]> git.openstreetmap.org Git - nominatim.git/log
nominatim.git
4 months agomove database setup to generic conftest.py
Sarah Hoffmann [Thu, 3 Apr 2025 09:20:38 +0000 (11:20 +0200)]
move database setup to generic conftest.py

4 months agoremove old behave osm2pgsql BDD tests
Sarah Hoffmann [Thu, 3 Apr 2025 08:59:54 +0000 (10:59 +0200)]
remove old behave osm2pgsql BDD tests

4 months agoimplement BDD osm2pgsql tests with pytest-bdd
Sarah Hoffmann [Mon, 31 Mar 2025 07:39:01 +0000 (09:39 +0200)]
implement BDD osm2pgsql tests with pytest-bdd

4 months agoenable python-bdd for github actions
Sarah Hoffmann [Tue, 18 Mar 2025 15:20:52 +0000 (16:20 +0100)]
enable python-bdd for github actions

4 months agorestrict coordinate output to 7 digits
Sarah Hoffmann [Tue, 18 Mar 2025 14:00:35 +0000 (15:00 +0100)]
restrict coordinate output to 7 digits

4 months agoreplace behave BDD API tests with pytest-bdd tests
Sarah Hoffmann [Tue, 18 Mar 2025 10:19:06 +0000 (11:19 +0100)]
replace behave BDD API tests with pytest-bdd tests

4 months agoMerge pull request #3687 from asharmalik19/test-linked-places-language
Sarah Hoffmann [Fri, 4 Apr 2025 08:58:53 +0000 (10:58 +0200)]
Merge pull request #3687 from asharmalik19/test-linked-places-language

test: linked places expand default language names

4 months agoUpdate linked places name test
Ashar [Thu, 3 Apr 2025 18:20:00 +0000 (14:20 -0400)]
Update linked places name test

Clean up test scenario by removing extra language variations and
improving table readability.

4 months agotest: linked places expand default language names
Ashar [Thu, 27 Mar 2025 23:03:33 +0000 (19:03 -0400)]
test: linked places expand default language names

Add failing test for issue #2714 to verify default language expansion

4 months agoMerge pull request #3702 from lonvia/remove-tokenizer-dir
Sarah Hoffmann [Thu, 3 Apr 2025 07:04:48 +0000 (09:04 +0200)]
Merge pull request #3702 from lonvia/remove-tokenizer-dir

Remove automatic setup of tokenizer directory

So far the tokenizer factory would create a directory for private data for the tokenizer and then hand in the directory location to the tokenizer.

ICU tokenizer doesn't need any extra data anymore, so it doesn't make sense to create a directory which then remains empty. If a tokenizer needs such a directory in the future, it needs to create it on its own and make sure to handle the situation correctly where no project directory is used at all.

4 months agoremove automatic setup of tokenizer directory
Sarah Hoffmann [Wed, 2 Apr 2025 18:20:04 +0000 (20:20 +0200)]
remove automatic setup of tokenizer directory

ICU tokenizer doesn't need any extra data anymore, so it doesn't
make sense to create a directory which then remains empty. If a
tokenizer needs such a directory in the future, it needs to create
it on its own and make sure to handle the situation correctly where
no project directory is used at all.

4 months agoMerge pull request #3700 from lonvia/ignore-inherited-addresses
Sarah Hoffmann [Wed, 2 Apr 2025 10:00:45 +0000 (12:00 +0200)]
Merge pull request #3700 from lonvia/ignore-inherited-addresses

Ignore POIs with inherited addresses for the address layer

4 months agoignore POIs with inherited addresses for the address layer
Sarah Hoffmann [Wed, 2 Apr 2025 08:09:49 +0000 (10:09 +0200)]
ignore POIs with inherited addresses for the address layer

We know that there is a building which describes the address as a
polygon and is therefore more suitable.

4 months agoMerge pull request #3686 from astridx/output_names
Sarah Hoffmann [Tue, 1 Apr 2025 18:16:15 +0000 (20:16 +0200)]
Merge pull request #3686 from astridx/output_names

Output names as setting

4 months agoMerge pull request #3675 from TuringVerified/generic-preprocessors
Sarah Hoffmann [Tue, 1 Apr 2025 18:14:43 +0000 (20:14 +0200)]
Merge pull request #3675 from TuringVerified/generic-preprocessors

Add generic preprocessors

4 months agoRemove unnecessary assert statement, Fix regex_replace docstring and simplify regex_r...
TuringVerified [Mon, 31 Mar 2025 19:57:45 +0000 (01:27 +0530)]
Remove unnecessary assert statement, Fix regex_replace docstring and simplify regex_replace

4 months agoUpdate documentation, optimise regex_replace, add tests
TuringVerified [Thu, 20 Mar 2025 14:30:34 +0000 (20:00 +0530)]
Update documentation, optimise regex_replace, add tests

4 months agoAdd generic preprocessor
TuringVerified [Thu, 13 Mar 2025 14:31:21 +0000 (20:01 +0530)]
Add generic preprocessor

4 months agoadd instructions for pip package upload
Sarah Hoffmann [Tue, 1 Apr 2025 09:59:03 +0000 (11:59 +0200)]
add instructions for pip package upload

4 months agoprepare release 5.1.0
Sarah Hoffmann [Tue, 1 Apr 2025 08:16:35 +0000 (10:16 +0200)]
prepare release 5.1.0

4 months agoMerge pull request #3695 from TuringVerified/doc-dependencies
Sarah Hoffmann [Tue, 1 Apr 2025 07:34:08 +0000 (09:34 +0200)]
Merge pull request #3695 from TuringVerified/doc-dependencies

[Small fix] Add documentation to install extras for mkdocstrings

4 months agoAdd mkdocstrings extra
TuringVerified [Tue, 1 Apr 2025 05:36:46 +0000 (11:06 +0530)]
Add mkdocstrings extra

4 months agoMerge pull request #3693 from lonvia/remove-unused-sql
Sarah Hoffmann [Mon, 31 Mar 2025 15:11:39 +0000 (17:11 +0200)]
Merge pull request #3693 from lonvia/remove-unused-sql

Remove SQL function for address lookup

4 months agooutput names as setting
astridx [Fri, 14 Mar 2025 21:03:36 +0000 (22:03 +0100)]
output names as setting

4 months agoMerge pull request #3692 from lonvia/word-lookup-variants
Sarah Hoffmann [Mon, 31 Mar 2025 14:38:31 +0000 (16:38 +0200)]
Merge pull request #3692 from lonvia/word-lookup-variants

Avoid matching penalty for abbreviated search terms

4 months agoremove SQL function for address lookup
Sarah Hoffmann [Mon, 31 Mar 2025 12:34:46 +0000 (14:34 +0200)]
remove SQL function for address lookup

This is now done in Python.

4 months agoMerge pull request #3691 from lonvia/more-search-tweaks
Sarah Hoffmann [Mon, 31 Mar 2025 13:06:09 +0000 (15:06 +0200)]
Merge pull request #3691 from lonvia/more-search-tweaks

More tweaks to search wights

4 months agoadapt tests to extended results
Sarah Hoffmann [Fri, 21 Mar 2025 19:24:09 +0000 (20:24 +0100)]
adapt tests to extended results

4 months agoadd lookup word to variants in word table
Sarah Hoffmann [Fri, 21 Mar 2025 08:02:52 +0000 (09:02 +0100)]
add lookup word to variants in word table

4 months agodo not write any word counts on initial word insert
Sarah Hoffmann [Thu, 20 Mar 2025 20:33:27 +0000 (21:33 +0100)]
do not write any word counts on initial word insert

4 months agodo not save word counts of 1
Sarah Hoffmann [Thu, 20 Mar 2025 19:01:46 +0000 (20:01 +0100)]
do not save word counts of 1

This is the default setting, which will be assumed when the count is
missing.

4 months agoalways use lookup when requested
Sarah Hoffmann [Mon, 31 Mar 2025 09:15:26 +0000 (11:15 +0200)]
always use lookup when requested

Doesn't seem to cause any issues in production.

4 months agoincrease allowable address counts
Sarah Hoffmann [Mon, 31 Mar 2025 09:12:08 +0000 (11:12 +0200)]
increase allowable address counts

4 months agomore agressively reduce expected count for multi-word terms
Sarah Hoffmann [Mon, 31 Mar 2025 09:06:21 +0000 (11:06 +0200)]
more agressively reduce expected count for multi-word terms

Improves searching of non-latin scripts with forced token spaces.

4 months agoMerge pull request #3690 from lonvia/fix-signature
Sarah Hoffmann [Mon, 31 Mar 2025 09:17:03 +0000 (11:17 +0200)]
Merge pull request #3690 from lonvia/fix-signature

Fix function signature for newer SQLAlchemy

4 months agofix function signature for newer SQLAlchemy
Sarah Hoffmann [Mon, 31 Mar 2025 07:42:29 +0000 (09:42 +0200)]
fix function signature for newer SQLAlchemy

4 months agoMerge pull request #3684 from lonvia/compact-en-variants
Sarah Hoffmann [Mon, 24 Mar 2025 14:15:13 +0000 (15:15 +0100)]
Merge pull request #3684 from lonvia/compact-en-variants

Clean up English variants

4 months agorestrict some English variants to end of word
Sarah Hoffmann [Fri, 21 Mar 2025 20:22:38 +0000 (21:22 +0100)]
restrict some English variants to end of word

4 months agomake English variant list more compact
Sarah Hoffmann [Fri, 21 Mar 2025 20:13:34 +0000 (21:13 +0100)]
make English variant list more compact

4 months agoMerge pull request #3682 from lonvia/fix-postcode-case
Sarah Hoffmann [Fri, 21 Mar 2025 10:41:24 +0000 (11:41 +0100)]
Merge pull request #3682 from lonvia/fix-postcode-case

Fix case issues when parsing postcodes

4 months agomake query upper-case when parsing postcodes
Sarah Hoffmann [Fri, 21 Mar 2025 08:44:15 +0000 (09:44 +0100)]
make query upper-case when parsing postcodes

The postcode patterns expect upper-case letters.

4 months agoMerge pull request #3679 from lonvia/output-fixes
Sarah Hoffmann [Wed, 19 Mar 2025 20:56:28 +0000 (21:56 +0100)]
Merge pull request #3679 from lonvia/output-fixes

Minor fixes for v1 frontend code

4 months agofix layer setting for structured search
Sarah Hoffmann [Tue, 18 Mar 2025 10:20:20 +0000 (11:20 +0100)]
fix layer setting for structured search

4 months agofix debug output for lookup type
Sarah Hoffmann [Tue, 18 Mar 2025 10:21:54 +0000 (11:21 +0100)]
fix debug output for lookup type

4 months agoMerge pull request #3678 from lonvia/search-tweaks
Sarah Hoffmann [Wed, 19 Mar 2025 15:00:52 +0000 (16:00 +0100)]
Merge pull request #3678 from lonvia/search-tweaks

Some minor tweaks to postcode parsing in query

4 months agopenalize postcode searches with multiple name qualifiers
Sarah Hoffmann [Wed, 19 Mar 2025 09:05:36 +0000 (10:05 +0100)]
penalize postcode searches with multiple name qualifiers

4 months agouse word_token length when penalizing against postcodes
Sarah Hoffmann [Tue, 18 Mar 2025 22:07:27 +0000 (23:07 +0100)]
use word_token length when penalizing against postcodes

4 months agopostcode token should have transliterated term in word_token
Sarah Hoffmann [Tue, 18 Mar 2025 21:55:33 +0000 (22:55 +0100)]
postcode token should have transliterated term in word_token

4 months agoimprove handling of leading postcodes
Sarah Hoffmann [Tue, 18 Mar 2025 21:32:58 +0000 (22:32 +0100)]
improve handling of leading postcodes

Setting the direction of the query while yielding assignments is
a bad idea because it may override a direction already set.

