Experiments with using adrf for a few views

.pre-commit-config.yaml

@@ -37,7 +37,7 @@ repos:
           - json
   # See https://github.com/prettier/prettier/issues/15742 for the fork reason
   - repo: https://github.com/rbubley/mirrors-prettier
-    rev: 'v3.8.1'
+    rev: 'v3.6.2'
     hooks:
       - id: prettier
         types_or:
@@ -49,7 +49,7 @@ repos:
           - 'prettier-plugin-organize-imports@4.1.0'
   # Python hooks
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.14.14
+    rev: v0.14.5
     hooks:
       - id: ruff-check
       - id: ruff-format
@@ -76,7 +76,7 @@ repos:
     hooks:
       - id: shellcheck
   - repo: https://github.com/google/yamlfmt
-    rev: v0.21.0
+    rev: v0.20.0
     hooks:
       - id: yamlfmt
         exclude: "^src-ui/pnpm-lock.yaml"

pyproject.toml

@@ -16,17 +16,18 @@ classifiers = [
 # This will allow testing to not install a webserver, mysql, etc
 
 dependencies = [
+  "adrf~=0.1.12",
   "azure-ai-documentintelligence>=1.0.2",
   "babel>=2.17",
   "bleach~=6.3.0",
-  "celery[redis]~=5.6.2",
+  "celery[redis]~=5.5.1",
   "channels~=4.2",
   "channels-redis~=4.2",
   "concurrent-log-handler~=0.9.25",
   "dateparser~=1.2",
   # WARNING: django does not use semver.
   #          Only patch versions are guaranteed to not introduce breaking changes.
-  "django~=5.2.10",
+  "django~=5.2.5",
   "django-allauth[mfa,socialaccount]~=65.13.1",
   "django-auditlog~=3.4.1",
   "django-cachalot~=2.8.0",
@@ -79,7 +80,7 @@ dependencies = [
   "torch~=2.9.1",
   "tqdm~=4.67.1",
   "watchfiles>=1.1.1",
-  "whitenoise~=6.11",
+  "whitenoise~=6.9",
   "whoosh-reloaded>=2.7.5",
   "zxing-cpp~=2.3.0",
 ]
@@ -88,13 +89,13 @@ optional-dependencies.mariadb = [
   "mysqlclient~=2.2.7",
 ]
 optional-dependencies.postgres = [
-  "psycopg[c,pool]==3.3",
+  "psycopg[c,pool]==3.2.12",
   # Direct dependency for proper resolution of the pre-built wheels
-  "psycopg-c==3.3",
+  "psycopg-c==3.2.12",
   "psycopg-pool==3.3",
 ]
 optional-dependencies.webserver = [
-  "granian[uvloop]~=2.6.0",
+  "granian[uvloop]~=2.5.1",
 ]
 
 [dependency-groups]
@@ -152,7 +153,7 @@ typing = [
 ]
 
 [tool.uv]
-required-version = ">=0.9.0"
+required-version = ">=0.5.14"
 package = false
 environments = [
   "sys_platform == 'darwin'",
@@ -162,8 +163,8 @@ environments = [
 [tool.uv.sources]
 # Markers are chosen to select these almost exclusively when building the Docker image
 psycopg-c = [
-  { url = "https://github.com/paperless-ngx/builder/releases/download/psycopg-trixie-3.3.0/psycopg_c-3.3.0-cp312-cp312-linux_x86_64.whl", marker = "sys_platform == 'linux' and platform_machine == 'x86_64' and python_version == '3.12'" },
-  { url = "https://github.com/paperless-ngx/builder/releases/download/psycopg-trixie-3.3.0/psycopg_c-3.3.0-cp312-cp312-linux_aarch64.whl", marker = "sys_platform == 'linux' and platform_machine == 'aarch64' and python_version == '3.12'" },
+  { url = "https://github.com/paperless-ngx/builder/releases/download/psycopg-bookworm-3.2.12/psycopg_c-3.2.12-cp312-cp312-linux_x86_64.whl", marker = "sys_platform == 'linux' and platform_machine == 'x86_64' and python_version == '3.12'" },
+  { url = "https://github.com/paperless-ngx/builder/releases/download/psycopg-bookworm-3.2.12/psycopg_c-3.2.12-cp312-cp312-linux_aarch64.whl", marker = "sys_platform == 'linux' and platform_machine == 'aarch64' and python_version == '3.12'" },
 ]
 zxing-cpp = [
   { url = "https://github.com/paperless-ngx/builder/releases/download/zxing-2.3.0/zxing_cpp-2.3.0-cp312-cp312-linux_x86_64.whl", marker = "sys_platform == 'linux' and platform_machine == 'x86_64' and python_version == '3.12'" },
@@ -306,7 +307,6 @@ markers = [
   "gotenberg: Tests requiring Gotenberg service",
   "tika: Tests requiring Tika service",
   "greenmail: Tests requiring Greenmail service",
-  "date_parsing: Tests which cover date parsing from content or filename",
 ]
 
 [tool.pytest_env]
@@ -333,10 +333,6 @@ exclude_also = [
 
 [tool.mypy]
 mypy_path = "src"
-files = [
-  "src/documents/plugins/date_parsing",
-  "src/documents/tests/date_parsing",
-]
 plugins = [
   "mypy_django_plugin.main",
   "mypy_drf_plugin.main",
@@ -348,28 +344,5 @@ disallow_untyped_defs = true
 warn_redundant_casts = true
 warn_unused_ignores = true
 
-# This prevents errors from imports, but allows type-checking logic to work
-follow_imports = "silent"
-
-[[tool.mypy.overrides]]
-module = [
-  "documents.*",
-  "paperless.*",
-  "paperless_ai.*",
-  "paperless_mail.*",
-  "paperless_tesseract.*",
-  "paperless_remote.*",
-  "paperless_text.*",
-  "paperless_tika.*",
-]
-ignore_errors = true
-
-[[tool.mypy.overrides]]
-module = [
-  "documents.plugins.date_parsing.*",
-  "documents.tests.date_parsing.*",
-]
-ignore_errors = false
-
 [tool.django-stubs]
 django_settings_module = "paperless.settings"

src/documents/consumer.py

@@ -32,12 +32,12 @@ from documents.models import WorkflowTrigger
 from documents.parsers import DocumentParser
 from documents.parsers import ParseError
 from documents.parsers import get_parser_class_for_mime_type
+from documents.parsers import parse_date
 from documents.permissions import set_permissions_for_object
 from documents.plugins.base import AlwaysRunPluginMixin
 from documents.plugins.base import ConsumeTaskPlugin
 from documents.plugins.base import NoCleanupPluginMixin
 from documents.plugins.base import NoSetupPluginMixin
-from documents.plugins.date_parsing import get_date_parser
 from documents.plugins.helpers import ProgressManager
 from documents.plugins.helpers import ProgressStatusOptions
 from documents.signals import document_consumption_finished
@@ -426,8 +426,7 @@ class ConsumerPlugin(
                     ProgressStatusOptions.WORKING,
                     ConsumerStatusShortMessage.PARSE_DATE,
                 )
-                with get_date_parser() as date_parser:
-                    date = next(date_parser.parse(self.filename, text), None)
+                date = parse_date(self.filename, text)
             archive_path = document_parser.get_archive_path()
             page_count = document_parser.get_page_count(self.working_copy, mime_type)
 

src/documents/parsers.py

@@ -9,17 +9,22 @@ import subprocess
 import tempfile
 from functools import lru_cache
 from pathlib import Path
+from re import Match
 from typing import TYPE_CHECKING
 
 from django.conf import settings
+from django.utils import timezone
 
 from documents.loggers import LoggingMixin
 from documents.signals import document_consumer_declaration
 from documents.utils import copy_file_with_basic_stats
 from documents.utils import run_subprocess
+from paperless.config import OcrConfig
+from paperless.utils import ocr_to_dateparser_languages
 
 if TYPE_CHECKING:
     import datetime
+    from collections.abc import Iterator
 
 # This regular expression will try to find dates in the document at
 # hand and will match the following formats:
@@ -254,6 +259,75 @@ def make_thumbnail_from_pdf(in_path: Path, temp_dir: Path, logging_group=None) -
     return out_path
 
