Squid url redirector

Diff
anonymous

Diff

Differences From Artifact [035ca6f621]:

To Artifact [ab921f1976]:


161
162
163
164
165
166
167



















































168
169
170
171
172
173
174
			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 around database
class tagDB(object):
	__slots__ = frozenset(['_cursor', '_db'])

	def __init__(self):
		config.section('database')
		if config['host'] == None:







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







161
162
163
164
165
166
167
168
169
170
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
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
			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
		# initialising class
		gevent.queue.JoinableQueue.__init__(self)
		# storing fileno
		self._fileno = fd.fileno()
		# creating interface
		self._io = io.FileIO(self._fileno, 'w', closefd)
		# using empty tail
		self._tail = None
		# putting file to nonblocking mode
		fcntl.fcntl(self._fileno, fcntl.F_SETFL, fcntl.fcntl(self._fileno, fcntl.F_GETFL)  | os.O_NONBLOCK)

	def __del__(self):
		# purge queue before deleting
		if not self.empty():
			self.join()

	def put(self, item, block=True, timeout=None):
		# calling real put
		gevent.queue.JoinableQueue.put(self, item, block, timeout)
		# installing event handler
		gevent.core.write_event(self._fileno, self._wait_helper)

	def _wait_helper(self, ev, evtype):
		# XXX ev, evtype checking?
		# checking leftover
		while True:
			if self._tail == None:
				try:
					self._tail = str(self.get_nowait()).encode('utf-8') + '\n'
				except gevent.queue.Empty:
					self._tail = None
					return
			# writing tail
			written = self._io.write(self._tail)
			length = len(self._tail)
			if written == length:
				self._tail = None
			elif written < length:
				self._tail = self._tail[written:]
				break
			else:
				break
		# reinstalling event handler
		gevent.core.write_event(self._fileno, self._wait_helper)

# wrapper around database
class tagDB(object):
	__slots__ = frozenset(['_cursor', '_db'])

	def __init__(self):
		config.section('database')
		if config['host'] == None:
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237

238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
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

	def dump_conf(self):
		self._cursor.execute("select netmask, redirect_url, from_weekday, to_weekday, from_time, to_time, tag::text from rules")
		return(self._field_names(), self._cursor.fetchall())

# abstract class with basic checking functionality
class Checker(object):
	__slots__ = frozenset(['_db', '_log', '_queue', '_request'])

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


	def process(self, id, site, ip_address, url_path, line = None):
		#self._log.info('trying {}\n'.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 "{}"\n'.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('{} {}\n'.format(id, reply))
				return(True)
		self.writeline('{}\n'.format(id))

	def check(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\n')
				self.writeline(line + '\n')

	def writeline(self, string):
		self._log.info('sending: ' + string)
		sys.stdout.write(string)
		sys.stdout.flush()

	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:







|




|


>


|






|










|

|

















|



<
|







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
332
333

334
335
336
337
338
339
340
341

	def dump_conf(self):
		self._cursor.execute("select netmask, redirect_url, from_weekday, to_weekday, from_time, to_time, tag::text from rules")
		return(self._field_names(), self._cursor.fetchall())

# 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 check(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\n')
				self.writeline(line)

	def writeline(self, string):
		self._log.info('sending: ' + 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: