3 * fix crash on update when addr:interpolation receives an illegal value
4 * fix minimum number of retrived results to be at least 10
5 * fix search for combinations of special term + name (e.g Hotel Bellevue)
6 * do not return interpolations without a parent street on reverse search
7 * improve invalidation of linked places on updates
8 * fix address parsing for interpolation lines
9 * make sure socket timeouts are respected during replication
10 (working around a bug in some versions of pyosmium)
11 * update bundled osm2pgsql to 1.7.1
12 * typing fixes to work with latest type annotations from typeshed
13 * smaller improvements to documention (thanks to @mausch)
17 * switch to ICU tokenizer as default
18 * add housenumber normalization and support optional spaces during search
19 * add postcode format checking and support optional spaces during search
20 * add function for cleaning housenumbers in word table
21 * add updates/deletion of country names imported from OSM
22 * linked places no longer overwrite names from a place permanently
23 * move default country name configuration into yaml file (thanks @tareqpi)
24 * more compact layout for interpolation and TIGER tables
25 * introduce mutations to ICU tokenizer (used for German umlauts)
26 * support reinitializing a full project directory with refresh --website
27 * fix various issues with linked places on updates
28 * add support for external sanitizers and token analyzers
29 * add CLI commands for forced indexing
30 * add CLI command for version report
31 * add offline import mode
32 * change geocodejson to return a feature class in the 'type' field
33 * add ISO3166-2 to address output (thanks @I70l0teN4ik)
34 * improve parsing and matching of addr: tags
35 * support relations as street members of associatedStreet
36 * better ranking for address results from TIGER data
37 * adapt rank classification to changed tag usage in OSM
38 * update bundled osm2pgsql to 1.6.0
39 * add typing information to Python code
40 * improve unit test coverage
41 * reorganise and speed up code for BDD tests, drop support for scenes
42 * move PHP unit tests to PHP 9.5
43 * extensive typo fixes in documentation (thanks @woodpeck,@StephanGeorg,
44 @amandasaurus, @nslxndr, @stefkiourk, @Luflosi, @kianmeng)
45 * drop official support for installation on CentOS
46 * add installation instructions for Ubuntu 22.04
47 * add support for PHP8
48 * add setup instructions for updates and systemd
49 * drop support for PostgreSQL 9.5
53 * fix initialisation error in replication script
54 * ICU tokenizer: avoid any special characters in word tokens
55 * better error message when API php script does not exist
56 * fix quoting of house numbers in SQL queries
57 * small fixes and improvements in search query parsing
58 * add documentation for moving the database to a different machine
62 * refactor name token computation and introduce ICU tokenizer
63 * name processing now happens in the indexer outside the DB
64 * reorganizes abbreviation handling and moves it to the indexing phases
65 * adds preprocessing of names
66 * add country-specific ranking for Spain, Slovakia
67 * partially switch to using SP-GIST indexes
68 * better updating of dependent addresses for name changes in streets
69 * remove unused/broken tables for external housenumbers
70 * move external postcodes to CSV format and no longer save them in tables
71 (adds support for postcodes for arbitrary countries)
72 * remove postcode helper entries from placex (thanks @AntoJvlt)
73 * change required format for TIGER data to CSV
74 * move configuration of default languages from wiki into config file
75 * expect customized configuration files in project directory by default
76 * disable search API for reverse-only import (thanks @darkshredder)
77 * port most of maintenance/import code to Python and remove PHP utils
78 * add catch-up mode for replication
79 * add updating of special phrases (thanks @AntoJvlt)
80 * add support for special phrases in CSV files (thanks @AntoJvlt)
81 * switch to case-independent matching between place and boundary names
82 * remove disabling of reverse query parsing
83 * minor tweaks to search algorithm to avoid more false positives
84 * major overhaul of the administrator and developer documentation
85 * add security disclosure policy
86 * add testing of installation scripts via CI
87 * drop support for Python < 3.6 and Postgresql < 9.5
91 * fix database check for reverse-only imports
92 * do not error out in status API result when import date is missing
93 * add array_key_last function for PHP < 7.3 (thanks to @woodpeck)