 
+def parse_date(filename, text) -> datetime.datetime | None:
+    return next(parse_date_generator(filename, text), None)
+
+
+def parse_date_generator(filename, text) -> Iterator[datetime.datetime]:
+    """
+    Returns the date of the document.
+    """
+
+    def __parser(ds: str, date_order: str) -> datetime.datetime:
+        """
+        Call dateparser.parse with a particular date ordering
+        """
+        import dateparser
+
+        ocr_config = OcrConfig()
+        languages = settings.DATE_PARSER_LANGUAGES or ocr_to_dateparser_languages(
+            ocr_config.language,
+        )
+
+        return dateparser.parse(
+            ds,
+            settings={
+                "DATE_ORDER": date_order,
+                "PREFER_DAY_OF_MONTH": "first",
+                "RETURN_AS_TIMEZONE_AWARE": True,
+                "TIMEZONE": settings.TIME_ZONE,
+            },
+            locales=languages,
+        )
+
+    def __filter(date: datetime.datetime) -> datetime.datetime | None:
+        if (
+            date is not None
+            and date.year > 1900
+            and date <= timezone.now()
+            and date.date() not in settings.IGNORE_DATES
+        ):
+            return date
+        return None
+
+    def __process_match(
+        match: Match[str],
+        date_order: str,
+    ) -> datetime.datetime | None:
+        date_string = match.group(0)
+
+        try:
+            date = __parser(date_string, date_order)
+        except Exception:
+            # Skip all matches that do not parse to a proper date
+            date = None
+
+        return __filter(date)
+
+    def __process_content(content: str, date_order: str) -> Iterator[datetime.datetime]:
+        for m in re.finditer(DATE_REGEX, content):
+            date = __process_match(m, date_order)
+            if date is not None:
+                yield date
+
+    # if filename date parsing is enabled, search there first:
+    if settings.FILENAME_DATE_ORDER:
+        yield from __process_content(filename, settings.FILENAME_DATE_ORDER)
+
+    # Iterate through all regex matches in text and try to parse the date
+    yield from __process_content(text, settings.DATE_ORDER)
+
+
 class ParseError(Exception):
     pass
 

src/documents/plugins/date_parsing/__init__.py

@@ -1,92 +0,0 @@
-import logging
-from functools import lru_cache
-from importlib.metadata import EntryPoint
-from importlib.metadata import entry_points
-from typing import Final
-
-from django.conf import settings
-from django.utils import timezone
-
-from documents.plugins.date_parsing.base import DateParserConfig
-from documents.plugins.date_parsing.base import DateParserPluginBase
-from documents.plugins.date_parsing.regex_parser import RegexDateParserPlugin
-from paperless.utils import ocr_to_dateparser_languages
-
-logger = logging.getLogger(__name__)
-
-DATE_PARSER_ENTRY_POINT_GROUP: Final = "paperless_ngx.date_parsers"
-
-
-@lru_cache(maxsize=1)
-def _discover_parser_class() -> type[DateParserPluginBase]:
-    """
-    Discovers the date parser plugin class to use.
-
-    - If one or more plugins are found, sorts them by name and returns the first.
-    - If no plugins are found, returns the default RegexDateParser.
-    """
-
-    eps: tuple[EntryPoint, ...]
-    try:
-        eps = entry_points(group=DATE_PARSER_ENTRY_POINT_GROUP)
-    except Exception as e:
-        # Log a warning
-        logger.warning(f"Could not query entry points for date parsers: {e}")
-        eps = ()
-
-    valid_plugins: list[EntryPoint] = []
-    for ep in eps:
-        try:
-            plugin_class = ep.load()
-            if plugin_class and issubclass(plugin_class, DateParserPluginBase):
-                valid_plugins.append(ep)
-            else:
-                logger.warning(f"Plugin {ep.name} does not subclass DateParser.")
-        except Exception as e:
-            logger.error(f"Unable to load date parser plugin {ep.name}: {e}")
-
-    if not valid_plugins:
-        return RegexDateParserPlugin
-
-    valid_plugins.sort(key=lambda ep: ep.name)
-
-    if len(valid_plugins) > 1:
-        logger.warning(
-            f"Multiple date parsers found: "
-            f"{[ep.name for ep in valid_plugins]}. "
-            f"Using the first one by name: '{valid_plugins[0].name}'.",
-        )
-
-    return valid_plugins[0].load()
-
-
-def get_date_parser() -> DateParserPluginBase:
-    """
-    Factory function to get an initialized date parser instance.
-
-    This function is responsible for:
-    1. Discovering the correct parser class (plugin or default).
-    2. Loading configuration from Django settings.
-    3. Instantiating the parser with the configuration.
-    """
-    # 1. Discover the class (this is cached)
-    parser_class = _discover_parser_class()
-
-    # 2. Load configuration from settings
-    # TODO: Get the language from the settings and/or configuration object, depending
-    languages = languages = (
-        settings.DATE_PARSER_LANGUAGES
-        or ocr_to_dateparser_languages(settings.OCR_LANGUAGE)
-    )
-
-    config = DateParserConfig(
-        languages=languages,
-        timezone_str=settings.TIME_ZONE,
-        ignore_dates=settings.IGNORE_DATES,
-        reference_time=timezone.now(),
-        filename_date_order=settings.FILENAME_DATE_ORDER,
-        content_date_order=settings.DATE_ORDER,
-    )
-
-    # 3. Instantiate the discovered class with the config
-    return parser_class(config=config)

src/documents/plugins/date_parsing/base.py

@@ -1,124 +0,0 @@
-import datetime
-import logging
-from abc import ABC
-from abc import abstractmethod
-from collections.abc import Iterator
-from dataclasses import dataclass
-from types import TracebackType
-
-try:
-    from typing import Self
-except ImportError:
-    from typing_extensions import Self
-
-import dateparser
-
-logger = logging.getLogger(__name__)
-
-
-@dataclass(frozen=True, slots=True)
-class DateParserConfig:
-    """
-    Configuration for a DateParser instance.
-
-    This object is created by the factory and passed to the
-    parser's constructor, decoupling the parser from settings.
-    """
-
-    languages: list[str]
-    timezone_str: str
-    ignore_dates: set[datetime.date]
-
-    # A "now" timestamp for filtering future dates.
-    # Passed in by the factory.
-    reference_time: datetime.datetime
-
-    # Settings for the default RegexDateParser
-    # Other plugins should use or consider these, but it is not required
-    filename_date_order: str | None
-    content_date_order: str
-
-
-class DateParserPluginBase(ABC):
-    """
-    Abstract base class for date parsing strategies.
-
-    Instances are configured via a DateParserConfig object.
-    """
-
-    def __init__(self, config: DateParserConfig):
-        """
-        Initializes the parser with its configuration.
-        """
-        self.config = config
-
-    def __enter__(self) -> Self:
-        """
-        Enter the runtime context related to this object.
-
-        Subclasses can override this to acquire resources (connections, handles).
-        """
-        return self
-
-    def __exit__(
-        self,
-        exc_type: type[BaseException] | None,
-        exc_val: BaseException | None,
-        exc_tb: TracebackType | None,
-    ) -> None:
-        """
-        Exit the runtime context related to this object.
-
-        Subclasses can override this to release resources.
-        """
-        # Default implementation does nothing.
-        # Returning None implies exceptions are propagated.
-
-    def _parse_string(
-        self,
-        date_string: str,
-        date_order: str,
-    ) -> datetime.datetime | None:
-        """
-        Helper method to parse a single date string using dateparser.
-
-        Uses configuration from `self.config`.
-        """
-        try:
-            return dateparser.parse(
-                date_string,
-                settings={
-                    "DATE_ORDER": date_order,
-                    "PREFER_DAY_OF_MONTH": "first",
-                    "RETURN_AS_TIMEZONE_AWARE": True,
-                    "TIMEZONE": self.config.timezone_str,
-                },
-                locales=self.config.languages,
-            )
-        except Exception as e:
-            logger.error(f"Error while parsing date string '{date_string}': {e}")
-            return None
-
-    def _filter_date(
-        self,
-        date: datetime.datetime | None,
-    ) -> datetime.datetime | None:
-        """
-        Helper method to validate a parsed datetime object.
-
-        Uses configuration from `self.config`.
-        """
-        if (
-            date is not None
-            and date.year > 1900
-            and date <= self.config.reference_time
-            and date.date() not in self.config.ignore_dates
-        ):
-            return date
-        return None
-
-    @abstractmethod
-    def parse(self, filename: str, content: str) -> Iterator[datetime.datetime]:
-        """
-        Parses a document's filename and content, yielding valid datetime objects.
-        """

src/documents/plugins/date_parsing/regex_parser.py

@@ -1,65 +0,0 @@
-import datetime
-import re
-from collections.abc import Iterator
-from re import Match
-
-from documents.plugins.date_parsing.base import DateParserPluginBase
-
-
-class RegexDateParserPlugin(DateParserPluginBase):
-    """
-    The default date parser, using a series of regular expressions.
-
-    It is configured entirely by the DateParserConfig object
-    passed to its constructor.
-    """
-
-    DATE_REGEX = re.compile(
-        r"(\b|(?!=([_-])))(\d{1,2})[\.\/-](\d{1,2})[\.\/-](\d{4}|\d{2})(\b|(?=([_-])))|"
-        r"(\b|(?!=([_-])))(\d{4}|\d{2})[\.\/-](\d{1,2})[\.\/-](\d{1,2})(\b|(?=([_-])))|"
-        r"(\b|(?!=([_-])))(\d{1,2}[\. ]+[a-zéûäëčžúřěáíóńźçŞğü]{3,9} \d{4}|[a-zéûäëčžúřěáíóńźçŞğü]{3,9} \d{1,2}, \d{4})(\b|(?=([_-])))|"
-        r"(\b|(?!=([_-])))([^\W\d_]{3,9} \d{1,2}, (\d{4}))(\b|(?=([_-])))|"
-        r"(\b|(?!=([_-])))([^\W\d_]{3,9} \d{4})(\b|(?=([_-])))|"
-        r"(\b|(?!=([_-])))(\d{1,2}[^ 0-9]{2}[\. ]+[^ ]{3,9}[ \.\/-]\d{4})(\b|(?=([_-])))|"
-        r"(\b|(?!=([_-])))(\b\d{1,2}[ \.\/-][a-zéûäëčžúřěáíóńźçŞğü]{3}[ \.\/-]\d{4})(\b|(?=([_-])))",
-        re.IGNORECASE,
-    )
-
-    def _process_match(
-        self,
-        match: Match[str],
-        date_order: str,
-    ) -> datetime.datetime | None:
-        """
-        Processes a single regex match using the base class helpers.
-        """
-        date_string = match.group(0)
-        date = self._parse_string(date_string, date_order)
-        return self._filter_date(date)
-
-    def _process_content(
-        self,
-        content: str,
-        date_order: str,
-    ) -> Iterator[datetime.datetime]:
-        """
-        Finds all regex matches in content and yields valid dates.
-        """
-        for m in re.finditer(self.DATE_REGEX, content):
-            date = self._process_match(m, date_order)
-            if date is not None:
-                yield date
-
-    def parse(self, filename: str, content: str) -> Iterator[datetime.datetime]:
-        """
-        Implementation of the abstract parse method.
-
-        Reads its configuration from `self.config`.
-        """
-        if self.config.filename_date_order:
-            yield from self._process_content(
-                filename,
-                self.config.filename_date_order,
-            )
-
-        yield from self._process_content(content, self.config.content_date_order)

src/documents/tests/date_parsing/conftest.py

@@ -1,82 +0,0 @@
-import datetime
-from collections.abc import Generator
-from typing import Any
-
-import pytest
-import pytest_django
-
-from documents.plugins.date_parsing import _discover_parser_class
-from documents.plugins.date_parsing.base import DateParserConfig
-from documents.plugins.date_parsing.regex_parser import RegexDateParserPlugin
-
-
-@pytest.fixture
-def base_config() -> DateParserConfig:
-    """Basic configuration for date parser testing."""
-    return DateParserConfig(
-        languages=["en"],
-        timezone_str="UTC",
-        ignore_dates=set(),
-        reference_time=datetime.datetime(
-            2024,
-            1,
-            15,
-            12,
-            0,
-            0,
-            tzinfo=datetime.timezone.utc,
-        ),
-        filename_date_order="YMD",
-        content_date_order="DMY",
-    )
-
-
-@pytest.fixture
-def config_with_ignore_dates() -> DateParserConfig:
-    """Configuration with dates to ignore."""
-    return DateParserConfig(
-        languages=["en", "de"],
-        timezone_str="America/New_York",
-        ignore_dates={datetime.date(2024, 1, 1), datetime.date(2024, 12, 25)},
-        reference_time=datetime.datetime(
-            2024,
-            1,
-            15,
-            12,
-            0,
-            0,
-            tzinfo=datetime.timezone.utc,
-        ),
-        filename_date_order="DMY",
-        content_date_order="MDY",
-    )
-
-
-@pytest.fixture
-def regex_parser(base_config: DateParserConfig) -> RegexDateParserPlugin:
-    """Instance of RegexDateParser with base config."""
-    return RegexDateParserPlugin(base_config)
-
-
-@pytest.fixture
-def clear_lru_cache() -> Generator[None, None, None]:
-    """
-    Ensure the LRU cache for _discover_parser_class is cleared
-    before and after any test that depends on it.
-    """
-    _discover_parser_class.cache_clear()
-    yield
-    _discover_parser_class.cache_clear()
-
-
-@pytest.fixture
-def mock_date_parser_settings(settings: pytest_django.fixtures.SettingsWrapper) -> Any:
-    """
-    Override Django settings for the duration of date parser tests.
-    """
-    settings.DATE_PARSER_LANGUAGES = ["en", "de"]
-    settings.TIME_ZONE = "UTC"
-    settings.IGNORE_DATES = [datetime.date(1900, 1, 1)]
-    settings.FILENAME_DATE_ORDER = "YMD"
-    settings.DATE_ORDER = "DMY"
-    return settings

src/documents/tests/date_parsing/test_date_parser_plugin_loading.py

@@ -1,228 +0,0 @@
-import datetime
-import logging
-from collections.abc import Iterator
-from importlib.metadata import EntryPoint
-
-import pytest
-import pytest_mock
-from django.utils import timezone
-
-from documents.plugins.date_parsing import DATE_PARSER_ENTRY_POINT_GROUP
-from documents.plugins.date_parsing import _discover_parser_class
-from documents.plugins.date_parsing import get_date_parser
-from documents.plugins.date_parsing.base import DateParserConfig
-from documents.plugins.date_parsing.base import DateParserPluginBase
-from documents.plugins.date_parsing.regex_parser import RegexDateParserPlugin
-
-
-class AlphaParser(DateParserPluginBase):
-    def parse(self, filename: str, content: str) -> Iterator[datetime.datetime]:
-        yield timezone.now()
-
-
-class BetaParser(DateParserPluginBase):
-    def parse(self, filename: str, content: str) -> Iterator[datetime.datetime]:
-        yield timezone.now()
-
-
-@pytest.mark.date_parsing
-@pytest.mark.usefixtures("clear_lru_cache")
-class TestDiscoverParserClass:
-    """Tests for the _discover_parser_class() function."""
-
-    def test_returns_default_when_no_plugins_found(
-        self,
-        mocker: pytest_mock.MockerFixture,
-    ) -> None:
-        mocker.patch(
-            "documents.plugins.date_parsing.entry_points",
-            return_value=(),
-        )
-        result = _discover_parser_class()
-        assert result is RegexDateParserPlugin
-
-    def test_returns_default_when_entrypoint_query_fails(
-        self,
-        mocker: pytest_mock.MockerFixture,
-        caplog: pytest.LogCaptureFixture,
-    ) -> None:
-        mocker.patch(
-            "documents.plugins.date_parsing.entry_points",
-            side_effect=RuntimeError("boom"),
-        )
-        result = _discover_parser_class()
-        assert result is RegexDateParserPlugin
-        assert "Could not query entry points" in caplog.text
-
-    def test_filters_out_invalid_plugins(
-        self,
-        mocker: pytest_mock.MockerFixture,
-        caplog: pytest.LogCaptureFixture,
-    ) -> None:
-        fake_ep = mocker.MagicMock(spec=EntryPoint)
-        fake_ep.name = "bad_plugin"
-        fake_ep.load.return_value = object  # not subclass of DateParser
-
-        mocker.patch(
-            "documents.plugins.date_parsing.entry_points",
-            return_value=(fake_ep,),
-        )
-
-        result = _discover_parser_class()
-        assert result is RegexDateParserPlugin
-        assert "does not subclass DateParser" in caplog.text
-
-    def test_skips_plugins_that_fail_to_load(
-        self,
-        mocker: pytest_mock.MockerFixture,
-        caplog: pytest.LogCaptureFixture,
-    ) -> None:
-        fake_ep = mocker.MagicMock(spec=EntryPoint)
-        fake_ep.name = "failing_plugin"
-        fake_ep.load.side_effect = ImportError("cannot import")
-
-        mocker.patch(
-            "documents.plugins.date_parsing.entry_points",
-            return_value=(fake_ep,),
-        )
-
-        result = _discover_parser_class()
-        assert result is RegexDateParserPlugin
-        assert "Unable to load date parser plugin failing_plugin" in caplog.text
-
-    def test_returns_single_valid_plugin_without_warning(
-        self,
-        mocker: pytest_mock.MockerFixture,
-        caplog: pytest.LogCaptureFixture,
-    ) -> None:
-        """If exactly one valid plugin is discovered, it should be returned without logging a warning."""
-
-        ep = mocker.MagicMock(spec=EntryPoint)
-        ep.name = "alpha"
-        ep.load.return_value = AlphaParser
-
-        mock_entry_points = mocker.patch(
-            "documents.plugins.date_parsing.entry_points",
-            return_value=(ep,),
-        )
-
-        with caplog.at_level(
-            logging.WARNING,
-            logger="documents.plugins.date_parsing",
-        ):
-            result = _discover_parser_class()
-
-        # It should have called entry_points with the correct group
-        mock_entry_points.assert_called_once_with(group=DATE_PARSER_ENTRY_POINT_GROUP)
-
-        # The discovered class should be exactly our AlphaParser
-        assert result is AlphaParser
-
-        # No warnings should have been logged
-        assert not any(
-            "Multiple date parsers found" in record.message for record in caplog.records
-        ), "Unexpected warning logged when only one plugin was found"
-
-    def test_returns_first_valid_plugin_by_name(
-        self,
-        mocker: pytest_mock.MockerFixture,
-    ) -> None:
-        ep_a = mocker.MagicMock(spec=EntryPoint)
-        ep_a.name = "alpha"
-        ep_a.load.return_value = AlphaParser
-
-        ep_b = mocker.MagicMock(spec=EntryPoint)
-        ep_b.name = "beta"
-        ep_b.load.return_value = BetaParser
-
-        mocker.patch(
-            "documents.plugins.date_parsing.entry_points",
-            return_value=(ep_b, ep_a),
-        )
-
-        result = _discover_parser_class()
-        assert result is AlphaParser
-
-    def test_logs_warning_if_multiple_plugins_found(
-        self,
-        mocker: pytest_mock.MockerFixture,
-        caplog: pytest.LogCaptureFixture,
-    ) -> None:
-        ep1 = mocker.MagicMock(spec=EntryPoint)
-        ep1.name = "a"
-        ep1.load.return_value = AlphaParser
-
-        ep2 = mocker.MagicMock(spec=EntryPoint)
-        ep2.name = "b"
-        ep2.load.return_value = BetaParser
-
-        mocker.patch(
-            "documents.plugins.date_parsing.entry_points",
-            return_value=(ep1, ep2),
-        )
-
-        with caplog.at_level(
-            logging.WARNING,
-            logger="documents.plugins.date_parsing",
-        ):
-            result = _discover_parser_class()
-
-        # Should select alphabetically first plugin ("a")
-        assert result is AlphaParser
-
-        # Should log a warning mentioning multiple parsers
-        assert any(
-            "Multiple date parsers found" in record.message for record in caplog.records
-        ), "Expected a warning about multiple date parsers"
-
-    def test_cache_behavior_only_runs_once(
-        self,
-        mocker: pytest_mock.MockerFixture,
-    ) -> None:
-        mock_entry_points = mocker.patch(
-            "documents.plugins.date_parsing.entry_points",
-            return_value=(),
-        )
-
-        # First call populates cache
-        _discover_parser_class()
-        # Second call should not re-invoke entry_points
-        _discover_parser_class()
-        mock_entry_points.assert_called_once()
-
-
-@pytest.mark.date_parsing
-@pytest.mark.usefixtures("mock_date_parser_settings")
-class TestGetDateParser:
-    """Tests for the get_date_parser() factory function."""
-
-    def test_returns_instance_of_discovered_class(
-        self,
-        mocker: pytest_mock.MockerFixture,
-    ) -> None:
-        mocker.patch(
-            "documents.plugins.date_parsing._discover_parser_class",
-            return_value=AlphaParser,
-        )
-        parser = get_date_parser()
-        assert isinstance(parser, AlphaParser)
-        assert isinstance(parser.config, DateParserConfig)
-        assert parser.config.languages == ["en", "de"]
-        assert parser.config.timezone_str == "UTC"
-        assert parser.config.ignore_dates == [datetime.date(1900, 1, 1)]
-        assert parser.config.filename_date_order == "YMD"
-        assert parser.config.content_date_order == "DMY"
-        # Check reference_time near now
-        delta = abs((parser.config.reference_time - timezone.now()).total_seconds())
-        assert delta < 2
-
-    def test_uses_default_regex_parser_when_no_plugins(
-        self,
-        mocker: pytest_mock.MockerFixture,
-    ) -> None:
-        mocker.patch(
-            "documents.plugins.date_parsing._discover_parser_class",
-            return_value=RegexDateParserPlugin,
-        )
-        parser = get_date_parser()
-        assert isinstance(parser, RegexDateParserPlugin)

src/documents/tests/date_parsing/test_date_parsing.py

@@ -1,433 +0,0 @@
-import datetime
-import logging
-from typing import Any
-
-import pytest
-import pytest_mock
-
-from documents.plugins.date_parsing.base import DateParserConfig
-from documents.plugins.date_parsing.regex_parser import RegexDateParserPlugin
-
-
-@pytest.mark.date_parsing
-class TestParseString:
-    """Tests for DateParser._parse_string method via RegexDateParser."""
-
-    @pytest.mark.parametrize(
-        ("date_string", "date_order", "expected_year"),
-        [
-            pytest.param("15/01/2024", "DMY", 2024, id="dmy_slash"),
-            pytest.param("01/15/2024", "MDY", 2024, id="mdy_slash"),
-            pytest.param("2024/01/15", "YMD", 2024, id="ymd_slash"),
-            pytest.param("January 15, 2024", "DMY", 2024, id="month_name_comma"),
-            pytest.param("15 Jan 2024", "DMY", 2024, id="day_abbr_month_year"),
-            pytest.param("15.01.2024", "DMY", 2024, id="dmy_dot"),
-            pytest.param("2024-01-15", "YMD", 2024, id="ymd_dash"),
-        ],
-    )
-    def test_parse_string_valid_formats(
-        self,
-        regex_parser: RegexDateParserPlugin,
-        date_string: str,
-        date_order: str,
-        expected_year: int,
-    ) -> None:
-        """Should correctly parse various valid date formats."""
-        result = regex_parser._parse_string(date_string, date_order)
-
-        assert result is not None
-        assert result.year == expected_year
-
-    @pytest.mark.parametrize(
-        "invalid_string",
-        [
-            pytest.param("not a date", id="plain_text"),
-            pytest.param("32/13/2024", id="invalid_day_month"),
-            pytest.param("", id="empty_string"),
-            pytest.param("abc123xyz", id="alphanumeric_gibberish"),
-            pytest.param("99/99/9999", id="out_of_range"),
-        ],
-    )
-    def test_parse_string_invalid_input(
-        self,
-        regex_parser: RegexDateParserPlugin,
-        invalid_string: str,
-    ) -> None:
-        """Should return None for invalid date strings."""
-        result = regex_parser._parse_string(invalid_string, "DMY")
-
-        assert result is None
-
-    def test_parse_string_handles_exceptions(
-        self,
-        caplog: pytest.LogCaptureFixture,
-        mocker: pytest_mock.MockerFixture,
-        regex_parser: RegexDateParserPlugin,
-    ) -> None:
-        """Should handle and log exceptions from dateparser gracefully."""
-        with caplog.at_level(
-            logging.ERROR,
-            logger="documents.plugins.date_parsing.base",
-        ):
-            # We still need to mock dateparser.parse to force the exception
-            mocker.patch(
-                "documents.plugins.date_parsing.base.dateparser.parse",
-                side_effect=ValueError(
-                    "Parsing error: 01/01/2024",
-                ),
-            )
-
-            # 1. Execute the function under test
-            result = regex_parser._parse_string("01/01/2024", "DMY")
-
-            assert result is None
-
-            # Check if an error was logged
-            assert len(caplog.records) == 1
-            assert caplog.records[0].levelname == "ERROR"
-
-            # Check if the specific error message is present
-            assert "Error while parsing date string" in caplog.text
-            # Optional: Check for the exact exception message if it's included in the log
-            assert "Parsing error: 01/01/2024" in caplog.text
-
-
-@pytest.mark.date_parsing
-class TestFilterDate:
-    """Tests for DateParser._filter_date method via RegexDateParser."""
-
-    @pytest.mark.parametrize(
-        ("date", "expected_output"),
-        [
-            # Valid Dates
-            pytest.param(
-                datetime.datetime(2024, 1, 10, tzinfo=datetime.timezone.utc),
-                datetime.datetime(2024, 1, 10, tzinfo=datetime.timezone.utc),
-                id="valid_past_date",
-            ),
-            pytest.param(
-                datetime.datetime(2024, 1, 15, 12, 0, 0, tzinfo=datetime.timezone.utc),
-                datetime.datetime(2024, 1, 15, 12, 0, 0, tzinfo=datetime.timezone.utc),
-                id="exactly_at_reference",
-            ),
-            pytest.param(
-                datetime.datetime(1901, 1, 1, tzinfo=datetime.timezone.utc),
-                datetime.datetime(1901, 1, 1, tzinfo=datetime.timezone.utc),
-                id="year_1901_valid",
-            ),
-            # Date is > reference_time
-            pytest.param(
-                datetime.datetime(2024, 1, 16, tzinfo=datetime.timezone.utc),
-                None,
-                id="future_date_day_after",
-            ),
-            # date.date() in ignore_dates
-            pytest.param(
-                datetime.datetime(2024, 1, 1, 0, 0, 0, tzinfo=datetime.timezone.utc),
-                None,
-                id="ignored_date_midnight_jan1",
-            ),
-            pytest.param(
-                datetime.datetime(2024, 1, 1, 10, 30, 0, tzinfo=datetime.timezone.utc),
-                None,
-                id="ignored_date_midday_jan1",
-            ),
-            pytest.param(
-                datetime.datetime(2024, 12, 25, 15, 0, 0, tzinfo=datetime.timezone.utc),
-                None,
-                id="ignored_date_dec25_future",
-            ),
-            # date.year <= 1900
-            pytest.param(
-                datetime.datetime(1899, 12, 31, tzinfo=datetime.timezone.utc),
-                None,
-                id="year_1899",
-            ),
-            pytest.param(
-                datetime.datetime(1900, 1, 1, tzinfo=datetime.timezone.utc),
-                None,
-                id="year_1900_boundary",
-            ),
-            # date is None
-            pytest.param(None, None, id="none_input"),
-        ],
-    )
-    def test_filter_date_validation_rules(
-        self,
-        config_with_ignore_dates: DateParserConfig,
-        date: datetime.datetime | None,
-        expected_output: datetime.datetime | None,
-    ) -> None:
-        """Should correctly validate dates against various rules."""
-        parser = RegexDateParserPlugin(config_with_ignore_dates)
-        result = parser._filter_date(date)
-        assert result == expected_output
-
-    def test_filter_date_respects_ignore_dates(
-        self,
-        config_with_ignore_dates: DateParserConfig,
-    ) -> None:
-        """Should filter out dates in the ignore_dates set."""
-        parser = RegexDateParserPlugin(config_with_ignore_dates)
-
-        ignored_date = datetime.datetime(
-            2024,
-            1,
-            1,
-            12,
-            0,
-            tzinfo=datetime.timezone.utc,
-        )
-        another_ignored = datetime.datetime(
-            2024,
-            12,
-            25,
-            15,
-            30,
-            tzinfo=datetime.timezone.utc,
-        )
-        allowed_date = datetime.datetime(
-            2024,
-            1,
-            2,
-            12,
-            0,
-            tzinfo=datetime.timezone.utc,
-        )
-
-        assert parser._filter_date(ignored_date) is None
-        assert parser._filter_date(another_ignored) is None
-        assert parser._filter_date(allowed_date) == allowed_date
-
-    def test_filter_date_timezone_aware(
-        self,
-        regex_parser: RegexDateParserPlugin,
-    ) -> None:
-        """Should work with timezone-aware datetimes."""
-        date_utc = datetime.datetime(2024, 1, 10, 12, 0, tzinfo=datetime.timezone.utc)
-
-        result = regex_parser._filter_date(date_utc)
-
-        assert result is not None
-        assert result.tzinfo is not None
-
-
-@pytest.mark.date_parsing
-class TestRegexDateParser:
-    @pytest.mark.parametrize(
-        ("filename", "content", "expected"),
-        [
-            pytest.param(
-                "report-2023-12-25.txt",
-                "Event recorded on 25/12/2022.",
-                [
-                    datetime.datetime(2023, 12, 25, tzinfo=datetime.timezone.utc),
-                    datetime.datetime(2022, 12, 25, tzinfo=datetime.timezone.utc),
-                ],