4 months agoMerge pull request #3673 from otbutz/parallel_safe
Sarah Hoffmann [Tue, 18 Mar 2025 20:46:53 +0000 (21:46 +0100)]
Merge pull request #3673 from otbutz/parallel_safe

Mark functions as PARALLEL SAFE

4 months agoMerge pull request #3676 from lonvia/adjust-place-levels-sa
Sarah Hoffmann [Tue, 18 Mar 2025 17:31:48 +0000 (18:31 +0100)]
Merge pull request #3676 from lonvia/adjust-place-levels-sa

Adjust place ranks for Saudi-Arabia

4 months agoadjust for special use of province and municipality in Saudi-Arabia
Sarah Hoffmann [Tue, 18 Mar 2025 15:38:10 +0000 (16:38 +0100)]
adjust for special use of province and municipality in Saudi-Arabia

4 months agoMark functions as PARALLEL SAFE
Thomas Butz [Thu, 13 Mar 2025 09:04:40 +0000 (10:04 +0100)]
Mark functions as PARALLEL SAFE

4 months agoadapt typing for latest SQLAlchemy version
Sarah Hoffmann [Thu, 13 Mar 2025 09:49:08 +0000 (10:49 +0100)]
adapt typing for latest SQLAlchemy version

4 months agoMerge pull request #3671 from lonvia/remove-osm2pgsql-libdir
Sarah Hoffmann [Tue, 11 Mar 2025 10:22:46 +0000 (11:22 +0100)]
Merge pull request #3671 from lonvia/remove-osm2pgsql-libdir

Remove code for setting osm2pgsql location via config.lib_dir

4 months agoremove code for setting osm2pgsql via config.lib_dir
Sarah Hoffmann [Tue, 11 Mar 2025 07:45:10 +0000 (08:45 +0100)]
remove code for setting osm2pgsql via config.lib_dir

With the internal osm2pgsql gone, configuration of the binary location
via settings is the only option left that makes sense.

5 months agoMerge pull request #3670 from lonvia/flake-for-tests
Sarah Hoffmann [Mon, 10 Mar 2025 08:35:24 +0000 (09:35 +0100)]
Merge pull request #3670 from lonvia/flake-for-tests

Extend linting with flake to tests

5 months agoremove unused bdd util functions
Sarah Hoffmann [Sun, 9 Mar 2025 16:34:40 +0000 (17:34 +0100)]
remove unused bdd util functions

5 months agoenable flake for bdd test code
Sarah Hoffmann [Sun, 9 Mar 2025 16:34:04 +0000 (17:34 +0100)]
enable flake for bdd test code

5 months agoalso enable flake for tests in github actions
Sarah Hoffmann [Sun, 9 Mar 2025 15:03:02 +0000 (16:03 +0100)]
also enable flake for tests in github actions

5 months agoenable flake for Python tests
Sarah Hoffmann [Sun, 9 Mar 2025 14:33:24 +0000 (15:33 +0100)]
enable flake for Python tests

5 months agoMerge pull request #3667 from eumiro/simplify-int-float
Sarah Hoffmann [Sun, 9 Mar 2025 08:44:15 +0000 (09:44 +0100)]
Merge pull request #3667 from eumiro/simplify-int-float

Simplify  int/float manipulation

5 months agoSimplify int/float manipulation
Miroslav Šedivý [Wed, 5 Mar 2025 19:55:53 +0000 (20:55 +0100)]
Simplify  int/float manipulation

5 months agoMerge pull request #3666 from eumiro/math-isclose
Sarah Hoffmann [Thu, 6 Mar 2025 16:53:01 +0000 (17:53 +0100)]
Merge pull request #3666 from eumiro/math-isclose

Replace custom Almost with stdlib math.isclose

5 months agoMerge pull request #3664 from eumiro/consolidate-random
Sarah Hoffmann [Thu, 6 Mar 2025 16:52:19 +0000 (17:52 +0100)]
Merge pull request #3664 from eumiro/consolidate-random

Consolidate usage of random module

5 months agoReplace custom Almost with stdlib math.isclose
Miroslav Šedivý [Wed, 5 Mar 2025 19:35:01 +0000 (20:35 +0100)]
Replace custom Almost with stdlib math.isclose