94 * fix more url when server name is unknown (thanks to @mogita)
95 * commit changes to replication log table
99 * fix smaller issues with special phrases import (thanks @AntoJvlt)
100 * add index to speed up continued indexing during import
101 * fix index on location_property_tiger(parent_place_id) (thanks @changpingc)
102 * make sure Python code is backward-compatible with Python 3.5
103 * various documentation fixes
107 * switch to dotenv for configuration file
108 * introduce 'make install' (reorganising most of the code)
109 * introduce nominatim tool as replacement for various php scripts
110 * introduce project directories and allow multiple installations from same build
111 * clean up BDD tests: drop nose, reorganise step code
112 * simplify test database for API BDD tests and autoinstall database
113 * port most of the code for command-line tools to Python
114 (thanks to @darkshredder and @AntoJvlt)
115 * add tests for all tooling
116 * replace pyosmium-get-changes with custom internal implementation using
118 * improve search for queries with housenumber and partial terms
119 * add database versioning
120 * use jinja2 for preprocessing SQL files
121 * introduce automatic migrations
122 * reverse fix preference of interpolations over housenumbers
123 * parallelize indexing of postcodes
124 * add non-key indexes to speed up housenumber + street searches
125 * switch housenumber field in placex to save transliterated names
129 * add full support for searching by and displaying of addr:* tags
130 * improve address output for large-area objects
131 * better use of country names from OSM data for search and display
132 * better debug output for reverse call
133 * add support for addr:place links without an place equivalent in OSM
134 * improve finding postcodes with normalisation artefacts
135 * batch object to index for rank 30, avoiding a wrap-around of transaction
137 * introduce dynamic address rank computation for administrative boundaries
138 depending on linked objects and their place in the admin level hierarchy
139 * add country-specific address ranking for Indonesia, Russia, Belgium and
140 the Netherlands (thanks @hendrikmoree)
141 * make sure wikidata/wikipedia tags are imported for all styles
142 * make POIs searchable by name and housenumber (thanks @joy-yyd)
143 * reverse geocoding now ignores places without an address rank (rivers etc.)
144 * installation of a webserver is no longer mandatory, for development
145 use the php internal webserver via 'make serve
146 * reduce the influence of place nodes in addresses
147 * drop support for the unspecific is_in tag
148 * various minor tweaks to supplied styles
149 * move HTML web frontend into its own project
150 * move scripts for processing external data sources into separate directories
151 * introduce separate configuration for website (thanks @krahulreddy)
152 * update documentation, in particular, clean up development docs
153 * update osm2pgsql to 1.4.0
157 * ensure that wikipedia tags are imported for all styles
158 * reinstate verbosity for indexing during updates
159 * make house number reappear in display name on named POIs
160 * introduce batch processing in indexer to avoid transaction ID overrun
161 * increase splitting for large geometries to improve indexing speed
162 * remove deprecated get_magic_quotes_gpc() function
163 * make sure that all postcodes have an entry in word and are thus searchable
164 * remove use of ST_Covers in conjunction woth ST_Intersects,
165 causes bad query planning and slow updates in Postgis3
170 * disable jit and parallel processing in PostgreSQL for osm2pgsql
171 * update libosmium to 2.15.6 (fixes an issue with processing hanging
172 on large multipolygons)
176 * structured select on HTML search page
177 * new PHP Nominatim\Shell class to wrap shell escaping
178 * remove polygon parameter from all API calls
179 * improve handling of postcode areas
180 * reorganise place linking algorithm, now using wikidata tag as well
181 * remove linkees from search_name and larger_area tables
182 * introduce country-specific address ranks
183 * reorganise rank address computation
184 * cleanup of partition function
185 * improve parenting for large POIs
186 * add support for Postgresql 12 and Postgis 3
187 * add earlier cleanup when --drop is given, to reduce memory usage
188 * remove use of place_id in URLs
189 * replace C nominatim indexer with a simpler Python implementation
190 * split up the huge sql/functions.sql file
191 * move osm2pgsql tests to osm2pgsql
192 * add new extratags style which imports all tags from OSM
193 * add new script for checking the import after completion
