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