1 # SPDX-License-Identifier: GPL-2.0-only
3 # This file is part of Nominatim. (https://nominatim.org)
5 # Copyright (C) 2022 by the Nominatim developer community.
6 # For a full list of authors see the git log.
8 Specialized processor for postcodes. Supports a 'lookup' variant of the
9 token, which produces variants with optional spaces.
11 from typing import Mapping, Any, List
13 from nominatim.tokenizer.token_analysis.generic_mutation import MutationVariantGenerator
15 ### Configuration section
17 def configure(rules: Mapping[str, Any], normalization_rules: str) -> None: # pylint: disable=W0613
18 """ All behaviour is currently hard-coded.
24 def create(normalizer: Any, transliterator: Any, config: None) -> 'PostcodeTokenAnalysis': # pylint: disable=W0613
25 """ Create a new token analysis instance for this module.
27 return PostcodeTokenAnalysis(normalizer, transliterator)
30 class PostcodeTokenAnalysis:
31 """ Special normalization and variant generation for postcodes.
33 This analyser must not be used with anything but postcodes as
34 it follows some special rules: `normalize` doesn't necessarily
35 need to return a standard form as per normalization rules. It
36 needs to return the canonical form of the postcode that is also
37 used for output. `get_variants_ascii` then needs to ensure that
38 the generated variants once more follow the standard normalization
39 and transliteration, so that postcodes are correctly recognised by
42 def __init__(self, norm: Any, trans: Any) -> None:
46 self.mutator = MutationVariantGenerator(' ', (' ', ''))
49 def normalize(self, name: str) -> str:
50 """ Return the standard form of the postcode.
52 return name.strip().upper()
55 def get_variants_ascii(self, norm_name: str) -> List[str]:
56 """ Compute the spelling variants for the given normalized postcode.
58 Takes the canonical form of the postcode, normalizes it using the
59 standard rules and then creates variants of the result where
60 all spaces are optional.
62 # Postcodes follow their own transliteration rules.
63 # Make sure at this point, that the terms are normalized in a way
64 # that they are searchable with the standard transliteration rules.
65 return [self.trans.transliterate(term) for term in
66 self.mutator.generate([self.norm.transliterate(norm_name)]) if term]