Squid url redirector

Check-in [88c03b5440]
anonymous

Check-in [88c03b5440]

Overview
Comment:Main loop reworked. With regexp support there could be possibilities of passing x level domain on regexp fail and not looking into x-1 level domain. Now all matches are selected from database ordered by domain level and checked until match is found.
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | master | trunk
Files: files | file ages | folders
SHA3-256: 88c03b544029127c9c6b8bdbe75450e0bfd5f999493dfb79616a0d1656880aea
User & Date: c.kworr@d4daf22a-8aaf-11de-a64d-234b64dd91b4 on 2009-10-09 06:40:07.000
Other Links: branch diff | manifest | tags
Context
2009-10-10
10:26
function to return site id check-in: e74427953f user: c.kworr@d4daf22a-8aaf-11de-a64d-234b64dd91b4 tags: master, trunk
2009-10-09
06:40
Main loop reworked. With regexp support there could be possibilities of passing x level domain on regexp fail and not looking into x-1 level domain. Now all matches are selected from database ordered by domain level and checked until match is found. check-in: 88c03b5440 user: c.kworr@d4daf22a-8aaf-11de-a64d-234b64dd91b4 tags: master, trunk
06:20
minor naming unification check-in: 33e72616c9 user: c.kworr@d4daf22a-8aaf-11de-a64d-234b64dd91b4 tags: master, trunk
Changes
Side-by-Side Diff Ignore Whitespace Patch
Modified squid-tagger.py from [859fd273be] to [0cfe66321a].
︙
24
25
26
27
28
29
30
31

32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

47
48
49
50
51
52
53
54
55
56
57




58
59
60
61
62
63
64
65
66
67

68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84














85
86
87
88
89
90
91

24
25
26
27
28
29
30

31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

46




47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73












74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94








-
+














-
+
-
-
-
-







+
+
+
+










+





-
-
-
-
-
-
-
-
-
-
-
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+









class tagDB:
	__slots__ = frozenset(['_prepared', '_check_stmt', '_db'])

	def __init__(self):
		self._prepared = set()
		self._db = False
		self._check_stmt = self._curs().prepare("select redirect_url, regexp from site_rule where site <@ tripdomain($1) and netmask >> $2::text::inet order by array_length(site, 1) desc limit 1")
		self._check_stmt = self._curs().prepare("select redirect_url, regexp from site_rule where site <@ tripdomain($1) and netmask >> $2::text::inet order by array_length(site, 1) desc")

	def _curs(self):
		if not self._db:
			config.section('database')
			self._db = postgresql.open(
				'pq://{0}:{1}@{2}/{3}'.format(
					config['user'],
					config['password'],
					config['host'],
					config['database'],
			) )
		return(self._db)

	def check(self, ip_address, site):
		result = self._check_stmt(site, ip_address)
		return self._check_stmt(site, ip_address)
		if len(result) > 0:
			return result[0]
		else:
			return None

class CheckerThread:
	__slots__ = frozenset(['_db', '_lock', '_lock_queue', '_log', '_queue'])

	def __init__(self, db, log):
		self._db = db
		self._log = log
		# Spin lock. Loop acquires it on start then releases it when holding queue
		# lock. This way the thread proceeds without stops while queue has data and
		# gets stalled when no data present. The lock is released by queue writer
		# after storing something into the queue
		self._lock = _thread.allocate_lock()
		self._lock_queue = _thread.allocate_lock()
		self._lock.acquire()
		self._queue = []
		_thread.start_new_thread(self._start, ())

	def _start(self):
		while True:
			self._lock.acquire()
			self._lock_queue.acquire()
			# yes this should be written this way, and yes, this is why I hate threading
			if len(self._queue) > 1 and self._lock.locked():
				self._lock.release()
			req = self._queue.pop(0)
			self._lock_queue.release()
			self._log.info('trying %s\n'%req[1])
			row = self._db.check(req[2], req[1])
			if row != None and row[0] != None:
				if row[1] != None:
					self._log.info('trying regexp "{0}" versus "{1}"\n'.format(row[1], req[3]))
					if re.compile(row[1]).match(req[3]):
						writeline('%s 302:%s\n'%(req[0], row[0]))
					else:
						writeline('%s -\n'%req[0])
				else:
					writeline('%s 302:%s\n'%(req[0], row[0]))
			else:
				writeline('%s -\n'%req[0])
			result = self._db.check(req[2], req[1])
			for row in result:
				if row != None and row[0] != None:
					if row[1] != None:
						self._log.info('trying regexp "{0}" versus "{1}"\n'.format(row[1], req[3]))
						if re.compile(row[1]).match(req[3]):
							writeline('%s 302:%s\n'%(req[0], row[0]))
							break
						else:
							continue
					else:
						writeline('%s 302:%s\n'%(req[0], row[0]))
						break
			writeline('%s -\n'%req[0])

	def check(self, line):
		request = re.compile('^([0-9]+)\ (http|ftp):\/\/([-\w.:]+)\/([^ ]*)\ ([0-9.]+)\/(-|[\w\.]+)\ (-|\w+)\ (-|GET|HEAD|POST).*$').match(line)
		if request:
			id = request.group(1)
			site = request.group(3)
			url_path = request.group(4)
︙