
print """
Imports publishers from the specified CSV file.
Usage:
python publisher_categories.py <CKAN config ini filepath> export pub_cats.csv
- produces a list of publishers and their categories
python publisher_categories.py <CKAN config ini filepath> import pub_cats.csv
- import an amended list of publishers and their categories
"""
if __name__ == '__main__':
if len(sys.argv) != 4:
print 'Wrong number of arguments %i' % len(sys.argv)
usage()
sys.exit(0)
cmd, config_ini, action, filepath = sys.argv
common.load_config(config_ini)
PublisherCategories.setup_logging(config_ini)
common.register_translator()
if action == 'export':
PublisherCategories.export(filepath)
elif action == 'import':
PublisherCategories.import_(filepath)
else:
raise NotImplementedError
def transfer(read_from, save_to):
click.echo('%s --> %s' % (read_from, save_to))
if read_from not in OPTIONS or save_to not in OPTIONS:
print 'Should be %s or %s' % (LOCAL, FIREBASE)
sys.exit(-1)
if read_from == save_to:
print 'Saving data to where it is from does not make sense.'
sys.exit(-2)
click.echo('This will OVERWRITE data in "%s". Are you sure? [y/N]' %
save_to)
confirm = sys.stdin.readline()
if confirm.strip() != 'y':
print 'byebye~'
return
common.READ_FROM = common.LOCAL if read_from == LOCAL else common.FIREBASE
common.SAVE_TO = (common.LOCAL,)\
if save_to == LOCAL else (common.FIREBASE,)
print 'Transfering catalog...'
catalog = common.load_catalog()
common.save_catalog(catalog)
print 'Transfering categories...'
catalog = common.load_catalog()
categories = common.load_categories()
common.save_categories(categories)
print 'Transfering filter results...'
f_results = common.load_filter_results()
common.save_filter_results(f_results)
print 'Transfering indicator results...'
i_results = common.load_indicator_results()
common.save_indicator_results(i_results)
print 'Transfering config...'
config = common.load_config()
common.save_config(config)
todo = []
for stocks in catalog.values():
todo.extend(stocks)
total = len(todo)
print 'Transfering sotcks...'
widgets = [
FormatLabel(
'Processed: %(value)d / {0} (in: %(elapsed)s)'.format(total))
]
pbar = ProgressBar(widgets=widgets, maxval=total)
count = 0
pbar.start()
for s in todo:
data = common.load_stock(s)
common.save_stock(s, data)
pbar.update(count)
count += 1
pbar.finish()
print 'Transfering state...'
catalog = common.load_catalog()
state = common.load_state()
common.save_state(state)
#!/usr/bin/python
import sys
import logging
import time
import common
import message_consumer
logger = logging.getLogger(__name__)
if __name__ == '__main__':
logging.basicConfig(stream=sys.stdout, level=logging.INFO)
logging.getLogger("pika").setLevel(logging.ERROR)
logging.getLogger("requests").setLevel(logging.WARNING)
logging.getLogger("urllib3").setLevel(logging.WARNING)
common.load_config()
while True:
try:
message_consumer.initiate_consumer()
except Exception as e:
logger.error(e)
time.sleep(60)
def command(cls, config_ini, options):
common.load_config(config_ini)
common.register_translator()
from ckan import model
from ckanext.dgu.lib.theme import (categorize_package, PRIMARY_THEME,
SECONDARY_THEMES)
rev = model.repo.new_revision()
rev.author = 'script-fix_themes.py'
datasets = common.get_datasets(state='active',
dataset_name=options.dataset,
organization_ref=options.organization)
def fix_theme(theme_str):
'''Returns (fixed_theme_str, outcome)'''
if not theme_str:
return '', 'Blank'
elif theme_str == 'null':
return '', '"null"->""'
elif theme_str in THEMES:
return theme_str, 'Ok'
else:
fixed_theme = THEME_MAP.get(theme_str)
if fixed_theme is None:
