tommy/libpostal
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[fix] adding Taiwan, Hong Kong, and Macao to the CJK patterns since language affects the order

Browse Source
This commit is contained in:
Al
2016-12-25 01:20:59 -05:00
parent 6da092e144
commit 5b5a3fe235
1 changed files with 4 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
scripts/geodata/addresses/components.py
View File

@@ -57,6 +57,9 @@ PARSER_DEFAULT_CONFIG = os.path.join(this_dir, os.pardir, os.pardir, os.pardir,
CHINA = 'cn' CHINA = 'cn'
JAPAN = 'jp' JAPAN = 'jp'
KOREA = 'kr' KOREA = 'kr'
TAIWAN = 'tw'
HONG_KONG = 'hk'
MACAO = 'mo'
JAPANESE_ROMAJI = 'ja_rm' JAPANESE_ROMAJI = 'ja_rm'
ENGLISH = 'en' ENGLISH = 'en'
@@ -66,7 +69,7 @@ CHINESE = 'zh'
KOREAN = 'ko' KOREAN = 'ko'
CJK_LANGUAGES = set([CHINESE, JAPANESE, KOREAN]) CJK_LANGUAGES = set([CHINESE, JAPANESE, KOREAN])
CJK_COUNTRIES = set([CHINA, JAPAN, KOREA]) CJK_COUNTRIES = set([CHINA, JAPAN, KOREA, TAIWAN, HONG_KONG, MACAO])
class AddressComponents(object): class AddressComponents(object):