forked from osm2pgsql-dev/osm2pgsql
-
Notifications
You must be signed in to change notification settings - Fork 0
/
middle-ram.cpp
224 lines (178 loc) · 5.56 KB
/
middle-ram.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
/* Implements the mid-layer processing for osm2pgsql
* using several arrays in RAM. This is fastest if you
* have sufficient RAM+Swap.
*
* This layer stores data read in from the planet.osm file
* and is then read by the backend processing code to
* emit the final geometry-enabled output formats
*/
#include <stdexcept>
#include <cassert>
#include <cstdio>
#include "id-tracker.hpp"
#include "middle-ram.hpp"
#include "node-ram-cache.hpp"
#include "options.hpp"
/* Object storage now uses 2 levels of storage arrays.
*
* - Low level storage of 2^16 (~65k) objects in an indexed array
* These are allocated dynamically when we need to first store data with
* an ID in this block
*
* - Fixed array of 2^(32 - 16) = 65k pointers to the dynamically allocated arrays.
*
* This allows memory usage to be efficient and scale dynamically without needing to
* hard code maximum IDs. We now support an ID range of -2^31 to +2^31.
* The negative IDs often occur in non-uploaded JOSM data or other data import scripts.
*
*/
void middle_ram_t::nodes_set(osmid_t id, double lat, double lon, const taglist_t &tags) {
cache->set(id, lat, lon, tags);
}
void middle_ram_t::ways_set(osmid_t id, const idlist_t &nds, const taglist_t &tags)
{
ways.set(id, new ramWay(tags, nds));
}
void middle_ram_t::relations_set(osmid_t id, const memberlist_t &members, const taglist_t &tags)
{
rels.set(id, new ramRel(tags, members));
}
size_t middle_ram_t::nodes_get_list(nodelist_t &out, const idlist_t nds) const
{
for (idlist_t::const_iterator it = nds.begin(); it != nds.end(); ++it) {
osmNode n;
if (!cache->get(&n, *it))
out.push_back(n);
}
return int(out.size());
}
void middle_ram_t::iterate_relations(pending_processor& pf)
{
//TODO: just dont do anything
//let the outputs enqueue everything they have the non slim middle
//has nothing of its own to enqueue as it doesnt have pending anything
pf.enqueue_relations(id_tracker::max());
//let the threads process the relations
pf.process_relations();
}
size_t middle_ram_t::pending_count() const
{
return 0;
}
void middle_ram_t::iterate_ways(middle_t::pending_processor& pf)
{
//let the outputs enqueue everything they have the non slim middle
//has nothing of its own to enqueue as it doesnt have pending anything
pf.enqueue_ways(id_tracker::max());
//let the threads process the ways
pf.process_ways();
}
void middle_ram_t::release_relations()
{
rels.clear();
}
void middle_ram_t::release_ways()
{
ways.clear();
}
bool middle_ram_t::ways_get(osmid_t id, taglist_t &tags, nodelist_t &nodes) const
{
if (simulate_ways_deleted) {
return false;
}
auto const *ele = ways.get(id);
if (!ele) {
return false;
}
tags = ele->tags;
nodes_get_list(nodes, ele->ndids);
return true;
}
size_t middle_ram_t::ways_get_list(const idlist_t &ids, idlist_t &way_ids,
multitaglist_t &tags, multinodelist_t &nodes) const
{
if (ids.empty())
{
return 0;
}
assert(way_ids.empty());
tags.assign(ids.size(), taglist_t());
nodes.assign(ids.size(), nodelist_t());
size_t count = 0;
for (idlist_t::const_iterator it = ids.begin(); it != ids.end(); ++it) {
if (ways_get(*it, tags[count], nodes[count])) {
way_ids.push_back(*it);
count++;
} else {
tags[count].clear();
nodes[count].clear();
}
}
if (count < ids.size()) {
tags.resize(count);
nodes.resize(count);
}
return int(count);
}
bool middle_ram_t::relations_get(osmid_t id, memberlist_t &members, taglist_t &tags) const
{
auto const *ele = rels.get(id);
if (!ele) {
return false;
}
tags = ele->tags;
members = ele->members;
return true;
}
void middle_ram_t::analyze(void)
{
/* No need */
}
void middle_ram_t::end(void)
{
/* No need */
}
void middle_ram_t::start(const options_t *out_options_)
{
out_options = out_options_;
/* latlong has a range of +-180, mercator +-20000
The fixed poing scaling needs adjusting accordingly to
be stored accurately in an int */
cache.reset(new node_ram_cache(out_options->alloc_chunkwise, out_options->cache, out_options->scale));
fprintf( stderr, "Mid: Ram, scale=%d\n", out_options->scale );
}
void middle_ram_t::stop(void)
{
cache.reset(nullptr);
release_ways();
release_relations();
}
void middle_ram_t::commit(void) {
}
middle_ram_t::middle_ram_t():
ways(), rels(), cache(), simulate_ways_deleted(false)
{
}
middle_ram_t::~middle_ram_t() {
//instance.reset();
}
std::vector<osmid_t> middle_ram_t::relations_using_way(osmid_t way_id) const
{
// this function shouldn't be called - relations_using_way is only used in
// slim mode, and a middle_ram_t shouldn't be constructed if the slim mode
// option is set.
throw std::runtime_error("middle_ram_t::relations_using_way is unimlpemented, and "
"should not have been called. This is probably a bug, please "
"report it at https://github.com/openstreetmap/osm2pgsql/issues");
}
namespace {
void no_delete(const middle_ram_t * middle) {
// boost::shared_ptr thinks we are going to delete
// the middle object, but we are not. Heh heh heh.
// So yeah, this is a hack...
}
}
std::shared_ptr<const middle_query_t> middle_ram_t::get_instance() const {
//shallow copy here because readonly access is thread safe
return std::shared_ptr<const middle_query_t>(this, no_delete);
}