Last active
February 4, 2022 15:35
-
-
Save ryukinix/c2541bdaa831c8bde49aa809883e1a30 to your computer and use it in GitHub Desktop.
An EBNF grammar based in the Lark Parser designed to parse multivariate date formats. (PT_BR)
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python3 | |
# coding: utf-8 | |
# | |
# Copyright © Neoway Business Solutions | |
# | |
# @project: Diário Oficial | |
# @author: Manoel Vilela | |
# | |
""" | |
An EBNF grammar based in the Lark Parser designed to | |
parse multivariate date formats. Grammar written for Portuguese in Mind. | |
""" | |
from lark import exceptions as LarkExceptions | |
from lark import Lark, InlineTransformer # pip install lark-parser | |
from datetime import datetime | |
default_date_format = "%d/%m/%Y" | |
grammar = '''\ | |
?date: day "de" month "de" year | |
| day "/" month "/" year | |
| day "-" month "-" year | |
| day "." month "." year | |
| day month year | |
day: INT | |
year: INT | |
month: month_name | month_number | |
month_name: | |
| ("janeiro" | "jan") -> jan | |
| ("fevereiro" | "fev") -> fev | |
| ("março" | "mar") -> mar | |
| ("abril" | "abr") -> abr | |
| ("maio" | "mai") -> mai | |
| ("junho" | "jun") -> jun | |
| ("julho" | "jul") -> jul | |
| ("agosto" | "ago") -> ago | |
| ("setembro" | "set") -> set | |
| ("outubro" | "out") -> out | |
| ("novembro" | "nov") -> nov | |
| ("dezembro" | "dez") -> dez | |
month_number: | |
| ["0"] "1" -> jan | |
| ["0"] "2" -> fev | |
| ["0"] "3" -> mar | |
| ["0"] "4" -> abr | |
| ["0"] "5" -> mai | |
| ["0"] "6" -> jun | |
| ["0"] "7" -> jul | |
| ["0"] "8" -> ago | |
| ["0"] "9" -> set | |
| "10" -> out | |
| "11" -> nov | |
| "12" -> dez | |
%import common.WORD | |
%import common.INT | |
%import common.DIGIT | |
%import common.WS | |
%ignore WS | |
''' | |
month_dict = { | |
"jan": 1, | |
"fev": 2, | |
"mar": 3, | |
"abr": 4, | |
"mai": 5, | |
"jun": 6, | |
"jul": 7, | |
"ago": 8, | |
"set": 9, | |
"out": 10, | |
"nov": 11, | |
"dez": 12 | |
} | |
parser = Lark(grammar, start='date') | |
class NaturalDateTree(InlineTransformer): | |
day = int | |
year = int | |
def month(self, tree): | |
month_label = tree._pretty_label() # jan, fev | |
return month_dict[month_label] | |
def date(self, day, month, year): | |
if year < 30: | |
year += 2000 | |
elif year > 30 and year < 100: | |
year += 1900 | |
try: | |
return datetime(year, month, day) | |
except ValueError: | |
return None | |
def parse_date(expr): | |
return NaturalDateTree().transform(parser.parse(expr)) | |
def parse(date_string, date_format=default_date_format): | |
try: | |
parsed_date = parse_date(date_string.lower()) | |
if parsed_date is not None: | |
return parsed_date.strftime(date_format) | |
except LarkExceptions.UnexpectedCharacters: | |
pass | |
return None | |
def run_tests(): | |
tests = [ | |
('02-08-2018', '02/08/2018'), | |
('1.03.2018', '01/03/2018'), | |
('1 de fevereiro de 2018', "01/02/2018"), | |
('02 fevereiro 2018', "02/02/2018"), | |
('28FEV2017', '28/02/2017'), | |
('31fev2019', None), | |
('shitty string', None), | |
('10 Mai 17', '10/05/2017'), | |
('08 Jun 98', '08/06/1998') | |
] | |
for entry, expected in tests: | |
parsed = parse(entry) | |
assert parsed == expected, f"Expected: {expected!r}, but: {parsed!r}" | |
if __name__ == '__main__': | |
run_tests() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment