d500448801 2009-10-05 c.kworr@d4daf: #!/usr/bin/env python3.1
d500448801 2009-10-05 c.kworr@d4daf:
ae30851739 2010-08-12 c.kworr@d4daf: import postgresql.api, re, sys
d500448801 2009-10-05 c.kworr@d4daf:
b93dc49210 2009-10-13 c.kworr@d4daf: # wrapper around syslog, can be muted
d500448801 2009-10-05 c.kworr@d4daf: class Logger:
d500448801 2009-10-05 c.kworr@d4daf: __slots__ = frozenset(['_syslog'])
d500448801 2009-10-05 c.kworr@d4daf:
d500448801 2009-10-05 c.kworr@d4daf: def __init__(self):
d500448801 2009-10-05 c.kworr@d4daf: config.section('log')
d500448801 2009-10-05 c.kworr@d4daf: if config['silent'] == 'yes':
d500448801 2009-10-05 c.kworr@d4daf: self._syslog = None
d500448801 2009-10-05 c.kworr@d4daf: else:
d500448801 2009-10-05 c.kworr@d4daf: import syslog
d500448801 2009-10-05 c.kworr@d4daf: self._syslog = syslog
d500448801 2009-10-05 c.kworr@d4daf: self._syslog.openlog('squidTag')
d500448801 2009-10-05 c.kworr@d4daf:
d500448801 2009-10-05 c.kworr@d4daf: def info(self, message):
4b22e25f24 2009-10-07 c.kworr@d4daf: if self._syslog:
d500448801 2009-10-05 c.kworr@d4daf: self._syslog.syslog(self._syslog.LOG_INFO, message)
d500448801 2009-10-05 c.kworr@d4daf:
d500448801 2009-10-05 c.kworr@d4daf: def notice(self, message):
4b22e25f24 2009-10-07 c.kworr@d4daf: if self._syslog:
d500448801 2009-10-05 c.kworr@d4daf: self._syslog.syslog(self._syslog.LOG_NOTICE, message)
d500448801 2009-10-05 c.kworr@d4daf:
b93dc49210 2009-10-13 c.kworr@d4daf: # wrapper around database
d500448801 2009-10-05 c.kworr@d4daf: class tagDB:
d301d9adc6 2010-08-13 c.kworr@d4daf: __slots__ = frozenset(('_check_stmt', '_db'))
b93dc49210 2009-10-13 c.kworr@d4daf:
b93dc49210 2009-10-13 c.kworr@d4daf: def __init__(self):
9450c03d41 2010-08-07 c.kworr@d4daf: config.section('database')
9450c03d41 2010-08-07 c.kworr@d4daf: self._db = postgresql.open(
9450c03d41 2010-08-07 c.kworr@d4daf: 'pq://{}:{}@{}/{}'.format(
9450c03d41 2010-08-07 c.kworr@d4daf: config['user'],
9450c03d41 2010-08-07 c.kworr@d4daf: config['password'],
9450c03d41 2010-08-07 c.kworr@d4daf: config['host'],
9450c03d41 2010-08-07 c.kworr@d4daf: config['database'],
9450c03d41 2010-08-07 c.kworr@d4daf: ) )
ae30851739 2010-08-12 c.kworr@d4daf: self._check_stmt = None
b93dc49210 2009-10-13 c.kworr@d4daf:
b93dc49210 2009-10-13 c.kworr@d4daf: def check(self, site, ip_address):
ae30851739 2010-08-12 c.kworr@d4daf: if self._check_stmt == None:
ae30851739 2010-08-12 c.kworr@d4daf: self._check_stmt = self._db.prepare("select redirect_url, regexp from site_rule where site <@ tripdomain($1) and netmask >> $2::text::inet order by array_length(site, 1) desc")
b93dc49210 2009-10-13 c.kworr@d4daf: return(self._check_stmt(site, ip_address))
ae30851739 2010-08-12 c.kworr@d4daf:
ae30851739 2010-08-12 c.kworr@d4daf: def dump(self):
d301d9adc6 2010-08-13 c.kworr@d4daf: return(self._db.prepare("select untrip(site), tag, regexp from urls natural join site natural join tag order by site, tag")())
31e69c4237 2010-08-12 c.kworr@d4daf:
31e69c4237 2010-08-12 c.kworr@d4daf: def load(self, csv_data):
31e69c4237 2010-08-12 c.kworr@d4daf: with self._db.xact():
31e69c4237 2010-08-12 c.kworr@d4daf: if config.options.flush_db:
31e69c4237 2010-08-12 c.kworr@d4daf: self._db.execute('delete from urls;')