return theme_str, 'Unknown theme %s - recategorizing' % theme_str
else:
assert (fixed_theme != theme_str)
return fixed_theme, 'Changed to long form'
package.extras[PRIMARY_THEME] = new_primary
def recategorize(pkg):
themes = categorize_package(pkg, stats_recategorize)
print 'Recategorize: %s' % themes
if themes:
pkg.extras[PRIMARY_THEME] = themes[0]
elif PRIMARY_THEME in pkg.extras:
pkg.extras[PRIMARY_THEME] = ''
if len(themes) > 1:
pkg.extras[SECONDARY_THEMES] = '["%s"]' % themes[1]
elif SECONDARY_THEMES in pkg.extras:
pkg.extras[SECONDARY_THEMES] = '[]'
for package in datasets:
if PRIMARY_THEME in package.extras:
primary = package.extras.get(PRIMARY_THEME)
new_primary, outcome = fix_theme(primary)
if new_primary != primary:
package.extras[PRIMARY_THEME] = new_primary
output = stats_primary.add(outcome, package.name)
if outcome != 'Ok':
print output
if outcome.startswith('Unknown theme'):
recategorize(package)
continue
else:
stats_primary.add('No theme', package.name)
if SECONDARY_THEMES in package.extras:
secondary = package.extras.get(SECONDARY_THEMES)
try:
secondary = json.loads(secondary)
except ValueError:
if secondary.startswith('{') and secondary.endswith('}'):
# '{Crime}' -> 'Crime'
secondary = secondary[1:-1].strip('\"')
print stats_secondary.add('Tidied {}', package.name)
else:
print stats_secondary.add('Error decoding JSON',
package.name)
if secondary == {}:
secondary = []
new_secondary = []
do_recategorize = False
if not isinstance(secondary, list):
secondary = [secondary]
for theme_str in secondary:
if not isinstance(theme_str, basestring):
print stats_secondary.add(
'Not a list of strings %s' % type(theme_str),
package.name)
continue
new_theme, outcome = fix_theme(theme_str)
if new_theme:
new_secondary.append(new_theme)
if outcome != 'Ok':
print stats_secondary.add(outcome, package.name)
if outcome.startswith('Unknown theme'):
do_recategorize = True
if do_recategorize:
recategorize(package)
continue
if json.dumps(new_secondary) != package.extras.get(
SECONDARY_THEMES):
stats_secondary.add('Fixed', package.name)
package.extras[SECONDARY_THEMES] = json.dumps(
new_secondary)
else:
stats_secondary.add('Ok', package.name)
else:
stats_secondary.add('No theme', package.name)
if 'themes-secondary' in package.extras:
print stats_secondary.add(
'Old key removed: themes-secondary', '%s %s' %
(package.name, package.extras['themes-secondary']))
del package.extras['themes-secondary']
print "\nPrimary theme:"
print stats_primary.report()
print "\nSecondary theme:"
print stats_secondary.report()
print "\nRecategorizations:"
print stats_recategorize.report()
if options.write:
print 'Writing'
model.Session.commit()
def main():
config = common.load_config()
update_assembly_info(config)
generate_version_file(config)
update_readme(config)
current_value = a * b / c + value_breakpoint.value_min
return current_value
def get_breakpoints(self, value_type, sensor_value):
return Breakpoints.query.filter(
Breakpoints.value_min <= sensor_value,
Breakpoints.value_max >= sensor_value,
Breakpoints.sensor_value_type_id == value_type).first()
def save_area_data(self, record_aqi, values, sensor_id):
v = {}
for key in values:
if not isinstance(key, str):
v[key.type] = values[key]
else:
v[key] = values[key]
area = AreaModel(aqi=record_aqi, sensor_id=sensor_id, **v)
self.db.session.add(area)
self.db.session.commit()
if __name__ == '__main__':
load_config()
app = create_app()
app.app_context().push()
calc = AQICalculator(db)
result = calc.execute()
def upload(channel_directory, filename, filepath):
config = load_config()
dbx = dropbox.Dropbox(config['dropbox_api_access_token'])