194 * update osm2pgsql, reducing memory usage
195 * use new wikipedia importance and add processing of wikidata tags
196 * add search form for details page
197 * use ExtraDataPath for country_grid table
198 * remove short_name from list of names to be displayed
199 * split up CMakeFile, so that all parts can be built separately
200 * update installation instructions for CentOS and Ubuntu
201 * add script for importing/updating multiple country extracts
202 * various documentation improvements
206 * fix security bug in /details endpoint where user input was not
211 * update osm2pgsql to fix hans during updates and lost address numbers
216 * increase required version for PostgreSQL(9.3), PostGIS(2.2) and PHP(7.0)
217 * better error reporting for out-of-memory errors
218 * exclude postcode ranges separated by colon from centre point calculation
219 * update osm2pgsql, better handling of imports without flatnode file
220 * switch to more efficient algorithm for word set computation
221 * use only boundries for country and state parts of addresses
222 * improve updates of addresses with housenumbers and interpolations
223 * remove country from place_addressline table and use country_code instead
224 * optimise indexes on search_name partition tables
225 * improve searching of attached streets for large objects like airports
226 * drop support for python 2
227 * new scripts for importing Wikidata for importance
228 * create and drop indexes concurrently to not clash with auto vacuum
229 * various documentation improvements
234 * zoom 17 in reverse now zooms in on minor streets
235 * fix use of postcode relations in address
236 * support for housenumber 0 on interpolations
237 * replace database abstraction DB with PDO and switch to using exceptions
238 * exclude line features at rank 30 from reverse geocoding
239 * remove self-reference and country from place_addressline
240 * make json output more readable (less escaping)
241 * update conversion scripts for postcodes
242 * scripts in utils/ are no longer executable (always use scripts in build dir)
243 * remove Natural Earth country fallback (OSM is complete enough)
244 * make rank assignments configurable
245 * allow accept languages with underscore
246 * new reverse-only import mode (without search index table)
247 * rely on boundaries only for states and countries
248 * update osm2pgsql, now using a configurable style
249 * provide multiple import styles
250 * improve search when house number and postcodes are dropped
251 * overhaul of setup code
252 * add support for PHPUnit 6
253 * update test database
254 * various documentation improvements
258 * complete rewrite of reverse search algorithm
259 * add new geojson and geocodejson output formats
260 * add simple export script to exprot addresses to CSV
261 * remove is_in terms from address computation
262 * remove unused search_name_country tables
263 * various smaller fixes to query parsing
264 * convert Tokens and token types to class types
265 * correctly handle update when boundary object type is changed
266 * improve debug output for /search endpoint
267 * update to latest osm2pgsql and leaflet.js
268 * overhaul of /details endpoint:
269 * new class parameter when using osmtype/osmid parameters
270 * permalink to instance-independent osmtype/osmid parameter format
271 * new json output format
272 * update CentOS vagrant machine to use SELinux
273 * add vagrant scripts for Ubuntu 18.04
274 * fix build process for BSD
275 * enable running the database on a different host than the setup scripts
276 * allow to configure use of custom PHP binaries (PHP_BIN)
277 * extensive coding style improvements to PHP code
278 * more PHP unit tests for new classes
279 * increase coverage for API tests
280 * add documentation for API
284 * rework postcode handling and introduce location_postcode table
285 * make setup less verbose and print a summary in the end
286 * setup: error out when web site user does not exist
287 * add more API tests to complete code coverage
288 * reinstate key-value amenity search (special term [key=value])
289 * fix detection of coordinates in query
290 * various smaller tweaks to ranking of search interpretations
291 * complete overhaul of PHP frontend code using OOP
292 * add address rank to details page
293 * update Tiger scripts for 2017 data and clean up unused code
294 * various bug fixes and improvements to UI
295 * improve reverse geocoding performance close to coasts
296 * more PHP style cleanup (quoting)
297 * allow unnamed road in reverse geocoding to avoid too far off results
298 * add function to recalculate counts for full-word search term
