Moved KANJI_MAP to icu-rules

This commit is contained in:
miku0
2023-07-31 11:57:49 +00:00
parent 4d61cc87cf
commit 67e1c7dc72
3 changed files with 11 additions and 44 deletions

View File

@@ -18,25 +18,6 @@ from nominatim.tokenizer.sanitizers.base import ProcessInfo
from nominatim.tokenizer.sanitizers.config import SanitizerConfig
from nominatim.data.place_name import PlaceName
KANJI_MAP = {
ord(''): '0',
ord(''): '1',
ord(''): '2',
ord(''): '3',
ord(''): '4',
ord(''): '5',
ord(''): '6',
ord(''): '7',
ord(''): '8',
ord(''): '9'
}
def convert_kanji_sequence_to_number(sequence: str) -> str:
"""Converts Kanji numbers to Arabic numbers
"""
converted = sequence.translate(KANJI_MAP)
return converted
def create(_: SanitizerConfig) -> Callable[[ProcessInfo], None]:
"""Set up the sanitizer
"""
@@ -49,11 +30,6 @@ def reconbine_housenumber(
) -> List[PlaceName]:
""" Recombine the tag of housenumber by using housenumber and blocknumber
"""
if tmp_blocknumber:
tmp_blocknumber = convert_kanji_sequence_to_number(tmp_blocknumber)
if tmp_housenumber:
tmp_housenumber = convert_kanji_sequence_to_number(tmp_housenumber)
if tmp_blocknumber and tmp_housenumber:
new_address.append(
PlaceName(
@@ -87,11 +63,6 @@ def reconbine_place(
) -> List[PlaceName]:
""" Recombine the tag of place by using neighbourhood and quarter
"""
if tmp_neighbourhood:
tmp_neighbourhood = convert_kanji_sequence_to_number(tmp_neighbourhood)
if tmp_quarter:
tmp_quarter = convert_kanji_sequence_to_number(tmp_quarter)
if tmp_neighbourhood and tmp_quarter:
new_address.append(
PlaceName(