upload_chunk_size = config['dropbox_chunk_size_mb'] * 1024 * 1024
# Dropbox doesn't support characters defined outside the BMP. This includes most, but not all emoji.
filename = re.sub(r'[^\u0000-\uffff]', '', filename)
full_path = DROPBOX_ROOT / channel_directory / filename
log.info(f'Full upload path is {full_path}')
total_size = filepath.stat().st_size
total_chunks = (total_size // upload_chunk_size) + 1
log.info(f'Uploading in {total_chunks} chunks')
log.info('Starting session')
session = dbx.files_upload_session_start(b'')
uploaded = 0
hasher = DropboxContentHasher()
for chunk_num in range(total_chunks):
log.info(f'Uploading chunk {chunk_num}')
is_last_chunk = chunk_num == total_chunks - 1
cursor = dropbox.files.UploadSessionCursor(
session_id=session.session_id,
offset=uploaded,
)
upload_chunk(dbx, cursor, filepath, chunk_num, upload_chunk_size,
hasher, is_last_chunk)
uploaded += total_size % upload_chunk_size if is_last_chunk else upload_chunk_size
log.info('Finishing session')
file_metadata = dbx.files_upload_session_finish(
b'',
dropbox.files.UploadSessionCursor(
session_id=session.session_id,
offset=uploaded,
),
dropbox.files.CommitInfo(path=str(full_path), ),
)
local_hash = hasher.hexdigest()
remote_hash = file_metadata.content_hash
# TODO: Actually take some sort of action based on this,
# especially considering that downloaders log to some file
# no one will ever see. Just retry reuploading it maybe?
assert local_hash == remote_hash, f'Local hash {local_hash} and remote hash {remote_hash} do not match'
# Get the shared link and thumbnail
# XXX: Maybe we should fetch the Youtube thumbnail way earlier?
shared_link = dbx.sharing_create_shared_link(str(full_path))
_, thumbnail_resp = dbx.files_get_thumbnail(
str(full_path),
format=dropbox.files.ThumbnailFormat.png,
size=dropbox.files.ThumbnailSize.w1024h768,
)
# This is probably pretty brittle
url = shared_link.url.replace("www.dropbox", "dl.dropboxusercontent")
return (url, thumbnail_resp.content)
def setUp(self):
config = load_config()
config['io_threads'] = 1
self.ctx = FedMsgContext(**config)
def command(cls, config_ini, options, submissions_csv_filepath):
# Inventive CSV. Columns:
# applicationnumber, applicationdate, jobrole, laname, officerauthorised, theme, responsedate, acceptancestatus, odicertificateurl, dguurl, inventoryurl, localcodes, dataseturl, schemaurl, guidanceurl, frequencyofpublishing, foinumberest, submissioncomplete, lastlaupdate, techreviewstatus, lasttechupdate, adminreviewstatus, paymentamount, closed, lastadminupdate, applicantnotes, administrationnotes, technicalnotes, lastupdated
with open(submissions_csv_filepath, 'rb') as f:
csv = UnicodeCsvReader(f, encoding='iso-8859-1')
header = csv.next()
header = [col_name.strip().lower().replace(' ', '_') for col_name in header]
Submission = namedtuple('Submission', header)
submissions = [Submission(*row) for row in csv]
if config_ini:
# this is only for when running from the command-line
#print 'Loading CKAN config...'
common.load_config(config_ini)
common.register_translator()
#print '...done'
from ckan import model
from ckan.plugins import toolkit
from ckanext.dgu.lib import helpers as dgu_helpers
from ckanext.dgu.model.schema_codelist import Schema
log = __import__('logging').getLogger(__name__)
# Match the organizations in the submissions
lga_orgs_by_dgu_org_name = {}
accepted_submission_dgu_orgs = set()
for submission in submissions:
la_title = la_map.get(submission.laname, submission.laname)
org = model.Session.query(model.Group) \
.filter_by(title=la_title) \
.first()
assert org, 'Submission org title not found: %r' % la_title
lga_orgs_by_dgu_org_name[org.name] = submission.laname
if submission.acceptancestatus == 'Accepted':
accepted_submission_dgu_orgs.add(org.name)
stats = Stats()
stats_incentive = Stats()
results = []
if options.write:
rev = model.repo.new_revision()
rev.author = 'script-%s.py' % __file__
# Iterate over organizations
if options.dataset:
dataset = toolkit.get_action('package_show')(data_dict={'id': options.dataset})
org_names = [dataset['organization']['name']]
elif options.organization:
org_names = [options.organization]
elif options.incentive_only:
org_names = sorted(accepted_submission_dgu_orgs)
else:
org_names = dgu_helpers.all_la_org_names()
#print '%s organizations' % len(org_names)
for org_name in org_names:
org_title = model.Group.by_name(org_name).title
lga_org = lga_orgs_by_dgu_org_name.get(org_name)
# Iterate over the schemas
if options.schema:
schema = all_schemas_by_dgu_name[options.schema]
if options.incentive_only and not schema.lga_name:
# not an incentive schema, so no results
schemas = []
elif options.incentive_only:
schemas = [all_schemas_by_lga_name[submission.theme]
for submission in submissions
if submission.laname == lga_org
and submission.theme == schema.lga_name
and submission.acceptancestatus == 'Accepted']
else:
schemas = [all_schemas_by_lga_name.get(
options.schema,
schema)]
elif options.incentive_only:
schemas = [all_schemas_by_lga_name[submission.theme]
for submission in submissions
if submission.laname == lga_org
and submission.acceptancestatus == 'Accepted']
else:
schemas = all_schemas
#print '%s schemas' % len(schemas)
for schema in schemas:
# Find the relevant incentive submission
if lga_org:
for submission in submissions:
if submission.laname == lga_org and \
submission.theme == schema.lga_name:
break
else:
submission = None
else:
submission = None
result = dict(
org_name=org_name,
org_title=org_title,
org_name_lga=submission.laname if submission else '',
schema_dgu_title=schema.dgu_schema_name,
schema_lga=schema.lga_name,
lga_application_number=submission.applicationnumber if submission else '',
lga_application_acceptance_status=submission.acceptancestatus if submission else '',
dataset_names=[],
dataset_titles=[],
dataset_schema_applied=[],
)
stat_id = '%s %s' % (org_name, schema.lga_name)
if submission:
stat_id += ' %s' % submission.applicationnumber
def add_datasets_to_results(datasets, result):
for dataset in datasets:
if dataset['name'] not in result['dataset_names']:
result['dataset_names'].append(dataset['name'])
result['dataset_titles'].append(dataset['title'])
schema_applied = True if schema.dgu_schema_name in \
[s['title'] for s in dataset.get('schema', [])] \
else False
result['dataset_schema_applied'].append(schema_applied)
if not schema_applied and options.write:
pkg = model.Package.get(dataset['name'])
schema_obj = Schema.by_title(schema.dgu_schema_name)
assert schema_obj, schema.dgu_schema_name
try:
schema_ids = json.loads(pkg.extras.get('schema') or '[]')
except ValueError:
log.error('Not valid JSON in schema field: %s %r',
dataset['name'], pkg.extras.get('schema'))
schema_ids = []
schema_ids.append(schema_obj.id)
pkg.extras['schema'] = json.dumps(schema_ids)
# Already a schema?
data_dict = {'fq': 'publisher:%s ' % org_name +
'schema_multi:"%s"' % schema.dgu_schema_name}
datasets = toolkit.get_action('package_search')(data_dict=data_dict)
if datasets['count'] > 0:
add_datasets_to_results(datasets['results'], result)
stats.add('OK - Dataset with schema',
stat_id + ' %s' % ';'.join(result['dataset_names']))
found_schema = True
else:
found_schema = False
# Submission specifies DGU dataset
if submission and submission.dguurl:
match = re.match('http://data.gov.uk/dataset/(.*)', submission.dguurl)
if match:
dataset_name = dataset_name_original = match.groups()[0]
# some have trailing /
dataset_name = dataset_name.strip('/')
# hampshire have a hash appended
if '#' in dataset_name:
dataset_name = dataset_name.split('#')[0]
# poole have a resource name appended
if '/resource' in dataset_name:
dataset_name = dataset_name.split('/resource')[0]
# manual corrections
if dataset_name in dataset_name_corrections:
dataset_name = dataset_name_corrections[dataset_name]
dataset = model.Package.by_name(dataset_name)
# salford ones added a '1'
if not dataset:
dataset = model.Package.by_name(dataset_name + '1')
if dataset:
dataset_name += '1'
if dataset and dataset.state == 'active':
dataset_dict = toolkit.get_action('package_show')(data_dict={'id': dataset.id})
add_datasets_to_results([dataset_dict], result)
if dataset_name != dataset_name_original:
stats_incentive.add('OK - DGU Dataset listed and with corrections it checks out',
stat_id + ' %s' % dataset_name)
else:
stats_incentive.add('OK - DGU Dataset listed and it checks out',
stat_id + ' %s' % dataset_name)
elif dataset:
stats_incentive.add('ERROR - DGU Dataset listed BUT it is deleted!',
'%s %s' % (stat_id, submission.dguurl))
else:
stats_incentive.add('ERROR - DGU Dataset listed BUT it is not found',
'%s %s' % (stat_id, submission.dguurl))
else:
stats_incentive.add('ERROR - DGU Dataset listed BUT the URL is not the correct format',
'%s %s' % (stat_id, submission.dguurl))
# Submission mentions dataset on LA site - maybe it is in DGU already?
elif submission and submission.dataseturl:
datasets = model.Session.query(model.Package) \
.join(model.ResourceGroup) \
.join(model.Resource) \
.filter(model.Resource.url==submission.dataseturl) \
.filter(model.Package.state=='active') \
.filter(model.Resource.state=='active') \
.all()
dataset_dicts = [
toolkit.get_action('package_show')(data_dict={'id': dataset.id})
for dataset in datasets]
add_datasets_to_results(dataset_dicts, result)
if len(datasets) > 1:
stats_incentive.add('No DGU Dataset, but Dataset URL matches multiple DGU datasets',
'%s %s' % (stat_id, datasets[0].name))
elif len(datasets) == 0:
stats_incentive.add('No DGU Dataset and Dataset URL not found on DGU',
stat_id)
else:
stats_incentive.add('No DGU Dataset, but Dataset URL matches DGU dataset',
'%s %s' % (stat_id, datasets[0].name))
# Search for datasets in the catalogue
datasets = cls.find_dataset_for_schema(schema=schema, org_name=org_name)
if datasets is None:
if not found_schema:
stats.add('Search revealed none', stat_id)
elif len(datasets) > 1:
add_datasets_to_results(datasets, result)
if not found_schema:
stats.add('Found datasets (multiple) in search', '%s %r' % (stat_id, [d['name'] for d in datasets]))
elif datasets:
add_datasets_to_results(datasets, result)
if not found_schema:
stats.add('Found dataset in search', '%s %s' % (stat_id, datasets[0]['name']))
else:
if not found_schema:
stats.add('No dataset for submission', stat_id)
results.append(result)
rows_with_datasets_count = \
len([result for result in results
if any(result['dataset_schema_applied'])])
rows_with_datasets_or_candidate_datasets_count = \
len([result for result in results
if result['dataset_schema_applied']])
if options.print_:
print '\n Incentive stats\n' + stats_incentive.report()
print '\n Overall stats\n' + stats.report()
if options.write:
print 'Writing'
model.Session.commit()
return {'table': results,
'rows_with_datasets_count': rows_with_datasets_count,
'rows_with_datasets_or_candidate_datasets_count': rows_with_datasets_or_candidate_datasets_count}
def init():
global config
config = common.load_config()
finished = True
start = time.time()
for iso in get_isos():
if iso.match() and fileset.tas(iso.name):
if not os.path.exists(iso.name):
iso.download()
# links are time-sensitive; if get_isos() is out-of-date will need
# to re-fetch
if time.time() > start + 300:
finished = False
break
if __name__ == "__main__":
config = common.load_config()
args = parse_args()
if args["list"]:
for iso in get_isos():
print("[%c] %s" % ([" ", "*"][iso.match()], iso.name))
sys.exit(0)
common.mkdirs(config["rhn-dumps-base"])
os.chdir(config["rhn-dumps-base"])
lock = common.Lock(".lock")
threads = []
for i in range(int(config["rhn-dumps-threads"])):
t = threading.Thread(target = worker, name = i)
if __name__ == '__main__':
usage = """Tool to migrate QA data from TaskStatus to QA table
usage: %prog [options] <ckan.ini>
"""
parser = OptionParser(usage=usage)
parser.add_option("-w", "--write",
action="store_true", dest="write",
help="write the changes")
parser.add_option('-p', '--publisher', dest='publisher')
parser.add_option('-d', '--dataset', dest='dataset')
parser.add_option('-r', '--resource', dest='resource')
(options, args) = parser.parse_args()
if len(args) != 1:
parser.error('Wrong number of arguments (%i)' % len(args))
config_ini = args[0]
print 'Loading CKAN config...'
common.load_config(config_ini)
common.register_translator()
print 'Done'
# Setup logging to print debug out for local only
rootLogger = logging.getLogger()
rootLogger.setLevel(logging.WARNING)
localLogger = logging.getLogger(__name__)
localLogger.setLevel(logging.DEBUG)
handler = logging.StreamHandler()
handler.setFormatter(logging.Formatter('%(message)s'))
localLogger.addHandler(handler)
migrate(options)
def command(cls, config_ini, dataset_names, options):
common.load_config(config_ini)
common.register_translator()
from pylons import config
apikey = config['dgu.merge_datasets.apikey']
ckan = ckanapi.RemoteCKAN('https://data.gov.uk', apikey=apikey)
#ckan = ckanapi.LocalCKAN()
if options.publisher:
org_name = common.name_stripped_of_url(options.publisher)
if options.search:
results = ckan.action.package_search(q=options.search, fq='publisher:%s' % org_name, rows=100)
dataset_names.extend([dataset['name']
for dataset in results['results']])
else:
org = ckan.action.organization_show(id=org_name,
include_datasets=True)
dataset_names.extend([d['name'] for d in org['packages']])
datasets = []
datasets_by_name = {}
def get_extra(dataset, key):
for extra in dataset['extras']:
if extra['key'] == key:
return extra['value']
for dataset_name in dataset_names:
print 'Dataset: %s' % dataset_name
for dataset_name in dataset_names:
# strip off the url part of the dataset name, if there is one
dataset_name = common.name_stripped_of_url(dataset_name)
dataset = ckan.action.package_show(id=dataset_name)
harvest_source_ref = get_extra(dataset, 'harvest_source_reference')
if harvest_source_ref:
print '** Discarding dataset %s due to harvest source: %s **' \
% (dataset_name, harvest_source_ref)
continue
datasets.append(dataset)
datasets_by_name[dataset['name']] = dataset
datasets.sort(key=lambda x: x['metadata_modified'])
# aggregate resources
def resource_identity(res_dict, dataset_name):
return (res_dict.get('date'), res_dict['url'],
res_dict.get('title') or res_dict['description'],
res_dict.get('format'),
dataset_name)
combined_resources = {} # identity
res_stats = Stats()
for dataset in datasets:
for resource in dataset['resources']:
identity = resource_identity(resource, dataset['name'])
resource['dataset_name'] = dataset['name']
if identity in combined_resources:
print res_stats.add('Discarding duplicate', '\n%s duplicate of \n%s' % (resource, combined_resources[identity]))
else:
combined_resources[identity] = resource
resources = combined_resources.values()
# find dates for resources
# NB This has been pulled out into timeseries_convert.py -
# TODO call that instead of having the code here too.
if options.frequency:
url_munge_re = re.compile('(%20|-|_|\.)')
def fields_to_hunt_for_date(res):
date = res.get('date')
if date:
yield 'date', date
title = res.get('title')
if title:
yield 'title', title
yield 'description', res['description']
yield 'url', url_munge_re.sub(' ', res['url'])
if not options.update:
dataset = datasets_by_name[res['dataset_name']]
yield 'dataset-title', dataset['title']
yield 'dataset-notes', dataset['notes']
ensure_regexes_are_initialized()
global regexes
for resource in resources:
for field_name, field_value in fields_to_hunt_for_date(resource):
if options.frequency in ('monthly', 'quarterly', 'twice annually'):
month, year = hunt_for_month_and_year(field_value)
if year and month:
resource['date'] = '%02d/%s' % (month, year)
res_stats.add('Found date in %s' % field_name,
'%s %r' %
(resource['date'], resource))
if resource.get('resource_type') == 'documentation':
resource['resource_type'] = 'file'
res_stats.add('Converted additional resource', resource)
break
elif options.frequency == 'annually':
year = regexes['year'].search(field_value)
if year:
resource['date'] = year.groups()[0]
res_stats.add('Found date in %s' % field_name,
'%s %r' %
(resource['date'], resource))
if resource.get('resource_type') == 'documentation':
resource['resource_type'] = 'file'
res_stats.add('Converted additional resource', resource)
break
else:
if resource.get('resource_type') == 'documentation':
print res_stats.add('Could not find date but it\'s Additional Resource', resource)
continue
print res_stats.add('Could not find date', resource)
continue
print 'Resources: \n', res_stats
resources_without_date = [res for res in resources
if not res.get('date') and
res.get('resource_type') != 'documentation']
for i, res in enumerate(resources_without_date):
print 'Resources without dates %s/%s' % (i+1, len(resources_without_date))
for field_name, field_value in fields_to_hunt_for_date(res):
print ' %s: %s' % (field_name, field_value.encode('latin-1', 'ignore'))
print 'https://data.gov.uk/dataset/%s/resource/%s' % (res['dataset_name'], res['id'])
date_format = {'annually': 'YYYY',
'monthly': 'MM/YYYY',
'twice annually': 'MM/YYYY',
'quarterly': 'MM/YYYY'}
input_ = raw_input('Date (%s) or DOCS to make it an Additional Resource: ' %
date_format[options.frequency])
if input_.strip().lower() == 'docs':
res['date'] = ''
res['resource_type'] = 'documentation'
else:
res['date'] = input_
resources.sort(key=lambda x: x.get('date', '').split('/')[::-1])
# Ensure there is not a mixture of resources with and without a date
have_dates = None
for res in resources:
if res.get('resource_type') == 'documentation':
continue
if have_dates is None:
have_dates = bool(res.get('date'))
else:
has_date = bool(res.get('date'))
if has_date != have_dates:
print [res.get('date') for res in resources]
print 'Cannot mix resources with dates and others without!'
import pdb
pdb.set_trace()
import socket
import json
import logging
from common import load_config, file_default_config
from datanode import DataNode
if __name__ == "__main__":
# socket.setdefaulttimeout(20)
config = load_config("config/datanode.json")
client_datanode_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
host = ""
port = config["client_comm_port"]
client_datanode_socket.bind((host, port))
client_datanode_socket.listen(config["max_client_num"])
datanode_instance = DataNode(config)
while True:
client, addr = client_datanode_socket.accept()
print("Connect from {}".format(addr))
# receive command from client
client_command = client.recv(2048).decode('utf-8')
# accept command successfully
client.send(bytes("ack", encoding="utf-8"))
client_command = json.loads(client_command, encoding="utf-8")
blockinfo = client_command.get("block_info", None)
block = None
ret_block = None
def main():
"""
Main function: run Infernal, filter results and flag RNA genes in TRAPID db.