31e69c4237 2010-08-12 c.kworr@d4daf: if config.options.flush_site:
31e69c4237 2010-08-12 c.kworr@d4daf: self._db.execute('delete from site;');
31e69c4237 2010-08-12 c.kworr@d4daf: insertreg = self._db.prepare("select set($1, $2, $3)")
31e69c4237 2010-08-12 c.kworr@d4daf: insert = self._db.prepare("select set($1, $2)")
31e69c4237 2010-08-12 c.kworr@d4daf: for row in csv_data:
31e69c4237 2010-08-12 c.kworr@d4daf: if len(row[2]) > 0:
31e69c4237 2010-08-12 c.kworr@d4daf: insertreg(row[0], row[1], row[2])
31e69c4237 2010-08-12 c.kworr@d4daf: else:
31e69c4237 2010-08-12 c.kworr@d4daf: insert(row[0], row[1])
31e69c4237 2010-08-12 c.kworr@d4daf: self._db.execute('vacuum analyze site;')
31e69c4237 2010-08-12 c.kworr@d4daf: self._db.execute('vacuum analyze urls;')
d301d9adc6 2010-08-13 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: def load_conf(self, csv_data):
d301d9adc6 2010-08-13 c.kworr@d4daf: with self._db.xact():
d301d9adc6 2010-08-13 c.kworr@d4daf: self._db.execute('delete from rules;')
d301d9adc6 2010-08-13 c.kworr@d4daf: insertconf = self._db.prepare("insert into rules (netmask, redirect_url, from_weekday, to_weekday, from_time, to_time, id_tag) values ($1::text::cidr, $2, $3, $4, $5::text::time, $6::text::time, get_tag($7::text::text[]))")
d301d9adc6 2010-08-13 c.kworr@d4daf: for row in csv_data:
d301d9adc6 2010-08-13 c.kworr@d4daf: insertconf(row[0], row[1], int(row[2]), int(row[3]), row[4], row[5], row[6])
d301d9adc6 2010-08-13 c.kworr@d4daf: self._db.execute('vacuum analyze rules;')
d301d9adc6 2010-08-13 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: def dump_conf(self):
d301d9adc6 2010-08-13 c.kworr@d4daf: return(self._db.prepare("select netmask, redirect_url, from_weekday, to_weekday, from_time, to_time, tag from rules natural join tag")())
b93dc49210 2009-10-13 c.kworr@d4daf:
b93dc49210 2009-10-13 c.kworr@d4daf: # abstract class with basic checking functionality
b93dc49210 2009-10-13 c.kworr@d4daf: class Checker:
ed7808827d 2009-10-14 c.kworr@d4daf: __slots__ = frozenset(['_db', '_log'])
7e3418d94f 2009-10-12 c.kworr@d4daf:
7e3418d94f 2009-10-12 c.kworr@d4daf: def __init__(self):
b93dc49210 2009-10-13 c.kworr@d4daf: self._db = tagDB()
b93dc49210 2009-10-13 c.kworr@d4daf: self._log = Logger()
7c13294e9f 2010-08-07 c.kworr@d4daf: self._log.info('started\n')
b93dc49210 2009-10-13 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: def process(self, id, site, ip_address, url_path, line = None):
b93dc49210 2009-10-13 c.kworr@d4daf: self._log.info('trying {}\n'.format(site))
b93dc49210 2009-10-13 c.kworr@d4daf: result = self._db.check(site, ip_address)
b93dc49210 2009-10-13 c.kworr@d4daf: reply = '-'
b93dc49210 2009-10-13 c.kworr@d4daf: for row in result:
b93dc49210 2009-10-13 c.kworr@d4daf: if row != None and row[0] != None:
b93dc49210 2009-10-13 c.kworr@d4daf: if row[1] != None:
b93dc49210 2009-10-13 c.kworr@d4daf: self._log.info('trying regexp "{}" versus "{}"\n'.format(row[1], url_path))
d2c54d0451 2010-03-01 c.kworr@d4daf: try:
d2c54d0451 2010-03-01 c.kworr@d4daf: if re.compile(row[1]).match(url_path):
1fa8a88371 2010-07-14 c.kworr@d4daf: reply = row[0].format(url_path)
d2c54d0451 2010-03-01 c.kworr@d4daf: break
d2c54d0451 2010-03-01 c.kworr@d4daf: else:
d2c54d0451 2010-03-01 c.kworr@d4daf: continue
d2c54d0451 2010-03-01 c.kworr@d4daf: except:
d2c54d0451 2010-03-01 c.kworr@d4daf: self._log.info("can't compile regexp")
b93dc49210 2009-10-13 c.kworr@d4daf: else:
1fa8a88371 2010-07-14 c.kworr@d4daf: reply = row[0].format(url_path)
b93dc49210 2009-10-13 c.kworr@d4daf: break
b93dc49210 2009-10-13 c.kworr@d4daf: self.writeline('{} {}\n'.format(id, reply))
7e3418d94f 2009-10-12 c.kworr@d4daf:
7e3418d94f 2009-10-12 c.kworr@d4daf: def check(self, line):
7e3418d94f 2009-10-12 c.kworr@d4daf: request = re.compile('^([0-9]+)\ (http|ftp):\/\/([-\w.:]+)\/([^ ]*)\ ([0-9.]+)\/(-|[\w\.]+)\ (-|\w+)\ (-|GET|HEAD|POST).*$').match(line)
7e3418d94f 2009-10-12 c.kworr@d4daf: if request:
7e3418d94f 2009-10-12 c.kworr@d4daf: id = request.group(1)
7e3418d94f 2009-10-12 c.kworr@d4daf: #proto = request.group(2)
7e3418d94f 2009-10-12 c.kworr@d4daf: site = request.group(3)
7e3418d94f 2009-10-12 c.kworr@d4daf: url_path = request.group(4)
7e3418d94f 2009-10-12 c.kworr@d4daf: ip_address = request.group(5)
ed7808827d 2009-10-14 c.kworr@d4daf: self.process(id, site, ip_address, url_path, line)
26fc9b34d9 2010-08-07 c.kworr@d4daf: return(True)
7e3418d94f 2009-10-12 c.kworr@d4daf: else:
7e3418d94f 2009-10-12 c.kworr@d4daf: self._log.info('bad request\n')
b93dc49210 2009-10-13 c.kworr@d4daf: self.writeline(line)
26fc9b34d9 2010-08-07 c.kworr@d4daf: return(False)
b93dc49210 2009-10-13 c.kworr@d4daf:
b93dc49210 2009-10-13 c.kworr@d4daf: def writeline(self, string):
b93dc49210 2009-10-13 c.kworr@d4daf: self._log.info('sending: ' + string)
b93dc49210 2009-10-13 c.kworr@d4daf: sys.stdout.write(string)
b93dc49210 2009-10-13 c.kworr@d4daf: sys.stdout.flush()
b93dc49210 2009-10-13 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: def loop(self):
ed7808827d 2009-10-14 c.kworr@d4daf: while True:
ed7808827d 2009-10-14 c.kworr@d4daf: line = sys.stdin.readline()
ed7808827d 2009-10-14 c.kworr@d4daf: if len(line) == 0:
ed7808827d 2009-10-14 c.kworr@d4daf: break
ed7808827d 2009-10-14 c.kworr@d4daf: self.check(line)
ed7808827d 2009-10-14 c.kworr@d4daf:
b93dc49210 2009-10-13 c.kworr@d4daf: # threaded checking facility
b93dc49210 2009-10-13 c.kworr@d4daf: class CheckerThread(Checker):
ed7808827d 2009-10-14 c.kworr@d4daf: __slots__ = frozenset(['_lock', '_lock_exit', '_lock_queue', '_queue'])
b93dc49210 2009-10-13 c.kworr@d4daf:
b93dc49210 2009-10-13 c.kworr@d4daf: def __init__(self):
ae30851739 2010-08-12 c.kworr@d4daf: import _thread
ae30851739 2010-08-12 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: # basic initialisation
b93dc49210 2009-10-13 c.kworr@d4daf: Checker.__init__(self)
ed7808827d 2009-10-14 c.kworr@d4daf:
b93dc49210 2009-10-13 c.kworr@d4daf: # Spin lock. Loop acquires it on start then releases it when holding queue
b93dc49210 2009-10-13 c.kworr@d4daf: # lock. This way the thread proceeds without stops while queue has data and
b93dc49210 2009-10-13 c.kworr@d4daf: # gets stalled when no data present. The lock is released by queue writer
b93dc49210 2009-10-13 c.kworr@d4daf: # after storing something into the queue
b93dc49210 2009-10-13 c.kworr@d4daf: self._lock = _thread.allocate_lock()
ed7808827d 2009-10-14 c.kworr@d4daf: self._lock_exit = _thread.allocate_lock()
