export more data for the tokenizer name preparation

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

View File

@@ -38,7 +38,31 @@ class PlaceInfo:
@property
def country_feature(self):
""" Return the country code if the place is a valid country boundary.
def country_code(self):
""" The country code of the country the place is in. Guaranteed
to be a two-letter lower-case string or None, if no country
could be found.
"""
return self._info.get('country_feature')
return self._info.get('country_code')
@property
def rank_address(self):
""" The computed rank address before rank correction.
"""
return self._info.get('rank_address')
def is_a(self, key, value):
""" Check if the place's primary tag corresponds to the given
key and value.
"""
return self._info.get('class') == key and self._info.get('type') == value
def is_country(self):
""" Check if the place is a valid country boundary.
"""
return self.rank_address == 4 \
and self.is_a('boundary', 'administrative') \
and self.country_code is not None

View File

@@ -39,7 +39,7 @@ class AbstractPlacexRunner:
@staticmethod
def get_place_details(worker, ids):
worker.perform("""SELECT place_id, (placex_prepare_update(placex)).*
worker.perform("""SELECT place_id, (placex_indexing_prepare(placex)).*
FROM placex WHERE place_id IN %s""",
(tuple((p[0] for p in ids)), ))

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:

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: