def __init__(self, db_path, browser_params, num_browsers, log_file = '~/openwpm.log', process_watchdog=False, task_description=None):
# Flow control
self.closing = False
self.failure_flag = False
self.threadlock = threading.Lock()
self.failurecount = 0
# sets up the information needed to write to the database
self.desc = task_description
self.db_path = db_path
self.log_file = log_file
self.process_watchdog = process_watchdog
# sets up the crawl data database
self.db = sqlite3.connect(db_path)
with open(os.path.join(os.path.dirname(__file__), 'schema.sql'), 'r') as f:
self.db.executescript(f.read())
# prepares browser settings
self.num_browsers = num_browsers
# special case: for singleton dictionary, we perform deep copies so that number of dicts is <num_browsers>
if type(browser_params) is not list:
browser_params = [copy.deepcopy(browser_params) for i in xrange(0, num_browsers)]
if len(browser_params) != num_browsers:
raise Exception("Number of browser parameter dictionaries is not the same as <num_browsers>")
# sets up logging server + connect a client
self.logging_status_queue = None
self.loggingserver = self._launch_loggingserver()
self.logger_address = self.logging_status_queue.get() # socket location: (address, port)
self.logger = MPLogger.loggingclient(*self.logger_address)
# sets up the DataAggregator + associated queues
self.aggregator_status_queue = None # queue used for sending graceful KILL command to DataAggregator
self.data_aggregator = self._launch_data_aggregator()
self.aggregator_address = self.aggregator_status_queue.get() # socket location: (address, port)
# open client socket
self.sock = clientsocket()
self.sock.connect(self.aggregator_address[0], self.aggregator_address[1])
# update task table
cur = self.db.cursor()
cur.execute("INSERT INTO task (description) VALUES (?)", (self.desc,))
self.db.commit()
self.task_id = cur.lastrowid
# sets up the BrowserManager(s) + associated queues
self.browsers = self._initialize_browsers(browser_params) # List of the Browser(s)
self._launch_browsers()
# start the manager watchdog
thread = threading.Thread(target=self._manager_watchdog, args=())
thread.daemon = True
thread.start()
def __init__(self, manager_params, browser_params, process_watchdog=False):
# Make paths absolute in manager_params
for path in ['data_directory', 'log_directory']:
if manager_params[path] is not None:
manager_params[path] = os.path.expanduser(manager_params[path])
manager_params['database_name'] = os.path.join(
manager_params['data_directory'], manager_params['database_name'])
manager_params['log_file'] = os.path.join(
manager_params['log_directory'], manager_params['log_file'])
manager_params['screenshot_path'] = os.path.join(
manager_params['data_directory'], 'screenshots')
manager_params['source_dump_path'] = os.path.join(
manager_params['data_directory'], 'sources')
self.manager_params = manager_params
# Create data directories if they do not exist
if not os.path.exists(manager_params['screenshot_path']):
os.makedirs(manager_params['screenshot_path'])
if not os.path.exists(manager_params['source_dump_path']):
os.makedirs(manager_params['source_dump_path'])
# check size of parameter dictionary
self.num_browsers = manager_params['num_browsers']
if len(browser_params) != self.num_browsers:
raise Exception(
"Number of <browser_params> dicts is not the same as manager_params['num_browsers']"
)
# Flow control
self.closing = False
self.failure_status = None
self.threadlock = threading.Lock()
self.failurecount = 0
if manager_params['failure_limit'] is not None:
self.failure_limit = manager_params['failure_limit']
else:
self.failure_limit = self.num_browsers * 2 + 10
self.process_watchdog = process_watchdog
# sets up the crawl data database
db_path = manager_params['database_name']
if not os.path.exists(manager_params['data_directory']):
os.mkdir(manager_params['data_directory'])
self.db = sqlite3.connect(db_path)
with open(os.path.join(os.path.dirname(__file__), 'schema.sql'),
'r') as f:
self.db.executescript(f.read())
self.db.commit()
# sets up logging server + connect a client
self.logging_status_queue = None
self.loggingserver = self._launch_loggingserver()
# socket location: (address, port)
self.manager_params['logger_address'] = self.logging_status_queue.get()
self.logger = MPLogger.loggingclient(
*self.manager_params['logger_address'])
# Mark if LDBAggregator is needed (if js is enabled on any browser)
self.ldb_enabled = False
for params in browser_params:
if params['save_javascript'] or params['save_javascript_proxy']:
self.ldb_enabled = True
break
# Initialize the data aggregators
self._launch_aggregators()
# open client socket
self.sock = clientsocket(serialization='dill')
self.sock.connect(*self.manager_params['aggregator_address'])
print 1
self._save_configuration(browser_params)
print 2
# read the last used site visit id
cur = self.db.cursor()
cur.execute("SELECT MAX(visit_id) from site_visits")
last_visit_id = cur.fetchone()[0]
if last_visit_id is None:
last_visit_id = 0
self.next_visit_id = last_visit_id + 1
print 3
# sets up the BrowserManager(s) + associated queues
self.browsers = self._initialize_browsers(
browser_params) # List of the Browser(s)
print 5
self._launch_browsers()
print 4
# start the manager watchdog
thread = threading.Thread(target=self._manager_watchdog, args=())
thread.daemon = True
thread.start()
def __init__(self, manager_params, browser_params, process_watchdog=False, task_description=None):
# Make paths absolute in manager_params
manager_params['data_directory'] = os.path.expanduser(manager_params['data_directory'])
manager_params['log_directory'] = os.path.expanduser(manager_params['log_directory'])
manager_params['database_name'] = os.path.join(manager_params['data_directory'],manager_params['database_name'])
manager_params['log_file'] = os.path.join(manager_params['log_directory'],manager_params['log_file'])
self.manager_params = manager_params
# Flow control
self.closing = False
self.failure_flag = False
self.threadlock = threading.Lock()
self.failurecount = 0
self.desc = task_description
self.process_watchdog = process_watchdog
# sets up the crawl data database
db_path = manager_params['database_name']
if not os.path.exists(manager_params['data_directory']):
os.mkdir(manager_params['data_directory'])
self.db = sqlite3.connect(db_path)
with open(os.path.join(os.path.dirname(__file__), 'schema.sql'), 'r') as f:
self.db.executescript(f.read())
# check size of parameter dictionary
self.num_browsers = manager_params['num_browsers']
if len(browser_params) != self.num_browsers:
raise Exception("Number of <browser_params> dicts is not the same as manager_params['num_browsers']")
# sets up logging server + connect a client
self.logging_status_queue = None
self.loggingserver = self._launch_loggingserver()
# socket location: (address, port)
self.manager_params['logger_address'] = self.logging_status_queue.get()
self.logger = MPLogger.loggingclient(*self.manager_params['logger_address'])
# Mark if LDBAggregator is needed (if js is enabled on any browser)
self.ldb_enabled = False
for params in browser_params:
if params['save_javascript']:
self.ldb_enabled = True
break
# Initialize the data aggregators
self._launch_aggregators()
# open client socket
self.sock = clientsocket()
self.sock.connect(*self.manager_params['aggregator_address'])
# update task table
cur = self.db.cursor()
cur.execute("INSERT INTO task (description) VALUES (?)", (self.desc,))
self.db.commit()
self.task_id = cur.lastrowid
# sets up the BrowserManager(s) + associated queues
self.browsers = self._initialize_browsers(browser_params) # List of the Browser(s)
self._launch_browsers()
# start the manager watchdog
thread = threading.Thread(target=self._manager_watchdog, args=())
thread.daemon = True
thread.start()
def __init__(self, manager_params, browser_params, process_watchdog=False):
# Make paths absolute in manager_params
for path in ['data_directory','log_directory']:
if manager_params[path] is not None:
manager_params[path] = os.path.expanduser(manager_params[path])
manager_params['database_name'] = os.path.join(manager_params['data_directory'],manager_params['database_name'])
manager_params['log_file'] = os.path.join(manager_params['log_directory'],manager_params['log_file'])
manager_params['screenshot_path'] = os.path.join(manager_params['data_directory'], 'screenshots')
manager_params['source_dump_path'] = os.path.join(manager_params['data_directory'], 'sources')
self.manager_params = manager_params
# Create data directories if they do not exist
if not os.path.exists(manager_params['screenshot_path']):
os.makedirs(manager_params['screenshot_path'])
if not os.path.exists(manager_params['source_dump_path']):
os.makedirs(manager_params['source_dump_path'])
# check size of parameter dictionary
self.num_browsers = manager_params['num_browsers']
if len(browser_params) != self.num_browsers:
raise Exception("Number of <browser_params> dicts is not the same as manager_params['num_browsers']")
# Flow control
self.closing = False
self.failure_status = None
self.threadlock = threading.Lock()
self.failurecount = 0
if manager_params['failure_limit'] is not None:
self.failure_limit = manager_params['failure_limit']
else:
self.failure_limit = self.num_browsers * 2 + 10
self.process_watchdog = process_watchdog
# sets up the crawl data database
db_path = manager_params['database_name']
if not os.path.exists(manager_params['data_directory']):
os.mkdir(manager_params['data_directory'])
self.db = sqlite3.connect(db_path)
with open(os.path.join(os.path.dirname(__file__), 'schema.sql'), 'r') as f:
self.db.executescript(f.read())
self.db.commit()
# sets up logging server + connect a client
self.logging_status_queue = None
self.loggingserver = self._launch_loggingserver()
# socket location: (address, port)
self.manager_params['logger_address'] = self.logging_status_queue.get()
self.logger = MPLogger.loggingclient(*self.manager_params['logger_address'])
# Mark if LDBAggregator is needed (if js is enabled on any browser)
self.ldb_enabled = False
for params in browser_params:
if params['save_javascript'] or params['save_javascript_proxy']:
self.ldb_enabled = True
break
# Initialize the data aggregators
self._launch_aggregators()
# open client socket
self.sock = clientsocket(serialization='dill')
self.sock.connect(*self.manager_params['aggregator_address'])
