Enables prek workspace mode and moves hooks to specific locations as possible, moving to mdformat instead of prettier

Bumps [nltk](https://github.com/nltk/nltk) from 3.9.2 to 3.9.3.
- [Changelog](https://github.com/nltk/nltk/blob/develop/ChangeLog)
- [Commits](https://github.com/nltk/nltk/compare/3.9.2...3.9.3)

---
updated-dependencies:
- dependency-name: nltk
  dependency-version: 3.9.3
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

.github/workflows/ci-backend.yml

@@ -31,7 +31,7 @@ jobs:
     runs-on: ubuntu-24.04
     strategy:
       matrix:
-        python-version: ['3.11', '3.12', '3.13', '3.14']
+        python-version: ['3.10', '3.11', '3.12']
       fail-fast: false
     steps:
       - name: Checkout

.pre-commit-config.yaml

@@ -1,9 +1,4 @@
-# This file configures pre-commit hooks.
-# See https://pre-commit.com/ for general information
-# See https://pre-commit.com/hooks.html for a listing of possible hooks
-# We actually run via https://github.com/j178/prek which is compatible
 repos:
-  # General hooks
   - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v6.0.0
     hooks:
@@ -11,21 +6,16 @@ repos:
       - id: check-json
         exclude: "tsconfig.*json"
       - id: check-yaml
-        args:
+        args: ["--unsafe"]
-          - "--unsafe"
       - id: check-toml
       - id: check-executables-have-shebangs
       - id: end-of-file-fixer
-        exclude_types:
+        exclude_types: [svg, pofile]
-          - svg
-          - pofile
         exclude: "(^LICENSE$|^src/documents/static/bootstrap.min.css$)"
       - id: mixed-line-ending
-        args:
+        args: ["--fix=lf"]
-          - "--fix=lf"
       - id: trailing-whitespace
-        exclude_types:
+        exclude_types: [svg]
-          - svg
       - id: check-case-conflict
       - id: detect-private-key
   - repo: https://github.com/codespell-project/codespell
@@ -33,53 +23,29 @@ repos:
     hooks:
       - id: codespell
         additional_dependencies: [tomli]
-        exclude_types:
+        exclude_types: [pofile, json]
-          - pofile
+  - repo: https://github.com/google/yamlfmt
-          - json
+    rev: v0.21.0
-  # See https://github.com/prettier/prettier/issues/15742 for the fork reason
-  - repo: https://github.com/rbubley/mirrors-prettier
-    rev: 'v3.8.1'
     hooks:
-      - id: prettier
+      - id: yamlfmt
-        types_or:
+        exclude: "^src-ui/pnpm-lock.yaml"
-          - javascript
+        types: [yaml]
-          - ts
-          - markdown
-        additional_dependencies:
-          - prettier@3.3.3
-          - 'prettier-plugin-organize-imports@4.1.0'
-  # Python hooks
-  - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.15.0
-    hooks:
-      - id: ruff-check
-      - id: ruff-format
-  - repo: https://github.com/tox-dev/pyproject-fmt
-    rev: "v2.12.1"
-    hooks:
-      - id: pyproject-fmt
-  # Dockerfile hooks
   - repo: https://github.com/AleksaC/hadolint-py
     rev: v2.14.0
     hooks:
       - id: hadolint
-  # Shell script hooks
   - repo: https://github.com/lovesegfault/beautysh
     rev: v6.4.2
     hooks:
       - id: beautysh
         types: [file]
         files: (\.sh$|/run$|/finish$)
-        args:
+        args: ["--tab"]
-          - "--tab"
   - repo: https://github.com/shellcheck-py/shellcheck-py
     rev: "v0.11.0.1"
     hooks:
       - id: shellcheck
-  - repo: https://github.com/google/yamlfmt
+  - repo: https://github.com/tox-dev/pyproject-fmt
-    rev: v0.21.0
+    rev: "v2.16.2"
     hooks:
-      - id: yamlfmt
+      - id: pyproject-fmt
-        exclude: "^src-ui/pnpm-lock.yaml"
-        types:
-          - yaml

docker/.pre-commit-config.yaml

@@ -0,0 +1,6 @@
+repos:
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.15.4
+    hooks:
+      - id: ruff-check
+      - id: ruff-format

docs/.markdownlint.yaml

@@ -0,0 +1,6 @@
+default: true
+MD013: false # line length -- mdformat handles this
+MD033: false # inline HTML -- MkDocs uses it
+MD041: false # first line heading -- not always true in MkDocs
+MD046: # code block style
+  style: fenced

docs/.pre-commit-config.yaml

@@ -0,0 +1,13 @@
+repos:
+  - repo: https://github.com/executablebooks/mdformat
+    rev: 1.0.0
+    hooks:
+      - id: mdformat
+        additional_dependencies:
+          - mdformat-mkdocs
+          - mdformat-ruff
+  - repo: https://github.com/DavidAnson/markdownlint-cli2
+    rev: v0.21.0
+    hooks:
+      - id: markdownlint-cli2
+        args: ["--config", ".markdownlint.yaml"]

pyproject.toml

@@ -3,9 +3,10 @@ name = "paperless-ngx"
 version = "2.20.8"
 description = "A community-supported supercharged document management system: scan, index and archive all your physical documents"
 readme = "README.md"
-requires-python = ">=3.11"
+requires-python = ">=3.10"
 classifiers = [
   "Programming Language :: Python :: 3 :: Only",
+  "Programming Language :: Python :: 3.10",
   "Programming Language :: Python :: 3.11",
   "Programming Language :: Python :: 3.12",
   "Programming Language :: Python :: 3.13",
@@ -176,7 +177,7 @@ torch = [
 ]
 
 [tool.ruff]
-target-version = "py311"
+target-version = "py310"
 line-length = 88
 src = [
   "src",

src-ui/.pre-commit-config.yaml

@@ -0,0 +1,9 @@
+repos:
+  - repo: https://github.com/rbubley/mirrors-prettier
+    rev: 'v3.8.1'
+    hooks:
+      - id: prettier
+        types_or: [javascript, ts]
+        additional_dependencies:
+          - prettier@3.3.3
+          - 'prettier-plugin-organize-imports@4.1.0'

src/.pre-commit-config.yaml

@@ -0,0 +1,10 @@
+orphan: true
+
+repos:
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.15.4
+    hooks:
+      - id: ruff-check
+      - id: ruff-format
+
+

src/documents/conditionals.py

@@ -1,5 +1,5 @@
-from datetime import UTC
 from datetime import datetime
+from datetime import timezone
 from typing import Any
 
 from django.conf import settings
@@ -139,7 +139,7 @@ def thumbnail_last_modified(request: Any, pk: int) -> datetime | None:
         # No cache, get the timestamp and cache the datetime
         last_modified = datetime.fromtimestamp(
             doc.thumbnail_path.stat().st_mtime,
-            tz=UTC,
+            tz=timezone.utc,
         )
         cache.set(doc_key, last_modified, CACHE_50_MINUTES)
         return last_modified

src/documents/consumer.py

@@ -2,7 +2,7 @@ import datetime
 import hashlib
 import os
 import tempfile
-from enum import StrEnum
+from enum import Enum
 from pathlib import Path
 from typing import TYPE_CHECKING
 from typing import Final
@@ -80,7 +80,7 @@ class ConsumerError(Exception):
     pass
 
 
-class ConsumerStatusShortMessage(StrEnum):
+class ConsumerStatusShortMessage(str, Enum):
     DOCUMENT_ALREADY_EXISTS = "document_already_exists"
     DOCUMENT_ALREADY_EXISTS_IN_TRASH = "document_already_exists_in_trash"
     ASN_ALREADY_EXISTS = "asn_already_exists"

src/documents/index.py

@@ -5,10 +5,10 @@ import math
 import re
 from collections import Counter
 from contextlib import contextmanager
-from datetime import UTC
 from datetime import datetime
 from datetime import time
 from datetime import timedelta
+from datetime import timezone
 from shutil import rmtree
 from time import sleep
 from typing import TYPE_CHECKING
@@ -437,7 +437,7 @@ class ManualResults:
 class LocalDateParser(English):
     def reverse_timezone_offset(self, d):
         return (d.replace(tzinfo=django_timezone.get_current_timezone())).astimezone(
-            UTC,
+            timezone.utc,
         )
 
     def date_from(self, *args, **kwargs):
@@ -641,8 +641,8 @@ def rewrite_natural_date_keywords(query_string: str) -> str:
                 end = datetime(local_now.year - 1, 12, 31, 23, 59, 59, tzinfo=tz)
 
         # Convert to UTC and format
-        start_str = start.astimezone(UTC).strftime("%Y%m%d%H%M%S")
+        start_str = start.astimezone(timezone.utc).strftime("%Y%m%d%H%M%S")
-        end_str = end.astimezone(UTC).strftime("%Y%m%d%H%M%S")
+        end_str = end.astimezone(timezone.utc).strftime("%Y%m%d%H%M%S")
         return f"{field}:[{start_str} TO {end_str}]"
 
     return re.sub(pattern, repl, query_string, flags=re.IGNORECASE)

src/documents/plugins/date_parsing/base.py

@@ -9,7 +9,7 @@ from types import TracebackType
 try:
     from typing import Self
 except ImportError:
-    from typing import Self
+    from typing_extensions import Self
 
 import dateparser
 

src/documents/plugins/helpers.py

@@ -8,7 +8,7 @@ if TYPE_CHECKING:
     from channels_redis.pubsub import RedisPubSubChannelLayer
 
 
-class ProgressStatusOptions(enum.StrEnum):
+class ProgressStatusOptions(str, enum.Enum):
     STARTED = "STARTED"
     WORKING = "WORKING"
     SUCCESS = "SUCCESS"

src/documents/tests/date_parsing/conftest.py

@@ -24,7 +24,7 @@ def base_config() -> DateParserConfig:
             12,
             0,
             0,
-            tzinfo=datetime.UTC,
+            tzinfo=datetime.timezone.utc,
         ),
         filename_date_order="YMD",
         content_date_order="DMY",
@@ -45,7 +45,7 @@ def config_with_ignore_dates() -> DateParserConfig:
             12,
             0,
             0,
-            tzinfo=datetime.UTC,
+            tzinfo=datetime.timezone.utc,
         ),
         filename_date_order="DMY",
         content_date_order="MDY",

src/documents/tests/date_parsing/test_date_parsing.py

@@ -101,50 +101,50 @@ class TestFilterDate:
         [
             # Valid Dates
             pytest.param(
-                datetime.datetime(2024, 1, 10, tzinfo=datetime.UTC),
+                datetime.datetime(2024, 1, 10, tzinfo=datetime.timezone.utc),
-                datetime.datetime(2024, 1, 10, tzinfo=datetime.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.UTC),
+                datetime.datetime(2024, 1, 15, 12, 0, 0, tzinfo=datetime.timezone.utc),
-                datetime.datetime(2024, 1, 15, 12, 0, 0, tzinfo=datetime.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.UTC),
+                datetime.datetime(1901, 1, 1, tzinfo=datetime.timezone.utc),
-                datetime.datetime(1901, 1, 1, tzinfo=datetime.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.UTC),
+                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.UTC),
+                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.UTC),
+                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.UTC),
+                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.UTC),
+                datetime.datetime(1899, 12, 31, tzinfo=datetime.timezone.utc),
                 None,
                 id="year_1899",
             ),
             pytest.param(
-                datetime.datetime(1900, 1, 1, tzinfo=datetime.UTC),
+                datetime.datetime(1900, 1, 1, tzinfo=datetime.timezone.utc),
                 None,
                 id="year_1900_boundary",
             ),
@@ -176,7 +176,7 @@ class TestFilterDate:
             1,
             12,
             0,
-            tzinfo=datetime.UTC,
+            tzinfo=datetime.timezone.utc,
         )
         another_ignored = datetime.datetime(
             2024,
@@ -184,7 +184,7 @@ class TestFilterDate:
             25,
             15,
             30,
-            tzinfo=datetime.UTC,
+            tzinfo=datetime.timezone.utc,
         )
         allowed_date = datetime.datetime(
             2024,
@@ -192,7 +192,7 @@ class TestFilterDate:
             2,
             12,
             0,
-            tzinfo=datetime.UTC,
+            tzinfo=datetime.timezone.utc,
         )
 
         assert parser._filter_date(ignored_date) is None
@@ -204,7 +204,7 @@ class TestFilterDate:
         regex_parser: RegexDateParserPlugin,
     ) -> None:
         """Should work with timezone-aware datetimes."""
-        date_utc = datetime.datetime(2024, 1, 10, 12, 0, tzinfo=datetime.UTC)
+        date_utc = datetime.datetime(2024, 1, 10, 12, 0, tzinfo=datetime.timezone.utc)
 
         result = regex_parser._filter_date(date_utc)
 
@@ -221,8 +221,8 @@ class TestRegexDateParser:
                 "report-2023-12-25.txt",
                 "Event recorded on 25/12/2022.",
                 [
-                    datetime.datetime(2023, 12, 25, tzinfo=datetime.UTC),
+                    datetime.datetime(2023, 12, 25, tzinfo=datetime.timezone.utc),
-                    datetime.datetime(2022, 12, 25, tzinfo=datetime.UTC),
+                    datetime.datetime(2022, 12, 25, tzinfo=datetime.timezone.utc),
                 ],
                 id="filename-y-m-d_and_content-d-m-y",
             ),
@@ -230,8 +230,8 @@ class TestRegexDateParser:
                 "img_2023.01.02.jpg",
                 "Taken on 01/02/2023",
                 [
-                    datetime.datetime(2023, 1, 2, tzinfo=datetime.UTC),
+                    datetime.datetime(2023, 1, 2, tzinfo=datetime.timezone.utc),
-                    datetime.datetime(2023, 2, 1, tzinfo=datetime.UTC),
+                    datetime.datetime(2023, 2, 1, tzinfo=datetime.timezone.utc),
                 ],
                 id="ambiguous-dates-respect-orders",
             ),
@@ -239,7 +239,7 @@ class TestRegexDateParser:
                 "notes.txt",
                 "bad date 99/99/9999 and 25/12/2022",
                 [
-                    datetime.datetime(2022, 12, 25, tzinfo=datetime.UTC),
+                    datetime.datetime(2022, 12, 25, tzinfo=datetime.timezone.utc),
                 ],
                 id="parse-exception-skips-bad-and-yields-good",
             ),
@@ -275,24 +275,24 @@ class TestRegexDateParser:
                 or "2023.12.25" in date_string
                 or "2023-12-25" in date_string
             ):
-                return datetime.datetime(2023, 12, 25, tzinfo=datetime.UTC)
+                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.UTC)
+                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.UTC)
+                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.UTC)
+                    return datetime.datetime(2023, 2, 1, tzinfo=datetime.timezone.utc)
                 if date_order == "YMD":
-                    return datetime.datetime(2023, 1, 2, tzinfo=datetime.UTC)
+                    return datetime.datetime(2023, 1, 2, tzinfo=datetime.timezone.utc)
                 # fallback
-                return datetime.datetime(2023, 2, 1, tzinfo=datetime.UTC)
+                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:
@@ -328,7 +328,7 @@ class TestRegexDateParser:
                 12,
                 0,
                 0,
-                tzinfo=datetime.UTC,
+                tzinfo=datetime.timezone.utc,
             ),
             filename_date_order="YMD",
             content_date_order="DMY",
@@ -344,13 +344,13 @@ class TestRegexDateParser:
         ) -> 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.UTC)
+                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.UTC)
+                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.UTC)
+                return datetime.datetime(2023, 1, 5, tzinfo=datetime.timezone.utc)
             return None
 
         mocker.patch(target, side_effect=fake_parse)
@@ -358,7 +358,7 @@ class TestRegexDateParser:
         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.UTC)]
+        assert results == [datetime.datetime(2023, 1, 5, tzinfo=datetime.timezone.utc)]
 
     def test_parse_handles_no_matches_and_returns_empty_list(
         self,
@@ -392,7 +392,7 @@ class TestRegexDateParser:
                 12,
                 0,
                 0,
-                tzinfo=datetime.UTC,
+                tzinfo=datetime.timezone.utc,
             ),
             filename_date_order=None,
             content_date_order="DMY",
@@ -409,9 +409,9 @@ class TestRegexDateParser:
         ) -> 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.UTC)
+                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.UTC)
+                return datetime.datetime(2023, 12, 25, tzinfo=datetime.timezone.utc)
             return None
 
         mock = mocker.patch(target, side_effect=fake_parse)
@@ -429,5 +429,5 @@ class TestRegexDateParser:
         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.UTC),
+            datetime.datetime(2022, 12, 25, tzinfo=datetime.timezone.utc),
         ]

src/documents/tests/test_file_handling.py

@@ -336,11 +336,7 @@ class TestFileHandling(DirectoriesMixin, FileSystemAssertsMixin, TestCase):
             added=d1,
         )
 
-        # Account for 3.14 padding changes
+        self.assertEqual(generate_filename(doc1), Path("232-01-09.pdf"))
-        expected_year: str = d1.strftime("%Y")
-        expected_filename: Path = Path(f"{expected_year}-01-09.pdf")
-
-        self.assertEqual(generate_filename(doc1), expected_filename)
 
         doc1.added = timezone.make_aware(datetime.datetime(2020, 11, 16, 1, 1, 1))
 

src/documents/tests/test_filters.py

@@ -21,7 +21,7 @@ class TestDateLocalization:
         14,
         30,
         5,
-        tzinfo=datetime.UTC,
+        tzinfo=datetime.timezone.utc,
     )
 
     TEST_DATETIME_STRING: str = "2023-10-26T14:30:05+00:00"

src/documents/tests/test_workflows.py

@@ -4570,7 +4570,7 @@ class TestDateWorkflowLocalization(
         14,
         30,
         5,
-        tzinfo=datetime.UTC,
+        tzinfo=datetime.timezone.utc,
     )
 
     @pytest.mark.parametrize(

src/documents/versioning.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 from dataclasses import dataclass
-from enum import StrEnum
+from enum import Enum
 from typing import TYPE_CHECKING
 from typing import Any
 
@@ -11,7 +11,7 @@ if TYPE_CHECKING:
     from django.http import HttpRequest
 
 
-class VersionResolutionError(StrEnum):
+class VersionResolutionError(str, Enum):
     INVALID = "invalid"
     NOT_FOUND = "not_found"