Merge remote-tracking branch 'upstream/master'

[nominatim.git] / nominatim / api / search / query.py

diff --git a/nominatim/api/search/query.py b/nominatim/api/search/query.py
index f2b18f873a8121fbdac79ea3c67b682826316e6b..a0d7add1b70118e32d628b4894a893386d09d996 100644 (file)
--- a/nominatim/api/search/query.py
+++ b/nominatim/api/search/query.py
@@ -7,7 +7,7 @@
 """
 Datastructures for a tokenized query.
 """
 """
 Datastructures for a tokenized query.
 """

-from typing import List, Tuple, Optional, NamedTuple, Iterator
+from typing import List, Tuple, Optional, Iterator

 from abc import ABC, abstractmethod
 import dataclasses
 import enum
 from abc import ABC, abstractmethod
 import dataclasses
 import enum


@@ -46,7 +46,7 @@ class TokenType(enum.Enum):
     """ Country name or reference. """
     QUALIFIER = enum.auto()
     """ Special term used together with name (e.g. _Hotel_ Bellevue). """
     """ Country name or reference. """
     QUALIFIER = enum.auto()
     """ Special term used together with name (e.g. _Hotel_ Bellevue). """

-    CATEGORY = enum.auto()
+    NEAR_ITEM = enum.auto()

     """ Special term used as searchable object(e.g. supermarket in ...). """
 
 
     """ Special term used as searchable object(e.g. supermarket in ...). """
 
 


@@ -70,14 +70,16 @@ class PhraseType(enum.Enum):
     COUNTRY = enum.auto()
     """ Contains the country name or code. """
 
     COUNTRY = enum.auto()
     """ Contains the country name or code. """
 

-    def compatible_with(self, ttype: TokenType) -> bool:
+    def compatible_with(self, ttype: TokenType,
+                        is_full_phrase: bool) -> bool:

         """ Check if the given token type can be used with the phrase type.
         """
         if self == PhraseType.NONE:
         """ Check if the given token type can be used with the phrase type.
         """
         if self == PhraseType.NONE:

-            return True
+            return not is_full_phrase or ttype != TokenType.QUALIFIER

         if self == PhraseType.AMENITY:
         if self == PhraseType.AMENITY:

-            return ttype in (TokenType.WORD, TokenType.PARTIAL,
-                             TokenType.QUALIFIER, TokenType.CATEGORY)
+            return ttype in (TokenType.WORD, TokenType.PARTIAL)\
+                   or (is_full_phrase and ttype == TokenType.NEAR_ITEM)\
+                   or (not is_full_phrase and ttype == TokenType.QUALIFIER)

         if self == PhraseType.STREET:
             return ttype in (TokenType.WORD, TokenType.PARTIAL, TokenType.HOUSENUMBER)
         if self == PhraseType.POSTCODE:
         if self == PhraseType.STREET:
             return ttype in (TokenType.WORD, TokenType.PARTIAL, TokenType.HOUSENUMBER)
         if self == PhraseType.POSTCODE:


@@ -97,6 +99,7 @@ class Token(ABC):
     penalty: float
     token: int
     count: int
     penalty: float
     token: int
     count: int

+    addr_count: int

     lookup_word: str
     is_indexed: bool
 
     lookup_word: str
     is_indexed: bool
 


@@ -107,13 +110,29 @@ class Token(ABC):
             category objects.
         """
 
             category objects.
         """
 

-
-class TokenRange(NamedTuple):
+@dataclasses.dataclass
+class TokenRange:

     """ Indexes of query nodes over which a token spans.
     """
     start: int
     end: int
 
     """ Indexes of query nodes over which a token spans.
     """
     start: int
     end: int
 

+    def __lt__(self, other: 'TokenRange') -> bool:
+        return self.end <= other.start
+
+
+    def __le__(self, other: 'TokenRange') -> bool:
+        return NotImplemented
+
+
+    def __gt__(self, other: 'TokenRange') -> bool:
+        return self.start >= other.end
+
+
+    def __ge__(self, other: 'TokenRange') -> bool:
+        return NotImplemented
+
+

     def replace_start(self, new_start: int) -> 'TokenRange':
         """ Return a new token range with the new start.
         """
     def replace_start(self, new_start: int) -> 'TokenRange':
         """ Return a new token range with the new start.
         """


@@ -151,7 +170,7 @@ class TokenList:
 
 @dataclasses.dataclass
 class QueryNode:
 
 @dataclasses.dataclass
 class QueryNode:

-    """ A node of the querry representing a break between terms.
+    """ A node of the query representing a break between terms.

     """
     btype: BreakType
     ptype: PhraseType
     """
     btype: BreakType
     ptype: PhraseType


@@ -228,7 +247,9 @@ class QueryStruct:
             be added to, then the token is silently dropped.
         """
         snode = self.nodes[trange.start]
             be added to, then the token is silently dropped.
         """
         snode = self.nodes[trange.start]

-        if snode.ptype.compatible_with(ttype):
+        full_phrase = snode.btype in (BreakType.START, BreakType.PHRASE)\
+                      and self.nodes[trange.end].btype in (BreakType.PHRASE, BreakType.END)
+        if snode.ptype.compatible_with(ttype, full_phrase):

             tlist = snode.get_tokens(trange.end, ttype)
             if tlist is None:
                 snode.starting.append(TokenList(trange.end, ttype, [token]))
             tlist = snode.get_tokens(trange.end, ttype)
             if tlist is None:
                 snode.starting.append(TokenList(trange.end, ttype, [token]))