Squid url redirector

Check-in [2654b86697]
anonymous

Check-in [2654b86697]

Overview
Comment:added gevent wrapper for SysLogHandler automatically recode all log lines from utf-8 to str simplify invoking added some comment
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | master | trunk
Files: files | file ages | folders
SHA3-256: 2654b866976c1e1a160bdfafd3cae179b097bc07d45b7dc2d1b07f1c9b5b0478
User & Date: arcade@b1t.name on 2012-07-09 11:26:55.000
Other Links: branch diff | manifest | tags
Context
2012-07-09
14:25
switched sg_import to python3.2 reworked SysLogHandlerQueue to stick to one socket reworked FReadlineQueue to use io.FileIO for file operation check-in: 67e8b3309d user: arcade@b1t.name tags: master, trunk
11:26
added gevent wrapper for SysLogHandler automatically recode all log lines from utf-8 to str simplify invoking added some comment check-in: 2654b86697 user: arcade@b1t.name tags: master, trunk
2012-07-07
15:08
logging fully rewritten to use sockets instead of syscalls check-in: fad48b740c user: arcade@b1t.name tags: master, trunk
Changes
92
93
94
95
96
97
98




































99
100
101
102
103
104
105
106
107
108
				self._config.set(self._section, name, None)
		return(self._config.get(self._section, name))

# initializing and reading in config file
config = Config()

import logging, logging.handlers




































logger = logging.getLogger('squidTag')
logger.setLevel(logging.INFO)
handler = logging.handlers.SysLogHandler('/dev/log')
handler.setFormatter(logging.Formatter(str('squidTag[%(process)s]: %(message)s')))
logger.addHandler(handler)

# tiny wrapper around a file to make reads from it geventable
# or should i move this somewhere?

class FReadlineQueue(gevent.queue.Queue):







>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>


|







92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
				self._config.set(self._section, name, None)
		return(self._config.get(self._section, name))

# initializing and reading in config file
config = Config()

import logging, logging.handlers

# wrapper around logging handler to make it queue records and don't stall when sending them

class SysLogHandlerQueue(logging.handlers.SysLogHandler):
	__slots__ = frozenset(['_event', '_tail', '_workers'])

	def __init__(self):
		logging.handlers.SysLogHandler.__init__(self, '/dev/log')
		self._event = gevent.event.Event()
		self._event.set()
		self._tail = gevent.queue.Queue()
		self._workers = set()

	def emit(self, record):
		# my syslog is broken and cannot into UTF-8 BOM
		record.msg = str(record.msg)
		self._tail.put(record)
		if self._tail.qsize() != 0:
			# in case queue is empty we will spawn new worker
			# all workers are logged so we can kill them on close()
			self._workers.add(gevent.spawn(self._writer))

	def _writer(self):
		# here we are locking the queue so we can be sure we are the only one
		self._event.wait()
		self._event.clear()
		while not self._tail.empty():
			logging.handlers.SysLogHandler.emit(self, self._tail.get())
		self._event.set()
		self._workers.remove(gevent.getcurrent())

	def close(self):
		for worker in self._workers:
			gevent.kill(worker)
		logging.handlers.SysLogHandler.close(self)

logger = logging.getLogger('squidTag')
logger.setLevel(logging.INFO)
handler = SysLogHandlerQueue()
handler.setFormatter(logging.Formatter(str('squidTag[%(process)s]: %(message)s')))
logger.addHandler(handler)

# tiny wrapper around a file to make reads from it geventable
# or should i move this somewhere?

class FReadlineQueue(gevent.queue.Queue):
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150


151
152
153
154
155
156
157
		if len(self._tail) > 0:
			rows[0] = self._tail + rows[0]
		# popping out last (incomplete) element
		self._tail = rows.pop(-1)
		# dropping all complete elements to the queue
		for row in rows:
			self.put_nowait(row)
			logger.info(str('< ' + row))
		if len(buf) > 0:
			# no EOF, reinstalling event handler
			gevent.core.read_event(self._fd.fileno(), self._wait_helper)
		else:
			# EOF found, sending EOF to queue
			self.put_nowait(None)

stdin = FReadlineQueue(sys.stdin)



class FWritelineQueue(gevent.queue.JoinableQueue):
	# storing fileno, io interface, leftover
	__slots__ = frozenset(['_fileno', '_io', '_tail'])

	def __init__(self, fd, closefd = True):
		import io







|








>
>







171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
		if len(self._tail) > 0:
			rows[0] = self._tail + rows[0]
		# popping out last (incomplete) element
		self._tail = rows.pop(-1)
		# dropping all complete elements to the queue
		for row in rows:
			self.put_nowait(row)
			logger.info('< ' + row)
		if len(buf) > 0:
			# no EOF, reinstalling event handler
			gevent.core.read_event(self._fd.fileno(), self._wait_helper)
		else:
			# EOF found, sending EOF to queue
			self.put_nowait(None)

stdin = FReadlineQueue(sys.stdin)

