1 # SPDX-License-Identifier: GPL-3.0-or-later
3 # This file is part of Nominatim. (https://nominatim.org)
5 # Copyright (C) 2024 by the Nominatim developer community.
6 # For a full list of authors see the git log.
8 Datastructures for a tokenized query.
10 from typing import List, Tuple, Optional, Iterator
11 from abc import ABC, abstractmethod
15 class BreakType(enum.Enum):
16 """ Type of break between tokens.
19 """ Begin of the query. """
21 """ End of the query. """
23 """ Break between two phrases. """
25 """ Break between words. """
27 """ Break inside a word, for example a hyphen or apostrophe. """
29 """ Break created as a result of tokenization.
30 This may happen in languages without spaces between words.
34 class TokenType(enum.Enum):
38 """ Full name of a place. """
40 """ Word term without breaks, does not necessarily represent a full name. """
41 HOUSENUMBER = enum.auto()
42 """ Housenumber term. """
43 POSTCODE = enum.auto()
44 """ Postal code term. """
46 """ Country name or reference. """
47 QUALIFIER = enum.auto()
48 """ Special term used together with name (e.g. _Hotel_ Bellevue). """
49 NEAR_ITEM = enum.auto()
50 """ Special term used as searchable object(e.g. supermarket in ...). """
53 class PhraseType(enum.Enum):
54 """ Designation of a phrase.
57 """ No specific designation (i.e. source is free-form query). """
59 """ Contains name or type of a POI. """
61 """ Contains a street name optionally with a housenumber. """
63 """ Contains the postal city. """
65 """ Contains the equivalent of a county. """
67 """ Contains a state or province. """
68 POSTCODE = enum.auto()
69 """ Contains a postal code. """
71 """ Contains the country name or code. """
73 def compatible_with(self, ttype: TokenType,
74 is_full_phrase: bool) -> bool:
75 """ Check if the given token type can be used with the phrase type.
77 if self == PhraseType.NONE:
78 return not is_full_phrase or ttype != TokenType.QUALIFIER
79 if self == PhraseType.AMENITY:
80 return ttype in (TokenType.WORD, TokenType.PARTIAL)\
81 or (is_full_phrase and ttype == TokenType.NEAR_ITEM)\
82 or (not is_full_phrase and ttype == TokenType.QUALIFIER)
83 if self == PhraseType.STREET:
84 return ttype in (TokenType.WORD, TokenType.PARTIAL, TokenType.HOUSENUMBER)
85 if self == PhraseType.POSTCODE:
86 return ttype == TokenType.POSTCODE
87 if self == PhraseType.COUNTRY:
88 return ttype == TokenType.COUNTRY
90 return ttype in (TokenType.WORD, TokenType.PARTIAL)
93 @dataclasses.dataclass
95 """ Base type for tokens.
96 Specific query analyzers must implement the concrete token class.
108 def get_category(self) -> Tuple[str, str]:
109 """ Return the category restriction for qualifier terms and
113 @dataclasses.dataclass
115 """ Indexes of query nodes over which a token spans.
120 def __lt__(self, other: 'TokenRange') -> bool:
121 return self.end <= other.start
124 def __le__(self, other: 'TokenRange') -> bool:
125 return NotImplemented
128 def __gt__(self, other: 'TokenRange') -> bool:
129 return self.start >= other.end
132 def __ge__(self, other: 'TokenRange') -> bool:
133 return NotImplemented
136 def replace_start(self, new_start: int) -> 'TokenRange':
137 """ Return a new token range with the new start.
139 return TokenRange(new_start, self.end)
142 def replace_end(self, new_end: int) -> 'TokenRange':
143 """ Return a new token range with the new end.
145 return TokenRange(self.start, new_end)
148 def split(self, index: int) -> Tuple['TokenRange', 'TokenRange']:
149 """ Split the span into two spans at the given index.
150 The index must be within the span.
152 return self.replace_end(index), self.replace_start(index)
155 @dataclasses.dataclass
157 """ List of all tokens of a given type going from one breakpoint to another.