299 * add function to check if new updates are available
300 * update documentation and switch to mkdocs for generating HTML
304 * fix bug in geometry building algorithm in osm2pgsql
305 * fix typos in build instructions
309 * move to cmake build system
310 * various fixes to HTML output
311 * reverse endpoint now can return geometries
312 * add support for PHP7
313 * move to on-the-fly computation of interpolations
314 * improve handling of linked places (updating)
315 * improve test framework:
316 * replace lettuce with behave
317 * use smaller database for API tests
318 * drop support for postgres < 9.1, postgis < 2.0 and PHP < 5.4
319 * make external data use optional (useful for imports without US)
320 * detect postgres and postgis versions automatically
321 * clean up query logging and remove unused tables
322 * move installation documentation into this repo
323 * add self-documenting vagrant scripts
324 * remove --create-website, recommend to use website directory in build
325 * add accessor functions for URL parameters and improve erro checking
326 * remove IP blocking and rate-limiting code
327 * enable CI via travis
328 * reformatting for more consistent coding style
329 * make country search term creation part of setup
330 * update country names and country grid
331 * handle roads that cross boundaries better
332 * keep full information on address tags
333 * update to refactored osm2pgsql which use libosmium based types
334 * switch from osmosis to pyosmium for updates
335 * be more strict when matching against special search terms
336 * handle postcode entries with mutliple values correctly
340 * reverse geocoding includes looking up housenumbers from Tiger data
341 * added parameter to return simplified geometries
342 * new lookup call for getting address information for OSM objects
343 * new namedetails and extratags parameters that expose the name and extratags
344 fields of the placex table
350 * drop support for postgres 8.4
351 * rewrite address interpolation
352 * switch to C++ version of osm2pgsql and rewrite tag filtering
353 * support for bridge:name and tunnel:name, man_made, junction
354 * drop way-node index (reduces database size by about 15%)
355 * add support for configuring tablespaces and webserver user
356 * better evaluation of search queries in right-to-left notation
357 * improve house number search for streets with many duplicate entries
358 * code cleanup (remove unused functions and tables)
362 * further improve ordering of results
363 * support for more lat/lon formats in search-as-reverse
364 * fix handling of GB postcodes
365 * new functional test suite
366 * support for waterway relations
367 * inherit postcodes from street to poi
368 * fix housenumber normalisation to find non-latin house numbers
369 * take viewbox into account for ordering of results
370 * pois may now inherit address tags from surrounding buildings
371 * improve what objects may participate in an address
372 * clean up handled class/type combinations to current OSM usage
377 * correct database rights for www-data
378 * add timestamps for update output
379 * load postgis via extension for postgis >= 2.0
380 * remove non-admin boundaries from addresses
381 * further improve ordering of results with same importance
382 * merge addr:postcode tags into object addresses
383 * include rank and importance in reverse geocode output
384 * replace ST_Line_Interpolate_Point with ST_LineInterpolatePoint
386 * update osm2pgsql to latest version
387 * properly detect changes of admin_level
388 * remove landuses when name is removed
393 * update to openlayers 2.12 (now custom built)
394 * update fallback OSM boundaries
395 * add support for postgresql 9.2/9.3 and postgis 2.x
396 * add structured queries
397 * add options for polygon output in various formats
398 (geojson, svg, kml, postgis text)
399 * maintenance functions for deleting objects and updating regions
400 (plcae_force_update/place_force_delete)
401 * web view for polygons that need deleting
402 * rate limiting using memcache
403 * improve layout of details page
404 * add support for boundary:postal_code
406 * improve parenting of POIs
407 * support for extract daily diffs from Geofabrik
408 * support for addresses without a street
409 (addr:place and conscription number house numbers)
410 * improve layout of word and search_name_* tables
411 * support for US ZIP+4 codes
412 * refactoring of front-end PHP code
413 * lots of smaller bug fixes
417 * delete outdated entries from location_area_country
418 * remove remaining uses of INTEGER, to allow node ids larger than 2^31