jittok
is a Swiss Army Knife-like toolbox for data processing.
pip install jittok
>>> from jittok import jptext
>>> jptext.normalization("ゼンカクカナ")
'ゼンカクカナ'
>>> from jittok import jptext
>>> jptext.to_numeric("一二,三四五億2百十万987")
1234502100987
>>> from jittok import jpdatetime
>>> jpdatetime.strptime("令和元年10月3日", "%Y年%m月%d日")
datetime.datetime(2019, 10, 3, 0, 0)
>>> jpdatetime.strptime("昭和64年1月1日", "%Y年%m月%d日")
datetime.datetime(1989, 1, 1, 0, 0)
It parses "明治"
, "大正"
, "昭和"
, "平成"
, "令和"
and does not support consistency check:
>>> from jittok import jpdatetime
>>> jpdatetime.strptime("大正90年10月3日", "%Y年%m月%d日")
datetime.datetime(2001, 10, 3, 0, 0)