the category parameter in search should result in a qualifier

[nominatim.git] / 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 c0c55a18a428f299aaece8e869f32babc8ed5fb5..35d10e782e3d19026de775270c329c4db8dddb2b 100644 (file)
--- a/nominatim/api/search/db_search_builder.py
+++ b/nominatim/api/search/db_search_builder.py
@@ -7,15 +7,44 @@
 """
 Convertion from token assignment to an abstract DB search.
 """
 """
 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
 import heapq
 
 from nominatim.api.types import SearchDetails, DataLayer

-from nominatim.api.search.query import QueryStruct, TokenType, TokenRange, BreakType
+from nominatim.api.search.query import QueryStruct, Token, TokenType, TokenRange, BreakType

 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.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]],
+                     search: dbs.AbstractSearch) -> dbs.NearSearch:
+    """ Create a new search that wraps the given search in a search
+        for near places of the given category.
+    """
+    return dbs.NearSearch(penalty=search.penalty,
+                          categories=dbf.WeightedCategories(categories,
+                                                            [0.0] * len(categories)),
+                          search=search)
+
+
+def build_poi_search(category: List[Tuple[str, str]],
+                     countries: Optional[List[str]]) -> dbs.PoiSearch:
+    """ Create a new search for places by the given category, possibly
+        constraint to the given countries.
+    """
+    if countries:
+        ccs = dbf.WeightedStrings(countries, [0.0] * len(countries))
+    else:
+        ccs = dbf.WeightedStrings([], [])
+
+    class _PoiData(dbf.SearchData):
+        penalty = 0.0
+        qualifiers = dbf.WeightedCategories(category, [0.0] * len(category))
+        countries=ccs
+
+    return dbs.PoiSearch(_PoiData())
+

 
 class SearchBuilder:
     """ Build the abstract search queries from token assignments.
 
 class SearchBuilder:
     """ Build the abstract search queries from token assignments.


@@ -67,6 +96,10 @@ class SearchBuilder:
                 sdata.qualifiers = categories
                 categories = None
                 builder = self.build_poi_search(sdata)
                 sdata.qualifiers = categories
                 categories = None
                 builder = self.build_poi_search(sdata)

+            elif assignment.housenumber:
+                hnr_tokens = self.query.get_tokens(assignment.housenumber,
+                                                   TokenType.HOUSENUMBER)
+                builder = self.build_housenumber_search(sdata, hnr_tokens, assignment.address)

             else:
                 builder = self.build_special_search(sdata, assignment.address,
                                                     bool(categories))
             else:
                 builder = self.build_special_search(sdata, assignment.address,
                                                     bool(categories))


@@ -78,9 +111,11 @@ class SearchBuilder:
             penalty = min(categories.penalties)
             categories.penalties = [p - penalty for p in categories.penalties]
             for search in builder:
             penalty = min(categories.penalties)
             categories.penalties = [p - penalty for p in categories.penalties]
             for search in builder:

-                yield dbs.NearSearch(penalty, categories, search)
+                yield dbs.NearSearch(penalty + assignment.penalty, categories, search)

         else:
         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]:
 
 
     def build_poi_search(self, sdata: dbf.SearchData) -> Iterator[dbs.AbstractSearch]:


@@ -98,8 +133,8 @@ class SearchBuilder:
         """ Build abstract search queries for searches that do not involve
             a named place.
         """
         """ Build abstract search queries for searches that do not involve
             a named place.
         """

-        if sdata.qualifiers or sdata.housenumbers:
-            # No special searches over housenumbers or qualifiers supported.
+        if sdata.qualifiers:
+            # No special searches over qualifiers supported.

             return
 
         if sdata.countries and not address and not sdata.postcodes \
             return
 
         if sdata.countries and not address and not sdata.postcodes \


@@ -107,12 +142,38 @@ class SearchBuilder:
             yield dbs.CountrySearch(sdata)
 
         if sdata.postcodes and (is_category or self.configured_for_postcode):
             yield dbs.CountrySearch(sdata)
 
         if sdata.postcodes and (is_category or self.configured_for_postcode):

+            penalty = 0.0 if sdata.countries else 0.1

             if address:
                 sdata.lookups = [dbf.FieldLookup('nameaddress_vector',
                                                  [t.token for r in address
                                                   for t in self.query.get_partials_list(r)],
                                                  'restrict')]
             if address:
                 sdata.lookups = [dbf.FieldLookup('nameaddress_vector',
                                                  [t.token for r in address
                                                   for t in self.query.get_partials_list(r)],
                                                  'restrict')]

-            yield dbs.PostcodeSearch(0.4, sdata)
+                penalty += 0.2
+            yield dbs.PostcodeSearch(penalty, sdata)
+
+
+    def build_housenumber_search(self, sdata: dbf.SearchData, hnrs: List[Token],
+                                 address: List[TokenRange]) -> Iterator[dbs.AbstractSearch]:
+        """ Build a simple address search for special entries where the
+            housenumber is the main name token.
+        """
+        sdata.lookups = [dbf.FieldLookup('name_vector', [t.token for t in hnrs], 'lookup_any')]
+
+        partials = [t for trange in address
+                       for t in self.query.get_partials_list(trange)]
+
+        if 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.housenumbers = dbf.WeightedStrings([], [])
+        yield dbs.PlaceSearch(0.05, sdata, sum(t.count for t in hnrs))

 
 
     def build_name_search(self, sdata: dbf.SearchData,
 
 
     def build_name_search(self, sdata: dbf.SearchData,


@@ -121,8 +182,10 @@ class SearchBuilder:
         """ Build abstract search queries for simple name or address searches.
         """
         if is_category or not sdata.housenumbers or self.configured_for_housenumbers:
         """ Build abstract search queries for simple name or address searches.
         """
         if is_category or not sdata.housenumbers or self.configured_for_housenumbers:

-            sdata.rankings.append(self.get_name_ranking(name))
-            name_penalty = sdata.rankings[-1].normalize_penalty()
+            ranking = self.get_name_ranking(name)
+            name_penalty = ranking.normalize_penalty()
+            if ranking.rankings:
+                sdata.rankings.append(ranking)

             for penalty, count, lookup in self.yield_lookups(name, address):
                 sdata.lookups = lookup
                 yield dbs.PlaceSearch(penalty + name_penalty, sdata, count)
             for penalty, count, lookup in self.yield_lookups(name, address):
                 sdata.lookups = lookup
                 yield dbs.PlaceSearch(penalty + name_penalty, sdata, count)


@@ -134,65 +197,43 @@ class SearchBuilder:
             be searched for. This takes into account how frequent the terms
             are and tries to find a lookup that optimizes index use.
         """
             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)
         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]
         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)
         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.
-            yield penalty, 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)
             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))
+        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]
         # 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)
             penalty += 1.2 * sum(t.penalty for t in addr_partials if not t.is_indexed)

-            log().var_dump('after', penalty)
-        if rare_names:
-            # 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
+        # Any of the full names applies with all of the partials from the address
+        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
 
         # 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:
 
 
     def get_name_ranking(self, trange: TokenRange) -> dbf.FieldRanking:


@@ -280,8 +321,15 @@ class SearchBuilder:
                               self.query.get_tokens(assignment.postcode,
                                                     TokenType.POSTCODE))
         if assignment.qualifier:
                               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])
 
         if assignment.address:
             sdata.set_ranking([self.get_addr_ranking(r) for r in assignment.address])


@@ -298,16 +346,13 @@ class SearchBuilder:
             Returns None if no category search is requested.
         """
         if assignment.category:
             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))
+            tokens: Dict[Tuple[str, str], float] = {}
+            for t in self.query.get_tokens(assignment.category, TokenType.CATEGORY):
+                cat = t.get_category()
+                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
 
 
         return None