Merge pull request #721 from paperless-ngx/bug-fix-date-ignore

Fix Ignore Date Parsing
This commit is contained in:
shamoon
2022-05-10 16:45:58 -07:00
committed by GitHub
10 changed files with 373 additions and 49 deletions

View File

@@ -1,9 +1,11 @@
import datetime
import json
import math
import multiprocessing
import os
import re
from typing import Final
from typing import Set
from urllib.parse import urlparse
from concurrent_log_handler.queue import setup_logging_queues
@@ -603,16 +605,42 @@ PAPERLESS_TIKA_GOTENBERG_ENDPOINT = os.getenv(
if PAPERLESS_TIKA_ENABLED:
INSTALLED_APPS.append("paperless_tika.apps.PaperlessTikaConfig")
# List dates that should be ignored when trying to parse date from document text
IGNORE_DATES = set()
if os.getenv("PAPERLESS_IGNORE_DATES", ""):
def _parse_ignore_dates(
env_ignore: str,
date_order: str = DATE_ORDER,
) -> Set[datetime.datetime]:
"""
If the PAPERLESS_IGNORE_DATES environment variable is set, parse the
user provided string(s) into dates
Args:
env_ignore (str): The value of the environment variable, comma seperated dates
date_order (str, optional): The format of the date strings. Defaults to DATE_ORDER.
Returns:
Set[datetime.datetime]: The set of parsed date objects
"""
import dateparser
for s in os.getenv("PAPERLESS_IGNORE_DATES", "").split(","):
d = dateparser.parse(s)
ignored_dates = set()
for s in env_ignore.split(","):
d = dateparser.parse(
s,
settings={
"DATE_ORDER": date_order,
},
)
if d:
IGNORE_DATES.add(d.date())
ignored_dates.add(d.date())
return ignored_dates
# List dates that should be ignored when trying to parse date from document text
IGNORE_DATES: Set[datetime.date] = set()
if os.getenv("PAPERLESS_IGNORE_DATES") is not None:
IGNORE_DATES = _parse_ignore_dates(os.getenv("PAPERLESS_IGNORE_DATES"))
ENABLE_UPDATE_CHECK = os.getenv("PAPERLESS_ENABLE_UPDATE_CHECK", "default")
if ENABLE_UPDATE_CHECK != "default":

View File

@@ -0,0 +1,58 @@
import datetime
from unittest import TestCase
from paperless.settings import _parse_ignore_dates
class TestIgnoreDateParsing(TestCase):
"""
Tests the parsing of the PAPERLESS_IGNORE_DATES setting value
"""
def _parse_checker(self, test_cases):
"""
Helper function to check ignore date parsing
Args:
test_cases (_type_): _description_
"""
for env_str, date_format, expected_date_set in test_cases:
self.assertSetEqual(
_parse_ignore_dates(env_str, date_format),
expected_date_set,
)
def test_no_ignore_dates_set(self):
"""
GIVEN:
- No ignore dates are set
THEN:
- No ignore dates are parsed
"""
self.assertSetEqual(_parse_ignore_dates(""), set())
def test_single_ignore_dates_set(self):
"""
GIVEN:
- Ignore dates are set per certain inputs
THEN:
- All ignore dates are parsed
"""
test_cases = [
("1985-05-01", "YMD", {datetime.date(1985, 5, 1)}),
(
"1985-05-01,1991-12-05",
"YMD",
{datetime.date(1985, 5, 1), datetime.date(1991, 12, 5)},
),
("2010-12-13", "YMD", {datetime.date(2010, 12, 13)}),
("11.01.10", "DMY", {datetime.date(2010, 1, 11)}),
(
"11.01.2001,15-06-1996",
"DMY",
{datetime.date(2001, 1, 11), datetime.date(1996, 6, 15)},
),
]
self._parse_checker(test_cases)