b93dc49210 2009-10-13 c.kworr@d4daf: self._lock_queue = _thread.allocate_lock()
b93dc49210 2009-10-13 c.kworr@d4daf: self._lock.acquire()
b93dc49210 2009-10-13 c.kworr@d4daf: self._queue = []
b93dc49210 2009-10-13 c.kworr@d4daf: _thread.start_new_thread(self._start, ())
b93dc49210 2009-10-13 c.kworr@d4daf:
b93dc49210 2009-10-13 c.kworr@d4daf: def _start(self):
b93dc49210 2009-10-13 c.kworr@d4daf: while True:
b93dc49210 2009-10-13 c.kworr@d4daf: self._lock.acquire()
ed7808827d 2009-10-14 c.kworr@d4daf: with self._lock_queue:
ed7808827d 2009-10-14 c.kworr@d4daf: # yes this should be written this way, and yes, this is why I hate threading
ed7808827d 2009-10-14 c.kworr@d4daf: if len(self._queue) > 1:
ed7808827d 2009-10-14 c.kworr@d4daf: if self._lock.locked():
ed7808827d 2009-10-14 c.kworr@d4daf: self._lock.release()
ed7808827d 2009-10-14 c.kworr@d4daf: req = self._queue.pop(0)
ed7808827d 2009-10-14 c.kworr@d4daf: Checker.process(self, req[0], req[1], req[2], req[3])
ed7808827d 2009-10-14 c.kworr@d4daf: with self._lock_queue:
ed7808827d 2009-10-14 c.kworr@d4daf: if len(self._queue) == 0:
ed7808827d 2009-10-14 c.kworr@d4daf: if self._lock_exit.locked():
ed7808827d 2009-10-14 c.kworr@d4daf: self._lock_exit.release()
ed7808827d 2009-10-14 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: def process(self, id, site, ip_address, url_path, line):
ed7808827d 2009-10-14 c.kworr@d4daf: with self._lock_queue:
ed7808827d 2009-10-14 c.kworr@d4daf: self._queue.append((id, site, ip_address, url_path))
ed7808827d 2009-10-14 c.kworr@d4daf: self._log.info('request {} queued ({})\n'.format(id, line))
ed7808827d 2009-10-14 c.kworr@d4daf: if not self._lock_exit.locked():
ed7808827d 2009-10-14 c.kworr@d4daf: self._lock_exit.acquire()
ed7808827d 2009-10-14 c.kworr@d4daf: if self._lock.locked():
ed7808827d 2009-10-14 c.kworr@d4daf: self._lock.release()
ed7808827d 2009-10-14 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: def loop(self):
ed7808827d 2009-10-14 c.kworr@d4daf: while True:
ed7808827d 2009-10-14 c.kworr@d4daf: line = sys.stdin.readline()
ed7808827d 2009-10-14 c.kworr@d4daf: if len(line) == 0:
ed7808827d 2009-10-14 c.kworr@d4daf: break
ed7808827d 2009-10-14 c.kworr@d4daf: self.check(line)
ed7808827d 2009-10-14 c.kworr@d4daf: self._lock_exit.acquire()
ed7808827d 2009-10-14 c.kworr@d4daf:
26fc9b34d9 2010-08-07 c.kworr@d4daf: # kqueue enabled class for BSD's
ed7808827d 2009-10-14 c.kworr@d4daf: class CheckerKqueue(Checker):
ed7808827d 2009-10-14 c.kworr@d4daf: __slots__ = frozenset(['_kq', '_select', '_queue'])
ed7808827d 2009-10-14 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: def __init__(self):
ed7808827d 2009-10-14 c.kworr@d4daf: # basic initialisation
ed7808827d 2009-10-14 c.kworr@d4daf: Checker.__init__(self)
ed7808827d 2009-10-14 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: # importing select module
ed7808827d 2009-10-14 c.kworr@d4daf: import select
ed7808827d 2009-10-14 c.kworr@d4daf: self._select = select
ed7808827d 2009-10-14 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: # kreating kqueue
ed7808827d 2009-10-14 c.kworr@d4daf: self._kq = self._select.kqueue()
7c13294e9f 2010-08-07 c.kworr@d4daf: assert self._kq.fileno() != -1, "Fatal error: can't initialise kqueue."
ed7808827d 2009-10-14 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: # watching sys.stdin for data
ed7808827d 2009-10-14 c.kworr@d4daf: self._kq.control([self._select.kevent(sys.stdin, self._select.KQ_FILTER_READ, self._select.KQ_EV_ADD)], 0)
ed7808827d 2009-10-14 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: # creating data queue
ed7808827d 2009-10-14 c.kworr@d4daf: self._queue = []
ed7808827d 2009-10-14 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: def loop(self):
ed7808827d 2009-10-14 c.kworr@d4daf: # Wait for data by default
ed7808827d 2009-10-14 c.kworr@d4daf: timeout = None
26fc9b34d9 2010-08-07 c.kworr@d4daf: eof = False
26fc9b34d9 2010-08-07 c.kworr@d4daf: buffer = ''
ed7808827d 2009-10-14 c.kworr@d4daf: while True:
26fc9b34d9 2010-08-07 c.kworr@d4daf: # checking if there is any data or witing for data to arrive
ed7808827d 2009-10-14 c.kworr@d4daf: kevs = self._kq.control(None, 1, timeout)
7c13294e9f 2010-08-07 c.kworr@d4daf:
ae1c0114c1 2010-08-09 c.kworr@d4daf: for kev in kevs:
ae1c0114c1 2010-08-09 c.kworr@d4daf: if kev.filter == self._select.KQ_FILTER_READ and kev.data > 0:
ae1c0114c1 2010-08-09 c.kworr@d4daf: # reading data in
ae1c0114c1 2010-08-09 c.kworr@d4daf: new_buffer = sys.stdin.read(kev.data)
ae1c0114c1 2010-08-09 c.kworr@d4daf: # if no data was sent - we have reached end of file
ae1c0114c1 2010-08-09 c.kworr@d4daf: if len(new_buffer) == 0:
ae1c0114c1 2010-08-09 c.kworr@d4daf: eof = True
ae1c0114c1 2010-08-09 c.kworr@d4daf: else:
ae1c0114c1 2010-08-09 c.kworr@d4daf: # adding current buffer to old buffer remains
ae1c0114c1 2010-08-09 c.kworr@d4daf: buffer += new_buffer
ae1c0114c1 2010-08-09 c.kworr@d4daf: # splitting to lines
ae1c0114c1 2010-08-09 c.kworr@d4daf: lines = buffer.split('\n')
ae1c0114c1 2010-08-09 c.kworr@d4daf: # last line that was not terminate by newline returns to buffer
ae1c0114c1 2010-08-09 c.kworr@d4daf: buffer = lines[-1]
ae1c0114c1 2010-08-09 c.kworr@d4daf: # an only if there was at least one newline
ae1c0114c1 2010-08-09 c.kworr@d4daf: if len(lines) > 1:
ae1c0114c1 2010-08-09 c.kworr@d4daf: for line in lines[:-1]:
ae1c0114c1 2010-08-09 c.kworr@d4daf: # add data to the queue
ae1c0114c1 2010-08-09 c.kworr@d4daf: if self.check(line + '\n'):
ae1c0114c1 2010-08-09 c.kworr@d4daf: # don't wait for more data, start processing
ae1c0114c1 2010-08-09 c.kworr@d4daf: timeout = 0
ae1c0114c1 2010-08-09 c.kworr@d4daf:
ae1c0114c1 2010-08-09 c.kworr@d4daf: # detect end of stream and exit if possible
ae1c0114c1 2010-08-09 c.kworr@d4daf: if kev.flags >> 15 == 1:
ae1c0114c1 2010-08-09 c.kworr@d4daf: self._kq.control([self._select.kevent(sys.stdin, self._select.KQ_FILTER_READ, self._select.KQ_EV_DELETE)], 0)
ae1c0114c1 2010-08-09 c.kworr@d4daf: eof = True
ae1c0114c1 2010-08-09 c.kworr@d4daf:
ae1c0114c1 2010-08-09 c.kworr@d4daf: if len(kevs) == 0:
7c13294e9f 2010-08-07 c.kworr@d4daf: if len(self._queue) > 0:
7c13294e9f 2010-08-07 c.kworr@d4daf: # get one request and process it
26fc9b34d9 2010-08-07 c.kworr@d4daf: req = self._queue.pop(0)