# wrapper against file handler that makes possible to queue some writes without stalling

class FWritelineQueue(gevent.queue.JoinableQueue):
	# storing fileno, io interface, leftover
	__slots__ = frozenset(['_fileno', '_io', '_tail'])

	def __init__(self, fd, closefd = True):
		import io
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
# abstract class with basic checking functionality
class Checker(object):
	__slots__ = frozenset(['_db', '_log', '_queue', '_request', '_stdout'])

	def __init__(self, queue, logger):
		self._db = tagDB()
		self._log = logger
		self._log.info(str('started'))
		self._request = re.compile('^([0-9]+)\ (http|ftp):\/\/([-\w.:]+)\/([^ ]*)\ ([0-9.]+)\/(-|[\w\.]+)\ (-|\w+)\ (-|GET|HEAD|POST).*$')
		self._queue = queue
		self._stdout = FWritelineQueue(sys.stdout, False)

	def process(self, id, site, ip_address, url_path, line = None):
		#self._log.info(str('trying {}'.format(site)))
		result = self._db.check(site, ip_address)
		reply = None
		#self._log.info(str('got {} lines from database'.format(len(result))))
		for row in result:
			if row != None and row[0] != None:
				if row[1] != None:
					self._log.info(str('trying regexp "{}" versus "{}"'.format(row[1], url_path)))
					try:
						if re.compile(row[1]).match(url_path):
							reply = row[0].format(url_path)
						else:
							continue
					except:
						self._log.info(str("can't compile regexp"))
				else:
					reply = row[0].format(url_path)
			if reply != None:
				self.writeline('{} {}'.format(id, reply))
				return(True)
		self.writeline('{}'.format(id))

	def check(self):
		while True:
			line = self._queue.get()
			if line == None:
				break
			#self._log.info(str('request: ' + line))
			request = self._request.match(line)
			if request:
				id = request.group(1)
				#proto = request.group(2)
				site = request.group(3)
				url_path = request.group(4)
				ip_address = request.group(5)
				self.process(id, site, ip_address, url_path, line)
			else:
				self._log.info(str('bad request'))
				self.writeline(line)

	def writeline(self, string):
		self._log.info(str('> ' + string))
		self._stdout.put(string)

	def loop(self):
		pool = gevent.pool.Pool()
		pool.spawn(self.check)
		pool.join()

if config.options.dump or config.options.load or config.options.dump_conf or config.options.load_conf:
	import csv

	tagdb = tagDB()
	data_fields = ['site', 'tag', 'regexp']
	conf_fields = ['netmask', 'redirect_url', 'from_weekday', 'to_weekday', 'from_time', 'to_time', 'tag']








|





|


|



|






|







|




|









|



|


<
<
<
<
<







301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357





358
359
360
361
362
363
364
# abstract class with basic checking functionality
class Checker(object):
	__slots__ = frozenset(['_db', '_log', '_queue', '_request', '_stdout'])

	def __init__(self, queue, logger):
		self._db = tagDB()
		self._log = logger
		self._log.info('started')
		self._request = re.compile('^([0-9]+)\ (http|ftp):\/\/([-\w.:]+)\/([^ ]*)\ ([0-9.]+)\/(-|[\w\.]+)\ (-|\w+)\ (-|GET|HEAD|POST).*$')
		self._queue = queue
		self._stdout = FWritelineQueue(sys.stdout, False)

	def process(self, id, site, ip_address, url_path, line = None):
		#self._log.info('trying {}'.format(site))
		result = self._db.check(site, ip_address)
		reply = None
		#self._log.info('got {} lines from database'.format(len(result)))
		for row in result:
			if row != None and row[0] != None:
				if row[1] != None:
					self._log.info('trying regexp "{}" versus "{}"'.format(row[1], url_path))
					try:
						if re.compile(row[1]).match(url_path):
							reply = row[0].format(url_path)
						else:
							continue
					except:
						self._log.info("can't compile regexp")
				else:
					reply = row[0].format(url_path)
			if reply != None:
				self.writeline('{} {}'.format(id, reply))
				return(True)
		self.writeline('{}'.format(id))

	def loop(self):
		while True:
			line = self._queue.get()
			if line == None:
				break
			#self._log.info('request: ' + line)
			request = self._request.match(line)
			if request:
				id = request.group(1)
				#proto = request.group(2)
				site = request.group(3)
				url_path = request.group(4)
				ip_address = request.group(5)
				self.process(id, site, ip_address, url_path, line)
			else:
				self._log.info('bad request')
				self.writeline(line)

	def writeline(self, string):
		self._log.info('> ' + string)
		self._stdout.put(string)






if config.options.dump or config.options.load or config.options.dump_conf or config.options.load_conf:
	import csv

	tagdb = tagDB()
	data_fields = ['site', 'tag', 'regexp']
	conf_fields = ['netmask', 'redirect_url', 'from_weekday', 'to_weekday', 'from_time', 'to_time', 'tag']