hans/Nominatim
1
0
Fork 1
mirror of https://github.com/osm-search/Nominatim.git synced 2026-03-11 13:24:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

move postcode normalization into tokenizer

Browse Source
This commit is contained in:
Sarah Hoffmann
2021-04-25 18:26:36 +02:00
parent d8ed1bfc60
commit ffc2d82b0e
12 changed files with 181 additions and 74 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
nominatim/indexer/indexer.py
View File

@@ -147,7 +147,7 @@ class Indexer:
if maxrank == 30:
self._index(runners.RankRunner(0, analyzer))
self._index(runners.InterpolationRunner(), 20)
self._index(runners.InterpolationRunner(analyzer), 20)
self._index(runners.RankRunner(30, analyzer), 20)
else:
self._index(runners.RankRunner(maxrank, analyzer))

31
nominatim/indexer/runners.py
View File

@@ -25,7 +25,7 @@ class AbstractPlacexRunner:
SET indexed_status = 0, address = v.addr, token_info = v.ti
FROM (VALUES {}) as v(id, addr, ti)
WHERE place_id = v.id
""".format(','.join(["(%s, %s::hstore, %s::json)"] * num_places))
""".format(','.join(["(%s, %s::hstore, %s::jsonb)"] * num_places))
def index_places(self, worker, places):
@@ -82,6 +82,10 @@ class InterpolationRunner:
location_property_osmline.
"""
def __init__(self, analyzer):
self.analyzer = analyzer
@staticmethod
def name():
return "interpolation lines (location_property_osmline)"
@@ -93,15 +97,30 @@ class InterpolationRunner:
@staticmethod
def sql_get_objects():
return """SELECT place_id FROM location_property_osmline
return """SELECT place_id, get_interpolation_address(address, osm_id) as address
FROM location_property_osmline
WHERE indexed_status > 0
ORDER BY geometry_sector"""
@staticmethod
def index_places(worker, ids):
worker.perform(""" UPDATE location_property_osmline
SET indexed_status = 0 WHERE place_id IN ({})
""".format(','.join((str(i[0]) for i in ids))))
@functools.lru_cache(maxsize=1)
def _index_sql(num_places):
return """ UPDATE location_property_osmline
SET indexed_status = 0, address = v.addr, token_info = v.ti
FROM (VALUES {}) as v(id, addr, ti)
WHERE place_id = v.id
""".format(','.join(["(%s, %s::hstore, %s::jsonb)"] * num_places))
def index_places(self, worker, places):
values = []
for place in places:
values.extend((place[x] for x in ('place_id', 'address')))
values.append(psycopg2.extras.Json(self.analyzer.process_place(place)))
worker.perform(self._index_sql(len(places)), values)
class PostcodeRunner: