X-Git-Url: https://git.openstreetmap.org./nominatim.git/blobdiff_plain/cf98cff2a166eb35ed2c946e03f1610069bdd1d4..231250f2eb272b77d54e4b4b18bd85a80413ac34:/nominatim/tokenizer/legacy_tokenizer.py diff --git a/nominatim/tokenizer/legacy_tokenizer.py b/nominatim/tokenizer/legacy_tokenizer.py index a68b69ec..8bfb309d 100644 --- a/nominatim/tokenizer/legacy_tokenizer.py +++ b/nominatim/tokenizer/legacy_tokenizer.py @@ -16,6 +16,7 @@ from nominatim.db import properties from nominatim.db import utils as db_utils from nominatim.db.sql_preprocessor import SQLPreprocessor from nominatim.errors import UsageError +from nominatim.tokenizer.base import AbstractAnalyzer, AbstractTokenizer DBCFG_NORMALIZATION = "tokenizer_normalization" DBCFG_MAXWORDFREQ = "tokenizer_maxwordfreq" @@ -76,7 +77,7 @@ def _check_module(module_dir, conn): raise UsageError("Database module cannot be accessed.") from err -class LegacyTokenizer: +class LegacyTokenizer(AbstractTokenizer): """ The legacy tokenizer uses a special PostgreSQL module to normalize names and queries. The tokenizer thus implements normalization through calls to the database. @@ -238,7 +239,7 @@ class LegacyTokenizer: properties.set_property(conn, DBCFG_MAXWORDFREQ, config.MAX_WORD_FREQUENCY) -class LegacyNameAnalyzer: +class LegacyNameAnalyzer(AbstractAnalyzer): """ The legacy analyzer uses the special Postgresql module for splitting names. @@ -255,14 +256,6 @@ class LegacyNameAnalyzer: self._cache = _TokenCache(self.conn) - def __enter__(self): - return self - - - def __exit__(self, exc_type, exc_value, traceback): - self.close() - - def close(self): """ Free all resources used by the analyzer. """ @@ -370,8 +363,7 @@ class LegacyNameAnalyzer: to_delete = existing_phrases - norm_phrases if to_add: - psycopg2.extras.execute_values( - cur, + cur.execute_values( """ INSERT INTO word (word_id, word_token, word, class, type, search_name_count, operator) (SELECT nextval('seq_word'), ' ' || make_standard_name(name), name, @@ -381,8 +373,7 @@ class LegacyNameAnalyzer: to_add) if to_delete and should_replace: - psycopg2.extras.execute_values( - cur, + cur.execute_values( """ DELETE FROM word USING (VALUES %s) as v(name, in_class, in_type, op) WHERE word = name and class = in_class and type = in_type and ((op = '-' and operator is null) or op = operator)""", @@ -414,16 +405,16 @@ class LegacyNameAnalyzer: """ token_info = _TokenInfo(self._cache) - names = place.get('name') + names = place.name if names: token_info.add_names(self.conn, names) - country_feature = place.get('country_feature') + 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) - address = place.get('address') + address = place.address if address: self._process_place_address(token_info, address)