self._save_configuration(browser_params)
# read the last used site visit id
cur = self.db.cursor()
cur.execute("SELECT MAX(visit_id) from site_visits")
last_visit_id = cur.fetchone()[0]
if last_visit_id is None:
last_visit_id = 0
self.next_visit_id = last_visit_id + 1
# sets up the BrowserManager(s) + associated queues
self.browsers = self._initialize_browsers(browser_params) # List of the Browser(s)
self._launch_browsers()
# start the manager watchdog
thread = threading.Thread(target=self._manager_watchdog, args=())
thread.daemon = True
thread.start()
def __init__(self, manager_params, browser_params, process_watchdog=False):
# Make paths absolute in manager_params
for path in ['data_directory', 'log_directory']:
if manager_params[path] is not None:
manager_params[path] = os.path.expanduser(manager_params[path])
manager_params['database_name'] = os.path.join(
manager_params['data_directory'], manager_params['database_name'])
manager_params['log_file'] = os.path.join(
manager_params['log_directory'], manager_params['log_file'])
manager_params['screenshot_path'] = os.path.join(
manager_params['data_directory'], 'screenshots')
manager_params['source_dump_path'] = os.path.join(
manager_params['data_directory'], 'sources')
self.manager_params = manager_params
self.browser_params = browser_params
# Create data directories if they do not exist
if not os.path.exists(manager_params['screenshot_path']):
os.makedirs(manager_params['screenshot_path'])
if not os.path.exists(manager_params['source_dump_path']):
os.makedirs(manager_params['source_dump_path'])
# check size of parameter dictionary
self.num_browsers = manager_params['num_browsers']
if len(browser_params) != self.num_browsers:
raise Exception("Number of <browser_params> dicts is not the same "
"as manager_params['num_browsers']")
# Flow control
self.closing = False
self.failure_status = None
self.threadlock = threading.Lock()
self.failurecount = 0
if manager_params['failure_limit'] is not None:
self.failure_limit = manager_params['failure_limit']
else:
self.failure_limit = self.num_browsers * 2 + 10
self.process_watchdog = process_watchdog
# sets up logging server + connect a client
self.logging_status_queue = None
self.loggingserver = self._launch_loggingserver()
# socket location: (address, port)
self.manager_params['logger_address'] = self.logging_status_queue.get()
self.logger = MPLogger.loggingclient(
*self.manager_params['logger_address'])
# Initialize the data aggregators
self._launch_aggregators()
# sets up the BrowserManager(s) + associated queues
self.browsers = self._initialize_browsers(browser_params)
self._launch_browsers()
# start the manager watchdog
thread = threading.Thread(target=self._manager_watchdog, args=())
thread.daemon = True
thread.start()
# Save crawl config information to database
openwpm_v, browser_v = get_version()
self.data_aggregator.save_configuration(openwpm_v, browser_v)
self.logger.info(
get_configuration_string(self.manager_params, browser_params,
(openwpm_v, browser_v)))
def __init__(self,
db_path,
browser_params,
num_browsers,
log_file='~/openwpm.log',
process_watchdog=False,
task_description=None):
# Flow control
self.closing = False
self.failure_flag = False
self.threadlock = threading.Lock()
self.failurecount = 0
# sets up the information needed to write to the database
self.desc = task_description
self.db_path = db_path
self.log_file = log_file
self.process_watchdog = process_watchdog
# sets up the crawl data database
self.db = sqlite3.connect(db_path)
with open(os.path.join(os.path.dirname(__file__), 'schema.sql'),
'r') as f:
self.db.executescript(f.read())
# prepares browser settings
self.num_browsers = num_browsers
# special case: for singleton dictionary, we perform deep copies so that number of dicts is <num_browsers>
if type(browser_params) is not list:
browser_params = [
copy.deepcopy(browser_params) for i in xrange(0, num_browsers)
]
if len(browser_params) != num_browsers:
raise Exception(
"Number of browser parameter dictionaries is not the same as <num_browsers>"
)
# sets up logging server + connect a client
self.logging_status_queue = None
self.loggingserver = self._launch_loggingserver()
self.logger_address = self.logging_status_queue.get(
) # socket location: (address, port)
self.logger = MPLogger.loggingclient(*self.logger_address)
# sets up the DataAggregator + associated queues
self.aggregator_status_queue = None # queue used for sending graceful KILL command to DataAggregator
self.data_aggregator = self._launch_data_aggregator()
self.aggregator_address = self.aggregator_status_queue.get(
) # socket location: (address, port)
# open client socket
self.sock = clientsocket()
self.sock.connect(self.aggregator_address[0],
self.aggregator_address[1])
# update task table
cur = self.db.cursor()
cur.execute("INSERT INTO task (description) VALUES (?)", (self.desc, ))
self.db.commit()
self.task_id = cur.lastrowid
# sets up the BrowserManager(s) + associated queues
self.browsers = self._initialize_browsers(
browser_params) # List of the Browser(s)
self._launch_browsers()
# start the manager watchdog
thread = threading.Thread(target=self._manager_watchdog, args=())
thread.daemon = True
thread.start()
