-
Notifications
You must be signed in to change notification settings - Fork 3.4k
/
extraction_containers.cpp
1231 lines (1098 loc) · 49 KB
/
extraction_containers.cpp
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
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
#include "extractor/extraction_containers.hpp"
#include "extractor/extraction_segment.hpp"
#include "extractor/extraction_way.hpp"
#include "extractor/files.hpp"
#include "extractor/name_table.hpp"
#include "extractor/restriction.hpp"
#include "extractor/serialization.hpp"
#include "util/coordinate_calculation.hpp"
#include "util/exception.hpp"
#include "util/exception_utils.hpp"
#include "util/for_each_indexed.hpp"
#include "util/log.hpp"
#include "util/timing_util.hpp"
#include <boost/assert.hpp>
#include <boost/numeric/conversion/cast.hpp>
#include <tbb/parallel_sort.h>
#include <chrono>
#include <limits>
#include <mutex>
#include <sstream>
namespace
{
namespace oe = osrm::extractor;
struct CmpEdgeByOSMStartID
{
using value_type = oe::InternalExtractorEdge;
bool operator()(const value_type &lhs, const value_type &rhs) const
{
return lhs.result.osm_source_id < rhs.result.osm_source_id;
}
};
struct CmpEdgeByOSMTargetID
{
using value_type = oe::InternalExtractorEdge;
bool operator()(const value_type &lhs, const value_type &rhs) const
{
return lhs.result.osm_target_id < rhs.result.osm_target_id;
}
};
struct CmpEdgeByInternalSourceTargetAndName
{
using value_type = oe::InternalExtractorEdge;
bool operator()(const value_type &lhs, const value_type &rhs) const
{
if (lhs.result.source != rhs.result.source)
return lhs.result.source < rhs.result.source;
if (lhs.result.source == SPECIAL_NODEID)
return false;
if (lhs.result.target != rhs.result.target)
return lhs.result.target < rhs.result.target;
if (lhs.result.target == SPECIAL_NODEID)
return false;
auto const lhs_name_id = edge_annotation_data[lhs.result.annotation_data].name_id;
auto const rhs_name_id = edge_annotation_data[rhs.result.annotation_data].name_id;
if (lhs_name_id == rhs_name_id)
return false;
if (lhs_name_id == EMPTY_NAMEID)
return false;
if (rhs_name_id == EMPTY_NAMEID)
return true;
BOOST_ASSERT(!name_offsets.empty() && name_offsets.back() == name_data.size());
const oe::ExtractionContainers::NameCharData::const_iterator data = name_data.begin();
return std::lexicographical_compare(data + name_offsets[lhs_name_id],
data + name_offsets[lhs_name_id + 1],
data + name_offsets[rhs_name_id],
data + name_offsets[rhs_name_id + 1]);
}
const oe::ExtractionContainers::AnnotationDataVector &edge_annotation_data;
const oe::ExtractionContainers::NameCharData &name_data;
const oe::ExtractionContainers::NameOffsets &name_offsets;
};
template <typename Iter>
inline NodeID mapExternalToInternalNodeID(Iter first, Iter last, const OSMNodeID value)
{
const auto it = std::lower_bound(first, last, value);
return (it == last || value < *it) ? SPECIAL_NODEID
: static_cast<NodeID>(std::distance(first, it));
}
/**
* Here's what these properties represent on the node-based-graph
* way "ABCD" way "AB"
* -----------------------------------------------------------------
* ⬇ A first_segment_source_id
* ⬇ |
* ⬇︎ B first_segment_target_id A first_segment_source_id
* ⬇︎ | ⬇ | last_segment_source_id
* ⬇︎ | ⬇ |
* ⬇︎ | B first_segment_target_id
* ⬇︎ C last_segment_source_id last_segment_target_id
* ⬇︎ |
* ⬇︎ D last_segment_target_id
*
* Finds the point where two ways connect at the end, and returns the 3
* node-based nodes that describe the turn (the node just before, the
* node at the turn, and the next node after the turn)
**/
std::tuple<OSMNodeID, OSMNodeID, OSMNodeID> find_turn_nodes(const oe::NodesOfWay &from,
const oe::NodesOfWay &via,
const OSMNodeID &intersection_node)
{
// connection node needed to choose orientation if from and via are the same way. E.g. u-turns
if (intersection_node == SPECIAL_OSM_NODEID ||
intersection_node == from.first_segment_source_id())
{
if (from.first_segment_source_id() == via.first_segment_source_id())
{
return std::make_tuple(from.first_segment_target_id(),
via.first_segment_source_id(),
via.first_segment_target_id());
}
if (from.first_segment_source_id() == via.last_segment_target_id())
{
return std::make_tuple(from.first_segment_target_id(),
via.last_segment_target_id(),
via.last_segment_source_id());
}
}
if (intersection_node == SPECIAL_OSM_NODEID ||
intersection_node == from.last_segment_target_id())
{
if (from.last_segment_target_id() == via.first_segment_source_id())
{
return std::make_tuple(from.last_segment_source_id(),
via.first_segment_source_id(),
via.first_segment_target_id());
}
if (from.last_segment_target_id() == via.last_segment_target_id())
{
return std::make_tuple(from.last_segment_source_id(),
via.last_segment_target_id(),
via.last_segment_source_id());
}
}
return std::make_tuple(SPECIAL_OSM_NODEID, SPECIAL_OSM_NODEID, SPECIAL_OSM_NODEID);
}
// Via-node paths describe a relation between the two segments closest
// to the shared via-node on the from and to ways.