5 months agoConsolidate usage of random module
Miroslav Šedivý [Tue, 4 Mar 2025 19:01:47 +0000 (20:01 +0100)]
Consolidate usage of random module

5 months agoMerge pull request #3665 from lonvia/pattern-matching-postcodes
Sarah Hoffmann [Wed, 5 Mar 2025 15:02:03 +0000 (16:02 +0100)]
Merge pull request #3665 from lonvia/pattern-matching-postcodes

Add full parsing of postcodes in query

5 months agorestrict postcode parsing in typed phrases
Sarah Hoffmann [Wed, 5 Mar 2025 09:08:07 +0000 (10:08 +0100)]
restrict postcode parsing in typed phrases

Postcodes can only appear in postcode-type phrases and must then
cover the full phrase

5 months agoadd unit tests for postcode parser
Sarah Hoffmann [Tue, 4 Mar 2025 15:25:00 +0000 (16:25 +0100)]
add unit tests for postcode parser

5 months agoadapt BDD tests to new postcode handling
Sarah Hoffmann [Tue, 4 Mar 2025 14:18:46 +0000 (15:18 +0100)]
adapt BDD tests to new postcode handling

5 months agoadd support for country prefixes in postcodes
Sarah Hoffmann [Tue, 4 Mar 2025 14:18:27 +0000 (15:18 +0100)]
add support for country prefixes in postcodes

5 months agocache all info of ICUQueryAnalyser in a single object
Sarah Hoffmann [Sun, 2 Mar 2025 16:31:04 +0000 (17:31 +0100)]
cache all info of ICUQueryAnalyser in a single object

5 months agoremove postcode computation for word table during import
Sarah Hoffmann [Sat, 1 Mar 2025 09:20:33 +0000 (10:20 +0100)]
remove postcode computation for word table during import

5 months agoreenable ZIP+ test
Sarah Hoffmann [Fri, 28 Feb 2025 16:19:56 +0000 (17:19 +0100)]
reenable ZIP+ test

5 months agoshow token begin and end in debug output
Sarah Hoffmann [Fri, 28 Feb 2025 16:19:37 +0000 (17:19 +0100)]
show token begin and end in debug output

5 months agoadd postcode parser
Sarah Hoffmann [Tue, 25 Feb 2025 19:56:07 +0000 (20:56 +0100)]
add postcode parser

5 months agomake word generation from query a class method
Sarah Hoffmann [Wed, 26 Feb 2025 16:22:14 +0000 (17:22 +0100)]
make word generation from query a class method

5 months agosearch: merge QueryPart array with QueryNodes
Sarah Hoffmann [Wed, 26 Feb 2025 13:37:08 +0000 (14:37 +0100)]
search: merge QueryPart array with QueryNodes

The basic information on terms is pretty much always used together
with the node inforamtion. Merging them together saves some
allocation while making lookup easier at the same time.

5 months agoenable parsing of US ZIP+ codes
Sarah Hoffmann [Tue, 25 Feb 2025 19:29:06 +0000 (20:29 +0100)]
enable parsing of US ZIP+ codes

The four-digit part of these postcodes will simply be ignored.

5 months agoMerge pull request #3659 from lonvia/custom-datrie-structure
Sarah Hoffmann [Mon, 24 Feb 2025 15:49:42 +0000 (16:49 +0100)]
Merge pull request #3659 from lonvia/custom-datrie-structure

Replace datrie library with a simple custom Python implementation

5 months agoremove datrie dependency
Sarah Hoffmann [Tue, 18 Feb 2025 20:12:02 +0000 (21:12 +0100)]
remove datrie dependency

5 months agoreplace datrie library with a more simple pure-Python class
Sarah Hoffmann [Tue, 18 Feb 2025 20:09:12 +0000 (21:09 +0100)]
replace datrie library with a more simple pure-Python class

5 months agoMerge pull request #3658 from lonvia/minor-query-parsing-optimisations
Sarah Hoffmann [Mon, 24 Feb 2025 09:16:47 +0000 (10:16 +0100)]
Merge pull request #3658 from lonvia/minor-query-parsing-optimisations

Minor query parsing optimisations

5 months agoadapt documentation for PhraseType type
Sarah Hoffmann [Fri, 21 Feb 2025 16:16:42 +0000 (17:16 +0100)]
adapt documentation for PhraseType type

5 months agoreplace PhraseType enum with simple int constants
Sarah Hoffmann [Fri, 21 Feb 2025 15:44:12 +0000 (16:44 +0100)]
replace PhraseType enum with simple int constants

5 months agoreplace TokenType enum with simple char constants
Sarah Hoffmann [Fri, 21 Feb 2025 09:23:41 +0000 (10:23 +0100)]
replace TokenType enum with simple char constants

5 months agoreplace BreakType enum with simple char constants
Sarah Hoffmann [Fri, 21 Feb 2025 08:57:48 +0000 (09:57 +0100)]
replace BreakType enum with simple char constants

5 months agoconsistently use query module as qmod
Sarah Hoffmann [Fri, 21 Feb 2025 08:31:21 +0000 (09:31 +0100)]
consistently use query module as qmod

5 months agoavoid yielding when extracting words from query
Sarah Hoffmann [Thu, 20 Feb 2025 22:32:39 +0000 (23:32 +0100)]
avoid yielding when extracting words from query

5 months agoremove word_number counting for phrases
Sarah Hoffmann [Thu, 20 Feb 2025 16:36:50 +0000 (17:36 +0100)]
remove word_number counting for phrases

We can just examine the break types to know if we are dealing
with a partial token.

5 months agoMerge pull request #3655 from lonvia/remove-name-ranking-in-postcode-search
Sarah Hoffmann [Thu, 20 Feb 2025 13:32:43 +0000 (14:32 +0100)]
Merge pull request #3655 from lonvia/remove-name-ranking-in-postcode-search

Tweak penalties for postcode searches

5 months agoremove address penalty for postcode search
Sarah Hoffmann [Thu, 20 Feb 2025 10:05:28 +0000 (11:05 +0100)]
remove address penalty for postcode search

Searches of the form <postcode> <city> are in fact quite common.

5 months agodo not rerank address by full match in postcode search
Sarah Hoffmann [Thu, 20 Feb 2025 09:15:03 +0000 (10:15 +0100)]
do not rerank address by full match in postcode search

The reranking result will not be completely correct because
the address of a postcode refer to the address _and_ name
of the parent and reranking was only done against the
address. We assume here that the postcode is precise enough
as to not require a penalty to to partial matches.

5 months agoMerge pull request #3653 from lonvia/trailing-spaces-in-normalization
Sarah Hoffmann [Wed, 19 Feb 2025 16:25:59 +0000 (17:25 +0100)]
Merge pull request #3653 from lonvia/trailing-spaces-in-normalization

Strip leading and trailing space markers during normalization

5 months agostrip normalisation results of normal and special spaces
Sarah Hoffmann [Wed, 19 Feb 2025 09:26:06 +0000 (10:26 +0100)]
strip normalisation results of normal and special spaces

5 months agoadd test for structured query with leading spaces
Sarah Hoffmann [Wed, 19 Feb 2025 08:38:10 +0000 (09:38 +0100)]
add test for structured query with leading spaces

5 months agoMerge pull request #3652 from lonvia/update-variants
Sarah Hoffmann [Tue, 18 Feb 2025 18:47:45 +0000 (19:47 +0100)]
Merge pull request #3652 from lonvia/update-variants

Cleanup and updates of tokenizer variant configuration

5 months agoMerge pull request #3649 from lonvia/actions-move-to-ubuntu22
Sarah Hoffmann [Tue, 18 Feb 2025 12:21:09 +0000 (13:21 +0100)]
Merge pull request #3649 from lonvia/actions-move-to-ubuntu22

Move Github actions to Unbuntu-22 image

5 months agoremove e-ë mutation, this is taken care of by transliteration
Sarah Hoffmann [Tue, 18 Feb 2025 09:31:44 +0000 (10:31 +0100)]
remove e-ë mutation, this is taken care of by transliteration