/
Fetch2.py
503 lines (429 loc) · 17.5 KB
/
Fetch2.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
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
# Code attributions: Pavel Tyslacki
import argparse
import csv
import datetime
import gzip
import io
import itertools
import json
import lzma
import os
import pickle
import sys
import time
import urllib.error
import urllib.parse
import urllib.request
import lxml.etree
import mercantile
import overpass
import shapely.geometry
import shapely.wkt
LOGS_URL = 'http://planet.openstreetmap.org/tile_logs/'
MIN_DATE = '0000-00-00'
MAX_DATE = '9999-99-99'
EXT = '.txt.xz'
COUNTRIES_QUERY = ('[out:csv(::id, "ISO3166-1", "ISO3166-1:alpha2")];'
'('
'relation["ISO3166-1"];'
'relation["ISO3166-1:alpha2"];'
');'
'out tags;')
RELATION_QUERY = ('[out:csv(::id, "ISO3166-1", "ISO3166-1:alpha2")];'
'('
'relation(%s);'
');'
'out tags;')
PREFETCH_GEOMETRY_LINK = 'http://polygons.openstreetmap.fr/?id=%s'
FETCH_GEOMETRY_LINK = 'http://polygons.openstreetmap.fr/get_wkt.py?id=%s¶ms=0'
MAX_FETCH_ATTEMPTS = [1, 3, 10, 30]
COUNTRIES_IDS_SKIP = (
11980,
1111111,
1362232,
1401925,
5466491,
)
MIN_ZOOM = 0
MAX_ZOOM = 19
SPLIT_ZOOM = 8
MIN_INTERSECTION_AREA = 0.001
MIN_INTERSECTION_AREA_PER_PERIMETER = 0.01
MIN_GEOMETRY_AREA_PER_PERIMETER = 0.000005
DUMPS_CACHE_FOLDER = 'tile_logs'
COUNTRIES_GEOM_CACHE_FOLDER = 'countries'
GEOM_CACHE = 'cache_geoms.picle'
GROUPED_GEOM_CACHE = 'cache_grouped.picle'
GROUPED_GEOM_WITH_EMPTY_CACHE = 'cache_grouped_with_empty.picle'
SLICED_TO_TILES_GEOM_CACHE = 'cache_sliced_to_tiles.picle'
TILE_CACHE = 'cache_tile.json'
class Stat(object):
def __init__(self):
self.out = sys.stderr
self.start = datetime.datetime.now()
self.in_all = 0
self.in_no_cached = 0
self.in_child_cache = 0
self.in_direct_cache = 0
self.child_zoom_less = 0
self.child_zoom_equal = 0
self.filtered_bbox = 0
self.filtered_geom = 0
self.append_geom = 0
def log_stats(self, msg, cache):
cached_for_child = 0
for cached_item in cache.values():
if '|' not in cached_item:
cached_for_child += 1
self.log('%s - %s - ps: %s/%s - cc: %s/%s - zm: %s/%s - '
'fl: %s/%s - ap: %s - cs: %s/%s',
msg, datetime.datetime.now() - self.start,
self.in_all, self.in_no_cached,
self.in_child_cache, self.in_direct_cache,
self.child_zoom_less, self.child_zoom_equal,
self.filtered_bbox, self.filtered_geom,
self.append_geom,
cached_for_child, len(cache),
)
def log(self, msg, *args):
self.out.write('%s\n' % (msg % args))
def _fetch(link):
attempt = 0
while True:
try:
time.sleep(MAX_FETCH_ATTEMPTS[attempt])
request = urllib.request.Request(link, headers={
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Encoding': 'gzip, deflate',
'Accept-Language': 'en-US,en;q=0.5',
'Connection': 'keep-alive',
})
return gzip.open(urllib.request.urlopen(request)).read().decode()
except urllib.error.HTTPError:
attempt += 1
if attempt >= len(MAX_FETCH_ATTEMPTS):
raise
def get_country_geom(osm_id, iso):
prefetch_link = FETCH_GEOMETRY_LINK % osm_id
link = FETCH_GEOMETRY_LINK % osm_id
if not os.path.exists(COUNTRIES_GEOM_CACHE_FOLDER):
os.mkdir(COUNTRIES_GEOM_CACHE_FOLDER)
file_name_wkt = os.path.join(COUNTRIES_GEOM_CACHE_FOLDER,
'%s-%s.wkt' % (iso, osm_id))
file_name_geojson = os.path.join(COUNTRIES_GEOM_CACHE_FOLDER,
'%s-%s.geojson' % (iso, osm_id))
if os.path.exists(file_name_wkt):
with open(file_name_wkt, 'r') as file:
response = file.read()
geom = shapely.wkt.loads(response)
elif os.path.exists(file_name_geojson):
with open(file_name_geojson, 'r') as file:
response = file.read()
geom = shapely.geometry.shape(json.loads(response))
else:
_fetch(prefetch_link)
response = _fetch(link)
if response.startswith('SRID=4326;'):
response = response[len('SRID=4326;'):]
geom = shapely.wkt.loads(response)
with open(file_name_wkt, 'w') as file:
file.write(response)
return geom
def get_countries(rel=None, country=None, query=None):
countries = {}
if query:
pass
elif rel:
query = RELATION_QUERY % rel
else:
query = COUNTRIES_QUERY
response = overpass.API()._GetFromOverpass(query)
reader = csv.reader(io.StringIO(response), delimiter='\t',)
next(reader)
for osm_id, iso3166_1, iso3166_1_alpha2 in reader:
osm_id = int(osm_id)
iso = iso3166_1 or iso3166_1_alpha2
if not rel and not country and osm_id in COUNTRIES_IDS_SKIP:
continue
if not rel and (not iso or len(iso) != 2):
continue
if country and iso != country:
continue
if (osm_id, iso) in countries:
continue
Stat().log('%s-%s', iso, osm_id)
geom = get_country_geom(osm_id, iso)
countries[osm_id] = (iso, geom, geom.bounds)
return countries
def _clear_xml_element(element):
element.clear()
for ancestor in element.xpath('ancestor-or-self::*'):
while ancestor.getprevious() is not None:
del ancestor.getparent()[0]
def get_date_from_link(link):
return link[:-len(EXT)][-len(MIN_DATE):]
def get_tile_usage_dump_links(date_from=None, date_to=None):
date_from = date_from or MIN_DATE
date_to = date_to or MAX_DATE
links = set()
response = urllib.request.urlopen(LOGS_URL)
for action, element in lxml.etree.iterparse(response, tag='a', html=True):
link = element.attrib['href']
_clear_xml_element(element)
if not link.endswith(EXT):
continue
if not date_from <= get_date_from_link(link) <= date_to:
continue
links.add(urllib.parse.urljoin(LOGS_URL, link))
return sorted(links)
def get_tile_usage_dump(link):
dump_cache = os.path.join(DUMPS_CACHE_FOLDER,
link[-len('tiles-YYYY-MM-DD.txt.xz'):])
if os.path.exists(dump_cache):
return io.BytesIO(open(dump_cache, 'rb').read())
return urllib.request.urlopen(link)
def detect_country(b, x, y, z, part_zoom, tiled_countries, all_countries, stat):
twest, tsouth, teast, tnorth = b
tgeom = shapely.geometry.box(twest, tsouth, teast, tnorth)
result = None
if z < part_zoom:
countries = all_countries
else:
delta_zooms = z - part_zoom
delta = 2 ** delta_zooms
px = x // delta
py = y // delta
countries = tiled_countries['%s/%s/%s' % (part_zoom, px, py)]
for iso, boundary, outer_box in countries:
owest, osouth, oeast, onorth = outer_box
if not (twest <= oeast and teast >= owest and
tnorth >= osouth and tsouth <= onorth):
stat.filtered_bbox += 1
continue
if not boundary.intersects(tgeom):
stat.filtered_geom += 1
continue
if result is None:
result = {iso}
else:
result.add(iso)
stat.append_geom += 1
return result and '|'.join(sorted(result)) or '??'
