X-Git-Url: https://git.openstreetmap.org./nominatim.git/blobdiff_plain/3266daa8fde98acc1fe4c9929cb5be3aed662add..8a1af9b56659d4ef956f45da2928687a17dea20a:/nominatim/api/search/db_search_builder.py diff --git a/nominatim/api/search/db_search_builder.py b/nominatim/api/search/db_search_builder.py index 67db3247..c755f2a7 100644 --- a/nominatim/api/search/db_search_builder.py +++ b/nominatim/api/search/db_search_builder.py @@ -7,7 +7,7 @@ """ Convertion from token assignment to an abstract DB search. """ -from typing import Optional, List, Tuple, Iterator +from typing import Optional, List, Tuple, Iterator, Dict import heapq from nominatim.api.types import SearchDetails, DataLayer @@ -15,7 +15,6 @@ from nominatim.api.search.query import QueryStruct, Token, TokenType, TokenRange from nominatim.api.search.token_assignment import TokenAssignment import nominatim.api.search.db_search_fields as dbf import nominatim.api.search.db_searches as dbs -from nominatim.api.logging import log def wrap_near_search(categories: List[Tuple[str, str]], @@ -90,12 +89,14 @@ class SearchBuilder: if sdata is None: return - categories = self.get_search_categories(assignment) + near_items = self.get_near_items(assignment) + if near_items is not None and not near_items: + return # impossible compbination of near items and category parameter if assignment.name is None: - if categories and not sdata.postcodes: - sdata.qualifiers = categories - categories = None + if near_items and not sdata.postcodes: + sdata.qualifiers = near_items + near_items = None builder = self.build_poi_search(sdata) elif assignment.housenumber: hnr_tokens = self.query.get_tokens(assignment.housenumber, @@ -103,18 +104,23 @@ class SearchBuilder: builder = self.build_housenumber_search(sdata, hnr_tokens, assignment.address) else: builder = self.build_special_search(sdata, assignment.address, - bool(categories)) + bool(near_items)) else: builder = self.build_name_search(sdata, assignment.name, assignment.address, - bool(categories)) + bool(near_items)) - if categories: - penalty = min(categories.penalties) - categories.penalties = [p - penalty for p in categories.penalties] + if near_items: + penalty = min(near_items.penalties) + near_items.penalties = [p - penalty for p in near_items.penalties] for search in builder: - yield dbs.NearSearch(penalty, categories, search) + search_penalty = search.penalty + search.penalty = 0.0 + yield dbs.NearSearch(penalty + assignment.penalty + search_penalty, + near_items, search) else: - yield from builder + for search in builder: + search.penalty += assignment.penalty + yield search def build_poi_search(self, sdata: dbf.SearchData) -> Iterator[dbs.AbstractSearch]: @@ -156,14 +162,27 @@ class SearchBuilder: """ Build a simple address search for special entries where the housenumber is the main name token. """ - partial_tokens: List[int] = [] - for trange in address: - partial_tokens.extend(t.token for t in self.query.get_partials_list(trange)) + sdata.lookups = [dbf.FieldLookup('name_vector', [t.token for t in hnrs], 'lookup_any')] + expected_count = sum(t.count for t in hnrs) + + partials = [t for trange in address + for t in self.query.get_partials_list(trange)] + + if expected_count < 8000: + sdata.lookups.append(dbf.FieldLookup('nameaddress_vector', + [t.token for t in partials], 'restrict')) + elif len(partials) != 1 or partials[0].count < 10000: + sdata.lookups.append(dbf.FieldLookup('nameaddress_vector', + [t.token for t in partials], 'lookup_all')) + else: + sdata.lookups.append( + dbf.FieldLookup('nameaddress_vector', + [t.token for t + in self.query.get_tokens(address[0], TokenType.WORD)], + 'lookup_any')) - sdata.lookups = [dbf.FieldLookup('name_vector', [t.token for t in hnrs], 'lookup_any'), - dbf.FieldLookup('nameaddress_vector', partial_tokens, 'lookup_all') - ] - yield dbs.PlaceSearch(0.05, sdata, sum(t.count for t in hnrs)) + sdata.housenumbers = dbf.WeightedStrings([], []) + yield dbs.PlaceSearch(0.05, sdata, expected_count) def build_name_search(self, sdata: dbf.SearchData, @@ -187,67 +206,44 @@ class SearchBuilder: be searched for. This takes into account how frequent the terms are and tries to find a lookup that optimizes index use. """ - penalty = 0.0 # extra penalty currently unused - + penalty = 0.0 # extra penalty name_partials = self.query.get_partials_list(name) - exp_name_count = min(t.count for t in name_partials) - addr_partials = [] - for trange in address: - addr_partials.extend(self.query.get_partials_list(trange)) + name_tokens = [t.token for t in name_partials] + + addr_partials = [t for r in address for t in self.query.get_partials_list(r)] addr_tokens = [t.token for t in addr_partials] + partials_indexed = all(t.is_indexed for t in name_partials) \ and all(t.is_indexed for t in addr_partials) + exp_count = min(t.count for t in name_partials) / (2**(len(name_partials) - 1)) - if (len(name_partials) > 3 or exp_name_count < 1000) and partials_indexed: - # Lookup by name partials, use address partials to restrict results. - lookup = [dbf.FieldLookup('name_vector', - [t.token for t in name_partials], 'lookup_all')] - if addr_tokens: - lookup.append(dbf.FieldLookup('nameaddress_vector', addr_tokens, 'restrict')) - yield penalty, exp_name_count, lookup - return - - exp_addr_count = min(t.count for t in addr_partials) if addr_partials else exp_name_count - if exp_addr_count < 1000 and partials_indexed: - # Lookup by address partials and restrict results through name terms. - # Give this a small penalty because lookups in the address index are - # more expensive - yield penalty + exp_addr_count/5000, exp_addr_count,\ - [dbf.FieldLookup('name_vector', [t.token for t in name_partials], 'restrict'), - dbf.FieldLookup('nameaddress_vector', addr_tokens, 'lookup_all')] + if (len(name_partials) > 3 or exp_count < 8000) and partials_indexed: + yield penalty, exp_count, dbf.lookup_by_names(name_tokens, addr_tokens) return # Partial term to frequent. Try looking up by rare full names first. name_fulls = self.query.get_tokens(name, TokenType.WORD) - rare_names = list(filter(lambda t: t.count < 1000, name_fulls)) - # At this point drop unindexed partials from the address. - # This might yield wrong results, nothing we can do about that. - if not partials_indexed: - addr_tokens = [t.token for t in addr_partials if t.is_indexed] - log().var_dump('before', penalty) - penalty += 1.2 * sum(t.penalty for t in addr_partials if not t.is_indexed) - log().var_dump('after', penalty) - if rare_names: + if name_fulls: + fulls_count = sum(t.count for t in name_fulls) + # At this point drop unindexed partials from the address. + # This might yield wrong results, nothing we can do about that. + if not partials_indexed: + addr_tokens = [t.token for t in addr_partials if t.is_indexed] + penalty += 1.2 * sum(t.penalty for t in addr_partials if not t.is_indexed) # Any of the full names applies with all of the partials from the address - lookup = [dbf.FieldLookup('name_vector', [t.token for t in rare_names], 'lookup_any')] - if addr_tokens: - lookup.append(dbf.FieldLookup('nameaddress_vector', addr_tokens, 'restrict')) - yield penalty, sum(t.count for t in rare_names), lookup + yield penalty, fulls_count / (2**len(addr_partials)),\ + dbf.lookup_by_any_name([t.token for t in name_fulls], addr_tokens, + 'restrict' if fulls_count < 10000 else 'lookup_all') # To catch remaining results, lookup by name and address - if all(t.is_indexed for t in name_partials): - lookup = [dbf.FieldLookup('name_vector', - [t.token for t in name_partials], 'lookup_all')] - else: - # we don't have the partials, try with the non-rare names - non_rare_names = [t.token for t in name_fulls if t.count >= 1000] - if not non_rare_names: - return - lookup = [dbf.FieldLookup('name_vector', non_rare_names, 'lookup_any')] - if addr_tokens: - lookup.append(dbf.FieldLookup('nameaddress_vector', addr_tokens, 'lookup_all')) - yield penalty + 0.1 * max(0, 5 - len(name_partials) - len(addr_tokens)),\ - min(exp_name_count, exp_addr_count), lookup + # We only do this if there is a reasonable number of results expected. + exp_count = exp_count / (2**len(addr_partials)) if addr_partials else exp_count + if exp_count < 10000 and all(t.is_indexed for t in name_partials): + lookup = [dbf.FieldLookup('name_vector', name_tokens, 'lookup_all')] + if addr_tokens: + lookup.append(dbf.FieldLookup('nameaddress_vector', addr_tokens, 'lookup_all')) + penalty += 0.35 * max(0, 5 - len(name_partials) - len(addr_tokens)) + yield penalty, exp_count, lookup def get_name_ranking(self, trange: TokenRange) -> dbf.FieldRanking: @@ -335,8 +331,15 @@ class SearchBuilder: self.query.get_tokens(assignment.postcode, TokenType.POSTCODE)) if assignment.qualifier: - sdata.set_qualifiers(self.query.get_tokens(assignment.qualifier, - TokenType.QUALIFIER)) + tokens = self.query.get_tokens(assignment.qualifier, TokenType.QUALIFIER) + if self.details.categories: + tokens = [t for t in tokens if t.get_category() in self.details.categories] + if not tokens: + return None + sdata.set_qualifiers(tokens) + elif self.details.categories: + sdata.qualifiers = dbf.WeightedCategories(self.details.categories, + [0.0] * len(self.details.categories)) if assignment.address: sdata.set_ranking([self.get_addr_ranking(r) for r in assignment.address]) @@ -346,23 +349,22 @@ class SearchBuilder: return sdata - def get_search_categories(self, - assignment: TokenAssignment) -> Optional[dbf.WeightedCategories]: - """ Collect tokens for category search or use the categories + def get_near_items(self, assignment: TokenAssignment) -> Optional[dbf.WeightedCategories]: + """ Collect tokens for near items search or use the categories requested per parameter. Returns None if no category search is requested. """ - if assignment.category: - tokens = [t for t in self.query.get_tokens(assignment.category, - TokenType.CATEGORY) - if not self.details.categories - or t.get_category() in self.details.categories] - return dbf.WeightedCategories([t.get_category() for t in tokens], - [t.penalty for t in tokens]) - - if self.details.categories: - return dbf.WeightedCategories(self.details.categories, - [0.0] * len(self.details.categories)) + if assignment.near_item: + tokens: Dict[Tuple[str, str], float] = {} + for t in self.query.get_tokens(assignment.near_item, TokenType.NEAR_ITEM): + cat = t.get_category() + # The category of a near search will be that of near_item. + # Thus, if search is restricted to a category parameter, + # the two sets must intersect. + if (not self.details.categories or cat in self.details.categories)\ + and t.penalty < tokens.get(cat, 1000.0): + tokens[cat] = t.penalty + return dbf.WeightedCategories(list(tokens.keys()), list(tokens.values())) return None