nominatim/api/search/db_search_builder.py

   1 # SPDX-License-Identifier: GPL-3.0-or-later
   2 #
   3 # This file is part of Nominatim. (https://nominatim.org)
   4 #
   5 # Copyright (C) 2023 by the Nominatim developer community.
   6 # For a full list of authors see the git log.
   7 """
   8 Conversion from token assignment to an abstract DB search.
   9 """
  10 from typing import Optional, List, Tuple, Iterator, Dict
  11 import heapq
  12
  13 from nominatim.api.types import SearchDetails, DataLayer
  14 from nominatim.api.search.query import QueryStruct, Token, TokenType, TokenRange, BreakType
  15 from nominatim.api.search.token_assignment import TokenAssignment
  16 import nominatim.api.search.db_search_fields as dbf
  17 import nominatim.api.search.db_searches as dbs
  18 import nominatim.api.search.db_search_lookups as lookups
  19
  20
  21 def wrap_near_search(categories: List[Tuple[str, str]],
  22                      search: dbs.AbstractSearch) -> dbs.NearSearch:
  23     """ Create a new search that wraps the given search in a search
  24         for near places of the given category.
  25     """
  26     return dbs.NearSearch(penalty=search.penalty,
  27                           categories=dbf.WeightedCategories(categories,
  28                                                             [0.0] * len(categories)),
  29                           search=search)
  30
  31
  32 def build_poi_search(category: List[Tuple[str, str]],
  33                      countries: Optional[List[str]]) -> dbs.PoiSearch:
  34     """ Create a new search for places by the given category, possibly
  35         constraint to the given countries.
  36     """
  37     if countries:
  38         ccs = dbf.WeightedStrings(countries, [0.0] * len(countries))
  39     else:
  40         ccs = dbf.WeightedStrings([], [])
  41
  42     class _PoiData(dbf.SearchData):
  43         penalty = 0.0
  44         qualifiers = dbf.WeightedCategories(category, [0.0] * len(category))
  45         countries=ccs
  46
  47     return dbs.PoiSearch(_PoiData())
  48
  49
  50 class SearchBuilder:
  51     """ Build the abstract search queries from token assignments.
  52     """
  53
  54     def __init__(self, query: QueryStruct, details: SearchDetails) -> None:
  55         self.query = query
  56         self.details = details
  57
  58
  59     @property
  60     def configured_for_country(self) -> bool:
  61         """ Return true if the search details are configured to
  62             allow countries in the result.
  63         """
  64         return self.details.min_rank <= 4 and self.details.max_rank >= 4 \
  65                and self.details.layer_enabled(DataLayer.ADDRESS)
  66
  67
  68     @property
  69     def configured_for_postcode(self) -> bool:
  70         """ Return true if the search details are configured to
  71             allow postcodes in the result.
  72         """
  73         return self.details.min_rank <= 5 and self.details.max_rank >= 11\
  74                and self.details.layer_enabled(DataLayer.ADDRESS)
  75
  76
  77     @property
  78     def configured_for_housenumbers(self) -> bool:
  79         """ Return true if the search details are configured to
  80             allow addresses in the result.
  81         """
  82         return self.details.max_rank >= 30 \
  83                and self.details.layer_enabled(DataLayer.ADDRESS)
  84
  85
  86     def build(self, assignment: TokenAssignment) -> Iterator[dbs.AbstractSearch]:
  87         """ Yield all possible abstract searches for the given token assignment.
  88         """
  89         sdata = self.get_search_data(assignment)
  90         if sdata is None:
  91             return
  92
  93         near_items = self.get_near_items(assignment)
  94         if near_items is not None and not near_items:
  95             return # impossible compbination of near items and category parameter
  96
  97         if assignment.name is None:
  98             if near_items and not sdata.postcodes:
  99                 sdata.qualifiers = near_items
 100                 near_items = None
 101                 builder = self.build_poi_search(sdata)
 102             elif assignment.housenumber:
 103                 hnr_tokens = self.query.get_tokens(assignment.housenumber,
 104                                                    TokenType.HOUSENUMBER)
 105                 builder = self.build_housenumber_search(sdata, hnr_tokens, assignment.address)
 106             else:
 107                 builder = self.build_special_search(sdata, assignment.address,
 108                                                     bool(near_items))
 109         else:
 110             builder = self.build_name_search(sdata, assignment.name, assignment.address,
 111                                              bool(near_items))
 112
 113         if near_items:
 114             penalty = min(near_items.penalties)
 115             near_items.penalties = [p - penalty for p in near_items.penalties]
 116             for search in builder:
 117                 search_penalty = search.penalty
 118                 search.penalty = 0.0
 119                 yield dbs.NearSearch(penalty + assignment.penalty + search_penalty,
 120                                      near_items, search)
 121         else:
 122             for search in builder:
 123                 search.penalty += assignment.penalty
 124                 yield search
 125
 126
 127     def build_poi_search(self, sdata: dbf.SearchData) -> Iterator[dbs.AbstractSearch]:
 128         """ Build abstract search query for a simple category search.
 129             This kind of search requires an additional geographic constraint.
 130         """
 131         if not sdata.housenumbers \
 132            and ((self.details.viewbox and self.details.bounded_viewbox) or self.details.near):
 133             yield dbs.PoiSearch(sdata)
 134
 135
 136     def build_special_search(self, sdata: dbf.SearchData,
 137                              address: List[TokenRange],
 138                              is_category: bool) -> Iterator[dbs.AbstractSearch]:
 139         """ Build abstract search queries for searches that do not involve
 140             a named place.
 141         """
 142         if sdata.qualifiers:
 143             # No special searches over qualifiers supported.
 144             return
 145
 146         if sdata.countries and not address and not sdata.postcodes \
 147            and self.configured_for_country:
 148             yield dbs.CountrySearch(sdata)
 149
 150         if sdata.postcodes and (is_category or self.configured_for_postcode):
 151             penalty = 0.0 if sdata.countries else 0.1
 152             if address:
 153                 sdata.lookups = [dbf.FieldLookup('nameaddress_vector',
 154                                                  [t.token for r in address
 155                                                   for t in self.query.get_partials_list(r)],
 156                                                  lookups.Restrict)]
 157                 penalty += 0.2
 158             yield dbs.PostcodeSearch(penalty, sdata)
 159
 160
 161     def build_housenumber_search(self, sdata: dbf.SearchData, hnrs: List[Token],
 162                                  address: List[TokenRange]) -> Iterator[dbs.AbstractSearch]:
 163         """ Build a simple address search for special entries where the
 164             housenumber is the main name token.
 165         """
 166         sdata.lookups = [dbf.FieldLookup('name_vector', [t.token for t in hnrs], lookups.LookupAny)]
 167         expected_count = sum(t.count for t in hnrs)
 168
 169         partials = {t.token: t.count for trange in address
 170                        for t in self.query.get_partials_list(trange)}
 171
 172         if expected_count < 8000:
 173             sdata.lookups.append(dbf.FieldLookup('nameaddress_vector',
 174                                                  list(partials), lookups.Restrict))
 175         elif len(partials) != 1 or list(partials.values())[0] < 10000:
 176             sdata.lookups.append(dbf.FieldLookup('nameaddress_vector',
 177                                                  list(partials), lookups.LookupAll))
 178         else:
 179             addr_fulls = [t.token for t
 180                           in self.query.get_tokens(address[0], TokenType.WORD)]
 181             if len(addr_fulls) > 5:
 182                 return
 183             sdata.lookups.append(
 184                 dbf.FieldLookup('nameaddress_vector', addr_fulls, lookups.LookupAny))
 185
 186         sdata.housenumbers = dbf.WeightedStrings([], [])
 187         yield dbs.PlaceSearch(0.05, sdata, expected_count)
 188
 189
 190     def build_name_search(self, sdata: dbf.SearchData,
 191                           name: TokenRange, address: List[TokenRange],
 192                           is_category: bool) -> Iterator[dbs.AbstractSearch]:
 193         """ Build abstract search queries for simple name or address searches.
 194         """
 195         if is_category or not sdata.housenumbers or self.configured_for_housenumbers:
 196             ranking = self.get_name_ranking(name)
 197             name_penalty = ranking.normalize_penalty()
 198             if ranking.rankings:
 199                 sdata.rankings.append(ranking)
 200             for penalty, count, lookup in self.yield_lookups(name, address):
 201                 sdata.lookups = lookup
 202                 yield dbs.PlaceSearch(penalty + name_penalty, sdata, count)
 203
 204
 205     def yield_lookups(self, name: TokenRange, address: List[TokenRange])\
 206                           -> Iterator[Tuple[float, int, List[dbf.FieldLookup]]]:
 207         """ Yield all variants how the given name and address should best
 208             be searched for. This takes into account how frequent the terms
 209             are and tries to find a lookup that optimizes index use.
 210         """
 211         penalty = 0.0 # extra penalty
 212         name_partials = {t.token: t for t in self.query.get_partials_list(name)}
 213
 214         addr_partials = [t for r in address for t in self.query.get_partials_list(r)]
 215         addr_tokens = list({t.token for t in addr_partials})
 216
 217         partials_indexed = all(t.is_indexed for t in name_partials.values()) \
 218                            and all(t.is_indexed for t in addr_partials)
 219         exp_count = min(t.count for t in name_partials.values()) / (2**(len(name_partials) - 1))
 220
 221         if (len(name_partials) > 3 or exp_count < 8000) and partials_indexed:
 222             yield penalty, exp_count, dbf.lookup_by_names(list(name_partials.keys()), addr_tokens)
 223             return
 224
 225         # Partial term to frequent. Try looking up by rare full names first.
 226         name_fulls = self.query.get_tokens(name, TokenType.WORD)
 227         if name_fulls:
 228             fulls_count = sum(t.count for t in name_fulls)
 229             if len(name_partials) == 1:
 230                 penalty += min(1, max(0, (exp_count - 50 * fulls_count) / (1000 * fulls_count)))
 231             # At this point drop unindexed partials from the address.
 232             # This might yield wrong results, nothing we can do about that.
 233             if not partials_indexed:
 234                 addr_tokens = [t.token for t in addr_partials if t.is_indexed]
 235                 penalty += 1.2 * sum(t.penalty for t in addr_partials if not t.is_indexed)
 236             # Any of the full names applies with all of the partials from the address
 237             yield penalty, fulls_count / (2**len(addr_tokens)),\
 238                   dbf.lookup_by_any_name([t.token for t in name_fulls],
 239                                          addr_tokens,
 240                                          fulls_count > 30000 / max(1, len(addr_tokens)))
 241
 242         # To catch remaining results, lookup by name and address
 243         # We only do this if there is a reasonable number of results expected.
 244         exp_count = exp_count / (2**len(addr_tokens)) if addr_tokens else exp_count
 245         if exp_count < 10000 and all(t.is_indexed for t in name_partials.values()):
 246             lookup = [dbf.FieldLookup('name_vector', list(name_partials.keys()), lookups.LookupAll)]
 247             if addr_tokens:
 248                 lookup.append(dbf.FieldLookup('nameaddress_vector', addr_tokens, lookups.LookupAll))
 249             penalty += 0.35 * max(1 if name_fulls else 0.1,
 250                                   5 - len(name_partials) - len(addr_tokens))
 251             yield penalty, exp_count, lookup
 252
 253
 254     def get_name_ranking(self, trange: TokenRange,
 255                          db_field: str = 'name_vector') -> dbf.FieldRanking:
 256         """ Create a ranking expression for a name term in the given range.
 257         """
 258         name_fulls = self.query.get_tokens(trange, TokenType.WORD)
 259         ranks = [dbf.RankedTokens(t.penalty, [t.token]) for t in name_fulls]
 260         ranks.sort(key=lambda r: r.penalty)
 261         # Fallback, sum of penalty for partials
 262         name_partials = self.query.get_partials_list(trange)
 263         default = sum(t.penalty for t in name_partials) + 0.2
 264         return dbf.FieldRanking(db_field, default, ranks)
 265
 266
 267     def get_addr_ranking(self, trange: TokenRange) -> dbf.FieldRanking:
 268         """ Create a list of ranking expressions for an address term
 269             for the given ranges.
 270         """
 271         todo: List[Tuple[int, int, dbf.RankedTokens]] = []
 272         heapq.heappush(todo, (0, trange.start, dbf.RankedTokens(0.0, [])))
 273         ranks: List[dbf.RankedTokens] = []
 274
 275         while todo: # pylint: disable=too-many-nested-blocks
 276             neglen, pos, rank = heapq.heappop(todo)
 277             for tlist in self.query.nodes[pos].starting:
 278                 if tlist.ttype in (TokenType.PARTIAL, TokenType.WORD):
 279                     if tlist.end < trange.end:
 280                         chgpenalty = PENALTY_WORDCHANGE[self.query.nodes[tlist.end].btype]
 281                         if tlist.ttype == TokenType.PARTIAL:
 282                             penalty = rank.penalty + chgpenalty \
 283                                       + max(t.penalty for t in tlist.tokens)
 284                             heapq.heappush(todo, (neglen - 1, tlist.end,
 285                                                   dbf.RankedTokens(penalty, rank.tokens)))
 286                         else:
 287                             for t in tlist.tokens:
 288                                 heapq.heappush(todo, (neglen - 1, tlist.end,
 289                                                       rank.with_token(t, chgpenalty)))
 290                     elif tlist.end == trange.end:
 291                         if tlist.ttype == TokenType.PARTIAL:
 292                             ranks.append(dbf.RankedTokens(rank.penalty
 293                                                           + max(t.penalty for t in tlist.tokens),
 294                                                           rank.tokens))
 295                         else:
 296                             ranks.extend(rank.with_token(t, 0.0) for t in tlist.tokens)
 297                         if len(ranks) >= 10:
 298                             # Too many variants, bail out and only add
 299                             # Worst-case Fallback: sum of penalty of partials
 300                             name_partials = self.query.get_partials_list(trange)
 301                             default = sum(t.penalty for t in name_partials) + 0.2
 302                             ranks.append(dbf.RankedTokens(rank.penalty + default, []))
 303                             # Bail out of outer loop
 304                             todo.clear()
 305                             break
 306
 307         ranks.sort(key=lambda r: len(r.tokens))
 308         default = ranks[0].penalty + 0.3
 309         del ranks[0]
 310         ranks.sort(key=lambda r: r.penalty)
 311
 312         return dbf.FieldRanking('nameaddress_vector', default, ranks)
 313
 314
 315     def get_search_data(self, assignment: TokenAssignment) -> Optional[dbf.SearchData]:
 316         """ Collect the tokens for the non-name search fields in the
 317             assignment.
 318         """
 319         sdata = dbf.SearchData()
 320         sdata.penalty = assignment.penalty
 321         if assignment.country:
 322             tokens = self.get_country_tokens(assignment.country)
 323             if not tokens:
 324                 return None
 325             sdata.set_strings('countries', tokens)
 326         elif self.details.countries:
 327             sdata.countries = dbf.WeightedStrings(self.details.countries,
 328                                                   [0.0] * len(self.details.countries))
 329         if assignment.housenumber:
 330             sdata.set_strings('housenumbers',
 331                               self.query.get_tokens(assignment.housenumber,
 332                                                     TokenType.HOUSENUMBER))
 333         if assignment.postcode:
 334             sdata.set_strings('postcodes',
 335                               self.query.get_tokens(assignment.postcode,
 336                                                     TokenType.POSTCODE))
 337         if assignment.qualifier:
 338             tokens = self.get_qualifier_tokens(assignment.qualifier)
 339             if not tokens:
 340                 return None
 341             sdata.set_qualifiers(tokens)
 342         elif self.details.categories:
 343             sdata.qualifiers = dbf.WeightedCategories(self.details.categories,
 344                                                       [0.0] * len(self.details.categories))
 345
 346         if assignment.address:
 347             if not assignment.name and assignment.housenumber:
 348                 # housenumber search: the first item needs to be handled like
 349                 # a name in ranking or penalties are not comparable with
 350                 # normal searches.
 351                 sdata.set_ranking([self.get_name_ranking(assignment.address[0],
 352                                                          db_field='nameaddress_vector')]
 353                                   + [self.get_addr_ranking(r) for r in assignment.address[1:]])
 354             else:
 355                 sdata.set_ranking([self.get_addr_ranking(r) for r in assignment.address])
 356         else:
 357             sdata.rankings = []
 358
 359         return sdata
 360
 361
 362     def get_country_tokens(self, trange: TokenRange) -> List[Token]:
 363         """ Return the list of country tokens for the given range,
 364             optionally filtered by the country list from the details
 365             parameters.
 366         """
 367         tokens = self.query.get_tokens(trange, TokenType.COUNTRY)
 368         if self.details.countries:
 369             tokens = [t for t in tokens if t.lookup_word in self.details.countries]
 370
 371         return tokens
 372
 373
 374     def get_qualifier_tokens(self, trange: TokenRange) -> List[Token]:
 375         """ Return the list of qualifier tokens for the given range,
 376             optionally filtered by the qualifier list from the details
 377             parameters.
 378         """
 379         tokens = self.query.get_tokens(trange, TokenType.QUALIFIER)
 380         if self.details.categories:
 381             tokens = [t for t in tokens if t.get_category() in self.details.categories]
 382
 383         return tokens
 384
 385
 386     def get_near_items(self, assignment: TokenAssignment) -> Optional[dbf.WeightedCategories]:
 387         """ Collect tokens for near items search or use the categories
 388             requested per parameter.
 389             Returns None if no category search is requested.
 390         """
 391         if assignment.near_item:
 392             tokens: Dict[Tuple[str, str], float] = {}
 393             for t in self.query.get_tokens(assignment.near_item, TokenType.NEAR_ITEM):
 394                 cat = t.get_category()
 395                 # The category of a near search will be that of near_item.
 396                 # Thus, if search is restricted to a category parameter,
 397                 # the two sets must intersect.
 398                 if (not self.details.categories or cat in self.details.categories)\
 399                    and t.penalty < tokens.get(cat, 1000.0):
 400                     tokens[cat] = t.penalty
 401             return dbf.WeightedCategories(list(tokens.keys()), list(tokens.values()))
 402
 403         return None
 404
 405
 406 PENALTY_WORDCHANGE = {
 407     BreakType.START: 0.0,
 408     BreakType.END: 0.0,
 409     BreakType.PHRASE: 0.0,
 410     BreakType.WORD: 0.1,
 411     BreakType.PART: 0.2,
 412     BreakType.TOKEN: 0.4
 413 }