// from: [a, b, c, d, e]
// to: [f, g, h, i, j]
//
// The via node establishes the orientation of the from/to intersection when choosing the
// segments.
// via | node path
// a=f | b,a,g
// a=j | b,a,i
// e=f | d,e,g
// e=j | d,e,i
oe::ViaNodePath find_via_node_path(const std::string &turn_relation_type,
const oe::NodesOfWay &from_segment,
const oe::NodesOfWay &to_segment,
const OSMNodeID via_node,
const std::function<NodeID(OSMNodeID)> &to_internal_node)
{
OSMNodeID from, via, to;
std::tie(from, via, to) = find_turn_nodes(from_segment, to_segment, via_node);
if (via == SPECIAL_OSM_NODEID)
{
// unconnected
osrm::util::Log(logDEBUG) << turn_relation_type
<< " references unconnected way: " << from_segment.way_id;
return oe::ViaNodePath{SPECIAL_NODEID, SPECIAL_NODEID, SPECIAL_NODEID};
}
return oe::ViaNodePath{to_internal_node(from), to_internal_node(via), to_internal_node(to)};
}
// Via way paths are comprised of:
// 1. The segment in the from way that intersects with the via ways
// 2. All segments that make up the via path
// 3. The segment in the to way that intersects with the via path.
//
// from: [a, b, c, d, e]
// via: [[f, g, h, i, j], [k, l], [m, n, o]]
// to: [p, q, r, s]
//
// First establish the orientation of the from/via intersection by finding which end
// nodes both ways share. From this we can select the from segment.
//
// intersect | from segment | next_connection
// a=f | b,a | f
// a=j | b,a | j
// e=f | e,d | f
// e=j | e,d | j
//
// Use the next connection to inform the orientation of the first via
// way and the intersection between first and second via ways.
//
// next_connection | intersect | via result | next_next_connection
// f | j=k | [f,g,h,i,j] | k
// f | j=l | [f,g,h,i,j] | l
// j | f=k | [j,i,h,g,f] | k
// j | f=l | [j,i,h,g,f] | l
//
// This is continued for the remaining via ways, appending to the via result
//
// The final via/to intersection also uses the next_connection information in a similar fashion.
//
// next_connection | intersect | to_segment
// m | o=p | p,q
// m | o=s | s,r
// o | m=p | p,q
// o | m=s | s,r
//
// The final result is a list of nodes that represent a valid from->via->to path through the
// ways.
//
// E.g. if intersection nodes are a=j, f=l, k=o, m=s
// the result will be {e [d,c,b,a,i,h,g,f,k,n,m] r}
oe::ViaWayPath find_via_way_path(const std::string &turn_relation_type,
const oe::NodesOfWay &from_way,
const std::vector<oe::NodesOfWay> &via_ways,
const oe::NodesOfWay &to_way,
const std::function<NodeID(OSMNodeID)> &to_internal_node)
{
BOOST_ASSERT(!via_ways.empty());
oe::ViaWayPath way_path;
// Find the orientation of the connected ways starting with the from-via intersection.
OSMNodeID from, via;
std::tie(from, via, std::ignore) =
find_turn_nodes(from_way, via_ways.front(), SPECIAL_OSM_NODEID);
if (via == SPECIAL_OSM_NODEID)
{
osrm::util::Log(logDEBUG) << turn_relation_type
<< " has unconnected from and via ways: " << from_way.way_id
<< ", " << via_ways.front().way_id;
return oe::ViaWayPath{SPECIAL_NODEID, {}, SPECIAL_NODEID};
}
way_path.from = to_internal_node(from);
way_path.via.push_back(to_internal_node(via));
// Use the connection node from the previous intersection to inform our conversion of
// via ways into internal nodes.
OSMNodeID next_connection = via;
for (const auto &via_way : via_ways)
{
if (next_connection == via_way.first_segment_source_id())
{
std::transform(std::next(via_way.node_ids.begin()),
via_way.node_ids.end(),
std::back_inserter(way_path.via),
to_internal_node);
next_connection = via_way.last_segment_target_id();
}
else if (next_connection == via_way.last_segment_target_id())
{
std::transform(std::next(via_way.node_ids.rbegin()),
via_way.node_ids.rend(),
std::back_inserter(way_path.via),
to_internal_node);
next_connection = via_way.first_segment_source_id();
}
else
{
osrm::util::Log(logDEBUG)
<< turn_relation_type << " has unconnected via way: " << via_way.way_id
<< " to node " << next_connection;
return oe::ViaWayPath{SPECIAL_NODEID, {}, SPECIAL_NODEID};
}
}
// Add the final to node after the via-to intersection.
if (next_connection == to_way.first_segment_source_id())
{
way_path.to = to_internal_node(to_way.first_segment_target_id());
}
else if (next_connection == to_way.last_segment_target_id())
{
way_path.to = to_internal_node(to_way.last_segment_source_id());
}
else
{
osrm::util::Log(logDEBUG) << turn_relation_type
<< " has unconnected via and to ways: " << via_ways.back().way_id
<< ", " << to_way.way_id;
return oe::ViaWayPath{SPECIAL_NODEID, {}, SPECIAL_NODEID};
}
return way_path;
}
// Check if we were able to resolve all the involved OSM elements before translating to an
// internal via-way turn path
oe::ViaWayPath
get_via_way_path_from_OSM_ids(const std::string &turn_relation_type,
const std::unordered_map<OSMWayID, oe::NodesOfWay> &referenced_ways,
const OSMWayID from_id,
const OSMWayID to_id,
const std::vector<OSMWayID> &via_ids,
const std::function<NodeID(OSMNodeID)> &to_internal_node)
{
auto const from_way_itr = referenced_ways.find(from_id);
if (from_way_itr->second.way_id != from_id)
{
osrm::util::Log(logDEBUG) << turn_relation_type
<< " references invalid from way: " << from_id;
return oe::ViaWayPath{SPECIAL_NODEID, {}, SPECIAL_NODEID};
}
std::vector<oe::NodesOfWay> via_ways;
for (const auto &via_id : via_ids)
{
auto const via_segment_itr = referenced_ways.find(via_id);
if (via_segment_itr->second.way_id != via_id)
{
osrm::util::Log(logDEBUG)
<< turn_relation_type << " references invalid via way: " << via_id;
return oe::ViaWayPath{SPECIAL_NODEID, {}, SPECIAL_NODEID};
}
via_ways.push_back(via_segment_itr->second);
}
auto const to_way_itr = referenced_ways.find(to_id);
if (to_way_itr->second.way_id != to_id)
{
osrm::util::Log(logDEBUG) << turn_relation_type << " references invalid to way: " << to_id;
return oe::ViaWayPath{SPECIAL_NODEID, {}, SPECIAL_NODEID};
}
return find_via_way_path(
turn_relation_type, from_way_itr->second, via_ways, to_way_itr->second, to_internal_node);
}
// Check if we were able to resolve all the involved OSM elements before translating to an
// internal via-node turn path
oe::ViaNodePath
get_via_node_path_from_OSM_ids(const std::string &turn_relation_type,
const std::unordered_map<OSMWayID, oe::NodesOfWay> &referenced_ways,
const OSMWayID from_id,
const OSMWayID to_id,
const OSMNodeID via_node,
const std::function<NodeID(OSMNodeID)> &to_internal_node)
{
auto const from_segment_itr = referenced_ways.find(from_id);
if (from_segment_itr->second.way_id != from_id)
{
osrm::util::Log(logDEBUG) << turn_relation_type << " references invalid way: " << from_id;
return oe::ViaNodePath{SPECIAL_NODEID, SPECIAL_NODEID, SPECIAL_NODEID};
}
auto const to_segment_itr = referenced_ways.find(to_id);
if (to_segment_itr->second.way_id != to_id)
{
osrm::util::Log(logDEBUG) << turn_relation_type << " references invalid way: " << to_id;
return oe::ViaNodePath{SPECIAL_NODEID, SPECIAL_NODEID, SPECIAL_NODEID};
}
return find_via_node_path(turn_relation_type,
from_segment_itr->second,
to_segment_itr->second,
via_node,
to_internal_node);
}
} // namespace
namespace osrm
{
namespace extractor
{
ExtractionContainers::ExtractionContainers()
{
// Insert four empty strings offsets for name, ref, destination, pronunciation, and exits
name_offsets.push_back(0);
name_offsets.push_back(0);
name_offsets.push_back(0);
name_offsets.push_back(0);
name_offsets.push_back(0);
// Insert the total length sentinel (corresponds to the next name string offset)
name_offsets.push_back(0);
// Sentinel for offset into used_nodes
way_node_id_offsets.push_back(0);
}
/**
* Processes the collected data and serializes it.
* At this point nodes are still referenced by their OSM id.
*
* - Identify nodes of ways used in restrictions and maneuver overrides
* - Filter nodes list to nodes that are referenced by ways
* - Prepare edges and compute routing properties
* - Prepare and validate restrictions and maneuver overrides
*
*/
void ExtractionContainers::PrepareData(ScriptingEnvironment &scripting_environment,
const std::string &osrm_path,
const std::string &name_file_name)
{
storage::tar::FileWriter writer(osrm_path, storage::tar::FileWriter::GenerateFingerprint);
const auto restriction_ways = IdentifyRestrictionWays();
const auto maneuver_override_ways = IdentifyManeuverOverrideWays();
PrepareNodes();
WriteNodes(writer);
PrepareEdges(scripting_environment);
all_nodes_list.clear(); // free all_nodes_list before allocation of normal_edges
all_nodes_list.shrink_to_fit();
WriteEdges(writer);
WriteMetadata(writer);
PrepareManeuverOverrides(maneuver_override_ways);
PrepareRestrictions(restriction_ways);
WriteCharData(name_file_name);
}
void ExtractionContainers::WriteCharData(const std::string &file_name)
{
util::UnbufferedLog log;
log << "writing street name index ... ";
TIMER_START(write_index);
files::writeNames(file_name,
NameTable{NameTable::IndexedData(
name_offsets.begin(), name_offsets.end(), name_char_data.begin())});
TIMER_STOP(write_index);
log << "ok, after " << TIMER_SEC(write_index) << "s";
}
void ExtractionContainers::PrepareNodes()
{
{
util::UnbufferedLog log;
log << "Sorting used nodes ... " << std::flush;
TIMER_START(sorting_used_nodes);
tbb::parallel_sort(used_node_id_list.begin(), used_node_id_list.end());
TIMER_STOP(sorting_used_nodes);
log << "ok, after " << TIMER_SEC(sorting_used_nodes) << "s";
}
{
util::UnbufferedLog log;
log << "Erasing duplicate nodes ... " << std::flush;
TIMER_START(erasing_dups);
auto new_end = std::unique(used_node_id_list.begin(), used_node_id_list.end());
used_node_id_list.resize(new_end - used_node_id_list.begin());
TIMER_STOP(erasing_dups);
log << "ok, after " << TIMER_SEC(erasing_dups) << "s";
}
{
util::UnbufferedLog log;
log << "Sorting all nodes ... " << std::flush;
TIMER_START(sorting_nodes);
tbb::parallel_sort(
all_nodes_list.begin(), all_nodes_list.end(), [](const auto &left, const auto &right) {
return left.node_id < right.node_id;
});
TIMER_STOP(sorting_nodes);
log << "ok, after " << TIMER_SEC(sorting_nodes) << "s";
}
{
util::UnbufferedLog log;
log << "Building node id map ... " << std::flush;
TIMER_START(id_map);
auto node_iter = all_nodes_list.begin();
auto ref_iter = used_node_id_list.begin();
auto used_nodes_iter = used_node_id_list.begin();
const auto all_nodes_list_end = all_nodes_list.end();
const auto used_node_id_list_end = used_node_id_list.end();
// compute the intersection of nodes that were referenced and nodes we actually have
while (node_iter != all_nodes_list_end && ref_iter != used_node_id_list_end)
{
if (node_iter->node_id < *ref_iter)
{
node_iter++;
continue;
}
if (node_iter->node_id > *ref_iter)
{
ref_iter++;
continue;
}
BOOST_ASSERT(node_iter->node_id == *ref_iter);
*used_nodes_iter = *ref_iter;
used_nodes_iter++;
node_iter++;
ref_iter++;
}
// Remove unused nodes and check maximal internal node id
used_node_id_list.resize(std::distance(used_node_id_list.begin(), used_nodes_iter));
if (used_node_id_list.size() > std::numeric_limits<NodeID>::max())
{
throw util::exception("There are too many nodes remaining after filtering, OSRM only "
"supports 2^32 unique nodes, but there were " +
std::to_string(used_node_id_list.size()) + SOURCE_REF);
}
max_internal_node_id = boost::numeric_cast<std::uint64_t>(used_node_id_list.size());
TIMER_STOP(id_map);
log << "ok, after " << TIMER_SEC(id_map) << "s";
}
}
void ExtractionContainers::PrepareEdges(ScriptingEnvironment &scripting_environment)
{
// Sort edges by start.
{
util::UnbufferedLog log;
log << "Sorting edges by start ... " << std::flush;
TIMER_START(sort_edges_by_start);
tbb::parallel_sort(all_edges_list.begin(), all_edges_list.end(), CmpEdgeByOSMStartID());
TIMER_STOP(sort_edges_by_start);
log << "ok, after " << TIMER_SEC(sort_edges_by_start) << "s";
}
{
util::UnbufferedLog log;
log << "Setting start coords ... " << std::flush;
TIMER_START(set_start_coords);
// Traverse list of edges and nodes in parallel and set start coord
auto node_iterator = all_nodes_list.begin();
auto edge_iterator = all_edges_list.begin();
const auto all_edges_list_end = all_edges_list.end();
const auto all_nodes_list_end = all_nodes_list.end();
while (edge_iterator != all_edges_list_end && node_iterator != all_nodes_list_end)
{
if (edge_iterator->result.osm_source_id < node_iterator->node_id)
{
util::Log(logDEBUG)
<< "Found invalid node reference " << edge_iterator->result.source;
edge_iterator->result.source = SPECIAL_NODEID;
++edge_iterator;
continue;
}
if (edge_iterator->result.osm_source_id > node_iterator->node_id)
{
node_iterator++;
continue;
}
// remove loops
if (edge_iterator->result.osm_source_id == edge_iterator->result.osm_target_id)
{
edge_iterator->result.source = SPECIAL_NODEID;
edge_iterator->result.target = SPECIAL_NODEID;
++edge_iterator;
continue;
}
BOOST_ASSERT(edge_iterator->result.osm_source_id == node_iterator->node_id);
// assign new node id
const auto node_id = mapExternalToInternalNodeID(
used_node_id_list.begin(), used_node_id_list.end(), node_iterator->node_id);
BOOST_ASSERT(node_id != SPECIAL_NODEID);
edge_iterator->result.source = node_id;
edge_iterator->source_coordinate.lat = node_iterator->lat;
edge_iterator->source_coordinate.lon = node_iterator->lon;
++edge_iterator;
}
// Remove all remaining edges. They are invalid because there are no corresponding nodes for
// them. This happens when using osmosis with bbox or polygon to extract smaller areas.
auto markSourcesInvalid = [](InternalExtractorEdge &edge) {
util::Log(logDEBUG) << "Found invalid node reference " << edge.result.source;
edge.result.source = SPECIAL_NODEID;
edge.result.osm_source_id = SPECIAL_OSM_NODEID;
};
std::for_each(edge_iterator, all_edges_list_end, markSourcesInvalid);
TIMER_STOP(set_start_coords);
log << "ok, after " << TIMER_SEC(set_start_coords) << "s";
}
{
// Sort Edges by target
util::UnbufferedLog log;
log << "Sorting edges by target ... " << std::flush;
TIMER_START(sort_edges_by_target);
tbb::parallel_sort(all_edges_list.begin(), all_edges_list.end(), CmpEdgeByOSMTargetID());
TIMER_STOP(sort_edges_by_target);
log << "ok, after " << TIMER_SEC(sort_edges_by_target) << "s";
}
{
// Compute edge weights
util::UnbufferedLog log;
log << "Computing edge weights ... " << std::flush;
TIMER_START(compute_weights);
auto node_iterator = all_nodes_list.begin();
auto edge_iterator = all_edges_list.begin();
const auto all_edges_list_end_ = all_edges_list.end();
const auto all_nodes_list_end_ = all_nodes_list.end();
const auto weight_multiplier =
scripting_environment.GetProfileProperties().GetWeightMultiplier();
while (edge_iterator != all_edges_list_end_ && node_iterator != all_nodes_list_end_)
{
// skip all invalid edges
if (edge_iterator->result.source == SPECIAL_NODEID)
{
++edge_iterator;
continue;
}
if (edge_iterator->result.osm_target_id < node_iterator->node_id)
{
util::Log(logDEBUG) << "Found invalid node reference "
<< static_cast<uint64_t>(edge_iterator->result.osm_target_id);
edge_iterator->result.target = SPECIAL_NODEID;
++edge_iterator;
continue;
}
if (edge_iterator->result.osm_target_id > node_iterator->node_id)
{
++node_iterator;
continue;
}
BOOST_ASSERT(edge_iterator->result.osm_target_id == node_iterator->node_id);
BOOST_ASSERT(edge_iterator->source_coordinate.lat !=
util::FixedLatitude{std::numeric_limits<std::int32_t>::min()});
BOOST_ASSERT(edge_iterator->source_coordinate.lon !=
util::FixedLongitude{std::numeric_limits<std::int32_t>::min()});
util::Coordinate source_coord(edge_iterator->source_coordinate);
util::Coordinate target_coord{node_iterator->lon, node_iterator->lat};
// flip source and target coordinates if segment is in backward direction only
if (!edge_iterator->result.flags.forward && edge_iterator->result.flags.backward)
std::swap(source_coord, target_coord);
const auto distance =
util::coordinate_calculation::greatCircleDistance(source_coord, target_coord);
const auto weight = edge_iterator->weight_data(distance);
const auto duration = edge_iterator->duration_data(distance);
const auto accurate_distance =
util::coordinate_calculation::greatCircleDistance(source_coord, target_coord);
ExtractionSegment segment(source_coord, target_coord, distance, weight, duration);
scripting_environment.ProcessSegment(segment);
auto &edge = edge_iterator->result;
edge.weight = std::max<EdgeWeight>(1, std::round(segment.weight * weight_multiplier));
edge.duration = std::max<EdgeWeight>(1, std::round(segment.duration * 10.));
edge.distance = accurate_distance;
// assign new node id
const auto node_id = mapExternalToInternalNodeID(
used_node_id_list.begin(), used_node_id_list.end(), node_iterator->node_id);
BOOST_ASSERT(node_id != SPECIAL_NODEID);
edge.target = node_id;
// orient edges consistently: source id < target id
// important for multi-edge removal
if (edge.source > edge.target)
{
std::swap(edge.source, edge.target);
// std::swap does not work with bit-fields
bool temp = edge.flags.forward;
edge.flags.forward = edge.flags.backward;
edge.flags.backward = temp;
}
++edge_iterator;
}
// Remove all remaining edges. They are invalid because there are no corresponding nodes for
// them. This happens when using osmosis with bbox or polygon to extract smaller areas.
auto markTargetsInvalid = [](InternalExtractorEdge &edge) {
util::Log(logDEBUG) << "Found invalid node reference " << edge.result.target;
edge.result.target = SPECIAL_NODEID;
};
std::for_each(edge_iterator, all_edges_list_end_, markTargetsInvalid);
TIMER_STOP(compute_weights);
log << "ok, after " << TIMER_SEC(compute_weights) << "s";
}
// Sort edges by start.
{
util::UnbufferedLog log;
log << "Sorting edges by renumbered start ... ";
TIMER_START(sort_edges_by_renumbered_start);
tbb::parallel_sort(all_edges_list.begin(),
all_edges_list.end(),
CmpEdgeByInternalSourceTargetAndName{
all_edges_annotation_data_list, name_char_data, name_offsets});
TIMER_STOP(sort_edges_by_renumbered_start);
log << "ok, after " << TIMER_SEC(sort_edges_by_renumbered_start) << "s";
}
BOOST_ASSERT(all_edges_list.size() > 0);
for (std::size_t i = 0; i < all_edges_list.size();)
{
// only invalid edges left
if (all_edges_list[i].result.source == SPECIAL_NODEID)
{
break;
}
// skip invalid edges
if (all_edges_list[i].result.target == SPECIAL_NODEID)
{
++i;
continue;
}
std::size_t start_idx = i;
NodeID source = all_edges_list[i].result.source;
NodeID target = all_edges_list[i].result.target;
auto min_forward = std::make_pair(std::numeric_limits<EdgeWeight>::max(),
std::numeric_limits<EdgeWeight>::max());
auto min_backward = std::make_pair(std::numeric_limits<EdgeWeight>::max(),
std::numeric_limits<EdgeWeight>::max());
std::size_t min_forward_idx = std::numeric_limits<std::size_t>::max();
std::size_t min_backward_idx = std::numeric_limits<std::size_t>::max();
// find minimal edge in both directions
while (i < all_edges_list.size() && all_edges_list[i].result.source == source &&
all_edges_list[i].result.target == target)
{
const auto &result = all_edges_list[i].result;
const auto value = std::make_pair(result.weight, result.duration);
if (result.flags.forward && value < min_forward)
{
min_forward_idx = i;
min_forward = value;
}
if (result.flags.backward && value < min_backward)
{
min_backward_idx = i;
min_backward = value;
}
// this also increments the outer loop counter!
i++;
}
BOOST_ASSERT(min_forward_idx == std::numeric_limits<std::size_t>::max() ||
min_forward_idx < i);
BOOST_ASSERT(min_backward_idx == std::numeric_limits<std::size_t>::max() ||
min_backward_idx < i);
BOOST_ASSERT(min_backward_idx != std::numeric_limits<std::size_t>::max() ||
min_forward_idx != std::numeric_limits<std::size_t>::max());
if (min_backward_idx == min_forward_idx)
{
all_edges_list[min_forward_idx].result.flags.is_split = false;
all_edges_list[min_forward_idx].result.flags.forward = true;
all_edges_list[min_forward_idx].result.flags.backward = true;
}
else
{
bool has_forward = min_forward_idx != std::numeric_limits<std::size_t>::max();
bool has_backward = min_backward_idx != std::numeric_limits<std::size_t>::max();
if (has_forward)
{
all_edges_list[min_forward_idx].result.flags.forward = true;
all_edges_list[min_forward_idx].result.flags.backward = false;
all_edges_list[min_forward_idx].result.flags.is_split = has_backward;
}
if (has_backward)
{
std::swap(all_edges_list[min_backward_idx].result.source,
all_edges_list[min_backward_idx].result.target);
all_edges_list[min_backward_idx].result.flags.forward = true;
all_edges_list[min_backward_idx].result.flags.backward = false;
all_edges_list[min_backward_idx].result.flags.is_split = has_forward;
}
}
// invalidate all unused edges
for (std::size_t j = start_idx; j < i; j++)
{
if (j == min_forward_idx || j == min_backward_idx)
{
continue;
}
all_edges_list[j].result.source = SPECIAL_NODEID;
all_edges_list[j].result.target = SPECIAL_NODEID;
}
}
}
void ExtractionContainers::WriteEdges(storage::tar::FileWriter &writer) const
{
std::vector<NodeBasedEdge> normal_edges;
normal_edges.reserve(all_edges_list.size());
{
util::UnbufferedLog log;
log << "Writing used edges ... " << std::flush;
TIMER_START(write_edges);
// Traverse list of edges and nodes in parallel and set target coord
for (const auto &edge : all_edges_list)
{
if (edge.result.source == SPECIAL_NODEID || edge.result.target == SPECIAL_NODEID)
{
continue;
}
// IMPORTANT: here, we're using slicing to only write the data from the base
// class of NodeBasedEdgeWithOSM
normal_edges.push_back(edge.result);
}
if (normal_edges.size() > std::numeric_limits<uint32_t>::max())
{
throw util::exception("There are too many edges, OSRM only supports 2^32" + SOURCE_REF);
}
storage::serialization::write(writer, "/extractor/edges", normal_edges);
TIMER_STOP(write_edges);
log << "ok, after " << TIMER_SEC(write_edges) << "s";
log << " -- Processed " << normal_edges.size() << " edges";
}
}
void ExtractionContainers::WriteMetadata(storage::tar::FileWriter &writer) const
{
util::UnbufferedLog log;
log << "Writing way meta-data ... " << std::flush;
TIMER_START(write_meta_data);
storage::serialization::write(writer, "/extractor/annotations", all_edges_annotation_data_list);
TIMER_STOP(write_meta_data);
log << "ok, after " << TIMER_SEC(write_meta_data) << "s";
log << " -- Metadata contains << " << all_edges_annotation_data_list.size() << " entries.";
}
void ExtractionContainers::WriteNodes(storage::tar::FileWriter &writer) const
{
{
util::UnbufferedLog log;
log << "Confirming/Writing used nodes ... ";
TIMER_START(write_nodes);
// identify all used nodes by a merging step of two sorted lists
auto node_iterator = all_nodes_list.begin();
auto node_id_iterator = used_node_id_list.begin();
const auto all_nodes_list_end = all_nodes_list.end();
const std::function<QueryNode()> encode_function = [&]() -> QueryNode {
BOOST_ASSERT(node_id_iterator != used_node_id_list.end());
BOOST_ASSERT(node_iterator != all_nodes_list_end);
BOOST_ASSERT(*node_id_iterator >= node_iterator->node_id);
while (*node_id_iterator > node_iterator->node_id &&
node_iterator != all_nodes_list_end)
{
++node_iterator;
}
if (node_iterator == all_nodes_list_end || *node_id_iterator < node_iterator->node_id)
{
throw util::exception(
"Invalid OSM data: Referenced non-existing node with ID " +
std::to_string(static_cast<std::uint64_t>(*node_id_iterator)));
}
BOOST_ASSERT(*node_id_iterator == node_iterator->node_id);
++node_id_iterator;
return *node_iterator++;
};
writer.WriteElementCount64("/extractor/nodes", used_node_id_list.size());
writer.WriteStreaming<QueryNode>(
"/extractor/nodes",
boost::make_function_input_iterator(encode_function, boost::infinite()),
used_node_id_list.size());
TIMER_STOP(write_nodes);
log << "ok, after " << TIMER_SEC(write_nodes) << "s";
}
{
util::UnbufferedLog log;
log << "Writing barrier nodes ... ";
TIMER_START(write_nodes);
std::vector<NodeID> internal_barrier_nodes;
for (const auto osm_id : barrier_nodes)
{
const auto node_id = mapExternalToInternalNodeID(
used_node_id_list.begin(), used_node_id_list.end(), osm_id);
if (node_id != SPECIAL_NODEID)
{
internal_barrier_nodes.push_back(node_id);
}
}
storage::serialization::write(writer, "/extractor/barriers", internal_barrier_nodes);
log << "ok, after " << TIMER_SEC(write_nodes) << "s";
}
{
util::UnbufferedLog log;
log << "Writing traffic light nodes ... ";
TIMER_START(write_nodes);
std::vector<NodeID> internal_traffic_signals;
for (const auto osm_id : traffic_signals)
{
const auto node_id = mapExternalToInternalNodeID(
used_node_id_list.begin(), used_node_id_list.end(), osm_id);
if (node_id != SPECIAL_NODEID)
{
internal_traffic_signals.push_back(node_id);
}
}
storage::serialization::write(
writer, "/extractor/traffic_lights", internal_traffic_signals);
log << "ok, after " << TIMER_SEC(write_nodes) << "s";
}
util::Log() << "Processed " << max_internal_node_id << " nodes";
}
ExtractionContainers::ReferencedWays ExtractionContainers::IdentifyManeuverOverrideWays()
{
ReferencedWays maneuver_override_ways;
// prepare for extracting source/destination nodes for all maneuvers
util::UnbufferedLog log;
log << "Collecting way information on " << external_maneuver_overrides_list.size()
<< " maneuver overrides...";
TIMER_START(identify_maneuver_override_ways);
const auto mark_ids = [&](auto const &external_maneuver_override) {
NodesOfWay dummy_segment{MAX_OSM_WAYID, {MAX_OSM_NODEID, MAX_OSM_NODEID}};
const auto &turn_path = external_maneuver_override.turn_path;
maneuver_override_ways[turn_path.From()] = dummy_segment;
maneuver_override_ways[turn_path.To()] = dummy_segment;
if (external_maneuver_override.turn_path.Type() == TurnPathType::VIA_WAY_TURN_PATH)
{
const auto &way = turn_path.AsViaWayPath();
for (const auto &via : way.via)
{
maneuver_override_ways[via] = dummy_segment;
}
}
};
// First, make an empty hashtable keyed by the ways referenced
// by the maneuver overrides
std::for_each(
external_maneuver_overrides_list.begin(), external_maneuver_overrides_list.end(), mark_ids);
const auto set_ids = [&](size_t way_list_idx, auto const &way_id) {
auto itr = maneuver_override_ways.find(way_id);
if (itr != maneuver_override_ways.end())
{
auto node_start_itr = used_node_id_list.begin() + way_node_id_offsets[way_list_idx];
auto node_end_itr = used_node_id_list.begin() + way_node_id_offsets[way_list_idx + 1];
itr->second = NodesOfWay(way_id, std::vector<OSMNodeID>(node_start_itr, node_end_itr));
}
};
// Then, populate the values in that hashtable for only the ways
// referenced
util::for_each_indexed(ways_list, set_ids);
TIMER_STOP(identify_maneuver_override_ways);
log << "ok, after " << TIMER_SEC(identify_maneuver_override_ways) << "s";
return maneuver_override_ways;
}
void ExtractionContainers::PrepareManeuverOverrides(const ReferencedWays &maneuver_override_ways)
{
auto const osm_node_to_internal_nbn = [&](auto const osm_node) {
auto internal = mapExternalToInternalNodeID(
used_node_id_list.begin(), used_node_id_list.end(), osm_node);
if (internal == SPECIAL_NODEID)
{
util::Log(logDEBUG) << "Maneuver override references invalid node: " << osm_node;
}
return internal;
};
const auto strings_to_turn_type_and_direction = [](const std::string &turn_string,
const std::string &direction_string) {
auto result = std::make_pair(guidance::TurnType::MaxTurnType,