hans/Nominatim
1
0
Fork 1
mirror of https://github.com/osm-search/Nominatim.git synced 2026-03-12 05:44:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

adapt housenumber cleanup to new word table structure

Browse Source
This commit is contained in:
Sarah Hoffmann
2022-02-18 16:43:17 +01:00
parent 89e1446131
commit 4a3bbd0319
1 changed files with 4 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
nominatim/tokenizer/icu_tokenizer.py
View File

@@ -119,12 +119,12 @@ class LegacyICUTokenizer(AbstractTokenizer):
if not conn.table_exists('search_name'): if not conn.table_exists('search_name'):
return return
with conn.cursor(name="hnr_counter") as cur: with conn.cursor(name="hnr_counter") as cur:
cur.execute("""SELECT word_id, word_token FROM word cur.execute("""SELECT DISTINCT word_id, coalesce(info->>'lookup', word_token) FROM word
WHERE type = 'H' WHERE type = 'H'
AND NOT EXISTS(SELECT * FROM search_name AND NOT EXISTS(SELECT * FROM search_name
WHERE ARRAY[word.word_id] && name_vector) WHERE ARRAY[word.word_id] && name_vector)
AND (char_length(word_token) > 6 AND (char_length(coalesce(word, word_token)) > 6
OR word_token not similar to '\\d+') OR coalesce(word, word_token) not similar to '\\d+')
""") """)
candidates = {token: wid for wid, token in cur} candidates = {token: wid for wid, token in cur}
with conn.cursor(name="hnr_counter") as cur: with conn.cursor(name="hnr_counter") as cur:
@@ -137,6 +137,7 @@ class LegacyICUTokenizer(AbstractTokenizer):
for hnr in row[0].split(';'): for hnr in row[0].split(';'):
candidates.pop(hnr, None) candidates.pop(hnr, None)
LOG.info("There are %s outdated housenumbers.", len(candidates)) LOG.info("There are %s outdated housenumbers.", len(candidates))
LOG.debug("Outdated housenumbers: %s", candidates.keys())
if candidates: if candidates:
with conn.cursor() as cur: with conn.cursor() as cur:
cur.execute("""DELETE FROM word WHERE word_id = any(%s)""", cur.execute("""DELETE FROM word WHERE word_id = any(%s)""",