5 OSM data is initially imported using [osm2pgsql](https://osm2pgsql.org).
6 Nominatim uses a custom flex style to create the initial import tables.
8 The import process creates the following tables:
10 
12 The `planet_osm_*` tables are the usual backing tables for OSM data. Note
13 that Nominatim uses them to look up special relations and to find nodes on
16 The osm2pgsql import produces a single table `place` as output with the following
19 * `osm_type` - kind of OSM object (**N** - node, **W** - way, **R** - relation)
20 * `osm_id` - original OSM ID
21 * `class` - key of principal tag defining the object type
22 * `type` - value of principal tag defining the object type
23 * `name` - collection of tags that contain a name or reference
24 * `admin_level` - numerical value of the tagged administrative level
25 * `address` - collection of tags defining the address of an object
26 * `extratags` - collection of additional interesting tags that are not
27 directly relevant for searching
28 * `geometry` - geometry of the object (in WGS84)
30 A single OSM object may appear multiple times in this table when it is tagged
31 with multiple tags that may constitute a principal tag. Take for example a
32 motorway bridge. In OSM, this would be a way which is tagged with
33 `highway=motorway` and `bridge=yes`. This way would appear in the `place` table
34 once with `class` of `highway` and once with a `class` of `bridge`. Thus the
35 *unique key* for `place` is (`osm_type`, `osm_id`, `class`).
37 How raw OSM tags are mapped to the columns in the place table is to a certain
38 degree configurable. See [Customizing Import Styles](../customize/Import-Styles.md)
43 The following tables carry all information needed to do the search:
45 
47 The **placex** table is the central table that saves all information about the
48 searchable places in Nominatim. The basic columns are the same as for the
49 place table and have the same meaning. The placex tables adds the following
52 * `place_id` - the internal unique ID to identify the place
53 * `partition` - the id to use with partitioned tables (see below)
54 * `geometry_sector` - a location hash used for geographically close ordering
55 * `parent_place_id` - the next higher place in the address hierarchy, only
56 relevant for POI-type places (with rank 30)
57 * `linked_place_id` - place ID of the place this object has been merged with.
58 When this ID is set, then the place is invisible for search.
59 * `importance` - measure how well known the place is
60 * `rank_search`, `rank_address` - search and address rank (see [Customizing ranking](../customize/Ranking.md)
61 * `wikipedia` - the wikipedia page used for computing the importance of the place
62 * `country_code` - the country the place is located in
63 * `housenumber` - normalized housenumber, if the place has one
64 * `postcode` - computed postcode for the place
65 * `indexed_status` - processing status of the place (0 - ready, 1 - freshly inserted, 2 - needs updating, 100 - needs deletion)
66 * `indexed_date` - timestamp when the place was processed last
67 * `centroid` - a point feature for the place
69 The **location_property_osmline** table is a special table for
70 [address interpolations](https://wiki.openstreetmap.org/wiki/Addresses#Using_interpolation).
71 The columns have the same meaning and use as the columns with the same name in
72 the placex table. Only three columns are special:
74 * `startnumber` and `endnumber` - beginning and end of the number range
76 * `interpolationtype` - a string `odd`, `even` or `all` to indicate
77 the interval between the numbers
79 Address interpolations are always ways in OSM, which is why there is no column
82 The **location_postcode** table holds computed centroids of all postcodes that
83 can be found in the OSM data. The meaning of the columns is again the same
84 as that of the placex table.
86 Every place needs an address, a set of surrounding places that describe the
87 location of the place. The set of address places is made up of OSM places
88 themselves. The **place_addressline** table cross-references for each place
89 all the places that make up its address. Two columns define the address
92 * `place_id` - reference to the place being addressed
93 * `address_place_id` - reference to the place serving as an address part
95 The most of the columns cache information from the placex entry of the address
96 part. The exceptions are:
98 * `fromarea` - is true if the address part has an area geometry and can
99 therefore be considered preceise
100 * `isaddress` - is true if the address part should show up in the address
101 output. Sometimes there are multiple places competing for for same address
102 type (e.g. multiple cities) and this field resolves the tie.
104 The **search_name** table contains the search index proper. It saves for each
105 place the terms with which the place can be found. The terms are split into
106 the name itself and all terms that make up the address. The table mirrors some
107 of the columns from placex for faster lookup.
109 Search terms are not saved as strings. Each term is assigned an integer and those
110 integers are saved in the name and address vectors of the search_name table. The
111 **word** table serves as the lookup table from string to such a word ID. The
112 exact content of the word table depends on the [tokenizer](Tokenizers.md) used.
114 ## Address computation tables
116 Next to the main search tables, there is a set of secondary helper tables used
117 to compute the address relations between places. These tables are partitioned.
118 Each country is assigned a partition number in the country_name table (see
119 below) and the data is then split between a set of tables, one for each
120 partition. Note that Nominatim still manually manages partitioned tables.
121 Native support for partitions in PostgreSQL only became usable with version 13.
122 It will be a little while before Nominatim drops support for older versions.
124 
126 The **search_name_X** tables are used to look up streets that appear in the
129 The **location_area_large_X** tables are used to look up larger areas
130 (administrative boundaries and place nodes) either through their geographic
131 closeness or through `addr:*` entries.
133 The **location_road_X** tables are used to find the closest street for a
136 All three table cache specific information from the placex table for their
137 selected subset of places:
139 * `keywords` and `name_vector` contain lists of term ids (from the word table)
140 that the full name of the place should match against
141 * `isguess` is true for places that are not described by an area
143 All other columns reflect their counterpart in the placex table.
145 ## Static data tables
147 Nominatim also creates a number of static tables at import:
149 * `nominatim_properties` saves settings that must not be changed after
151 * `address_levels` save the rank information from the
152 [ranking configuration](../customize/Ranking.md)
153 * `country_name` contains a fallback of names for all countries, their
154 default languages and saves the assignment of countries to partitions.
155 * `country_osm_grid` provides a fallback for country geometries
157 ## Auxiliary data tables
159 Finally there are some table for auxiliary data:
161 * `location_property_tiger` - saves housenumber from the Tiger import. Its
162 layout is similar to that of `location_propoerty_osmline`.
163 * `place_class_*` tables are helper tables to facilitate lookup of POIs
164 by their class and type. They exist because it is not possible to create
165 combined indexes with geometries.