forked from sherpya/geolite2legacy
-
Notifications
You must be signed in to change notification settings - Fork 0
/
geolite2legacy.py
executable file
·475 lines (380 loc) · 15.6 KB
/
geolite2legacy.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
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
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
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
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
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
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
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
#!/usr/bin/env python
# The MIT License (MIT)
#
# Copyright (c) 2015 Mark Teodoro
# Copyright (c) 2018-2019 Gianluigi Tiesi
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from __future__ import print_function
import os
import re
import sys
import csv
import struct
import codecs
import logging
import argparse
# python 2 ipaddr
try:
from ipaddr import IPNetwork
# python3 built-in ipaddress
except:
from ipaddress import ip_network as IPNetwork
from time import time
from zipfile import ZipFile
from collections import defaultdict
from pygeoip_const import *
re_words = re.compile(r'\W+', re.U)
cc_idx = dict((cc.lower(), i) for i, cc in enumerate(COUNTRY_CODES))
cc_idx['cw'] = cc_idx['an'] # netherlands antilles / curacao
cc_idx['uk'] = cc_idx['gb'] # uk / great britain
cc_idx['sx'] = cc_idx['fx'] # st. martin?
cc_idx['xk'] = cc_idx['rs'] # kosovo -> serbia
continent_codes = {'AS': 'AP'}
geoname2fips = {}
output_encoding = 'utf-8'
datfilecomment = ''
def serialize_text(text):
try:
return text.encode(output_encoding)
except UnicodeEncodeError:
print('Warning cannot encode {!r} using {}'.format(text, output_encoding))
return text.encode(output_encoding, 'replace')
if sys.version_info[0] == 2:
# noinspection PyShadowingBuiltins,PyUnresolvedReferences
range = xrange
def decode_text(text):
return text.decode('utf-8')
# noinspection PyPep8Naming,PyUnusedLocal
def TextIOWrapper(f, encoding=None):
return f
else:
from io import TextIOWrapper
def decode_text(text):
return text
class RadixTreeNode(object):
__slots__ = ['segment', 'lhs', 'rhs']
def __init__(self, segment):
self.segment = segment
self.lhs = None
self.rhs = None
class RadixTree(object):
seek_depth = -1
edition = -1
reclen = -1
segreclen = -1
def __init__(self, debug=False):
self.debug = debug
self.netcount = 0
self.segments = [RadixTreeNode(0)]
self.data_offsets = {}
self.data_segments = []
self.cur_offset = 1
def __setitem__(self, net, data):
self.netcount += 1
# python2 ipaddr
try:
inet = int(net)
# python3 built-in ipaddress
except:
inet = int(net.network_address)
node = self.segments[0]
for depth in range(self.seek_depth, self.seek_depth - (net.prefixlen - 1), -1):
if inet & (1 << depth):
if not node.rhs:
node.rhs = RadixTreeNode(len(self.segments))
self.segments.append(node.rhs)
node = node.rhs
else:
if not node.lhs:
node.lhs = RadixTreeNode(len(self.segments))
self.segments.append(node.lhs)
node = node.lhs
if data not in self.data_offsets:
self.data_offsets[data] = self.cur_offset
enc_data = self.encode(*data)
self.data_segments.append(enc_data)
self.cur_offset += (len(enc_data))
if self.debug:
# store net after data for easier debugging
data = data, net
if inet & (1 << self.seek_depth - (net.prefixlen - 1)):
node.rhs = data
else:
node.lhs = data
def gen_nets(self, codes, outfile):
raise NotImplementedError
def load(self, locationsfile, outfile):
locations = {}
if locationsfile:
for row in csv.DictReader(locationsfile):
geoname_id = row['geoname_id']
# remap continent codes according to https://dev.maxmind.com/geoip/legacy/codes/iso3166/
continent_code = row['continent_code']
row['continent_code'] = continent_codes.get(continent_code, continent_code)
locations[geoname_id] = row
for nets, data in self.gen_nets(locations, outfile):
for net in nets:
self[net] = data
def dump_node(self, node):
if not node:
# empty leaf
return '--'
elif isinstance(node, RadixTreeNode):
# internal node
return node.segment
else:
# data leaf
data = node[0] if self.debug else node
return '%d %s' % (len(self.segments) + self.data_offsets[data], node)
def dump(self):
for node in self.segments:
print(node.segment, [self.dump_node(node.lhs), self.dump_node(node.rhs)])
def encode(self, *args):
raise NotImplementedError
def encode_rec(self, rec, reclen):
"""encode rec as 4-byte little-endian int, then truncate it to reclen"""
assert (reclen <= 4)
return struct.pack('<I', rec)[:reclen]
def serialize_node(self, node):
if not node:
# empty leaf
rec = len(self.segments)
elif isinstance(node, RadixTreeNode):
# internal node
rec = node.segment
else:
# data leaf
data = node[0] if self.debug else node
rec = len(self.segments) + self.data_offsets[data]
return self.encode_rec(rec, self.reclen)
def serialize(self, f):
if len(self.segments) >= 2 ** (8 * self.segreclen):
logging.warning('too many segments for final segment record size!')