"""
cmd_args = parse_arguments()
# Read experiment's initial processing configuration file
config = common.load_config(cmd_args.ini_file_initial,
{"infernal", "trapid_db", "experiment"})
# The web application sets the Rfam clan string to 'None' in case the user chose no clans
# If this is the case, exit the script with an information message
if config["infernal"]["rfam_clans"] == "None":
sys.stderr.write(
"[Message] No Rfam clans selected: skip ncRNA annotation step.\n")
sys.exit()
try:
# Run Infernal, parse and export results to DB
sys.stderr.write(
'[Message] Starting ncRNA annotation procedure: %s\n' %
time.strftime('%Y/%m/%d %H:%M:%S'))
exp_id = config["experiment"]["exp_id"]
tmp_exp_dir = config["experiment"]["tmp_exp_dir"]
rfam_dir = config["infernal"]["rfam_dir"]
exp_clans = config["infernal"]["rfam_clans"].split(",")
# Lists containing all needed parameters for `common.db_connect()` (TRAPID + reference DB)
trapid_db_data = common.get_db_connection_data(config, 'trapid_db')
reference_db_data = common.get_db_connection_data(
config, 'reference_db')
db_connection = common.db_connect(*trapid_db_data)
common.update_experiment_log(exp_id, 'start_nc_rna_search', 'Infernal',
2, db_connection)
db_connection.close()
create_infernal_files(exp_id, tmp_exp_dir, rfam_dir, exp_clans,
trapid_db_data)
# run_cmpress(exp_id=exp_id, tmp_exp_dir=tmp_exp_dir)
total_m_nts = get_infernal_z_value(exp_id, trapid_db_data)
infernal_tblout = run_infernal(exp_id, tmp_exp_dir, total_m_nts)
# Filter Infernal tabulated output (keep best non-ovelrapping matches)
# infernal_tblout_filtered = filter_out_overlaps(exp_id=exp_id, tmp_exp_dir=tmp_exp_dir, tblout_file=infernal_tblout)
infernal_tblout_filtered = keep_best_results(exp_id, tmp_exp_dir,
infernal_tblout)
# Get filtered results as list of dict and add clan information
# Read Rfam clan information from `clanin` file. Would it make more sense to retrieve it when creating it?
cm_clans = get_exp_cm_clans(exp_id, tmp_exp_dir)
filtered_infernal_results = infernal_tblout_to_list(
infernal_tblout_filtered, cm_clans)
infernal_results = infernal_tblout_to_list(infernal_tblout, cm_clans)
# Flag potential rna genes (set `is_rna_gene` value to 1 and `rf_ids` in `transcripts` table)
flag_rna_genes(exp_id, filtered_infernal_results, trapid_db_data)
# Store filtered results in `rna_similarities` ...
store_rna_similarities(exp_id, infernal_results, trapid_db_data)
# ... and `rna_families`
store_rna_families(exp_id, filtered_infernal_results, trapid_db_data)
# Annotate transcripts using GO terms from Rfam
rfam_go = retrieve_rfam_go_data(trapid_db_data)
go_data = get_go_data(reference_db_data)
# perform_go_annotation(exp_id, infernal_results, rfam_go, go_data, tmp_exp_dir)
perform_go_annotation(exp_id, filtered_infernal_results, rfam_go,
go_data, tmp_exp_dir)
# That's it for now
db_connection = common.db_connect(*trapid_db_data)
common.update_experiment_log(exp_id, 'stop_nc_rna_search', 'Infernal',
2, db_connection)
db_connection.close()
sys.stderr.write(
'[Message] Finished ncRNA annotation procedure: %s\n' %
time.strftime('%Y/%m/%d %H:%M:%S'))
# If any exception was raised, update the experiment's log, set status to 'error', and exit
except Exception:
print_exc()
common.stop_initial_processing_error(exp_id, trapid_db_data)
import socket
import json
from common import load_config
from namenoode import FileSystem
normal_message = {"success": True, "message": "ack"}
normal_message_bytes = bytes(json.dumps(normal_message).encode('utf-8'))
if __name__ == "__main__":
# socket.setdefaulttimeout(20)
config = load_config("config/namenode.json")
client_server_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
host = ""
port = config["client_comm_port"]
client_server_socket.bind((host, port))
client_server_socket.listen(config["max_client_num"])
file_system = FileSystem(config=config["file_system_config"])
while True:
client, addr = client_server_socket.accept()
print("client address: {}".format(addr))
# recv command from client
client_command = client.recv(4096).decode('utf-8')
client_command = json.loads(client_command, encoding='utf-8')
if client_command["command"] == "test_list":
file_system.test_out()
client.send(normal_message_bytes)
if client_command["command"] == "list":