26fc9b34d9 2010-08-07 c.kworr@d4daf: Checker.process(self, req[0], req[1], req[2], req[3])
26fc9b34d9 2010-08-07 c.kworr@d4daf: if len(self._queue) == 0:
26fc9b34d9 2010-08-07 c.kworr@d4daf: # wait for data - we have nothing to process
26fc9b34d9 2010-08-07 c.kworr@d4daf: timeout = None
7c13294e9f 2010-08-07 c.kworr@d4daf:
7c13294e9f 2010-08-07 c.kworr@d4daf: # if queue is empty and we reached end of stream - we can exit
7c13294e9f 2010-08-07 c.kworr@d4daf: if len(self._queue) == 0 and eof:
7c13294e9f 2010-08-07 c.kworr@d4daf: break
ed7808827d 2009-10-14 c.kworr@d4daf:
ed7808827d 2009-10-14 c.kworr@d4daf: def process(self, id, site, ip_address, url_path, line):
26fc9b34d9 2010-08-07 c.kworr@d4daf: # simply adding data to the queue
ed7808827d 2009-10-14 c.kworr@d4daf: self._queue.append((id, site, ip_address, url_path))
ed7808827d 2009-10-14 c.kworr@d4daf: self._log.info('request {} queued ({})\n'.format(id, line))
7e3418d94f 2009-10-12 c.kworr@d4daf:
fc934cead1 2009-10-13 c.kworr@d4daf: # this classes processes config file and substitutes default values
d500448801 2009-10-05 c.kworr@d4daf: class Config:
ae30851739 2010-08-12 c.kworr@d4daf: __slots__ = frozenset(['_config', '_default', '_section', 'options'])
b93dc49210 2009-10-13 c.kworr@d4daf: _default = {
b93dc49210 2009-10-13 c.kworr@d4daf: 'reactor': {
b93dc49210 2009-10-13 c.kworr@d4daf: 'reactor': 'thread',
b93dc49210 2009-10-13 c.kworr@d4daf: },
fc934cead1 2009-10-13 c.kworr@d4daf: 'log': {
fc934cead1 2009-10-13 c.kworr@d4daf: 'silent': 'no',
fc934cead1 2009-10-13 c.kworr@d4daf: },
fc934cead1 2009-10-13 c.kworr@d4daf: 'database': {
fc934cead1 2009-10-13 c.kworr@d4daf: 'host': 'localhost',
fc934cead1 2009-10-13 c.kworr@d4daf: 'database': 'squidTag',
fc934cead1 2009-10-13 c.kworr@d4daf: },}
d500448801 2009-10-05 c.kworr@d4daf:
fc934cead1 2009-10-13 c.kworr@d4daf: # function to read in config file
d500448801 2009-10-05 c.kworr@d4daf: def __init__(self):
ae30851739 2010-08-12 c.kworr@d4daf: import configparser, optparse, os
ae30851739 2010-08-12 c.kworr@d4daf:
d500448801 2009-10-05 c.kworr@d4daf: parser = optparse.OptionParser()
d500448801 2009-10-05 c.kworr@d4daf: parser.add_option('-c', '--config', dest = 'config',
d500448801 2009-10-05 c.kworr@d4daf: help = 'config file location', metavar = 'FILE',
d500448801 2009-10-05 c.kworr@d4daf: default = '/usr/local/etc/squid-tagger.conf')
ae30851739 2010-08-12 c.kworr@d4daf: parser.add_option('-d', '--dump', dest = 'dump',
ae30851739 2010-08-12 c.kworr@d4daf: help = 'dump database', action = 'store_true', metavar = 'bool',
ae30851739 2010-08-12 c.kworr@d4daf: default = False)
31e69c4237 2010-08-12 c.kworr@d4daf: parser.add_option('-f', '--flush-database', dest = 'flush_db',
31e69c4237 2010-08-12 c.kworr@d4daf: help = 'flush previous database on load', default = False,
31e69c4237 2010-08-12 c.kworr@d4daf: action = 'store_true', metavar = 'bool')
31e69c4237 2010-08-12 c.kworr@d4daf: parser.add_option('-F', '--flush-site', dest = 'flush_site',
31e69c4237 2010-08-12 c.kworr@d4daf: help = 'when flushing previous dtabase flush site index too',
31e69c4237 2010-08-12 c.kworr@d4daf: action = 'store_true', default = False, metavar = 'bool')
31e69c4237 2010-08-12 c.kworr@d4daf: parser.add_option('-l', '--load', dest = 'load',
31e69c4237 2010-08-12 c.kworr@d4daf: help = 'load database', action = 'store_true', metavar = 'bool',
31e69c4237 2010-08-12 c.kworr@d4daf: default = False)
d301d9adc6 2010-08-13 c.kworr@d4daf: parser.add_option('-D', '--dump-conf', dest = 'dump_conf',
d301d9adc6 2010-08-13 c.kworr@d4daf: help = 'dump filtering rules', default = False, metavar = 'bool',
d301d9adc6 2010-08-13 c.kworr@d4daf: action = 'store_true')
d301d9adc6 2010-08-13 c.kworr@d4daf: parser.add_option('-L', '--load-conf', dest = 'load_conf',
d301d9adc6 2010-08-13 c.kworr@d4daf: help = 'load filtering rules', default = False, metavar = 'bool',
d301d9adc6 2010-08-13 c.kworr@d4daf: action = 'store_true')
d500448801 2009-10-05 c.kworr@d4daf:
ae30851739 2010-08-12 c.kworr@d4daf: (self.options, args) = parser.parse_args()
d500448801 2009-10-05 c.kworr@d4daf:
ae30851739 2010-08-12 c.kworr@d4daf: assert os.access(self.options.config, os.R_OK), "Fatal error: can't read {}".format(self.options.config)
d500448801 2009-10-05 c.kworr@d4daf:
d500448801 2009-10-05 c.kworr@d4daf: self._config = configparser.ConfigParser()
ae30851739 2010-08-12 c.kworr@d4daf: self._config.readfp(open(self.options.config))
d500448801 2009-10-05 c.kworr@d4daf:
fc934cead1 2009-10-13 c.kworr@d4daf: # function to select config file section or create one
d500448801 2009-10-05 c.kworr@d4daf: def section(self, section):
fc934cead1 2009-10-13 c.kworr@d4daf: if not self._config.has_section(section):
fc934cead1 2009-10-13 c.kworr@d4daf: self._config.add_section(section)
d500448801 2009-10-05 c.kworr@d4daf: self._section = section
d500448801 2009-10-05 c.kworr@d4daf:
fc934cead1 2009-10-13 c.kworr@d4daf: # function to get config parameter, if parameter doesn't exists the default
fc934cead1 2009-10-13 c.kworr@d4daf: # value or None is substituted
d500448801 2009-10-05 c.kworr@d4daf: def __getitem__(self, name):
fc934cead1 2009-10-13 c.kworr@d4daf: if not self._config.has_option(self._section, name):
b93dc49210 2009-10-13 c.kworr@d4daf: if self._section in self._default:
b93dc49210 2009-10-13 c.kworr@d4daf: if name in self._default[self._section]:
fc934cead1 2009-10-13 c.kworr@d4daf: self._config.set(self._section, name, self._default[self._section][name])
fc934cead1 2009-10-13 c.kworr@d4daf: else:
fc934cead1 2009-10-13 c.kworr@d4daf: self._config.set(self._section, name, None)
fc934cead1 2009-10-13 c.kworr@d4daf: else:
fc934cead1 2009-10-13 c.kworr@d4daf: self._config.set(self._section, name, None)
b93dc49210 2009-10-13 c.kworr@d4daf: return(self._config.get(self._section, name))
d500448801 2009-10-05 c.kworr@d4daf:
fc934cead1 2009-10-13 c.kworr@d4daf: # initializing and reading in config file
d500448801 2009-10-05 c.kworr@d4daf: config = Config()
d500448801 2009-10-05 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: if config.options.dump or config.options.load or config.options.dump_conf or config.options.load_conf:
d301d9adc6 2010-08-13 c.kworr@d4daf: import csv
d301d9adc6 2010-08-13 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: tagdb = tagDB()
d301d9adc6 2010-08-13 c.kworr@d4daf: data_fields = ['site', 'tags', 'regexp']
d301d9adc6 2010-08-13 c.kworr@d4daf: conf_fields = ['netmask', 'redirect_url', 'from_weekday', 'to_weekday', 'from_time', 'to_time', 'tag']
d301d9adc6 2010-08-13 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: if config.options.dump or config.options.dump_conf:
d301d9adc6 2010-08-13 c.kworr@d4daf: csv_writer = csv.writer(sys.stdout)
d301d9adc6 2010-08-13 c.kworr@d4daf: if config.options.dump:
d301d9adc6 2010-08-13 c.kworr@d4daf: # dumping database
d301d9adc6 2010-08-13 c.kworr@d4daf: csv_writer.writerow(data_fields)
d301d9adc6 2010-08-13 c.kworr@d4daf: for row in tagdb.dump():
d301d9adc6 2010-08-13 c.kworr@d4daf: csv_writer.writerow([row[0], '{' + ','.join(row[1]) + '}', row[2]])
d301d9adc6 2010-08-13 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: elif config.options.dump_conf:
d301d9adc6 2010-08-13 c.kworr@d4daf: # dumping rules
d301d9adc6 2010-08-13 c.kworr@d4daf: csv_writer.writerow(conf_fields)
d301d9adc6 2010-08-13 c.kworr@d4daf: for row in tagdb.dump_conf():
d301d9adc6 2010-08-13 c.kworr@d4daf: csv_writer.writerow([row[0], row[1], row[2], row[3], row[4], row[5], '{' + ','.join(row[6]) + '}'])
d301d9adc6 2010-08-13 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: elif config.options.load or config.options.load_conf:
d301d9adc6 2010-08-13 c.kworr@d4daf: csv_reader = csv.reader(sys.stdin)
d301d9adc6 2010-08-13 c.kworr@d4daf: first_row = next(csv_reader)
d301d9adc6 2010-08-13 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: if config.options.load:
d301d9adc6 2010-08-13 c.kworr@d4daf: # loading database
13f90978c2 2010-08-13 c.kworr@d4daf: assert first_row == data_fields, 'File must contain csv data with theese columns: ' + repr(data_fields)
d301d9adc6 2010-08-13 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: tagdb.load(csv_reader)
d301d9adc6 2010-08-13 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: elif config.options.load_conf:
d301d9adc6 2010-08-13 c.kworr@d4daf: # loading database
13f90978c2 2010-08-13 c.kworr@d4daf: assert first_row == conf_fields, 'File must contain csv data with theese columns: ' + repr(conf_fields)
d301d9adc6 2010-08-13 c.kworr@d4daf:
d301d9adc6 2010-08-13 c.kworr@d4daf: tagdb.load_conf(csv_reader)
ae30851739 2010-08-12 c.kworr@d4daf:
ae30851739 2010-08-12 c.kworr@d4daf: else:
ae30851739 2010-08-12 c.kworr@d4daf: # main loop
ae30851739 2010-08-12 c.kworr@d4daf: config.section('reactor')
ae30851739 2010-08-12 c.kworr@d4daf: if config['reactor'] == 'thread':
ae30851739 2010-08-12 c.kworr@d4daf: checker = CheckerThread()
ae30851739 2010-08-12 c.kworr@d4daf: elif config['reactor'] == 'plain':
ae30851739 2010-08-12 c.kworr@d4daf: checker = Checker()
ae30851739 2010-08-12 c.kworr@d4daf: elif config['reactor'] == 'kqueue':
ae30851739 2010-08-12 c.kworr@d4daf: checker = CheckerKqueue()
ae30851739 2010-08-12 c.kworr@d4daf:
ae30851739 2010-08-12 c.kworr@d4daf: checker.loop()