add Python part for new ICU-based tokenizer

[nominatim.git] / lib-sql / tokenizer / legacy_tokenizer_indices.sql

CREATE INDEX {{sql.if_index_not_exists}} idx_word_word_id
  ON word USING BTREE (word_id) {{db.tablespace.search_index}};