-                id="filename-y-m-d_and_content-d-m-y",
-            ),
-            pytest.param(
-                "img_2023.01.02.jpg",
-                "Taken on 01/02/2023",
-                [
-                    datetime.datetime(2023, 1, 2, tzinfo=datetime.timezone.utc),
-                    datetime.datetime(2023, 2, 1, tzinfo=datetime.timezone.utc),
-                ],
-                id="ambiguous-dates-respect-orders",
-            ),
-            pytest.param(
-                "notes.txt",
-                "bad date 99/99/9999 and 25/12/2022",
-                [
-                    datetime.datetime(2022, 12, 25, tzinfo=datetime.timezone.utc),
-                ],
-                id="parse-exception-skips-bad-and-yields-good",
-            ),
-        ],
-    )
-    def test_parse_returns_expected_dates(
-        self,
-        base_config: DateParserConfig,
-        mocker: pytest_mock.MockerFixture,
-        filename: str,
-        content: str,
-        expected: list[datetime.datetime],
-    ) -> None:
-        """
-        High-level tests that exercise RegexDateParser.parse only.
-        dateparser.parse is mocked so tests are deterministic.
-        """
-        parser = RegexDateParserPlugin(base_config)
-
-        # Patch the dateparser.parse
-        target = "documents.plugins.date_parsing.base.dateparser.parse"
-
-        def fake_parse(
-            date_string: str,
-            settings: dict[str, Any] | None = None,
-            locales: None = None,
-        ) -> datetime.datetime | None:
-            date_order = settings.get("DATE_ORDER") if settings else None
-
-            # Filename-style YYYY-MM-DD / YYYY.MM.DD
-            if (
-                "2023-12-25" in date_string
-                or "2023.12.25" in date_string
-                or "2023-12-25" in date_string
-            ):
-                return datetime.datetime(2023, 12, 25, tzinfo=datetime.timezone.utc)
-
-            # content DMY 25/12/2022
-            if "25/12/2022" in date_string or "25-12-2022" in date_string:
-                return datetime.datetime(2022, 12, 25, tzinfo=datetime.timezone.utc)
-
-            # filename YMD 2023.01.02
-            if "2023.01.02" in date_string or "2023-01-02" in date_string:
-                return datetime.datetime(2023, 1, 2, tzinfo=datetime.timezone.utc)
-
-            # ambiguous 01/02/2023 -> respect DATE_ORDER setting
-            if "01/02/2023" in date_string:
-                if date_order == "DMY":
-                    return datetime.datetime(2023, 2, 1, tzinfo=datetime.timezone.utc)
-                if date_order == "YMD":
-                    return datetime.datetime(2023, 1, 2, tzinfo=datetime.timezone.utc)
-                # fallback
-                return datetime.datetime(2023, 2, 1, tzinfo=datetime.timezone.utc)
-
-            # simulate parse failure for malformed input
-            if "99/99/9999" in date_string or "bad date" in date_string:
-                raise Exception("parse failed for malformed date")
-
-            return None
-
-        mocker.patch(target, side_effect=fake_parse)
-
-        results = list(parser.parse(filename, content))
-
-        assert results == expected
-        for dt in results:
-            assert dt.tzinfo is not None
-
-    def test_parse_filters_future_and_ignored_dates(
-        self,
-        mocker: pytest_mock.MockerFixture,
-    ) -> None:
-        """
-        Ensure parser filters out:
-          - dates after reference_time
-          - dates whose .date() are in ignore_dates
-        """
-        cfg = DateParserConfig(
-            languages=["en"],
-            timezone_str="UTC",
-            ignore_dates={datetime.date(2023, 12, 10)},
-            reference_time=datetime.datetime(
-                2024,
-                1,
-                15,
-                12,
-                0,
-                0,
-                tzinfo=datetime.timezone.utc,
-            ),
-            filename_date_order="YMD",
-            content_date_order="DMY",
-        )
-        parser = RegexDateParserPlugin(cfg)
-
-        target = "documents.plugins.date_parsing.base.dateparser.parse"
-
-        def fake_parse(
-            date_string: str,
-            settings: dict[str, Any] | None = None,
-            locales: None = None,
-        ) -> datetime.datetime | None:
-            if "10/12/2023" in date_string or "10-12-2023" in date_string:
-                # ignored date
-                return datetime.datetime(2023, 12, 10, tzinfo=datetime.timezone.utc)
-            if "01/02/2024" in date_string or "01-02-2024" in date_string:
-                # future relative to reference_time -> filtered
-                return datetime.datetime(2024, 2, 1, tzinfo=datetime.timezone.utc)
-            if "05/01/2023" in date_string or "05-01-2023" in date_string:
-                # valid
-                return datetime.datetime(2023, 1, 5, tzinfo=datetime.timezone.utc)
-            return None
-
-        mocker.patch(target, side_effect=fake_parse)
-
-        content = "Ignored: 10/12/2023, Future: 01/02/2024, Keep: 05/01/2023"
-        results = list(parser.parse("whatever.txt", content))
-
-        assert results == [datetime.datetime(2023, 1, 5, tzinfo=datetime.timezone.utc)]
-
-    def test_parse_handles_no_matches_and_returns_empty_list(
-        self,
-        base_config: DateParserConfig,
-    ) -> None:
-        """
-        When there are no matching date-like substrings, parse should yield nothing.
-        """
-        parser = RegexDateParserPlugin(base_config)
-        results = list(
-            parser.parse("no-dates.txt", "this has no dates whatsoever"),
-        )
-        assert results == []
-
-    def test_parse_skips_filename_when_filename_date_order_none(
-        self,
-        mocker: pytest_mock.MockerFixture,
-    ) -> None:
-        """
-        When filename_date_order is None the parser must not attempt to parse the filename.
-        Only dates found in the content should be passed to dateparser.parse.
-        """
-        cfg = DateParserConfig(
-            languages=["en"],
-            timezone_str="UTC",
-            ignore_dates=set(),
-            reference_time=datetime.datetime(
-                2024,
-                1,
-                15,
-                12,
-                0,
-                0,
-                tzinfo=datetime.timezone.utc,
-            ),
-            filename_date_order=None,
-            content_date_order="DMY",
-        )
-        parser = RegexDateParserPlugin(cfg)
-
-        # Patch the module's dateparser.parse so we can inspect calls
-        target = "documents.plugins.date_parsing.base.dateparser.parse"
-
-        def fake_parse(
-            date_string: str,
-            settings: dict[str, Any] | None = None,
-            locales: None = None,
-        ) -> datetime.datetime | None:
-            # return distinct datetimes so we can tell which source was parsed
-            if "25/12/2022" in date_string:
-                return datetime.datetime(2022, 12, 25, tzinfo=datetime.timezone.utc)
-            if "2023-12-25" in date_string:
-                return datetime.datetime(2023, 12, 25, tzinfo=datetime.timezone.utc)
-            return None
-
-        mock = mocker.patch(target, side_effect=fake_parse)
-
-        filename = "report-2023-12-25.txt"
-        content = "Event recorded on 25/12/2022."
-
-        results = list(parser.parse(filename, content))
-
-        # Only the content date should have been parsed -> one call
-        assert mock.call_count == 1
-
-        # # first call, first positional arg
-        called_date_string = mock.call_args_list[0][0][0]
-        assert "25/12/2022" in called_date_string
-        # And the parser should have yielded the corresponding datetime
-        assert results == [
-            datetime.datetime(2022, 12, 25, tzinfo=datetime.timezone.utc),
-        ]

src/documents/tests/test_api_documents.py

@@ -1978,11 +1978,11 @@ class TestDocumentApi(DirectoriesMixin, DocumentConsumeDelayMixin, APITestCase):
         response = self.client.get(f"/api/documents/{doc.pk}/suggestions/")
         self.assertEqual(response.status_code, status.HTTP_200_OK)
 