for node in self.segments:
f.write(self.serialize_node(node.lhs))
f.write(self.serialize_node(node.rhs))
f.write(struct.pack('B', 42)) # So long, and thanks for all the fish!
f.write(b''.join(self.data_segments))
f.write(datfilecomment.encode('ascii')) # .dat file comment - can be anything
f.write(struct.pack('B', 0xff) * 3)
f.write(struct.pack('B', self.edition))
f.write(self.encode_rec(len(self.segments), self.segreclen))
class ASNRadixTree(RadixTree):
seek_depth = 31
edition = ASNUM_EDITION
reclen = STANDARD_RECORD_LENGTH
segreclen = SEGMENT_RECORD_LENGTH
def gen_nets(self, locations, infile):
for row in csv.DictReader(infile):
nets = [IPNetwork(row['network'])]
org = decode_text(row['autonomous_system_organization'])
asn = row['autonomous_system_number']
entry = u'AS{} {}'.format(asn, org)
yield nets, (serialize_text(entry),)
def encode(self, data):
return data + b'\0\0\0'
class ASNv6RadixTree(ASNRadixTree):
seek_depth = 127
edition = ASNUM_EDITION_V6
reclen = STANDARD_RECORD_LENGTH
segreclen = SEGMENT_RECORD_LENGTH
class CityRev1RadixTree(RadixTree):
seek_depth = 31
edition = CITY_EDITION_REV1
reclen = STANDARD_RECORD_LENGTH
segreclen = SEGMENT_RECORD_LENGTH
def gen_nets(self, locations, infile):
for row in csv.DictReader(infile):
location = locations.get(row['geoname_id'])
if location is None:
continue
nets = [IPNetwork(row['network'])]
country_iso_code = location['country_iso_code'] or location['continent_code']
fips_code = geoname2fips.get(location['geoname_id'])
if fips_code is None:
logging.debug('Missing fips-10-4 for {}'.format(location['subdivision_1_name']))
fips_code = '00'
else:
logging.debug('fips-10-4 for {} is {}'.format(location['subdivision_1_name'], fips_code))
yield nets, (country_iso_code,
serialize_text(fips_code), # region
serialize_text(decode_text(location['city_name'])),
serialize_text(row['postal_code']),
row['latitude'],
row['longitude'],
location['metro_code'],
'') # area_code
def encode(self, country, region, city, postal_code, lat, lon, metro_code, area_code):
def str2num(num, ntype):
return ntype(num) if num else ntype(0)
country = country.lower()
lat, lon = round(str2num(lat, float), 4), round(str2num(lon, float), 4)
metro_code, area_code = str2num(metro_code, int), str2num(area_code, int)
buf = []
try:
buf.append(struct.pack('B', cc_idx[country]))
except KeyError:
logging.warning("'%s': missing country. update const.COUNTRY_CODES?", country)
buf.append(struct.pack('B', cc_idx['']))
buf.append(b'\0'.join((region, city, postal_code)))
buf.append(b'\0')
buf.append(self.encode_rec(int((lat + 180) * 10000), 3))
buf.append(self.encode_rec(int((lon + 180) * 10000), 3))
if (metro_code or area_code) and country == 'us':
buf.append(self.encode_rec(metro_code * 1000 + area_code, 3))
else:
buf.append(b'\0\0\0')
return b''.join(buf)
class CityRev1v6RadixTree(CityRev1RadixTree):
seek_depth = 127
edition = CITY_EDITION_REV1_V6
reclen = STANDARD_RECORD_LENGTH
segreclen = SEGMENT_RECORD_LENGTH
class CountryRadixTree(RadixTree):
seek_depth = 31
edition = COUNTRY_EDITION
reclen = STANDARD_RECORD_LENGTH
segreclen = SEGMENT_RECORD_LENGTH
def gen_nets(self, locations, infile):
for row in csv.DictReader(infile):
location = locations.get(row['geoname_id'])
if location is None:
continue
nets = [IPNetwork(row['network'])]
country_iso_code = location['country_iso_code'] or location['continent_code']
yield nets, (country_iso_code,)
def encode(self, cc):
# unused
return ''
def serialize_node(self, node):
if not node:
# empty leaf
rec = COUNTRY_BEGIN
elif isinstance(node, RadixTreeNode):
# internal node
rec = node.segment
else:
# data leaf
data = node[0] if self.debug else node
cc = data[0]
try:
offset = cc_idx[cc.lower()]
except KeyError:
logging.warning("'%s': missing country. update const.COUNTRY_CODES?", cc)
offset = 0
# data leaves directly encode cc index as an offset
rec = COUNTRY_BEGIN + offset
return self.encode_rec(rec, self.reclen)
def serialize(self, f):
for node in self.segments:
f.write(self.serialize_node(node.lhs))
f.write(self.serialize_node(node.rhs))
f.write(struct.pack('B', 0x00) * 3)
f.write(datfilecomment.encode('ascii')) # .dat file comment - can be anything
f.write(struct.pack('B', 0xff) * 3)
f.write(struct.pack('B', self.edition))
f.write(self.encode_rec(len(self.segments), self.segreclen))
class Countryv6RadixTree(CountryRadixTree):
seek_depth = 127
edition = COUNTRY_EDITION_V6
reclen = STANDARD_RECORD_LENGTH
segreclen = SEGMENT_RECORD_LENGTH
RTree = {
'Country': {'IPv4': CountryRadixTree, 'IPv6': Countryv6RadixTree},
'City': {'IPv4': CityRev1RadixTree, 'IPv6': CityRev1v6RadixTree},
'ASN': {'IPv4': ASNRadixTree, 'IPv6': ASNv6RadixTree}
}
Filenames = {
'Country': {'IPv4': "GeoIP.dat", 'IPv6': "GeoIPv6.dat"},
'City': {'IPv4': "GeoIPCity.dat", 'IPv6': "GeoIPCityv6.dat"},
'ASN': {'IPv4': "GeoIPASNum.dat", 'IPv6': "GeoIPASNumv6.dat"}
}
def parse_fips(fipsfile):
with open(fipsfile) as f:
for row in csv.DictReader(f):
geoname2fips[row['geoname_id']] = row['region']
return geoname2fips
def main():
global output_encoding, datfilecomment
parser = argparse.ArgumentParser()
parser.add_argument('-i', '--input-file', required=True, help='input zip file containings csv databases')
parser.add_argument('-o', '--output-file', help='output GeoIP dat file')
parser.add_argument('-f', '--fips-file', help='geonameid to fips code mappings')
parser.add_argument('-e', '--encoding', help='encoding to use for the output rather than utf-8')
parser.add_argument('-d', '--debug', action='store_true', default=False, help='debug mode')
parser.add_argument('-6', '--ipv6', action='store_const', default='IPv4', const='IPv6', help='use ipv6 database')
opts = parser.parse_args()
if opts.encoding:
try:
codecs.lookup(opts.encoding)
except LookupError as e:
print(e)
sys.exit(1)
output_encoding = opts.encoding
re_entry = re.compile(r'.*?/Geo(?:Lite|IP)2-(?P<database>.*?)-(?P<filetype>.*?)-(?P<arg>.*)\.csv')
entries = defaultdict(lambda: defaultdict(dict))
ziparchive = ZipFile(opts.input_file)
for entry in ziparchive.filelist:
match = re_entry.match(entry.filename)
if match is None:
continue
db, filetype, arg = match.groups()
entries[db][filetype][arg] = entry
if len(entries) != 1:
print('More than one kind of database found, please check the archive')
sys.exit(1)
# noinspection PyUnboundLocalVariable
datfilecomment = '{} converted to legacy MaxMind DB with geolite2legacy'.format(os.path.dirname(entry.filename))
dbtype, entries = entries.popitem()
if dbtype == 'ASN':
locs = None
else:
if not {'Locations', 'Blocks'} <= set(entries.keys()):
print('Missing Locations or Block files, please check the archive')
sys.exit(1)
locs = entries['Locations'].get('en')
if locs is None:
print('Selected locale not found in archive')
sys.exit(1)
locs = TextIOWrapper(ziparchive.open(locs, 'r'), encoding='utf-8')
if dbtype not in RTree:
print('{} not supported'.format(dbtype))
sys.exit(1)
r = RTree[dbtype][opts.ipv6](debug=opts.debug)
blocks = entries['Blocks'].get(opts.ipv6)
if blocks is None:
print('The selected block file not found in archive')
sys.exit(1)
if dbtype != 'ASN':
fips_file = opts.fips_file or os.path.join(os.path.dirname(os.path.realpath(__file__)), 'geoname2fips.csv')
parse_fips(fips_file)
tstart = time()
print('Database type {} - Blocks {} - Encoding: {}'.format(dbtype, opts.ipv6, output_encoding))
r.load(locs, TextIOWrapper(ziparchive.open(blocks, 'r'), encoding='utf-8'))
if not opts.output_file:
opts.output_file = Filenames[dbtype][opts.ipv6]
print('Output file {}'.format(opts.output_file))
with open(opts.output_file, 'wb') as output:
r.serialize(output)
tstop = time()
print('wrote %d-node trie with %d networks (%d distinct labels) in %d seconds' % (
len(r.segments), r.netcount, len(r.data_offsets), tstop - tstart))
if __name__ == '__main__':
main()