164 def add_penalty(self, penalty: float) -> None:
165 """ Add the given penalty to all tokens in the list.
167 for token in self.tokens:
168 token.penalty += penalty
171 @dataclasses.dataclass
173 """ A node of the query representing a break between terms.
177 starting: List[TokenList] = dataclasses.field(default_factory=list)
179 def has_tokens(self, end: int, *ttypes: TokenType) -> bool:
180 """ Check if there are tokens of the given types ending at the
183 return any(tl.end == end and tl.ttype in ttypes for tl in self.starting)
186 def get_tokens(self, end: int, ttype: TokenType) -> Optional[List[Token]]:
187 """ Get the list of tokens of the given type starting at this node
188 and ending at the node 'end'. Returns 'None' if no such
191 for tlist in self.starting:
192 if tlist.end == end and tlist.ttype == ttype:
197 @dataclasses.dataclass
199 """ A normalized query part. Phrases may be typed which means that
200 they then represent a specific part of the address.
207 """ A tokenized search query together with the normalized source
208 from which the tokens have been parsed.
210 The query contains a list of nodes that represent the breaks
211 between words. Tokens span between nodes, which don't necessarily
212 need to be direct neighbours. Thus the query is represented as a
213 directed acyclic graph.
215 When created, a query contains a single node: the start of the
216 query. Further nodes can be added by appending to 'nodes'.
219 def __init__(self, source: List[Phrase]) -> None:
221 self.nodes: List[QueryNode] = \
222 [QueryNode(BreakType.START, source[0].ptype if source else PhraseType.NONE)]
225 def num_token_slots(self) -> int:
226 """ Return the length of the query in vertice steps.
228 return len(self.nodes) - 1
231 def add_node(self, btype: BreakType, ptype: PhraseType) -> None:
232 """ Append a new break node with the given break type.
233 The phrase type denotes the type for any tokens starting
236 self.nodes.append(QueryNode(btype, ptype))
239 def add_token(self, trange: TokenRange, ttype: TokenType, token: Token) -> None:
240 """ Add a token to the query. 'start' and 'end' are the indexes of the
241 nodes from which to which the token spans. The indexes must exist
242 and are expected to be in the same phrase.
243 'ttype' denotes the type of the token and 'token' the token to
246 If the token type is not compatible with the phrase it should
247 be added to, then the token is silently dropped.
249 snode = self.nodes[trange.start]
250 full_phrase = snode.btype in (BreakType.START, BreakType.PHRASE)\
251 and self.nodes[trange.end].btype in (BreakType.PHRASE, BreakType.END)
252 if snode.ptype.compatible_with(ttype, full_phrase):
253 tlist = snode.get_tokens(trange.end, ttype)
255 snode.starting.append(TokenList(trange.end, ttype, [token]))
260 def get_tokens(self, trange: TokenRange, ttype: TokenType) -> List[Token]:
261 """ Get the list of tokens of a given type, spanning the given
262 nodes. The nodes must exist. If no tokens exist, an
263 empty list is returned.
265 return self.nodes[trange.start].get_tokens(trange.end, ttype) or []
268 def get_partials_list(self, trange: TokenRange) -> List[Token]:
269 """ Create a list of partial tokens between the given nodes.
270 The list is composed of the first token of type PARTIAL
271 going to the subsequent node. Such PARTIAL tokens are
274 return [next(iter(self.get_tokens(TokenRange(i, i+1), TokenType.PARTIAL)))
275 for i in range(trange.start, trange.end)]
278 def iter_token_lists(self) -> Iterator[Tuple[int, QueryNode, TokenList]]:
279 """ Iterator over all token lists in the query.
281 for i, node in enumerate(self.nodes):
282 for tlist in node.starting:
286 def find_lookup_word_by_id(self, token: int) -> str:
287 """ Find the first token with the given token ID and return
288 its lookup word. Returns 'None' if no such token exists.
289 The function is very slow and must only be used for
292 for node in self.nodes:
293 for tlist in node.starting:
294 for t in tlist.tokens:
296 return f"[{tlist.ttype.name[0]}]{t.lookup_word}"