-    @mock.patch("documents.views.get_date_parser")
+    @mock.patch("documents.parsers.parse_date_generator")
     @override_settings(NUMBER_OF_SUGGESTED_DATES=0)
     def test_get_suggestions_dates_disabled(
         self,
-        mock_get_date_parser: mock.MagicMock,
+        parse_date_generator,
     ):
         """
         GIVEN:
@@ -1999,8 +1999,7 @@ class TestDocumentApi(DirectoriesMixin, DocumentConsumeDelayMixin, APITestCase):
         )
 
         self.client.get(f"/api/documents/{doc.pk}/suggestions/")
-
-        mock_get_date_parser.assert_not_called()
+        self.assertFalse(parse_date_generator.called)
 
     def test_saved_views(self):
         u1 = User.objects.create_superuser("user1")

src/documents/tests/test_date_parsing.py

@@ -0,0 +1,538 @@
+import datetime
+from zoneinfo import ZoneInfo
+
+import pytest
+from pytest_django.fixtures import SettingsWrapper
+
+from documents.parsers import parse_date
+from documents.parsers import parse_date_generator
+
+
+@pytest.mark.django_db()
+class TestDate:
+    def test_date_format_1(self):
+        text = "lorem ipsum 130218 lorem ipsum"
+        assert parse_date("", text) is None
+
+    def test_date_format_2(self):
+        text = "lorem ipsum 2018 lorem ipsum"
+        assert parse_date("", text) is None
+
+    def test_date_format_3(self):
+        text = "lorem ipsum 20180213 lorem ipsum"
+        assert parse_date("", text) is None
+
+    def test_date_format_4(self, settings_timezone: ZoneInfo):
+        text = "lorem ipsum 13.02.2018 lorem ipsum"
+        date = parse_date("", text)
+        assert date == datetime.datetime(2018, 2, 13, 0, 0, tzinfo=settings_timezone)
+
+    def test_date_format_5(self, settings_timezone: ZoneInfo):
+        text = "lorem ipsum 130218, 2018, 20180213 and lorem 13.02.2018 lorem ipsum"
+        date = parse_date("", text)
+        assert date == datetime.datetime(2018, 2, 13, 0, 0, tzinfo=settings_timezone)
+
+    def test_date_format_6(self):
+        text = (
+            "lorem ipsum\n"
+            "Wohnort\n"
+            "3100\n"
+            "IBAN\n"
+            "AT87 4534\n"
+            "1234\n"
+            "1234 5678\n"
+            "BIC\n"
+            "lorem ipsum"
+        )
+        assert parse_date("", text) is None
+
+    def test_date_format_7(
+        self,
+        settings: SettingsWrapper,
+        settings_timezone: ZoneInfo,
+    ):
+        settings.DATE_PARSER_LANGUAGES = ["de"]
+        text = "lorem ipsum\nMärz 2019\nlorem ipsum"
+        date = parse_date("", text)
+        assert date == datetime.datetime(2019, 3, 1, 0, 0, tzinfo=settings_timezone)
+
+    def test_date_format_8(
+        self,
+        settings: SettingsWrapper,
+        settings_timezone: ZoneInfo,
+    ):
+        settings.DATE_PARSER_LANGUAGES = ["de"]
+        text = (
+            "lorem ipsum\n"
+            "Wohnort\n"
+            "3100\n"
+            "IBAN\n"
+            "AT87 4534\n"
+            "1234\n"
+            "1234 5678\n"
+            "BIC\n"
+            "lorem ipsum\n"
+            "März 2020"
+        )
+        assert parse_date("", text) == datetime.datetime(
+            2020,
+            3,
+            1,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_9(
+        self,
+        settings: SettingsWrapper,
+        settings_timezone: ZoneInfo,
+    ):
+        settings.DATE_PARSER_LANGUAGES = ["de"]
+        text = "lorem ipsum\n27. Nullmonth 2020\nMärz 2020\nlorem ipsum"
+        assert parse_date("", text) == datetime.datetime(
+            2020,
+            3,
+            1,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_10(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 22-MAR-2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            22,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_11(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 22 MAR 2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            22,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_12(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 22/MAR/2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            22,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_13(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 22.MAR.2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            22,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_14(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 22.MAR 2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            22,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_15(self):
+        text = "Customer Number Currency 22.MAR.22 Credit Card 1934829304"
+        assert parse_date("", text) is None
+
+    def test_date_format_16(self):
+        text = "Customer Number Currency 22.MAR,22 Credit Card 1934829304"
+        assert parse_date("", text) is None
+
+    def test_date_format_17(self):
+        text = "Customer Number Currency 22,MAR,2022 Credit Card 1934829304"
+        assert parse_date("", text) is None
+
+    def test_date_format_18(self):
+        text = "Customer Number Currency 22 MAR,2022 Credit Card 1934829304"
+        assert parse_date("", text) is None
+
+    def test_date_format_19(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 21st MAR 2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            21,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_20(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 22nd March 2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            22,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_21(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 2nd MAR 2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            2,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_22(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 23rd MAR 2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            23,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_23(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 24th MAR 2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            24,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_24(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 21-MAR-2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            21,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_25(self, settings_timezone: ZoneInfo):
+        text = "Customer Number Currency 25TH MAR 2022 Credit Card 1934829304"
+        assert parse_date("", text) == datetime.datetime(
+            2022,
+            3,
+            25,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_date_format_26(self, settings_timezone: ZoneInfo):
+        text = "CHASE 0 September 25, 2019 JPMorgan Chase Bank, NA. P0 Box 182051"
+        assert parse_date("", text) == datetime.datetime(
+            2019,
+            9,
+            25,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_crazy_date_past(self):
+        assert parse_date("", "01-07-0590 00:00:00") is None
+
+    def test_crazy_date_future(self):
+        assert parse_date("", "01-07-2350 00:00:00") is None
+
+    def test_crazy_date_with_spaces(self):
+        assert parse_date("", "20 408000l 2475") is None
+
+    def test_utf_month_names(
+        self,
+        settings: SettingsWrapper,
+        settings_timezone: ZoneInfo,
+    ):
+        settings.DATE_PARSER_LANGUAGES = ["fr", "de", "hr", "cs", "pl", "tr"]
+        assert parse_date("", "13 décembre 2023") == datetime.datetime(
+            2023,
+            12,
+            13,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "13 août 2022") == datetime.datetime(
+            2022,
+            8,
+            13,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "11 März 2020") == datetime.datetime(
+            2020,
+            3,
+            11,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "17. ožujka 2018.") == datetime.datetime(
+            2018,
+            3,
+            17,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "1. veljače 2016.") == datetime.datetime(
+            2016,
+            2,
+            1,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "15. února 1985") == datetime.datetime(
+            1985,
+            2,
+            15,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "30. září 2011") == datetime.datetime(
+            2011,
+            9,
+            30,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "28. května 1990") == datetime.datetime(
+            1990,
+            5,
+            28,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "1. grudzień 1997") == datetime.datetime(
+            1997,
+            12,
+            1,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "17 Şubat 2024") == datetime.datetime(
+            2024,
+            2,
+            17,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "30 Ağustos 2012") == datetime.datetime(
+            2012,
+            8,
+            30,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "17 Eylül 2000") == datetime.datetime(
+            2000,
+            9,
+            17,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+        assert parse_date("", "5. október 1992") == datetime.datetime(
+            1992,
+            10,
+            5,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_multiple_dates(self, settings_timezone: ZoneInfo):
+        text = """This text has multiple dates.
+                  For example 02.02.2018, 22 July 2022 and December 2021.
+                  But not 24-12-9999 because it's in the future..."""
+        dates = list(parse_date_generator("", text))
+
+        assert dates == [
+            datetime.datetime(2018, 2, 2, 0, 0, tzinfo=settings_timezone),
+            datetime.datetime(
+                2022,
+                7,
+                22,
+                0,
+                0,
+                tzinfo=settings_timezone,
+            ),
+            datetime.datetime(
+                2021,
+                12,
+                1,
+                0,
+                0,
+                tzinfo=settings_timezone,
+            ),
+        ]
+
+    def test_filename_date_parse_valid_ymd(
+        self,
+        settings: SettingsWrapper,
+        settings_timezone: ZoneInfo,
+    ):
+        """
+        GIVEN:
+            - Date parsing from the filename is enabled
+            - Filename date format is with Year Month Day (YMD)
+            - Filename contains date matching the format
+
+        THEN:
+            - Should parse the date from the filename
+        """
+        settings.FILENAME_DATE_ORDER = "YMD"
+
+        assert parse_date(
+            "/tmp/Scan-2022-04-01.pdf",
+            "No date in here",
+        ) == datetime.datetime(2022, 4, 1, 0, 0, tzinfo=settings_timezone)
+
+    def test_filename_date_parse_valid_dmy(
+        self,
+        settings: SettingsWrapper,
+        settings_timezone: ZoneInfo,
+    ):
+        """
+        GIVEN:
+            - Date parsing from the filename is enabled
+            - Filename date format is with Day Month Year (DMY)
+            - Filename contains date matching the format
+
+        THEN:
+            - Should parse the date from the filename
+        """
+        settings.FILENAME_DATE_ORDER = "DMY"
+        assert parse_date(
+            "/tmp/Scan-10.01.2021.pdf",
+            "No date in here",
+        ) == datetime.datetime(2021, 1, 10, 0, 0, tzinfo=settings_timezone)
+
+    def test_filename_date_parse_invalid(self, settings: SettingsWrapper):
+        """
+        GIVEN:
+            - Date parsing from the filename is enabled
+            - Filename includes no date
+            - File content includes no date
+
+        THEN:
+            - No date is parsed
+        """
+        settings.FILENAME_DATE_ORDER = "YMD"
+        assert parse_date("/tmp/20 408000l 2475 - test.pdf", "No date in here") is None
+
+    def test_filename_date_ignored_use_content(
+        self,
+        settings: SettingsWrapper,
+        settings_timezone: ZoneInfo,
+    ):
+        """
+        GIVEN:
+            - Date parsing from the filename is enabled
+            - Filename date format is with Day Month Year (YMD)
+            - Date order is Day Month Year (DMY, the default)
+            - Filename contains date matching the format
+            - Filename date is an ignored date
+            - File content includes a date
+
+        THEN:
+            - Should parse the date from the content not filename
+        """
+        settings.FILENAME_DATE_ORDER = "YMD"
+        settings.IGNORE_DATES = (datetime.date(2022, 4, 1),)
+        assert parse_date(
+            "/tmp/Scan-2022-04-01.pdf",
+            "The matching date is 24.03.2022",
+        ) == datetime.datetime(2022, 3, 24, 0, 0, tzinfo=settings_timezone)
+
+    def test_ignored_dates_default_order(
+        self,
+        settings: SettingsWrapper,
+        settings_timezone: ZoneInfo,
+    ):
+        """
+        GIVEN:
+            - Ignore dates have been set
+            - File content includes ignored dates
+            - File content includes 1 non-ignored date
+
+        THEN:
+            - Should parse the date non-ignored date from content
+        """
+        settings.IGNORE_DATES = (datetime.date(2019, 11, 3), datetime.date(2020, 1, 17))
+        text = "lorem ipsum 110319, 20200117 and lorem 13.02.2018 lorem ipsum"
+        assert parse_date("", text) == datetime.datetime(
+            2018,
+            2,
+            13,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )
+
+    def test_ignored_dates_order_ymd(
+        self,
+        settings: SettingsWrapper,
+        settings_timezone: ZoneInfo,
+    ):
+        """
+        GIVEN:
+            - Ignore dates have been set
+            - Date order is Year Month Date (YMD)
+            - File content includes ignored dates
+            - File content includes 1 non-ignored date
+
+        THEN:
+            - Should parse the date non-ignored date from content
+        """
+
+        settings.FILENAME_DATE_ORDER = "YMD"
+        settings.IGNORE_DATES = (datetime.date(2019, 11, 3), datetime.date(2020, 1, 17))
+
+        text = "lorem ipsum 190311, 20200117 and lorem 13.02.2018 lorem ipsum"
+
+        assert parse_date("", text) == datetime.datetime(
+            2018,
+            2,
+            13,
+            0,
+            0,
+            tzinfo=settings_timezone,
+        )

src/documents/views.py

@@ -148,6 +148,7 @@ from documents.models import Workflow
 from documents.models import WorkflowAction
 from documents.models import WorkflowTrigger
 from documents.parsers import get_parser_class_for_mime_type
+from documents.parsers import parse_date_generator
 from documents.permissions import AcknowledgeTasksPermissions
 from documents.permissions import PaperlessAdminPermissions
 from documents.permissions import PaperlessNotePermissions
@@ -157,7 +158,6 @@ from documents.permissions import get_document_count_filter_for_user
 from documents.permissions import get_objects_for_user_owner_aware
 from documents.permissions import has_perms_owner_aware
 from documents.permissions import set_permissions_for_object
-from documents.plugins.date_parsing import get_date_parser
 from documents.schema import generate_object_with_permissions_schema
 from documents.serialisers import AcknowledgeTasksViewSerializer
 from documents.serialisers import BulkDownloadSerializer
@@ -1023,17 +1023,16 @@ class DocumentViewSet(
 
             dates = []
             if settings.NUMBER_OF_SUGGESTED_DATES > 0:
-                with get_date_parser() as date_parser:
-                    gen = date_parser.parse(doc.filename, doc.content)
-                    dates = sorted(
-                        {
-                            i
-                            for i in itertools.islice(
-                                gen,
-                                settings.NUMBER_OF_SUGGESTED_DATES,
-                            )
-                        },
-                    )
+                gen = parse_date_generator(doc.filename, doc.content)
+                dates = sorted(
+                    {
+                        i
+                        for i in itertools.islice(
+                            gen,
+                            settings.NUMBER_OF_SUGGESTED_DATES,
+                        )
+                    },
+                )
 
             resp_data = {
                 "correspondents": [

src/paperless_mail/views.py

@@ -1,7 +1,12 @@
 import datetime
 import logging
 from datetime import timedelta
+from typing import Any
 
+from adrf.views import APIView
+from adrf.viewsets import ModelViewSet
+from adrf.viewsets import ReadOnlyModelViewSet
+from asgiref.sync import sync_to_async
 from django.http import HttpResponseBadRequest
 from django.http import HttpResponseForbidden
 from django.http import HttpResponseRedirect
@@ -15,11 +20,9 @@ from httpx_oauth.oauth2 import GetAccessTokenError
 from rest_framework import serializers
 from rest_framework.decorators import action
 from rest_framework.filters import OrderingFilter
-from rest_framework.generics import GenericAPIView
 from rest_framework.permissions import IsAuthenticated
+from rest_framework.request import Request
 from rest_framework.response import Response
-from rest_framework.viewsets import ModelViewSet
-from rest_framework.viewsets import ReadOnlyModelViewSet
 
 from documents.filters import ObjectOwnedOrGrantedPermissionsFilter
 from documents.permissions import PaperlessObjectPermissions
@@ -39,6 +42,8 @@ from paperless_mail.serialisers import MailRuleSerializer
 from paperless_mail.serialisers import ProcessedMailSerializer
 from paperless_mail.tasks import process_mail_accounts
 
+logger: logging.Logger = logging.getLogger("paperless_mail")
+
 
 @extend_schema_view(
     test=extend_schema(
@@ -66,71 +71,75 @@ from paperless_mail.tasks import process_mail_accounts
     ),
 )
 class MailAccountViewSet(ModelViewSet, PassUserMixin):
-    model = MailAccount
-
     queryset = MailAccount.objects.all().order_by("pk")
     serializer_class = MailAccountSerializer
     pagination_class = StandardPagination
     permission_classes = (IsAuthenticated, PaperlessObjectPermissions)
     filter_backends = (ObjectOwnedOrGrantedPermissionsFilter,)
 
-    def get_permissions(self):
+    def get_permissions(self) -> list[Any]:
         if self.action == "test":
-            # Test action does not require object level permissions
-            self.permission_classes = (IsAuthenticated,)
+            return [IsAuthenticated()]
         return super().get_permissions()
 
     @action(methods=["post"], detail=False)
-    def test(self, request):
-        logger = logging.getLogger("paperless_mail")
+    async def test(self, request: Request) -> Response | HttpResponseBadRequest:
         request.data["name"] = datetime.datetime.now().isoformat()
         serializer = self.get_serializer(data=request.data)
-        serializer.is_valid(raise_exception=True)
 
-        # account exists, use the password from there instead of *** and refresh_token / expiration
+        # Validation must be wrapped because of sync DB validators
+        await sync_to_async(serializer.is_valid)(raise_exception=True)
+
+        validated_data: dict[str, Any] = serializer.validated_data
+
         if (
-            len(serializer.validated_data.get("password").replace("*", "")) == 0
-            and request.data["id"] is not None
+            len(str(validated_data.get("password", "")).replace("*", "")) == 0
+            and request.data.get("id") is not None
         ):
-            existing_account = MailAccount.objects.get(pk=request.data["id"])
-            serializer.validated_data["password"] = existing_account.password
-            serializer.validated_data["account_type"] = existing_account.account_type
-            serializer.validated_data["refresh_token"] = existing_account.refresh_token
-            serializer.validated_data["expiration"] = existing_account.expiration
+            existing_account = await MailAccount.objects.aget(pk=request.data["id"])
+            validated_data.update(
+                {
+                    "password": existing_account.password,
+                    "account_type": existing_account.account_type,
+                    "refresh_token": existing_account.refresh_token,
+                    "expiration": existing_account.expiration,
+                },
+            )
 
-        account = MailAccount(**serializer.validated_data)
-        with get_mailbox(
-            account.imap_server,
-            account.imap_port,
-            account.imap_security,
-        ) as M:
-            try:
+        account = MailAccount(**validated_data)
+
+        def _blocking_imap_test() -> bool:
+            with get_mailbox(
+                account.imap_server,
+                account.imap_port,
+                account.imap_security,
+            ) as m_box:
                 if (
                     account.is_token
-                    and account.expiration is not None
+                    and account.expiration
                     and account.expiration < timezone.now()
                 ):
                     oauth_manager = PaperlessMailOAuth2Manager()
                     if oauth_manager.refresh_account_oauth_token(existing_account):
                         # User is not changing password and token needs to be refreshed
-                        existing_account.refresh_from_db()
                         account.password = existing_account.password
                     else:
                         raise MailError("Unable to refresh oauth token")
+                mailbox_login(m_box, account)
+                return True
 
-                mailbox_login(M, account)
-                return Response({"success": True})
-            except MailError as e:
-                logger.error(
-                    f"Mail account {account} test failed: {e}",
-                )
-                return HttpResponseBadRequest("Unable to connect to server")
+        try:
+            await sync_to_async(_blocking_imap_test, thread_sensitive=False)()
+            return Response({"success": True})
+        except MailError as e:
+            logger.error(f"Mail account {account} test failed: {e}")
+            return HttpResponseBadRequest("Unable to connect to server")
 
     @action(methods=["post"], detail=True)
-    def process(self, request, pk=None):
-        account = self.get_object()
+    async def process(self, request: Request, pk: int | None = None) -> Response:
+        # FIX: Use aget_object() provided by adrf to avoid SynchronousOnlyOperation
+        account = await self.aget_object()
         process_mail_accounts.delay([account.pk])
-
         return Response({"result": "OK"})
 
 
@@ -144,21 +153,38 @@ class ProcessedMailViewSet(ReadOnlyModelViewSet, PassUserMixin):
         ObjectOwnedOrGrantedPermissionsFilter,
     )
     filterset_class = ProcessedMailFilterSet
-
     queryset = ProcessedMail.objects.all().order_by("-processed")
 
     @action(methods=["post"], detail=False)
-    def bulk_delete(self, request):
-        mail_ids = request.data.get("mail_ids", [])
+    async def bulk_delete(
+        self,
+        request: Request,
+    ) -> Response | HttpResponseBadRequest | HttpResponseForbidden:
+        mail_ids: list[int] = request.data.get("mail_ids", [])
         if not isinstance(mail_ids, list) or not all(
             isinstance(i, int) for i in mail_ids
         ):
             return HttpResponseBadRequest("mail_ids must be a list of integers")
-        mails = ProcessedMail.objects.filter(id__in=mail_ids)
-        for mail in mails:
-            if not has_perms_owner_aware(request.user, "delete_processedmail", mail):
+
+        # Store objects to delete after verification
+        to_delete: list[ProcessedMail] = []
+
+        # We must verify permissions for every requested ID
+        async for mail in ProcessedMail.objects.filter(id__in=mail_ids):
+            can_delete = await sync_to_async(has_perms_owner_aware)(
+                request.user,
+                "delete_processedmail",
+                mail,
+            )
+            if not can_delete:
+                # This is what the test is looking for: 403 on permission failure
                 return HttpResponseForbidden("Insufficient permissions")
-            mail.delete()
+            to_delete.append(mail)
+
+        # Only perform deletions if all items passed the permission check
+        for mail in to_delete:
+            await mail.adelete()
+
         return Response({"result": "OK", "deleted_mail_ids": mail_ids})
 
 
@@ -178,77 +204,74 @@ class MailRuleViewSet(ModelViewSet, PassUserMixin):
         responses={200: None},
     ),
 )
-class OauthCallbackView(GenericAPIView):
+class OauthCallbackView(APIView):
     permission_classes = (IsAuthenticated,)
 
-    def get(self, request, format=None):
-        if not (
-            request.user and request.user.has_perms(["paperless_mail.add_mailaccount"])
-        ):
+    async def get(
+        self,
+        request: Request,
+    ) -> Response | HttpResponseBadRequest | HttpResponseRedirect:
+        has_perm = await sync_to_async(request.user.has_perm)(
+            "paperless_mail.add_mailaccount",
+        )
+        if not has_perm:
             return HttpResponseBadRequest(
                 "You do not have permission to add mail accounts",
             )
 
-        logger = logging.getLogger("paperless_mail")
-        code = request.query_params.get("code")
-        # Gmail passes scope as a query param, Outlook does not
-        scope = request.query_params.get("scope")
+        code: str | None = request.query_params.get("code")
+        state: str | None = request.query_params.get("state")
+        scope: str | None = request.query_params.get("scope")
 
-        if code is None:
-            logger.error(
-                f"Invalid oauth callback request, code: {code}, scope: {scope}",
-            )
-            return HttpResponseBadRequest("Invalid request, see logs for more detail")
+        if not code or not state:
+            return HttpResponseBadRequest("Invalid request parameters")
 
         oauth_manager = PaperlessMailOAuth2Manager(
             state=request.session.get("oauth_state"),
         )
-
-        state = request.query_params.get("state", "")
         if not oauth_manager.validate_state(state):
-            logger.error(
-                f"Invalid oauth callback request received state: {state}, expected: {oauth_manager.state}",
-            )
-            return HttpResponseBadRequest("Invalid request, see logs for more detail")
+            return HttpResponseBadRequest("Invalid OAuth state")
 
         try:
-            if scope is not None and "google" in scope:
-                # Google
+            defaults: dict[str, Any] = {
+                "username": "",
+                "imap_security": MailAccount.ImapSecurity.SSL,
+                "imap_port": 993,
+            }
+
+            if scope and "google" in scope:
                 account_type = MailAccount.MailAccountType.GMAIL_OAUTH
                 imap_server = "imap.gmail.com"
-                defaults = {
-                    "name": f"Gmail OAuth {timezone.now()}",
-                    "username": "",
-                    "imap_security": MailAccount.ImapSecurity.SSL,
-                    "imap_port": 993,
-                    "account_type": account_type,
-                }
-                result = oauth_manager.get_gmail_access_token(code)
-
-            elif scope is None:
-                # Outlook
+                defaults.update(
+                    {
+                        "name": f"Gmail OAuth {timezone.now()}",
+                        "account_type": account_type,
+                    },
+                )
+                result = await sync_to_async(oauth_manager.get_gmail_access_token)(code)
+            else:
                 account_type = MailAccount.MailAccountType.OUTLOOK_OAUTH
                 imap_server = "outlook.office365.com"
-                defaults = {
-                    "name": f"Outlook OAuth {timezone.now()}",
-                    "username": "",
-                    "imap_security": MailAccount.ImapSecurity.SSL,
-                    "imap_port": 993,
-                    "account_type": account_type,
-                }
+                defaults.update(
+                    {
+                        "name": f"Outlook OAuth {timezone.now()}",
+                        "account_type": account_type,
+                    },
+                )
+                result = await sync_to_async(oauth_manager.get_outlook_access_token)(
+                    code,
+                )
 
-                result = oauth_manager.get_outlook_access_token(code)
-
-            access_token = result["access_token"]
-            refresh_token = result["refresh_token"]
-            expires_in = result["expires_in"]
-            account, _ = MailAccount.objects.update_or_create(
-                password=access_token,
-                is_token=True,
+            account, _ = await MailAccount.objects.aupdate_or_create(
                 imap_server=imap_server,
-                refresh_token=refresh_token,
-                expiration=timezone.now() + timedelta(seconds=expires_in),
-                defaults=defaults,
+                refresh_token=result["refresh_token"],
+                defaults={
+                    **defaults,
+                    "password": result["access_token"],
+                    "is_token": True,
+                    "expiration": timezone.now()
+                    + timedelta(seconds=result["expires_in"]),
+                },
             )
             return HttpResponseRedirect(
                 f"{oauth_manager.oauth_redirect_url}?oauth_success=1&account_id={account.pk}",