def detect_country_with_cache(k, b, x, y, z,
part_zoom, tiled_countries, all_countries,
min_cache_zoom, cache, stat):
stat.in_all += 1
if k in cache:
stat.in_direct_cache += 1
return cache[k]
for pz in range(min_cache_zoom, z):
delta_zooms = z - pz
delta = 2 ** delta_zooms
px = x // delta
py = y // delta
ck = '%s/%s/%s' % (pz, px, py)
potential_cc = False
if ck in cache:
country = cache[ck]
potential_cc = True
else:
stat.in_no_cached += 1
country = detect_country(
b, px, py, pz, part_zoom, tiled_countries, all_countries, stat)
cache[k] = country
if pz < z and '|' not in country:
stat.child_zoom_less += 1
if potential_cc:
stat.in_child_cache += 1
return country
stat.in_no_cached += 1
stat.child_zoom_equal += 1
country = detect_country(
b, x, y, z, part_zoom, tiled_countries, all_countries, stat)
cache[k] = country
return country
def process_item(out, min_cache_zoom, cache, link,
part_zoom, tiled_countries, all_countries,
min_zoom, max_zoom, skip_empty_tiles):
stat = Stat()
date = get_date_from_link(link)
for line in lzma.LZMAFile(get_tile_usage_dump(link)):
path, count = line.decode().strip().split()
z, x, y = path.split('/')
x = int(x)
y = int(y)
z = int(z)
if min_zoom is not None and z < min_zoom:
continue
if max_zoom is not None and z > max_zoom:
continue
twest, tsouth, teast, tnorth = b = mercantile.bounds(x, y, z)
country = detect_country_with_cache(
path, b, x, y, z, part_zoom, tiled_countries, all_countries,
min_cache_zoom, cache, stat)
if skip_empty_tiles and country == '??':
continue
lat = tnorth + (tnorth - tsouth) / 2
lon = twest + (teast - twest) / 2
out.write(('%s,%s,%s,%s,%s,%s,%s,%s\n' % (
date, z, x, y, count, lat, lon, country)).encode())
stat.log_stats(date, cache)
def create_cache(part_zoom, countries, splited_countries, min_cache_zoom):
stat = Stat()
cache = {}
if min_cache_zoom:
stat.log_stats('cache (%s)' % min_cache_zoom, cache)
return min_cache_zoom, cache
min_cache_zoom = MIN_ZOOM
for z in range(MAX_ZOOM + 1):
for x in range(2 ** z):
for y in range(2 ** z):
k = '%s/%s/%s' % (z, x, y)
b = mercantile.bounds(x, y, z)
country = detect_country_with_cache(
k, b, x, y, z, part_zoom, countries, splited_countries,
z, cache, stat)
if '|' not in country:
min_cache_zoom = z
if min_cache_zoom > 0:
break
stat.log_stats('cache (%s)' % min_cache_zoom, cache)
return min_cache_zoom, cache
def add_no_country_items(grouped_countries, tiled_countries):
no_country_geoms = {}
for tile, iso_geoms in tiled_countries.items():
z, x, y = tile.split('/')
group_divider = max(2 ** int(z) // 32, 1)
group_key = (int(x) // group_divider, int(y) // group_divider)
for iso, geom, _ in iso_geoms:
if iso != '??':
continue
if group_key not in no_country_geoms:
no_country_geoms[group_key] = geom
else:
no_country_geoms[group_key] = no_country_geoms[group_key].union(geom)
return grouped_countries + tuple(('??', geom, geom.bounds)
for geom in no_country_geoms.values())
def slice_geoms_to_tiles(grouped_countries):
tiles = {}
z = SPLIT_ZOOM
for x in range(2 ** z):
for y in range(2 ** z):
start = datetime.datetime.utcnow()
twest, tsouth, teast, tnorth = mercantile.bounds(x, y, z)
bound = shapely.geometry.box(twest, tsouth, teast, tnorth)
no_country_geom = shapely.geometry.box(twest, tsouth, teast, tnorth)
parts = []
for iso, geom, _ in grouped_countries:
if not bound.intersects(geom):
continue
geom_part = bound.intersection(geom)
no_country_geom = no_country_geom.difference(geom_part)
parts.append((iso, geom_part))
if not no_country_geom.buffer(0).is_empty:
parts.append(('??', no_country_geom))
time_spent = (datetime.datetime.utcnow() - start).total_seconds()
Stat().log('polygons in %s/%s/%s: %s - %s%s', z, x, y,
len(parts), '|'.join(iso for iso, _ in parts),
'' if time_spent < 1 else (' (%s sec.)' % time_spent))
tiles['%s/%s/%s' % (z, x, y)] = tuple((iso, geom, geom.bounds)
for iso, geom in parts)
return tiles
def group_geoms(countries):
grouped_countries = {}
for iso, geom, bbox in countries.values():
geom = geom.buffer(0)
if iso not in grouped_countries:
grouped_countries[iso] = geom
else:
grouped_countries[iso] = grouped_countries[iso].union(geom)
overlapped = []
for iso_geom_1, iso_geom_2 in itertools.combinations(grouped_countries.items(), 2):
iso_1, geom_1 = iso_geom_1
iso_2, geom_2 = iso_geom_2
west_1, south_1, east_1, north_1 = geom_1.bounds
west_2, south_2, east_2, north_2 = geom_2.bounds
if not (west_1 <= east_2 and east_1 >= west_2 and
north_1 >= south_2 and south_1 <= north_2):
continue
if not geom_1.intersects(geom_2):
continue
intersection = geom_1.intersection(geom_2)
if not intersection.area or not intersection.length:
continue
area_per_perimeter = intersection.area / intersection.length
if (intersection.area < MIN_INTERSECTION_AREA or
area_per_perimeter < MIN_INTERSECTION_AREA_PER_PERIMETER):
continue
overlapped.append([iso_1, iso_2])
Stat().log('intersects %s and %s', iso_1, iso_2)
# only 2 geometries overlapping processed,
# 3 and more ignored because not found for enough big intersection
grouped_countries_update = {}
for iso_1, iso_2 in overlapped:
geom_1 = grouped_countries[iso_1]
geom_2 = grouped_countries[iso_2]
iso_1_2 = '+'.join(sorted([iso_1, iso_2]))
grouped_countries_update[iso_1_2] = geom_1.intersection(geom_2)
for iso, geom, geom_other in ((iso_1, geom_1, geom_2), (iso_2, geom_2, geom_1)):
geom = grouped_countries_update.get(iso, geom)
if geom is None:
continue
geom = geom.difference(geom_other)
if (geom.is_empty or not geom.area or not geom.length or
geom.area / geom.length < MIN_GEOMETRY_AREA_PER_PERIMETER):
geom = None
grouped_countries_update[iso] = geom
grouped_countries.update(grouped_countries_update)
return tuple((iso, geom, geom.bounds)
for iso, geom in grouped_countries.items()
if geom and not geom.is_empty)
def _cached_op(op, *args, title=None, cache=None, loader=pickle, block=True):
if cache and os.path.exists(cache):
with open(cache, 'rb' if block else 'r') as cache_file:
result = loader.load(cache_file)
else:
result = op(*args)
if cache:
with open(cache, 'wb' if block else 'w') as cache_file:
loader.dump(result, cache_file)
if title:
Stat().log('%s: %s', title, len(result))
return result
def process_all(out, date_from=None, date_to=None,
min_zoom=None, max_zoom=None,
rel=None, country=None, min_cache_zoom=None):
use_cache = not rel and not country
full_countries = _cached_op(
get_countries, rel, country,
title='total countries', cache=use_cache and GEOM_CACHE)
grouped_countries = _cached_op(
group_geoms, full_countries,
title='grouped countries', cache=use_cache and GROUPED_GEOM_CACHE)
tiled_countries = _cached_op(
slice_geoms_to_tiles, grouped_countries,
title='total parts', cache=use_cache and SLICED_TO_TILES_GEOM_CACHE)
grouped_countries = _cached_op(
add_no_country_items, grouped_countries, tiled_countries,
title='grouped with empty', cache=use_cache and GROUPED_GEOM_WITH_EMPTY_CACHE)
min_cache_zoom, cache = _cached_op(
create_cache, SPLIT_ZOOM, tiled_countries, grouped_countries, min_cache_zoom,
cache=use_cache and TILE_CACHE, loader=json, block=False)
for link in get_tile_usage_dump_links(date_from, date_to):
process_item(out, min_cache_zoom, cache, link,
SPLIT_ZOOM, tiled_countries, grouped_countries,
min_zoom, max_zoom, rel or country)
if use_cache:
with open(TILE_CACHE, 'w') as tile_cache_file:
json.dump([min_cache_zoom, cache], tile_cache_file)
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='Fetch and concat OSM tiles access logs.')
parser.add_argument('--date_from', default=None, help='filter from date (min 2014-01-01)')
parser.add_argument('--date_to', default=None, help='filter to date (max today)')
parser.add_argument('--min_zoom', type=int, default=None, help='filter from zoom (min 0)')
parser.add_argument('--max_zoom', type=int, default=None, help='filter to zoom (max 19)')
parser.add_argument('--rel', type=int, default=None, help='filter by OSM relation id geometry')
parser.add_argument('--country', default=None, help='filter by country ISO3166 alpha 2 code OSM geometry')
parser.add_argument('--min_cache_zoom', type=int, default=None)
stdout = sys.stdout if sys.version_info.major == 2 else sys.stdout.buffer
process_all(stdout, **parser.parse_args().__dict__)