hans/Nominatim
1
0
Fork 1
mirror of https://github.com/osm-search/Nominatim.git synced 2026-03-09 11:34:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

export more data for the tokenizer name preparation

Browse Source 
Adds class, type, country and rank to the exported information
and removes the rather odd hack for countries. Whether a place
represents a country boundary can now be computed by the tokenizer.
This commit is contained in:
Sarah Hoffmann
2021-09-29 11:54:14 +02:00
parent 231250f2eb
commit be65c8303f
7 changed files with 85 additions and 45 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
nominatim/tokenizer/icu_tokenizer.py
View File

@@ -397,9 +397,8 @@ class LegacyICUNameAnalyzer(AbstractAnalyzer):
token_info.add_names(fulls, partials)
country_feature = place.country_feature
if country_feature and re.fullmatch(r'[A-Za-z][A-Za-z]', country_feature):
self.add_country_names(country_feature.lower(), names)
if place.is_country():
self.add_country_names(place.country_code, names)
address = place.address
if address:

5
nominatim/tokenizer/legacy_tokenizer.py
View File

@@ -410,9 +410,8 @@ class LegacyNameAnalyzer(AbstractAnalyzer):
if names:
token_info.add_names(self.conn, names)
country_feature = place.country_feature
if country_feature and re.fullmatch(r'[A-Za-z][A-Za-z]', country_feature):
self.add_country_names(country_feature.lower(), names)
if place.is_country():
self.add_country_names(place.country_code, names)
address = place.address
if address: