feat(profiling): add workflow trigger matching profiling

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

.github/workflows/ci-backend.yml

@@ -165,7 +165,6 @@ jobs:
       contents: read
     env:
       DEFAULT_PYTHON: "3.12"
-      PAPERLESS_SECRET_KEY: "ci-typing-not-a-real-secret"
     steps:
       - name: Checkout
         uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

.github/workflows/ci-release.yml

@@ -88,7 +88,6 @@ jobs:
           uv export --quiet --no-dev --all-extras --format requirements-txt --output-file requirements.txt
       - name: Compile messages
         env:
-          PAPERLESS_SECRET_KEY: "ci-release-not-a-real-secret"
           PYTHON_VERSION: ${{ steps.setup-python.outputs.python-version }}
         run: |
           cd src/
@@ -97,7 +96,6 @@ jobs:
             manage.py compilemessages
       - name: Collect static files
         env:
-          PAPERLESS_SECRET_KEY: "ci-release-not-a-real-secret"
           PYTHON_VERSION: ${{ steps.setup-python.outputs.python-version }}
         run: |
           cd src/

.github/workflows/translate-strings.yml

@@ -36,8 +36,6 @@ jobs:
             --group dev \
             --frozen
       - name: Generate backend translation strings
-        env:
-          PAPERLESS_SECRET_KEY: "ci-translate-not-a-real-secret"
         run: cd src/ && uv run manage.py makemessages -l en_US -i "samples*"
       - name: Install pnpm
         uses: pnpm/action-setup@fc06bc1257f339d1d5d8b3a19a8cae5388b55320 # v5.0.0

.gitignore

@@ -79,7 +79,6 @@ virtualenv
 /docker-compose.env
 /docker-compose.yml
 .ruff_cache/
-.mypy_cache/
 
 # Used for development
 scripts/import-for-development
@@ -112,6 +111,4 @@ celerybeat-schedule*
 
 # ignore pnpm package store folder created when setting up the devcontainer
 .pnpm-store/
-
-# Git worktree local folder
 .worktrees

docs/migration-v3.md

@@ -241,66 +241,3 @@ For example:
   }
 }
 ```
-
-## Consume Script Positional Arguments Removed
-
-Pre- and post-consumption scripts no longer receive positional arguments. All information is
-now passed exclusively via environment variables, which have been available since earlier versions.
-
-### Pre-consumption script
-
-Previously, the original file path was passed as `$1`. It is now only available as
-`DOCUMENT_SOURCE_PATH`.
-
-**Before:**
-
-```bash
-#!/usr/bin/env bash
-# $1 was the original file path
-process_document "$1"
-```
-
-**After:**
-
-```bash
-#!/usr/bin/env bash
-process_document "${DOCUMENT_SOURCE_PATH}"
-```
-
-### Post-consumption script
-
-Previously, document metadata was passed as positional arguments `$1` through `$8`:
-
-| Argument | Environment Variable Equivalent |
-| -------- | ------------------------------- |
-| `$1`     | `DOCUMENT_ID`                   |
-| `$2`     | `DOCUMENT_FILE_NAME`            |
-| `$3`     | `DOCUMENT_SOURCE_PATH`          |
-| `$4`     | `DOCUMENT_THUMBNAIL_PATH`       |
-| `$5`     | `DOCUMENT_DOWNLOAD_URL`         |
-| `$6`     | `DOCUMENT_THUMBNAIL_URL`        |
-| `$7`     | `DOCUMENT_CORRESPONDENT`        |
-| `$8`     | `DOCUMENT_TAGS`                 |
-
-**Before:**
-
-```bash
-#!/usr/bin/env bash
-DOCUMENT_ID=$1
-CORRESPONDENT=$7
-TAGS=$8
-```
-
-**After:**
-
-```bash
-#!/usr/bin/env bash
-# Use environment variables directly
-echo "Document ${DOCUMENT_ID} from ${DOCUMENT_CORRESPONDENT} tagged: ${DOCUMENT_TAGS}"
-```
-
-### Action Required
-
-Update any pre- or post-consumption scripts that read `$1`, `$2`, etc. to use the
-corresponding environment variables instead. Environment variables have been the preferred
-option since v1.8.0.

profiling.py

@@ -0,0 +1,150 @@
+"""
+Temporary profiling utilities for comparing implementations.
+
+Usage in a management command or shell::
+
+    from profiling import profile_block, profile_cpu, measure_memory
+
+    with profile_block("new check_sanity"):
+        messages = check_sanity()
+
+    with profile_block("old check_sanity"):
+        messages = check_sanity_old()
+
+Drop this file when done.
+"""
+
+from __future__ import annotations
+
+import tracemalloc
+from collections.abc import Callable  # noqa: TC003
+from collections.abc import Generator  # noqa: TC003
+from contextlib import contextmanager
+from time import perf_counter
+from typing import Any
+
+from django.db import connection
+from django.db import reset_queries
+from django.test.utils import override_settings
+
+
+@contextmanager
+def profile_block(label: str = "block") -> Generator[None, None, None]:
+    """Profile memory, wall time, and DB queries for a code block.
+
+    Prints a summary to stdout on exit. Requires no external packages.
+    Enables DEBUG temporarily to capture Django's query log.
+    """
+    tracemalloc.start()
+    snapshot_before = tracemalloc.take_snapshot()
+
+    with override_settings(DEBUG=True):
+        reset_queries()
+        start = perf_counter()
+
+        yield
+
+        elapsed = perf_counter() - start
+        queries = list(connection.queries)
+
+    snapshot_after = tracemalloc.take_snapshot()
+    _, peak = tracemalloc.get_traced_memory()
+    tracemalloc.stop()
+
+    # Compare snapshots for top allocations
+    stats = snapshot_after.compare_to(snapshot_before, "lineno")
+
+    query_time = sum(float(q["time"]) for q in queries)
+    mem_diff = sum(s.size_diff for s in stats)
+
+    print(f"\n{'=' * 60}")  # noqa: T201
+    print(f"  Profile: {label}")  # noqa: T201
+    print(f"{'=' * 60}")  # noqa: T201
+    print(f"  Wall time:    {elapsed:.4f}s")  # noqa: T201
+    print(f"  Queries:      {len(queries)} ({query_time:.4f}s)")  # noqa: T201
+    print(f"  Memory delta: {mem_diff / 1024:.1f} KiB")  # noqa: T201
+    print(f"  Peak memory:  {peak / 1024:.1f} KiB")  # noqa: T201
+    print("\n  Top 5 allocations:")  # noqa: T201
+    for stat in stats[:5]:
+        print(f"    {stat}")  # noqa: T201
+    print(f"{'=' * 60}\n")  # noqa: T201
+
+
+def profile_cpu(
+    fn: Callable[[], Any],
+    *,
+    label: str,
+    top: int = 30,
+    sort: str = "cumtime",
+) -> tuple[Any, float]:
+    """Run *fn()* under cProfile, print stats, return (result, elapsed_s).
+
+    Args:
+        fn: Zero-argument callable to profile.
+        label: Human-readable label printed in the header.
+        top: Number of cProfile rows to print.
+        sort: cProfile sort key (default: cumulative time).
+
+    Returns:
+        ``(result, elapsed_s)`` where *result* is the return value of *fn()*.
+    """
+    import cProfile
+    import io
+    import pstats
+
+    pr = cProfile.Profile()
+    t0 = perf_counter()
+    pr.enable()
+    result = fn()
+    pr.disable()
+    elapsed = perf_counter() - t0
+
+    buf = io.StringIO()
+    ps = pstats.Stats(pr, stream=buf).sort_stats(sort)
+    ps.print_stats(top)
+
+    print(f"\n{'=' * 72}")  # noqa: T201
+    print(f"  {label}")  # noqa: T201
+    print(f"  wall time: {elapsed * 1000:.1f} ms")  # noqa: T201
+    print(f"{'=' * 72}")  # noqa: T201
+    print(buf.getvalue())  # noqa: T201
+
+    return result, elapsed
+
+
+def measure_memory(fn: Callable[[], Any], *, label: str) -> tuple[Any, float, float]:
+    """Run *fn()* under tracemalloc, print allocation report.
+
+    Args:
+        fn: Zero-argument callable to profile.
+        label: Human-readable label printed in the header.
+
+    Returns:
+        ``(result, peak_kib, delta_kib)``.
+    """
+    tracemalloc.start()
+    snapshot_before = tracemalloc.take_snapshot()
+    t0 = perf_counter()
+    result = fn()
+    elapsed = perf_counter() - t0
+    snapshot_after = tracemalloc.take_snapshot()
+    _, peak = tracemalloc.get_traced_memory()
+    tracemalloc.stop()
+
+    stats = snapshot_after.compare_to(snapshot_before, "lineno")
+    delta_kib = sum(s.size_diff for s in stats) / 1024
+
+    print(f"\n{'=' * 72}")  # noqa: T201
+    print(f"  [memory] {label}")  # noqa: T201
+    print(f"  wall time:    {elapsed * 1000:.1f} ms")  # noqa: T201
+    print(f"  memory delta: {delta_kib:+.1f} KiB")  # noqa: T201
+    print(f"  peak traced:  {peak / 1024:.1f} KiB")  # noqa: T201
+    print(f"{'=' * 72}")  # noqa: T201
+    print("  Top allocation sites (by size_diff):")  # noqa: T201
+    for stat in stats[:20]:
+        if stat.size_diff != 0:
+            print(  # noqa: T201
+                f"    {stat.size_diff / 1024:+8.1f} KiB  {stat.traceback.format()[0]}",
+            )
+
+    return result, peak / 1024, delta_kib

pyproject.toml

@@ -24,10 +24,11 @@ dependencies = [
   "dateparser~=1.2",
   # WARNING: django does not use semver.
   #          Only patch versions are guaranteed to not introduce breaking changes.
-  "django~=5.2.13",
+  "django~=5.2.10",
   "django-allauth[mfa,socialaccount]~=65.15.0",
   "django-auditlog~=3.4.1",
   "django-cachalot~=2.9.0",
+  "django-celery-results~=2.6.0",
   "django-compression-middleware~=0.5.0",
   "django-cors-headers~=4.9.0",
   "django-extensions~=4.1",
@@ -112,7 +113,7 @@ testing = [
   "factory-boy~=3.3.1",
   "faker~=40.12.0",
   "imagehash",
-  "pytest~=9.0.3",
+  "pytest~=9.0.0",
   "pytest-cov~=7.1.0",
   "pytest-django~=4.12.0",
   "pytest-env~=1.6.0",
@@ -311,7 +312,7 @@ markers = [
   "date_parsing: Tests which cover date parsing from content or filename",
   "management: Tests which cover management commands/functionality",
   "search: Tests for the Tantivy search backend",
-  "api: Tests for REST API endpoints",
+  "profiling: Performance profiling tests — print measurements, no assertions",
 ]
 
 [tool.pytest_env]

src-ui/pnpm-lock.yaml

@@ -4363,8 +4363,8 @@ packages:
   flatted@3.4.2:
     resolution: {integrity: sha512-PjDse7RzhcPkIJwy5t7KPWQSZ9cAbzQXcafsetQoD7sOJRQlGikNbx7yZp2OotDnJyrDcbyRq3Ttb18iYOqkxA==}
 
-  follow-redirects@1.16.0:
-    resolution: {integrity: sha512-y5rN/uOsadFT/JfYwhxRS5R7Qce+g3zG97+JrtFZlC9klX/W5hD7iiLzScI4nZqUS7DNUdhPgw4xI8W2LuXlUw==}
+  follow-redirects@1.15.11:
+    resolution: {integrity: sha512-deG2P0JfjrTxl50XGCDyfI97ZGVCxIpfKYmfyrQ54n5FO/0gfIES8C/Psl6kWVDolizcaaxZJnTS0QSMxvnsBQ==}
     engines: {node: '>=4.0'}
     peerDependencies:
       debug: '*'
@@ -11427,7 +11427,7 @@ snapshots:
 
   flatted@3.4.2: {}
 
-  follow-redirects@1.16.0(debug@4.4.3):
+  follow-redirects@1.15.11(debug@4.4.3):
     optionalDependencies:
       debug: 4.4.3
 
@@ -11634,7 +11634,7 @@ snapshots:
   http-proxy@1.18.1(debug@4.4.3):
     dependencies:
       eventemitter3: 4.0.7
-      follow-redirects: 1.16.0(debug@4.4.3)
+      follow-redirects: 1.15.11(debug@4.4.3)
       requires-port: 1.0.0
     transitivePeerDependencies:
       - debug

src-ui/src/app/components/admin/tasks/tasks.component.html

@@ -76,27 +76,33 @@
               <label class="form-check-label" for="task{{task.id}}"></label>
             </div>
           </td>
-          <td class="overflow-auto name-col">{{ task.input_data?.filename }}</td>
+          <td class="overflow-auto name-col">{{ task.task_file_name }}</td>
           <td class="d-none d-lg-table-cell">{{ task.date_created | customDate:'short' }}</td>
           @if (activeTab !== 'started' && activeTab !== 'queued') {
             <td class="d-none d-lg-table-cell">
-              @if (task.result_message?.length > 50) {
+              @if (task.result?.length > 50) {
                 <div class="result" (click)="expandTask(task); $event.stopPropagation();"
                   [ngbPopover]="resultPopover" popoverClass="shadow small mobile" triggers="mouseenter:mouseleave" container="body">
-                  <span class="small d-none d-md-inline-block font-monospace text-muted">{{ task.result_message | slice:0:50 }}&hellip;</span>
+                  <span class="small d-none d-md-inline-block font-monospace text-muted">{{ task.result | slice:0:50 }}&hellip;</span>
                 </div>
               }
-              @if (task.result_message?.length <= 50) {
-                <span class="small d-none d-md-inline-block font-monospace text-muted">{{ task.result_message }}</span>
+              @if (task.result?.length <= 50) {
+                <span class="small d-none d-md-inline-block font-monospace text-muted">{{ task.result }}</span>
               }
               <ng-template #resultPopover>
-                <pre class="small mb-0">{{ task.result_message | slice:0:300 }}@if (task.result_message.length > 300) {
+                <pre class="small mb-0">{{ task.result | slice:0:300 }}@if (task.result.length > 300) {
                   &hellip;
                 }</pre>
-                @if (task.result_message?.length > 300) {
+                @if (task.result?.length > 300) {
                   <br/><em>(<ng-container i18n>click for full output</ng-container>)</em>
                 }
               </ng-template>
+              @if (task.duplicate_documents?.length > 0) {
+                <div class="small text-warning-emphasis d-flex align-items-center gap-1">
+                  <i-bs class="lh-1" width="1em" height="1em" name="exclamation-triangle"></i-bs>
+                  <span i18n>Duplicate(s) detected</span>
+                </div>
+              }
             </td>
           }
           <td class="d-lg-none">
@@ -110,7 +116,7 @@
                 <i-bs name="check" class="me-1"></i-bs><ng-container i18n>Dismiss</ng-container>
               </button>
               <ng-container *pngxIfPermissions="{ action: PermissionAction.View, type: PermissionType.Document }">
-                @if (task.related_document_ids?.[0]) {
+                @if (task.related_document) {
                   <button class="btn btn-sm btn-outline-primary" (click)="dismissAndGo(task); $event.stopPropagation();">
                     <i-bs name="file-text" class="me-1"></i-bs><ng-container i18n>Open Document</ng-container>
                   </button>
@@ -121,7 +127,7 @@
         </tr>
         <tr>
           <td class="p-0" [class.border-0]="expandedTask !== task.id" colspan="5">
-            <pre #collapse="ngbCollapse" [ngbCollapse]="expandedTask !== task.id" class="small mb-0"><div class="small p-1 p-lg-3 ms-lg-3">{{ task.result_message }}</div></pre>
+            <pre #collapse="ngbCollapse" [ngbCollapse]="expandedTask !== task.id" class="small mb-0"><div class="small p-1 p-lg-3 ms-lg-3">{{ task.result }}</div></pre>
           </td>
         </tr>
       }

src-ui/src/app/components/admin/tasks/tasks.component.spec.ts

@@ -20,8 +20,8 @@ import { throwError } from 'rxjs'
 import { routes } from 'src/app/app-routing.module'
 import {
   PaperlessTask,
+  PaperlessTaskName,
   PaperlessTaskStatus,
-  PaperlessTaskTriggerSource,
   PaperlessTaskType,
 } from 'src/app/data/paperless-task'
 import { IfPermissionsDirective } from 'src/app/directives/if-permissions.directive'
@@ -39,100 +39,81 @@ const tasks: PaperlessTask[] = [
   {
     id: 467,
     task_id: '11ca1a5b-9f81-442c-b2c8-7e4ae53657f1',
-    input_data: { filename: 'test.pdf' },
+    task_file_name: 'test.pdf',
     date_created: new Date('2023-03-01T10:26:03.093116Z'),
     date_done: new Date('2023-03-01T10:26:07.223048Z'),
-    task_type: PaperlessTaskType.ConsumeFile,
-    task_type_display: 'Consume File',
-    trigger_source: PaperlessTaskTriggerSource.FolderConsume,
-    trigger_source_display: 'Folder Consume',
-    status: PaperlessTaskStatus.Failure,
-    status_display: 'Failure',
-    result_message:
-      'test.pd: Not consuming test.pdf: It is a duplicate of test (#100)',
+    type: PaperlessTaskType.Auto,
+    task_name: PaperlessTaskName.ConsumeFile,
+    status: PaperlessTaskStatus.Failed,
+    result: 'test.pd: Not consuming test.pdf: It is a duplicate of test (#100)',
     acknowledged: false,
-    related_document_ids: [],
+    related_document: null,
   },
   {
     id: 466,
     task_id: '10ca1a5b-3c08-442c-b2c8-7e4ae53657f1',
-    input_data: { filename: '191092.pdf' },
+    task_file_name: '191092.pdf',
     date_created: new Date('2023-03-01T09:26:03.093116Z'),
     date_done: new Date('2023-03-01T09:26:07.223048Z'),
-    task_type: PaperlessTaskType.ConsumeFile,
-    task_type_display: 'Consume File',
-    trigger_source: PaperlessTaskTriggerSource.FolderConsume,
-    trigger_source_display: 'Folder Consume',
-    status: PaperlessTaskStatus.Failure,
-    status_display: 'Failure',
-    result_message:
+    type: PaperlessTaskType.Auto,
+    task_name: PaperlessTaskName.ConsumeFile,
+    status: PaperlessTaskStatus.Failed,
+    result:
       '191092.pd: Not consuming 191092.pdf: It is a duplicate of 191092 (#311)',
     acknowledged: false,
-    related_document_ids: [],
+    related_document: null,
   },
   {
     id: 465,
     task_id: '3612d477-bb04-44e3-985b-ac580dd496d8',
-    input_data: { filename: 'Scan Jun 6, 2023 at 3.19 PM.pdf' },
+    task_file_name: 'Scan Jun 6, 2023 at 3.19 PM.pdf',
     date_created: new Date('2023-06-06T15:22:05.722323-07:00'),
     date_done: new Date('2023-06-06T15:22:14.564305-07:00'),
-    task_type: PaperlessTaskType.ConsumeFile,
-    task_type_display: 'Consume File',
-    trigger_source: PaperlessTaskTriggerSource.FolderConsume,
-    trigger_source_display: 'Folder Consume',
+    type: PaperlessTaskType.Auto,
+    task_name: PaperlessTaskName.ConsumeFile,
     status: PaperlessTaskStatus.Pending,
-    status_display: 'Pending',
-    result_message: null,
+    result: null,
     acknowledged: false,
-    related_document_ids: [],
+    related_document: null,
   },
   {
     id: 464,
     task_id: '2eac4716-2aa6-4dcd-9953-264e11656d7e',
-    input_data: { filename: 'paperless-mail-l4dkg8ir' },
+    task_file_name: 'paperless-mail-l4dkg8ir',
     date_created: new Date('2023-06-04T11:24:32.898089-07:00'),
     date_done: new Date('2023-06-04T11:24:44.678605-07:00'),
-    task_type: PaperlessTaskType.ConsumeFile,
-    task_type_display: 'Consume File',
-    trigger_source: PaperlessTaskTriggerSource.EmailConsume,
-    trigger_source_display: 'Email Consume',
-    status: PaperlessTaskStatus.Success,
-    status_display: 'Success',
-    result_message: 'Success. New document id 422 created',
+    type: PaperlessTaskType.Auto,
+    task_name: PaperlessTaskName.ConsumeFile,
+    status: PaperlessTaskStatus.Complete,
+    result: 'Success. New document id 422 created',
     acknowledged: false,
-    related_document_ids: [422],
+    related_document: 422,
   },
   {
     id: 463,
     task_id: '28125528-1575-4d6b-99e6-168906e8fa5c',
-    input_data: { filename: 'onlinePaymentSummary.pdf' },
+    task_file_name: 'onlinePaymentSummary.pdf',
     date_created: new Date('2023-06-01T13:49:51.631305-07:00'),
     date_done: new Date('2023-06-01T13:49:54.190220-07:00'),
-    task_type: PaperlessTaskType.ConsumeFile,
-    task_type_display: 'Consume File',
-    trigger_source: PaperlessTaskTriggerSource.FolderConsume,
-    trigger_source_display: 'Folder Consume',
-    status: PaperlessTaskStatus.Success,
-    status_display: 'Success',
-    result_message: 'Success. New document id 421 created',
+    type: PaperlessTaskType.Auto,
+    task_name: PaperlessTaskName.ConsumeFile,
+    status: PaperlessTaskStatus.Complete,
+    result: 'Success. New document id 421 created',
     acknowledged: false,
-    related_document_ids: [421],
+    related_document: 421,
   },
   {
     id: 462,
     task_id: 'a5b9ca47-0c8e-490f-a04c-6db5d5fc09e5',
-    input_data: { filename: 'paperless-mail-_rrpmqk6' },
+    task_file_name: 'paperless-mail-_rrpmqk6',
     date_created: new Date('2023-06-07T02:54:35.694916Z'),
     date_done: null,
-    task_type: PaperlessTaskType.ConsumeFile,
-    task_type_display: 'Consume File',
-    trigger_source: PaperlessTaskTriggerSource.EmailConsume,
-    trigger_source_display: 'Email Consume',
+    type: PaperlessTaskType.Auto,
+    task_name: PaperlessTaskName.ConsumeFile,
     status: PaperlessTaskStatus.Started,
-    status_display: 'Started',
-    result_message: null,
+    result: null,
     acknowledged: false,
-    related_document_ids: [],
+    related_document: null,
   },
 ]
 
@@ -186,7 +167,7 @@ describe('TasksComponent', () => {
     fixture.detectChanges()
     httpTestingController
       .expectOne(
-        `${environment.apiBaseUrl}tasks/?task_type=consume_file&acknowledged=false`
+        `${environment.apiBaseUrl}tasks/?task_name=consume_file&acknowledged=false`
       )
       .flush(tasks)
   })
@@ -195,7 +176,7 @@ describe('TasksComponent', () => {
     const tabButtons = fixture.debugElement.queryAll(By.directive(NgbNavItem))
 
     let currentTasksLength = tasks.filter(
-      (t) => t.status === PaperlessTaskStatus.Failure
+      (t) => t.status === PaperlessTaskStatus.Failed
     ).length
     component.activeTab = TaskTab.Failed
     fixture.detectChanges()
@@ -207,7 +188,7 @@ describe('TasksComponent', () => {
     ).toHaveLength(currentTasksLength + 1)
 
     currentTasksLength = tasks.filter(
-      (t) => t.status === PaperlessTaskStatus.Success
+      (t) => t.status === PaperlessTaskStatus.Complete
     ).length
     component.activeTab = TaskTab.Completed
     fixture.detectChanges()
@@ -327,7 +308,7 @@ describe('TasksComponent', () => {
     expect(component.selectedTasks).toEqual(
       new Set(
         tasks
-          .filter((t) => t.status === PaperlessTaskStatus.Failure)
+          .filter((t) => t.status === PaperlessTaskStatus.Failed)
           .map((t) => t.id)
       )
     )
@@ -341,7 +322,7 @@ describe('TasksComponent', () => {
     component.dismissAndGo(tasks[3])
     expect(routerSpy).toHaveBeenCalledWith([
       'documents',
-      tasks[3].related_document_ids?.[0],
+      tasks[3].related_document,
     ])
   })
 

src-ui/src/app/components/admin/tasks/tasks.component.ts

@@ -175,7 +175,7 @@ export class TasksComponent
 
   dismissAndGo(task: PaperlessTask) {
     this.dismissTask(task)
-    this.router.navigate(['documents', task.related_document_ids?.[0]])
+    this.router.navigate(['documents', task.related_document])
   }
 
   expandTask(task: PaperlessTask) {
@@ -207,13 +207,11 @@ export class TasksComponent
     if (this._filterText.length) {
       tasks = tasks.filter((t) => {
         if (this.filterTargetID == TaskFilterTargetID.Name) {
-          return (t.input_data?.filename as string)
-            ?.toLowerCase()
+          return t.task_file_name
+            .toLowerCase()
             .includes(this._filterText.toLowerCase())
         } else if (this.filterTargetID == TaskFilterTargetID.Result) {
-          return t.result_message
-            ?.toLowerCase()
-            .includes(this._filterText.toLowerCase())
+          return t.result.toLowerCase().includes(this._filterText.toLowerCase())
         }
       })
     }

src-ui/src/app/components/common/system-status-dialog/system-status-dialog.component.html

@@ -169,10 +169,10 @@
                   }
                 </button>
                 @if (currentUserIsSuperUser) {
-                  @if (isRunning(PaperlessTaskType.IndexOptimize)) {
+                  @if (isRunning(PaperlessTaskName.IndexOptimize)) {
                     <div class="spinner-border spinner-border-sm ms-2" role="status"></div>
                   } @else {
-                    <button class="btn btn-sm d-flex align-items-center btn-dark small ms-2" (click)="runTask(PaperlessTaskType.IndexOptimize)">
+                    <button class="btn btn-sm d-flex align-items-center btn-dark small ms-2" (click)="runTask(PaperlessTaskName.IndexOptimize)">
                       <i-bs name="play-fill" class="me-1"></i-bs>
                       <ng-container i18n>Run Task</ng-container>
                     </button>
@@ -203,10 +203,10 @@
                   }
                 </button>
                 @if (currentUserIsSuperUser) {
-                  @if (isRunning(PaperlessTaskType.TrainClassifier)) {
+                  @if (isRunning(PaperlessTaskName.TrainClassifier)) {
                     <div class="spinner-border spinner-border-sm ms-2" role="status"></div>
                   } @else {
-                    <button class="btn btn-sm d-flex align-items-center btn-dark small ms-2" (click)="runTask(PaperlessTaskType.TrainClassifier)">
+                    <button class="btn btn-sm d-flex align-items-center btn-dark small ms-2" (click)="runTask(PaperlessTaskName.TrainClassifier)">
                       <i-bs name="play-fill" class="me-1"></i-bs>
                       <ng-container i18n>Run Task</ng-container>
                     </button>
@@ -237,10 +237,10 @@
                   }
                 </button>
                 @if (currentUserIsSuperUser) {
-                  @if (isRunning(PaperlessTaskType.SanityCheck)) {
+                  @if (isRunning(PaperlessTaskName.SanityCheck)) {
                     <div class="spinner-border spinner-border-sm ms-2" role="status"></div>
                   } @else {
-                    <button class="btn btn-sm d-flex align-items-center btn-dark small ms-2" (click)="runTask(PaperlessTaskType.SanityCheck)">
+                    <button class="btn btn-sm d-flex align-items-center btn-dark small ms-2" (click)="runTask(PaperlessTaskName.SanityCheck)">
                       <i-bs name="play-fill" class="me-1"></i-bs>
                       <ng-container i18n>Run Task</ng-container>
                     </button>
@@ -285,10 +285,10 @@
                     }
                   </button>
                   @if (currentUserIsSuperUser) {
-                    @if (isRunning(PaperlessTaskType.LlmIndex)) {
+                    @if (isRunning(PaperlessTaskName.LLMIndexUpdate)) {
                       <div class="spinner-border spinner-border-sm ms-2" role="status"></div>
                     } @else {
-                      <button class="btn btn-sm d-flex align-items-center btn-dark small ms-2" (click)="runTask(PaperlessTaskType.LlmIndex)">
+                      <button class="btn btn-sm d-flex align-items-center btn-dark small ms-2" (click)="runTask(PaperlessTaskName.LLMIndexUpdate)">
                         <i-bs name="play-fill" class="me-1"></i-bs>
                         <ng-container i18n>Run Task</ng-container>
                       </button>

src-ui/src/app/components/common/system-status-dialog/system-status-dialog.component.spec.ts

@@ -25,7 +25,7 @@ import {
 import { NgbActiveModal } from '@ng-bootstrap/ng-bootstrap'
 import { NgxBootstrapIconsModule, allIcons } from 'ngx-bootstrap-icons'
 import { Subject, of, throwError } from 'rxjs'
-import { PaperlessTaskType } from 'src/app/data/paperless-task'
+import { PaperlessTaskName } from 'src/app/data/paperless-task'
 import {
   InstallType,
   SystemStatus,
@@ -138,9 +138,9 @@ describe('SystemStatusDialogComponent', () => {
   })
 
   it('should check if task is running', () => {
-    component.runTask(PaperlessTaskType.IndexOptimize)
-    expect(component.isRunning(PaperlessTaskType.IndexOptimize)).toBeTruthy()
-    expect(component.isRunning(PaperlessTaskType.SanityCheck)).toBeFalsy()
+    component.runTask(PaperlessTaskName.IndexOptimize)
+    expect(component.isRunning(PaperlessTaskName.IndexOptimize)).toBeTruthy()
+    expect(component.isRunning(PaperlessTaskName.SanityCheck)).toBeFalsy()
   })
 
   it('should support running tasks, refresh status and show toasts', () => {
@@ -151,22 +151,22 @@ describe('SystemStatusDialogComponent', () => {
 
     // fail first
     runSpy.mockReturnValue(throwError(() => new Error('error')))
-    component.runTask(PaperlessTaskType.IndexOptimize)
-    expect(runSpy).toHaveBeenCalledWith(PaperlessTaskType.IndexOptimize)
+    component.runTask(PaperlessTaskName.IndexOptimize)
+    expect(runSpy).toHaveBeenCalledWith(PaperlessTaskName.IndexOptimize)
     expect(toastErrorSpy).toHaveBeenCalledWith(
-      `Failed to start task ${PaperlessTaskType.IndexOptimize}, see the logs for more details`,
+      `Failed to start task ${PaperlessTaskName.IndexOptimize}, see the logs for more details`,
       expect.any(Error)
     )
 
     // succeed
     runSpy.mockReturnValue(of({}))
     getStatusSpy.mockReturnValue(of(status))
-    component.runTask(PaperlessTaskType.IndexOptimize)
-    expect(runSpy).toHaveBeenCalledWith(PaperlessTaskType.IndexOptimize)
+    component.runTask(PaperlessTaskName.IndexOptimize)
+    expect(runSpy).toHaveBeenCalledWith(PaperlessTaskName.IndexOptimize)
 
     expect(getStatusSpy).toHaveBeenCalled()
     expect(toastSpy).toHaveBeenCalledWith(
-      `Task ${PaperlessTaskType.IndexOptimize} started`
+      `Task ${PaperlessTaskName.IndexOptimize} started`
     )
   })
 

src-ui/src/app/components/common/system-status-dialog/system-status-dialog.component.ts

@@ -8,7 +8,7 @@ import {
 } from '@ng-bootstrap/ng-bootstrap'
 import { NgxBootstrapIconsModule } from 'ngx-bootstrap-icons'
 import { Subject, takeUntil } from 'rxjs'
-import { PaperlessTaskType } from 'src/app/data/paperless-task'
+import { PaperlessTaskName } from 'src/app/data/paperless-task'
 import {
   SystemStatus,
   SystemStatusItemStatus,
@@ -49,14 +49,14 @@ export class SystemStatusDialogComponent implements OnInit, OnDestroy {
   private settingsService = inject(SettingsService)
 
   public SystemStatusItemStatus = SystemStatusItemStatus
-  public PaperlessTaskType = PaperlessTaskType
+  public PaperlessTaskName = PaperlessTaskName
   public status: SystemStatus
   public frontendVersion: string = environment.version
   public versionMismatch: boolean = false
 
   public copied: boolean = false
 
-  private runningTasks: Set<PaperlessTaskType> = new Set()
+  private runningTasks: Set<PaperlessTaskName> = new Set()
   private unsubscribeNotifier: Subject<any> = new Subject()
 
   get currentUserIsSuperUser(): boolean {
@@ -107,11 +107,11 @@ export class SystemStatusDialogComponent implements OnInit, OnDestroy {
     return now.getTime() - date.getTime() > hours * 60 * 60 * 1000
   }
 
-  public isRunning(taskName: PaperlessTaskType): boolean {
+  public isRunning(taskName: PaperlessTaskName): boolean {
     return this.runningTasks.has(taskName)
   }
 
-  public runTask(taskName: PaperlessTaskType) {
+  public runTask(taskName: PaperlessTaskName) {
     this.runningTasks.add(taskName)
     this.toastService.showInfo(`Task ${taskName} started`)
     this.tasksService.run(taskName).subscribe({

src-ui/src/app/data/paperless-task.ts

@@ -1,63 +1,49 @@
+import { Document } from './document'
 import { ObjectWithId } from './object-with-id'
 
 export enum PaperlessTaskType {
-  ConsumeFile = 'consume_file',
-  TrainClassifier = 'train_classifier',
-  SanityCheck = 'sanity_check',
-  IndexOptimize = 'index_optimize',
-  IndexRebuild = 'index_rebuild',
-  MailFetch = 'mail_fetch',
-  LlmIndex = 'llm_index',
+  Auto = 'auto_task',
+  ScheduledTask = 'scheduled_task',
+  ManualTask = 'manual_task',
 }
 
-export enum PaperlessTaskTriggerSource {
-  Scheduled = 'scheduled',
-  WebUI = 'web_ui',
-  ApiUpload = 'api_upload',
-  FolderConsume = 'folder_consume',
-  EmailConsume = 'email_consume',
-  System = 'system',
-  Manual = 'manual',
+export enum PaperlessTaskName {
+  ConsumeFile = 'consume_file',
+  TrainClassifier = 'train_classifier',
+  SanityCheck = 'check_sanity',
+  IndexOptimize = 'index_optimize',
+  LLMIndexUpdate = 'llmindex_update',
 }
 
 export enum PaperlessTaskStatus {
-  Pending = 'pending',
-  Started = 'started',
-  Success = 'success',
-  Failure = 'failure',
-  Revoked = 'revoked',
+  Pending = 'PENDING',
+  Started = 'STARTED',
+  Complete = 'SUCCESS',
+  Failed = 'FAILURE',
 }
 
 export interface PaperlessTask extends ObjectWithId {
-  task_id: string
-  task_type: PaperlessTaskType
-  task_type_display: string
-  trigger_source: PaperlessTaskTriggerSource
-  trigger_source_display: string
+  type: PaperlessTaskType
+
   status: PaperlessTaskStatus
-  status_display: string
-  date_created: Date
-  date_started?: Date
-  date_done?: Date
-  duration_seconds?: number
-  wait_time_seconds?: number
-  input_data: Record<string, unknown>
-  result_data?: Record<string, unknown>
-  result_message?: string
-  related_document_ids: number[]
+
   acknowledged: boolean
+
+  task_id: string
+
+  task_file_name: string
+
+  task_name: PaperlessTaskName
+
+  date_created: Date
+
+  date_done?: Date
+
+  result?: string
+
+  related_document?: number
+
+  duplicate_documents?: Document[]
+
   owner?: number
 }
-
-export interface PaperlessTaskSummary {
-  task_type: PaperlessTaskType
-  total_count: number
-  pending_count: number
-  success_count: number
-  failure_count: number
-  avg_duration_seconds: number | null
-  avg_wait_time_seconds: number | null
-  last_run: Date | null
-  last_success: Date | null
-  last_failure: Date | null
-}

src-ui/src/app/services/tasks.service.spec.ts

@@ -5,7 +5,11 @@ import {
 } from '@angular/common/http/testing'
 import { TestBed } from '@angular/core/testing'
 import { environment } from 'src/environments/environment'
-import { PaperlessTaskStatus, PaperlessTaskType } from '../data/paperless-task'
+import {
+  PaperlessTaskName,
+  PaperlessTaskStatus,
+  PaperlessTaskType,
+} from '../data/paperless-task'
 import { TasksService } from './tasks.service'
 
 describe('TasksService', () => {
@@ -33,7 +37,7 @@ describe('TasksService', () => {
   it('calls tasks api endpoint on reload', () => {
     tasksService.reload()
     const req = httpTestingController.expectOne(
-      `${environment.apiBaseUrl}tasks/?task_type=consume_file&acknowledged=false`
+      `${environment.apiBaseUrl}tasks/?task_name=consume_file&acknowledged=false`
     )
     expect(req.request.method).toEqual('GET')
   })
@@ -42,7 +46,7 @@ describe('TasksService', () => {
     tasksService.loading = true
     tasksService.reload()
     httpTestingController.expectNone(
-      `${environment.apiBaseUrl}tasks/?task_type=consume_file&acknowledged=false`
+      `${environment.apiBaseUrl}tasks/?task_name=consume_file&acknowledged=false`
     )
   })
 
@@ -59,7 +63,7 @@ describe('TasksService', () => {
     // reload is then called
     httpTestingController
       .expectOne(
-        `${environment.apiBaseUrl}tasks/?task_type=consume_file&acknowledged=false`
+        `${environment.apiBaseUrl}tasks/?task_name=consume_file&acknowledged=false`
       )
       .flush([])
   })
@@ -68,56 +72,56 @@ describe('TasksService', () => {
     expect(tasksService.total).toEqual(0)
     const mockTasks = [
       {
-        task_type: PaperlessTaskType.ConsumeFile,
-        status: PaperlessTaskStatus.Success,
+        type: PaperlessTaskType.Auto,
+        task_name: PaperlessTaskName.ConsumeFile,
+        status: PaperlessTaskStatus.Complete,
         acknowledged: false,
         task_id: '1234',
-        input_data: { filename: 'file1.pdf' },
+        task_file_name: 'file1.pdf',
         date_created: new Date(),
-        related_document_ids: [],
       },
       {
-        task_type: PaperlessTaskType.ConsumeFile,
-        status: PaperlessTaskStatus.Failure,
+        type: PaperlessTaskType.Auto,
+        task_name: PaperlessTaskName.ConsumeFile,
+        status: PaperlessTaskStatus.Failed,
         acknowledged: false,
         task_id: '1235',
-        input_data: { filename: 'file2.pdf' },
+        task_file_name: 'file2.pdf',
         date_created: new Date(),
-        related_document_ids: [],
       },
       {
-        task_type: PaperlessTaskType.ConsumeFile,
+        type: PaperlessTaskType.Auto,
+        task_name: PaperlessTaskName.ConsumeFile,
         status: PaperlessTaskStatus.Pending,
         acknowledged: false,
         task_id: '1236',
-        input_data: { filename: 'file3.pdf' },
+        task_file_name: 'file3.pdf',
         date_created: new Date(),
-        related_document_ids: [],
       },
       {
-        task_type: PaperlessTaskType.ConsumeFile,
+        type: PaperlessTaskType.Auto,
+        task_name: PaperlessTaskName.ConsumeFile,
         status: PaperlessTaskStatus.Started,
         acknowledged: false,
         task_id: '1237',
-        input_data: { filename: 'file4.pdf' },
+        task_file_name: 'file4.pdf',
         date_created: new Date(),
-        related_document_ids: [],
       },
       {
-        task_type: PaperlessTaskType.ConsumeFile,
-        status: PaperlessTaskStatus.Success,
+        type: PaperlessTaskType.Auto,
+        task_name: PaperlessTaskName.ConsumeFile,
+        status: PaperlessTaskStatus.Complete,
         acknowledged: false,
         task_id: '1238',
-        input_data: { filename: 'file5.pdf' },
+        task_file_name: 'file5.pdf',
         date_created: new Date(),
-        related_document_ids: [],
       },
     ]
 
     tasksService.reload()
 
     const req = httpTestingController.expectOne(
-      `${environment.apiBaseUrl}tasks/?task_type=consume_file&acknowledged=false`
+      `${environment.apiBaseUrl}tasks/?task_name=consume_file&acknowledged=false`
     )
 
     req.flush(mockTasks)
@@ -130,9 +134,9 @@ describe('TasksService', () => {
   })
 
   it('supports running tasks', () => {
-    tasksService.run(PaperlessTaskType.SanityCheck).subscribe((res) => {
+    tasksService.run(PaperlessTaskName.SanityCheck).subscribe((res) => {
       expect(res).toEqual({
-        task_id: 'abc-123',
+        result: 'success',
       })
     })
     const req = httpTestingController.expectOne(
@@ -140,7 +144,7 @@ describe('TasksService', () => {
     )
     expect(req.request.method).toEqual('POST')
     req.flush({
-      task_id: 'abc-123',
+      result: 'success',
     })
   })
 })

src-ui/src/app/services/tasks.service.ts

@@ -4,8 +4,8 @@ import { Observable, Subject } from 'rxjs'
 import { first, takeUntil, tap } from 'rxjs/operators'
 import {
   PaperlessTask,
+  PaperlessTaskName,
   PaperlessTaskStatus,
-  PaperlessTaskType,
 } from 'src/app/data/paperless-task'
 import { environment } from 'src/environments/environment'
 
@@ -18,7 +18,7 @@ export class TasksService {
   private baseUrl: string = environment.apiBaseUrl
   private endpoint: string = 'tasks'
 
-  public loading: boolean = false
+  public loading: boolean
 
   private fileTasks: PaperlessTask[] = []
 
@@ -33,27 +33,21 @@ export class TasksService {
   }
 
   public get queuedFileTasks(): PaperlessTask[] {
-    return this.fileTasks.filter(
-      (t) => t.status === PaperlessTaskStatus.Pending
-    )
+    return this.fileTasks.filter((t) => t.status == PaperlessTaskStatus.Pending)
   }
 
   public get startedFileTasks(): PaperlessTask[] {
-    return this.fileTasks.filter(
-      (t) => t.status === PaperlessTaskStatus.Started
-    )
+    return this.fileTasks.filter((t) => t.status == PaperlessTaskStatus.Started)
   }
 
   public get completedFileTasks(): PaperlessTask[] {
     return this.fileTasks.filter(
-      (t) => t.status === PaperlessTaskStatus.Success
+      (t) => t.status == PaperlessTaskStatus.Complete
     )
   }
 
   public get failedFileTasks(): PaperlessTask[] {
-    return this.fileTasks.filter(
-      (t) => t.status === PaperlessTaskStatus.Failure
-    )
+    return this.fileTasks.filter((t) => t.status == PaperlessTaskStatus.Failed)
   }
 
   public reload() {
@@ -62,16 +56,18 @@ export class TasksService {
 
     this.http
       .get<PaperlessTask[]>(
-        `${this.baseUrl}${this.endpoint}/?task_type=${PaperlessTaskType.ConsumeFile}&acknowledged=false`
+        `${this.baseUrl}${this.endpoint}/?task_name=consume_file&acknowledged=false`
       )
       .pipe(takeUntil(this.unsubscribeNotifer), first())
       .subscribe((r) => {
-        this.fileTasks = r
+        this.fileTasks = r.filter(
+          (t) => t.task_name == PaperlessTaskName.ConsumeFile
+        )
         this.loading = false
       })
   }
 
-  public dismissTasks(task_ids: Set<number>): Observable<any> {
+  public dismissTasks(task_ids: Set<number>) {
     return this.http
       .post(`${this.baseUrl}tasks/acknowledge/`, {
         tasks: [...task_ids],
@@ -85,24 +81,16 @@ export class TasksService {
       )
   }
 
-  public dismissAllTasks(): Observable<any> {
-    return this.http.post(`${this.baseUrl}tasks/acknowledge_all/`, {}).pipe(
-      first(),
-      takeUntil(this.unsubscribeNotifer),
-      tap(() => {
-        this.reload()
-      })
-    )
-  }
-
   public cancelPending(): void {
     this.unsubscribeNotifer.next(true)
   }
 
-  public run(taskType: PaperlessTaskType): Observable<{ task_id: string }> {
-    return this.http.post<{ task_id: string }>(
+  public run(taskName: PaperlessTaskName): Observable<any> {
+    return this.http.post<any>(
       `${environment.apiBaseUrl}${this.endpoint}/run/`,
-      { task_type: taskType }
+      {
+        task_name: taskName,
+      }
     )
   }
 }

src/documents/admin.py

@@ -144,30 +144,18 @@ class StoragePathAdmin(GuardedModelAdmin):
 
 
 class TaskAdmin(admin.ModelAdmin):
-    list_display = (
-        "task_id",
-        "task_type",
-        "trigger_source",
-        "status",
-        "date_created",
-        "date_done",
-        "duration_seconds",
-    )
-    list_filter = ("status", "task_type", "trigger_source", "date_done")
-    search_fields = ("task_id", "task_type", "status")
+    list_display = ("task_id", "task_file_name", "task_name", "date_done", "status")
+    list_filter = ("status", "date_done", "task_name")
+    search_fields = ("task_name", "task_id", "status", "task_file_name")
     readonly_fields = (
         "task_id",
-        "task_type",
-        "trigger_source",
+        "task_file_name",
+        "task_name",
         "status",
         "date_created",
         "date_started",
         "date_done",
-        "duration_seconds",
-        "wait_time_seconds",
-        "input_data",
-        "result_data",
-        "result_message",
+        "result",
     )
 
 

src/documents/consumer.py

@@ -313,6 +313,7 @@ class ConsumerPlugin(
             run_subprocess(
                 [
                     settings.PRE_CONSUME_SCRIPT,
+                    original_file_path,
                 ],
                 script_env,
                 self.log,
@@ -382,6 +383,14 @@ class ConsumerPlugin(
             run_subprocess(
                 [
                     settings.POST_CONSUME_SCRIPT,
+                    str(document.pk),
+                    document.get_public_filename(),
+                    os.path.normpath(document.source_path),
+                    os.path.normpath(document.thumbnail_path),
+                    reverse("document-download", kwargs={"pk": document.pk}),
+                    reverse("document-thumb", kwargs={"pk": document.pk}),
+                    str(document.correspondent),
+                    str(",".join(document.tags.all().values_list("name", flat=True))),
                 ],
                 script_env,
                 self.log,
@@ -641,10 +650,6 @@ class ConsumerPlugin(
                         # If we get here, it was successful. Proceed with post-consume
                         # hooks. If they fail, nothing will get changed.
 
-                        document = Document.objects.prefetch_related("versions").get(
-                            pk=document.pk,
-                        )
-
                         document_consumption_finished.send(
                             sender=self.__class__,
                             document=document,

src/documents/filters.py

@@ -26,10 +26,8 @@ from django.db.models.functions import Cast
 from django.utils.translation import gettext_lazy as _
 from django_filters import DateFilter
 from django_filters.rest_framework import BooleanFilter
-from django_filters.rest_framework import DateTimeFilter
 from django_filters.rest_framework import Filter
 from django_filters.rest_framework import FilterSet
-from django_filters.rest_framework import MultipleChoiceFilter
 from drf_spectacular.utils import extend_schema_field
 from guardian.utils import get_group_obj_perms_model
 from guardian.utils import get_user_obj_perms_model
@@ -864,56 +862,18 @@ class ShareLinkBundleFilterSet(FilterSet):
 
 
 class PaperlessTaskFilterSet(FilterSet):
-    task_type = MultipleChoiceFilter(
-        choices=PaperlessTask.TaskType.choices,
-        label="Task Type",
-    )
-
-    trigger_source = MultipleChoiceFilter(
-        choices=PaperlessTask.TriggerSource.choices,
-        label="Trigger Source",
-    )
-
-    status = MultipleChoiceFilter(
-        choices=PaperlessTask.Status.choices,
-        label="Status",
-    )
-
-    is_complete = BooleanFilter(
-        method="filter_is_complete",
-        label="Is Complete",
-    )
-
     acknowledged = BooleanFilter(
         label="Acknowledged",
         field_name="acknowledged",
     )
 
-    date_created_after = DateTimeFilter(
-        field_name="date_created",
-        lookup_expr="gte",
-        label="Created After",
-    )
-
-    date_created_before = DateTimeFilter(
-        field_name="date_created",
-        lookup_expr="lte",
-        label="Created Before",
-    )
-
     class Meta:
         model = PaperlessTask
-        fields = ["task_type", "trigger_source", "status", "acknowledged", "owner"]
-
-    def filter_is_complete(self, queryset, name, value):
-        complete = [
-            PaperlessTask.Status.SUCCESS,
-            PaperlessTask.Status.FAILURE,
-            PaperlessTask.Status.REVOKED,
-        ]
-        if value:
-            return queryset.filter(status__in=complete)
-        return queryset.exclude(status__in=complete)
+        fields = {
+            "type": ["exact"],
+            "task_name": ["exact"],
+            "status": ["exact"],
+        }
 
 
 class ObjectOwnedOrGrantedPermissionsFilter(ObjectPermissionsFilter):

src/documents/management/commands/document_create_classifier.py

@@ -22,6 +22,7 @@ class Command(PaperlessCommand):
             self.buffered_logging("paperless.classifier"),
         ):
             train_classifier(
+                scheduled=False,
                 status_callback=lambda msg: self.console.print(f"  {msg}"),
             )
 

src/documents/management/commands/document_llmindex.py

@@ -17,6 +17,7 @@ class Command(PaperlessCommand):
     def handle(self, *args: Any, **options: Any) -> None:
         llmindex_index(
             rebuild=options["command"] == "rebuild",
+            scheduled=False,
             iter_wrapper=lambda docs: self.track(
                 docs,
                 description="Indexing documents...",

src/documents/management/commands/document_sanity_checker.py

@@ -111,6 +111,7 @@ class Command(PaperlessCommand):
 
     def handle(self, *args: Any, **options: Any) -> None:
         messages = check_sanity(
+            scheduled=False,
             iter_wrapper=lambda docs: self.track(
                 docs,
                 description="Checking documents...",

src/documents/migrations/0019_task_system_redesign.py

@@ -1,213 +0,0 @@
-"""
-Drop and recreate the PaperlessTask table with the new structured schema.
-
-We intentionally drop all existing task data -- the old schema was
-string-based and incompatible with the new JSONField result storage.
-"""
-
-import django.db.models.deletion
-import django.utils.timezone
-from django.conf import settings
-from django.db import migrations
-from django.db import models
-
-
-class Migration(migrations.Migration):
-    dependencies = [
-        ("documents", "0018_saved_view_simple_search_rules"),
-        migrations.swappable_dependency(settings.AUTH_USER_MODEL),
-    ]
-
-    operations = [
-        migrations.DeleteModel(name="PaperlessTask"),
-        migrations.CreateModel(
-            name="PaperlessTask",
-            fields=[
-                (
-                    "id",
-                    models.AutoField(
-                        auto_created=True,
-                        primary_key=True,
-                        serialize=False,
-                        verbose_name="ID",
-                    ),
-                ),
-                (
-                    "owner",
-                    models.ForeignKey(
-                        blank=True,
-                        default=None,
-                        null=True,
-                        on_delete=django.db.models.deletion.SET_NULL,
-                        to=settings.AUTH_USER_MODEL,
-                        verbose_name="owner",
-                    ),
-                ),
-                (
-                    "task_id",
-                    models.CharField(
-                        help_text="Celery task ID",
-                        max_length=255,
-                        unique=True,
-                        verbose_name="Task ID",
-                    ),
-                ),
-                (
-                    "task_type",
-                    models.CharField(
-                        choices=[
-                            ("consume_file", "Consume File"),
-                            ("train_classifier", "Train Classifier"),
-                            ("sanity_check", "Sanity Check"),
-                            ("index_optimize", "Index Optimize"),
-                            ("index_rebuild", "Index Rebuild"),
-                            ("mail_fetch", "Mail Fetch"),
-                            ("llm_index", "LLM Index"),
-                        ],
-                        db_index=True,
-                        help_text="The kind of work being performed",
-                        max_length=50,
-                        verbose_name="Task Type",
-                    ),
-                ),
-                (
-                    "trigger_source",
-                    models.CharField(
-                        choices=[
-                            ("scheduled", "Scheduled"),
-                            ("web_ui", "Web UI"),
-                            ("api_upload", "API Upload"),
-                            ("folder_consume", "Folder Consume"),
-                            ("email_consume", "Email Consume"),
-                            ("system", "System"),
-                            ("manual", "Manual"),
-                        ],
-                        db_index=True,
-                        help_text="What initiated this task",
-                        max_length=50,
-                        verbose_name="Trigger Source",
-                    ),
-                ),
-                (
-                    "status",
-                    models.CharField(
-                        choices=[
-                            ("pending", "Pending"),
-                            ("started", "Started"),
-                            ("success", "Success"),
-                            ("failure", "Failure"),
-                            ("revoked", "Revoked"),
-                        ],
-                        db_index=True,
-                        default="pending",
-                        max_length=30,
-                        verbose_name="Status",
-                    ),
-                ),
-                (
-                    "date_created",
-                    models.DateTimeField(
-                        db_index=True,
-                        default=django.utils.timezone.now,
-                        verbose_name="Created",
-                    ),
-                ),
-                (
-                    "date_started",
-                    models.DateTimeField(
-                        blank=True,
-                        null=True,
-                        verbose_name="Started",
-                    ),
-                ),
-                (
-                    "date_done",
-                    models.DateTimeField(
-                        blank=True,
-                        db_index=True,
-                        null=True,
-                        verbose_name="Completed",
-                    ),
-                ),
-                (
-                    "duration_seconds",
-                    models.FloatField(
-                        blank=True,
-                        help_text="Elapsed time from start to completion",
-                        null=True,
-                        verbose_name="Duration (seconds)",
-                    ),
-                ),
-                (
-                    "wait_time_seconds",
-                    models.FloatField(
-                        blank=True,
-                        help_text="Time from task creation to worker pickup",
-                        null=True,
-                        verbose_name="Wait Time (seconds)",
-                    ),
-                ),
-                (
-                    "input_data",
-                    models.JSONField(
-                        blank=True,
-                        default=dict,
-                        help_text="Structured input parameters for the task",
-                        verbose_name="Input Data",
-                    ),
-                ),
-                (
-                    "result_data",
-                    models.JSONField(
-                        blank=True,
-                        help_text="Structured result data from task execution",
-                        null=True,
-                        verbose_name="Result Data",
-                    ),
-                ),
-                (
-                    "result_message",
-                    models.TextField(
-                        blank=True,
-                        help_text="Human-readable result message",
-                        null=True,
-                        verbose_name="Result Message",
-                    ),
-                ),
-                (
-                    "acknowledged",
-                    models.BooleanField(
-                        db_index=True,
-                        default=False,
-                        verbose_name="Acknowledged",
-                    ),
-                ),
-            ],
-            options={
-                "verbose_name": "Task",
-                "verbose_name_plural": "Tasks",
-                "ordering": ["-date_created"],
-            },
-        ),
-        migrations.AddIndex(
-            model_name="paperlesstask",
-            index=models.Index(
-                fields=["status", "date_created"],
-                name="documents_p_status_8aa687_idx",
-            ),
-        ),
-        migrations.AddIndex(
-            model_name="paperlesstask",
-            index=models.Index(
-                fields=["task_type", "status"],
-                name="documents_p_task_ty_e4a93f_idx",
-            ),
-        ),
-        migrations.AddIndex(
-            model_name="paperlesstask",
-            index=models.Index(
-                fields=["owner", "acknowledged", "date_created"],
-                name="documents_p_owner_i_62c545_idx",
-            ),
-        ),
-    ]

src/documents/migrations/0020_drop_celery_results.py

@@ -1,22 +0,0 @@
-from django.db import migrations
-
-
-class Migration(migrations.Migration):
-    dependencies = [
-        ("documents", "0019_task_system_redesign"),
-    ]
-
-    operations = [
-        migrations.RunSQL(
-            sql="DROP TABLE IF EXISTS django_celery_results_taskresult;",
-            reverse_sql=migrations.RunSQL.noop,
-        ),
-        migrations.RunSQL(
-            sql="DROP TABLE IF EXISTS django_celery_results_groupresult;",
-            reverse_sql=migrations.RunSQL.noop,
-        ),
-        migrations.RunSQL(
-            sql="DROP TABLE IF EXISTS django_celery_results_chordcounter;",
-            reverse_sql=migrations.RunSQL.noop,
-        ),
-    ]

src/documents/models.py

@@ -3,6 +3,7 @@ from pathlib import Path
 from typing import Final
 
 import pathvalidate
+from celery import states
 from django.conf import settings
 from django.contrib.auth.models import Group
 from django.contrib.auth.models import User
@@ -380,10 +381,7 @@ class Document(SoftDeleteModel, ModelWithOwner):  # type: ignore[django-manager-
             if isinstance(prefetched_cache, dict)
             else None
         )
-        if prefetched_versions is not None:
-            # Empty list means prefetch ran and found no versions — use own content.
-            if not prefetched_versions:
-                return self.content
+        if prefetched_versions:
             latest_prefetched = max(prefetched_versions, key=lambda doc: doc.id)
             return latest_prefetched.content
 
@@ -662,170 +660,97 @@ class UiSettings(models.Model):
 
 
 class PaperlessTask(ModelWithOwner):
-    """
-    Tracks background task execution for user visibility and debugging.
-
-    State transitions:
-        PENDING -> STARTED -> SUCCESS
-        PENDING -> STARTED -> FAILURE
-        PENDING -> REVOKED (if cancelled before starting)
-    """
-
-    class Status(models.TextChoices):
-        PENDING = "pending", _("Pending")
-        STARTED = "started", _("Started")
-        SUCCESS = "success", _("Success")
-        FAILURE = "failure", _("Failure")
-        REVOKED = "revoked", _("Revoked")
+    ALL_STATES = sorted(states.ALL_STATES)
+    TASK_STATE_CHOICES = sorted(zip(ALL_STATES, ALL_STATES))
 
     class TaskType(models.TextChoices):
-        CONSUME_FILE = "consume_file", _("Consume File")
-        TRAIN_CLASSIFIER = "train_classifier", _("Train Classifier")
-        SANITY_CHECK = "sanity_check", _("Sanity Check")
-        INDEX_OPTIMIZE = "index_optimize", _("Index Optimize")
-        INDEX_REBUILD = "index_rebuild", _("Index Rebuild")
-        MAIL_FETCH = "mail_fetch", _("Mail Fetch")
-        LLM_INDEX = "llm_index", _("LLM Index")
+        AUTO = ("auto_task", _("Auto Task"))
+        SCHEDULED_TASK = ("scheduled_task", _("Scheduled Task"))
+        MANUAL_TASK = ("manual_task", _("Manual Task"))
 
-    class TriggerSource(models.TextChoices):
-        SCHEDULED = "scheduled", _("Scheduled")  # Celery beat
-        WEB_UI = "web_ui", _("Web UI")  # Document uploaded via web
-        API_UPLOAD = "api_upload", _("API Upload")  # Document uploaded via API
-        FOLDER_CONSUME = "folder_consume", _("Folder Consume")  # Consume folder
-        EMAIL_CONSUME = "email_consume", _("Email Consume")  # Email attachment
-        SYSTEM = (
-            "system",
-            _("System"),
-        )  # Auto-triggered by system (self-heal, config side-effect)
-        MANUAL = "manual", _("Manual")  # User explicitly ran via /api/tasks/run/
+    class TaskName(models.TextChoices):
+        CONSUME_FILE = ("consume_file", _("Consume File"))
+        TRAIN_CLASSIFIER = ("train_classifier", _("Train Classifier"))
+        CHECK_SANITY = ("check_sanity", _("Check Sanity"))
+        INDEX_OPTIMIZE = ("index_optimize", _("Index Optimize"))
+        LLMINDEX_UPDATE = ("llmindex_update", _("LLM Index Update"))
 
-    # Identification
     task_id = models.CharField(
         max_length=255,
         unique=True,
         verbose_name=_("Task ID"),
-        help_text=_("Celery task ID"),
+        help_text=_("Celery ID for the Task that was run"),
     )
 
-    task_type = models.CharField(
-        max_length=50,
-        choices=TaskType.choices,
-        verbose_name=_("Task Type"),
-        help_text=_("The kind of work being performed"),
-        db_index=True,
+    acknowledged = models.BooleanField(
+        default=False,
+        verbose_name=_("Acknowledged"),
+        help_text=_("If the task is acknowledged via the frontend or API"),
     )
 
-    trigger_source = models.CharField(
-        max_length=50,
-        choices=TriggerSource.choices,
-        verbose_name=_("Trigger Source"),
-        help_text=_("What initiated this task"),
-        db_index=True,
+    task_file_name = models.CharField(
+        null=True,
+        max_length=255,
+        verbose_name=_("Task Filename"),
+        help_text=_("Name of the file which the Task was run for"),
+    )
+
+    task_name = models.CharField(
+        null=True,
+        max_length=255,
+        choices=TaskName.choices,
+        verbose_name=_("Task Name"),
+        help_text=_("Name of the task that was run"),
     )
 
-    # State tracking
     status = models.CharField(
         max_length=30,
-        choices=Status.choices,
-        default=Status.PENDING,
-        verbose_name=_("Status"),
-        db_index=True,
+        default=states.PENDING,
+        choices=TASK_STATE_CHOICES,
+        verbose_name=_("Task State"),
+        help_text=_("Current state of the task being run"),
     )
 
-    # Timestamps
     date_created = models.DateTimeField(
+        null=True,
         default=timezone.now,
-        verbose_name=_("Created"),
-        db_index=True,
+        verbose_name=_("Created DateTime"),
+        help_text=_("Datetime field when the task result was created in UTC"),
     )
 
     date_started = models.DateTimeField(
         null=True,
-        blank=True,
-        verbose_name=_("Started"),
+        default=None,
+        verbose_name=_("Started DateTime"),
+        help_text=_("Datetime field when the task was started in UTC"),
     )
 
     date_done = models.DateTimeField(
         null=True,
-        blank=True,
-        verbose_name=_("Completed"),
-        db_index=True,
+        default=None,
+        verbose_name=_("Completed DateTime"),
+        help_text=_("Datetime field when the task was completed in UTC"),
     )
 
-    # Duration fields -- populated by task_postrun signal handler
-    duration_seconds = models.FloatField(
+    result = models.TextField(
         null=True,
-        blank=True,
-        verbose_name=_("Duration (seconds)"),
-        help_text=_("Elapsed time from start to completion"),
-    )
-
-    wait_time_seconds = models.FloatField(
-        null=True,
-        blank=True,
-        verbose_name=_("Wait Time (seconds)"),
-        help_text=_("Time from task creation to worker pickup"),
-    )
-
-    # Input/Output data
-    input_data = models.JSONField(
-        default=dict,
-        blank=True,
-        verbose_name=_("Input Data"),
-        help_text=_("Structured input parameters for the task"),
-    )
-
-    result_data = models.JSONField(
-        null=True,
-        blank=True,
+        default=None,
         verbose_name=_("Result Data"),
-        help_text=_("Structured result data from task execution"),
+        help_text=_(
+            "The data returned by the task",
+        ),
     )
 
-    result_message = models.TextField(
-        null=True,
-        blank=True,
-        verbose_name=_("Result Message"),
-        help_text=_("Human-readable result message"),
+    type = models.CharField(
+        max_length=30,
+        choices=TaskType.choices,
+        default=TaskType.AUTO,
+        verbose_name=_("Task Type"),
+        help_text=_("The type of task that was run"),
     )
 
-    # Acknowledgment
-    acknowledged = models.BooleanField(
-        default=False,
-        verbose_name=_("Acknowledged"),
-        db_index=True,
-    )
-
-    class Meta:
-        verbose_name = _("Task")
-        verbose_name_plural = _("Tasks")
-        ordering = ["-date_created"]
-        indexes = [
-            models.Index(fields=["status", "date_created"]),
-            models.Index(fields=["task_type", "status"]),
-            models.Index(fields=["owner", "acknowledged", "date_created"]),
-        ]
-
     def __str__(self) -> str:
-        return f"{self.get_task_type_display()} [{self.task_id[:8]}]"
-
-    @property
-    def is_complete(self) -> bool:
-        return self.status in (
-            self.Status.SUCCESS,
-            self.Status.FAILURE,
-            self.Status.REVOKED,
-        )
-
-    @property
-    def related_document_ids(self) -> list[int]:
-        if not self.result_data:
-            return []
-        if doc_id := self.result_data.get("document_id"):
-            return [doc_id]
-        if dup_id := self.result_data.get("duplicate_of"):
-            return [dup_id]
-        return []
+        return f"Task {self.task_id}"
 
 
 class Note(SoftDeleteModel):

src/documents/sanity_checker.py

@@ -10,6 +10,7 @@ is an identity function that adds no overhead.
 """
 
 import logging
+import uuid
 from collections import defaultdict
 from collections.abc import Iterator
 from pathlib import Path
@@ -17,9 +18,12 @@ from typing import TYPE_CHECKING
 from typing import Final
 from typing import TypedDict
 
+from celery import states
 from django.conf import settings
+from django.utils import timezone
 
 from documents.models import Document
+from documents.models import PaperlessTask
 from documents.utils import IterWrapper
 from documents.utils import compute_checksum
 from documents.utils import identity
@@ -178,9 +182,8 @@ def _check_thumbnail(
     present_files: set[Path],
 ) -> None:
     """Verify the thumbnail exists and is readable."""
-    # doc.thumbnail_path already returns a resolved Path; no need to re-resolve.
-    thumbnail_path: Final[Path] = doc.thumbnail_path
-    if not thumbnail_path.is_file():
+    thumbnail_path: Final[Path] = Path(doc.thumbnail_path).resolve()
+    if not thumbnail_path.exists() or not thumbnail_path.is_file():
         messages.error(doc.pk, "Thumbnail of document does not exist.")
         return
 
@@ -197,9 +200,8 @@ def _check_original(
     present_files: set[Path],
 ) -> None:
     """Verify the original file exists, is readable, and has matching checksum."""
-    # doc.source_path already returns a resolved Path; no need to re-resolve.
-    source_path: Final[Path] = doc.source_path
-    if not source_path.is_file():
+    source_path: Final[Path] = Path(doc.source_path).resolve()
+    if not source_path.exists() or not source_path.is_file():
         messages.error(doc.pk, "Original of document does not exist.")
         return
 
@@ -235,9 +237,8 @@ def _check_archive(
     elif doc.has_archive_version:
         if TYPE_CHECKING:
             assert isinstance(doc.archive_path, Path)
-        # doc.archive_path already returns a resolved Path; no need to re-resolve.
-        archive_path: Final[Path] = doc.archive_path  # type: ignore[assignment]
-        if not archive_path.is_file():
+        archive_path: Final[Path] = Path(doc.archive_path).resolve()
+        if not archive_path.exists() or not archive_path.is_file():
             messages.error(doc.pk, "Archived version of document does not exist.")
             return
 
@@ -283,33 +284,59 @@ def _check_document(
 
 def check_sanity(
     *,
+    scheduled: bool = True,
     iter_wrapper: IterWrapper[Document] = identity,
 ) -> SanityCheckMessages:
     """Run a full sanity check on the document archive.
 
     Args:
+        scheduled: Whether this is a scheduled (automatic) or manual check.
+            Controls the task type recorded in the database.
         iter_wrapper: A callable that wraps the document iterable, e.g.,
             for progress bar display. Defaults to identity (no wrapping).
 
     Returns:
         A SanityCheckMessages instance containing all detected issues.
     """
+    paperless_task = PaperlessTask.objects.create(
+        task_id=uuid.uuid4(),
+        type=(
+            PaperlessTask.TaskType.SCHEDULED_TASK
+            if scheduled
+            else PaperlessTask.TaskType.MANUAL_TASK
+        ),
+        task_name=PaperlessTask.TaskName.CHECK_SANITY,
+        status=states.STARTED,
+        date_created=timezone.now(),
+        date_started=timezone.now(),
+    )
+
     messages = SanityCheckMessages()
     present_files = _build_present_files()
 
-    documents = Document.global_objects.only(
-        "pk",
-        "filename",
-        "mime_type",
-        "checksum",
-        "archive_checksum",
-        "archive_filename",
-        "content",
-    ).iterator(chunk_size=500)
+    documents = Document.global_objects.all()
     for doc in iter_wrapper(documents):
         _check_document(doc, messages, present_files)
 
     for extra_file in present_files:
         messages.warning(None, f"Orphaned file in media dir: {extra_file}")
 
+    paperless_task.status = states.SUCCESS if not messages.has_error else states.FAILURE
+    if messages.total_issue_count == 0:
+        paperless_task.result = "No issues found."
+    else:
+        parts: list[str] = []
+        if messages.document_error_count:
+            parts.append(f"{messages.document_error_count} document(s) with errors")
+        if messages.document_warning_count:
+            parts.append(f"{messages.document_warning_count} document(s) with warnings")
+        if messages.global_warning_count:
+            parts.append(f"{messages.global_warning_count} global warning(s)")
+        paperless_task.result = ", ".join(parts) + " found."
+        if messages.has_error:
+            paperless_task.result += " Check logs for details."
+
+    paperless_task.date_done = timezone.now()
+    paperless_task.save(update_fields=["status", "result", "date_done"])
+
     return messages

src/documents/serialisers.py

@@ -12,6 +12,7 @@ from typing import Literal
 from typing import TypedDict
 
 import magic
+from celery import states
 from django.conf import settings
 from django.contrib.auth.models import Group
 from django.contrib.auth.models import User
@@ -99,7 +100,7 @@ logger = logging.getLogger("paperless.serializers")
 
 
 # https://www.django-rest-framework.org/api-guide/serializers/#example
-class DynamicFieldsModelSerializer(serializers.ModelSerializer[Any]):
+class DynamicFieldsModelSerializer(serializers.ModelSerializer):
     """
     A ModelSerializer that takes an additional `fields` argument that
     controls which fields should be displayed.
@@ -120,7 +121,7 @@ class DynamicFieldsModelSerializer(serializers.ModelSerializer[Any]):
                 self.fields.pop(field_name)
 
 
-class MatchingModelSerializer(serializers.ModelSerializer[Any]):
+class MatchingModelSerializer(serializers.ModelSerializer):
     document_count = serializers.IntegerField(read_only=True)
 
     def get_slug(self, obj) -> str:
@@ -260,7 +261,7 @@ class SetPermissionsSerializer(serializers.DictField):
 
 class OwnedObjectSerializer(
     SerializerWithPerms,
-    serializers.ModelSerializer[Any],
+    serializers.ModelSerializer,
     SetPermissionsMixin,
 ):
     def __init__(self, *args, **kwargs) -> None:
@@ -468,7 +469,7 @@ class OwnedObjectSerializer(
         return super().update(instance, validated_data)
 
 
-class OwnedObjectListSerializer(serializers.ListSerializer[Any]):
+class OwnedObjectListSerializer(serializers.ListSerializer):
     def to_representation(self, documents):
         self.child.context["shared_object_pks"] = self.child.get_shared_object_pks(
             documents,
@@ -681,27 +682,27 @@ class TagSerializer(MatchingModelSerializer, OwnedObjectSerializer):
         return super().validate(attrs)
 
 
-class CorrespondentField(serializers.PrimaryKeyRelatedField[Correspondent]):
+class CorrespondentField(serializers.PrimaryKeyRelatedField):
     def get_queryset(self):
         return Correspondent.objects.all()
 
 
-class TagsField(serializers.PrimaryKeyRelatedField[Tag]):
+class TagsField(serializers.PrimaryKeyRelatedField):
     def get_queryset(self):
         return Tag.objects.all()
 
 
-class DocumentTypeField(serializers.PrimaryKeyRelatedField[DocumentType]):
+class DocumentTypeField(serializers.PrimaryKeyRelatedField):
     def get_queryset(self):
         return DocumentType.objects.all()
 
 
-class StoragePathField(serializers.PrimaryKeyRelatedField[StoragePath]):
+class StoragePathField(serializers.PrimaryKeyRelatedField):
     def get_queryset(self):
         return StoragePath.objects.all()
 
 
-class CustomFieldSerializer(serializers.ModelSerializer[CustomField]):
+class CustomFieldSerializer(serializers.ModelSerializer):
     data_type = serializers.ChoiceField(
         choices=CustomField.FieldDataType,
         read_only=False,
@@ -815,7 +816,7 @@ def validate_documentlink_targets(user, doc_ids):
         )
 
 
-class CustomFieldInstanceSerializer(serializers.ModelSerializer[CustomFieldInstance]):
+class CustomFieldInstanceSerializer(serializers.ModelSerializer):
     field = serializers.PrimaryKeyRelatedField(queryset=CustomField.objects.all())
     value = ReadWriteSerializerMethodField(allow_null=True)
 
@@ -921,14 +922,14 @@ class CustomFieldInstanceSerializer(serializers.ModelSerializer[CustomFieldInsta
         ]
 
 
-class BasicUserSerializer(serializers.ModelSerializer[User]):
+class BasicUserSerializer(serializers.ModelSerializer):
     # Different than paperless.serializers.UserSerializer
     class Meta:
         model = User
         fields = ["id", "username", "first_name", "last_name"]
 
 
-class NotesSerializer(serializers.ModelSerializer[Note]):
+class NotesSerializer(serializers.ModelSerializer):
     user = BasicUserSerializer(read_only=True)
 
     class Meta:
@@ -1255,7 +1256,7 @@ class DocumentSerializer(
         list_serializer_class = OwnedObjectListSerializer
 
 
-class SearchResultListSerializer(serializers.ListSerializer[Document]):
+class SearchResultListSerializer(serializers.ListSerializer):
     def to_representation(self, hits):
         document_ids = [hit["id"] for hit in hits]
         # Fetch all Document objects in the list in one SQL query.
@@ -1312,7 +1313,7 @@ class SearchResultSerializer(DocumentSerializer):
         list_serializer_class = SearchResultListSerializer
 
 
-class SavedViewFilterRuleSerializer(serializers.ModelSerializer[SavedViewFilterRule]):
+class SavedViewFilterRuleSerializer(serializers.ModelSerializer):
     class Meta:
         model = SavedViewFilterRule
         fields = ["rule_type", "value"]
@@ -2400,7 +2401,7 @@ class StoragePathSerializer(MatchingModelSerializer, OwnedObjectSerializer):
         return super().update(instance, validated_data)
 
 
-class UiSettingsViewSerializer(serializers.ModelSerializer[UiSettings]):
+class UiSettingsViewSerializer(serializers.ModelSerializer):
     settings = serializers.DictField(required=False, allow_null=True)
 
     class Meta:
@@ -2428,81 +2429,7 @@ class UiSettingsViewSerializer(serializers.ModelSerializer[UiSettings]):
         return ui_settings
 
 
-class TaskSerializerV10(OwnedObjectSerializer):
-    """Task serializer for API v10+ using new field names."""
-
-    related_document_ids = serializers.ListField(
-        child=serializers.IntegerField(),
-        read_only=True,
-    )
-    task_type_display = serializers.CharField(
-        source="get_task_type_display",
-        read_only=True,
-    )
-    trigger_source_display = serializers.CharField(
-        source="get_trigger_source_display",
-        read_only=True,
-    )
-    status_display = serializers.CharField(
-        source="get_status_display",
-        read_only=True,
-    )
-
-    class Meta:
-        model = PaperlessTask
-        fields = (
-            "id",
-            "task_id",
-            "task_type",
-            "task_type_display",
-            "trigger_source",
-            "trigger_source_display",
-            "status",
-            "status_display",
-            "date_created",
-            "date_started",
-            "date_done",
-            "duration_seconds",
-            "wait_time_seconds",
-            "input_data",
-            "result_data",
-            "result_message",
-            "related_document_ids",
-            "acknowledged",
-            "owner",
-        )
-        read_only_fields = fields
-
-
-class TaskSerializerV9(serializers.ModelSerializer):
-    """Task serializer for API v9 backwards compatibility.
-
-    Maps old field names to the new model fields so existing clients continue
-    to work unchanged.
-    """
-
-    # v9 field: task_name -> task_type
-    task_name = serializers.CharField(source="task_type", read_only=True)
-
-    # v9 field: task_file_name -> input_data.filename
-    task_file_name = serializers.SerializerMethodField()
-
-    # v9 field: type -> trigger_source (mapped to old enum labels)
-    type = serializers.SerializerMethodField()
-
-    # v9 field: result -> result_message (with legacy format fallback)
-    result = serializers.CharField(
-        source="result_message",
-        read_only=True,
-        allow_null=True,
-    )
-
-    # v9 field: related_document -> first document ID from result_data
-    related_document = serializers.SerializerMethodField()
-
-    # v9 field: duplicate_documents -> list of duplicate IDs from result_data
-    duplicate_documents = serializers.SerializerMethodField()
-
+class TasksViewSerializer(OwnedObjectSerializer):
     class Meta:
         model = PaperlessTask
         fields = (
@@ -2510,59 +2437,59 @@ class TaskSerializerV9(serializers.ModelSerializer):
             "task_id",
             "task_name",
             "task_file_name",
-            "type",
-            "status",
             "date_created",
             "date_done",
+            "type",
+            "status",
             "result",
             "acknowledged",
             "related_document",
             "duplicate_documents",
             "owner",
         )
-        read_only_fields = fields
 
-    def get_task_file_name(self, obj: PaperlessTask) -> str | None:
-        if not obj.input_data:
-            return None
-        return obj.input_data.get("filename")
+    related_document = serializers.SerializerMethodField()
+    duplicate_documents = serializers.SerializerMethodField()
+    created_doc_re = re.compile(r"New document id (\d+) created")
+    duplicate_doc_re = re.compile(r"It is a duplicate of .* \(#(\d+)\)")
 
-    def get_type(self, obj: PaperlessTask) -> str:
-        # Old type values: AUTO_TASK, SCHEDULED_TASK, MANUAL_TASK
-        source_to_old_type = {
-            PaperlessTask.TriggerSource.SCHEDULED: "SCHEDULED_TASK",
-            PaperlessTask.TriggerSource.SYSTEM: "AUTO_TASK",
-        }
-        return source_to_old_type.get(obj.trigger_source, "MANUAL_TASK")
+    def get_related_document(self, obj) -> str | None:
+        result = None
+        re = None
+        if obj.result:
+            match obj.status:
+                case states.SUCCESS:
+                    re = self.created_doc_re
+                case states.FAILURE:
+                    re = (
+                        self.duplicate_doc_re
+                        if "existing document is in the trash" not in obj.result
+                        else None
+                    )
+            if re is not None:
+                try:
+                    result = re.search(obj.result).group(1)
+                except Exception:
+                    pass
 
-    def get_related_document(self, obj: PaperlessTask) -> int | None:
-        ids = obj.related_document_ids
-        return ids[0] if ids else None
+        return result
 
-    def get_duplicate_documents(self, obj: PaperlessTask) -> list[int]:
-        if not obj.result_data:
+    @extend_schema_field(DuplicateDocumentSummarySerializer(many=True))
+    def get_duplicate_documents(self, obj):
+        related_document = self.get_related_document(obj)
+        request = self.context.get("request")
+        user = request.user if request else None
+        document = Document.global_objects.filter(pk=related_document).first()
+        if not related_document or not user or not document:
             return []
-        dup_of = obj.result_data.get("duplicate_of")
-        return [dup_of] if dup_of is not None else []
+        duplicates = _get_viewable_duplicates(document, user)
+        return list(duplicates.values("id", "title", "deleted_at"))
 
 
-class TaskSummarySerializer(serializers.Serializer):
-    task_type = serializers.CharField()
-    total_count = serializers.IntegerField()
-    pending_count = serializers.IntegerField()
-    success_count = serializers.IntegerField()
-    failure_count = serializers.IntegerField()
-    avg_duration_seconds = serializers.FloatField(allow_null=True)
-    avg_wait_time_seconds = serializers.FloatField(allow_null=True)
-    last_run = serializers.DateTimeField(allow_null=True)
-    last_success = serializers.DateTimeField(allow_null=True)
-    last_failure = serializers.DateTimeField(allow_null=True)
-
-
-class RunTaskSerializer(serializers.Serializer):
-    task_type = serializers.ChoiceField(
-        choices=PaperlessTask.TaskType.choices,
-        label="Task Type",
+class RunTaskViewSerializer(serializers.Serializer[dict[str, Any]]):
+    task_name = serializers.ChoiceField(
+        choices=PaperlessTask.TaskName.choices,
+        label="Task Name",
         write_only=True,
     )
 
@@ -2833,7 +2760,7 @@ class BulkEditObjectsSerializer(SerializerWithPerms, SetPermissionsMixin):
         return attrs
 
 
-class WorkflowTriggerSerializer(serializers.ModelSerializer[WorkflowTrigger]):
+class WorkflowTriggerSerializer(serializers.ModelSerializer):
     id = serializers.IntegerField(required=False, allow_null=True)
     sources = fields.MultipleChoiceField(
         choices=WorkflowTrigger.DocumentSourceChoices.choices,
@@ -2943,7 +2870,7 @@ class WorkflowTriggerSerializer(serializers.ModelSerializer[WorkflowTrigger]):
         return super().update(instance, validated_data)
 
 
-class WorkflowActionEmailSerializer(serializers.ModelSerializer[WorkflowActionEmail]):
+class WorkflowActionEmailSerializer(serializers.ModelSerializer):
     id = serializers.IntegerField(allow_null=True, required=False)
 
     class Meta:
@@ -2957,9 +2884,7 @@ class WorkflowActionEmailSerializer(serializers.ModelSerializer[WorkflowActionEm
         ]
 
 
-class WorkflowActionWebhookSerializer(
-    serializers.ModelSerializer[WorkflowActionWebhook],
-):
+class WorkflowActionWebhookSerializer(serializers.ModelSerializer):
     id = serializers.IntegerField(allow_null=True, required=False)
 
     def validate_url(self, url):
@@ -2980,7 +2905,7 @@ class WorkflowActionWebhookSerializer(
         ]
 
 
-class WorkflowActionSerializer(serializers.ModelSerializer[WorkflowAction]):
+class WorkflowActionSerializer(serializers.ModelSerializer):
     id = serializers.IntegerField(required=False, allow_null=True)
     assign_correspondent = CorrespondentField(allow_null=True, required=False)
     assign_tags = TagsField(many=True, allow_null=True, required=False)
@@ -3102,7 +3027,7 @@ class WorkflowActionSerializer(serializers.ModelSerializer[WorkflowAction]):
         return attrs
 
 
-class WorkflowSerializer(serializers.ModelSerializer[Workflow]):
+class WorkflowSerializer(serializers.ModelSerializer):
     order = serializers.IntegerField(required=False)
 
     triggers = WorkflowTriggerSerializer(many=True)

src/documents/signals/handlers.py

@@ -8,6 +8,7 @@ from typing import TYPE_CHECKING
 from typing import Any
 
 from celery import shared_task
+from celery import states
 from celery.signals import before_task_publish
 from celery.signals import task_failure
 from celery.signals import task_postrun
@@ -30,7 +31,6 @@ from documents import matching
 from documents.caching import clear_document_caches
 from documents.caching import invalidate_llm_suggestions_cache
 from documents.data_models import ConsumableDocument
-from documents.data_models import DocumentSource
 from documents.file_handling import create_source_path_directory
 from documents.file_handling import delete_empty_directories
 from documents.file_handling import generate_filename
@@ -999,175 +999,68 @@ def run_workflows(
         return overrides, "\n".join(messages)
 
 
-# ---------------------------------------------------------------------------
-# Task tracking -- Celery signal handlers
-# ---------------------------------------------------------------------------
-
-TRACKED_TASKS: dict[str, PaperlessTask.TaskType] = {
-    "documents.tasks.consume_file": PaperlessTask.TaskType.CONSUME_FILE,
-    "documents.tasks.train_classifier": PaperlessTask.TaskType.TRAIN_CLASSIFIER,
-    "documents.tasks.sanity_check": PaperlessTask.TaskType.SANITY_CHECK,
-    "documents.tasks.index_optimize": PaperlessTask.TaskType.INDEX_OPTIMIZE,
-    "documents.tasks.llmindex_index": PaperlessTask.TaskType.LLM_INDEX,
-    "paperless_mail.tasks.process_mail_accounts": PaperlessTask.TaskType.MAIL_FETCH,
-}
-
-_DOCUMENT_SOURCE_TO_TRIGGER: dict[Any, PaperlessTask.TriggerSource] = {
-    DocumentSource.ConsumeFolder: PaperlessTask.TriggerSource.FOLDER_CONSUME,
-    DocumentSource.ApiUpload: PaperlessTask.TriggerSource.API_UPLOAD,
-    DocumentSource.MailFetch: PaperlessTask.TriggerSource.EMAIL_CONSUME,
-    DocumentSource.WebUI: PaperlessTask.TriggerSource.WEB_UI,
-}
-
-
-def _extract_input_data(
-    task_type: PaperlessTask.TaskType,
-    args: tuple,
-    task_kwargs: dict,
-) -> dict:
-    if task_type == PaperlessTask.TaskType.CONSUME_FILE:
-        input_doc = args[0] if args else task_kwargs.get("input_doc")
-        overrides = args[1] if len(args) >= 2 else task_kwargs.get("overrides")
-        if input_doc is None:
-            return {}
-        data: dict = {
-            "filename": input_doc.original_file.name,
-            "mime_type": input_doc.mime_type,
-        }
-        if input_doc.original_path:
-            data["source_path"] = str(input_doc.original_path)
-        if input_doc.mailrule_id:
-            data["mailrule_id"] = input_doc.mailrule_id
-        if overrides:
-            override_dict = {
-                k: v
-                for k, v in vars(overrides).items()
-                if v is not None and not k.startswith("_")
-            }
-            if override_dict:
-                data["overrides"] = override_dict
-        return data
-
-    if task_type == PaperlessTask.TaskType.MAIL_FETCH:
-        account_ids = args[0] if args else task_kwargs.get("account_ids")
-        return {"account_ids": account_ids}
-
-    return {}
-
-
-def _determine_trigger_source(
-    task_type: PaperlessTask.TaskType,
-    args: tuple,
-    task_kwargs: dict,
-    headers: dict,
-) -> PaperlessTask.TriggerSource:
-    # Explicit header takes priority -- covers beat ("scheduled") and system auto-runs ("system")
-    header_source = headers.get("trigger_source")
-    if header_source == "scheduled":
-        return PaperlessTask.TriggerSource.SCHEDULED
-    if header_source == "system":
-        return PaperlessTask.TriggerSource.SYSTEM
-
-    if task_type == PaperlessTask.TaskType.CONSUME_FILE:
-        input_doc = args[0] if args else task_kwargs.get("input_doc")
-        if input_doc is not None:
-            return _DOCUMENT_SOURCE_TO_TRIGGER.get(
-                input_doc.source,
-                PaperlessTask.TriggerSource.API_UPLOAD,
-            )
-
-    return PaperlessTask.TriggerSource.MANUAL
-
-
-def _extract_owner_id(
-    task_type: PaperlessTask.TaskType,
-    args: tuple,
-    task_kwargs: dict,
-) -> int | None:
-    if task_type != PaperlessTask.TaskType.CONSUME_FILE:
-        return None
-    overrides = args[1] if len(args) >= 2 else task_kwargs.get("overrides")
-    if overrides and hasattr(overrides, "owner_id"):
-        return overrides.owner_id
-    return None
-
-
-def _parse_legacy_result(result: str) -> dict | None:
-    import re as _re
-
-    if match := _re.search(r"New document id (\d+) created", result):
-        return {"document_id": int(match.group(1))}
-    if match := _re.search(r"It is a duplicate of .* \(#(\d+)\)", result):
-        return {
-            "duplicate_of": int(match.group(1)),
-            "duplicate_in_trash": "existing document is in the trash" in result,
-        }
-    return None
-
-
 @before_task_publish.connect
-def before_task_publish_handler(
-    sender=None,
-    headers=None,
-    body=None,
-    **kwargs,
-) -> None:
+def before_task_publish_handler(sender=None, headers=None, body=None, **kwargs) -> None:
     """
-    Creates the PaperlessTask record when the task is published to broker.
+    Creates the PaperlessTask object in a pending state.  This is sent before
+    the task reaches the broker, but before it begins executing on a worker.
 
     https://docs.celeryq.dev/en/stable/userguide/signals.html#before-task-publish
-    https://docs.celeryq.dev/en/stable/internals/protocol.html#version-2
-    """
-    if headers is None or body is None:
-        return
 
-    task_name = headers.get("task", "")
-    task_type = TRACKED_TASKS.get(task_name)
-    if task_type is None:
+    https://docs.celeryq.dev/en/stable/internals/protocol.html#version-2
+
+    """
+    if "task" not in headers or headers["task"] != "documents.tasks.consume_file":
+        # Assumption: this is only ever a v2 message
         return
 
     try:
         close_old_connections()
-        args, task_kwargs, _ = body
-        task_id = headers["id"]
 
-        input_data = _extract_input_data(task_type, args, task_kwargs)
-        trigger_source = _determine_trigger_source(
-            task_type,
-            args,
-            task_kwargs,
-            headers,
-        )
-        owner_id = _extract_owner_id(task_type, args, task_kwargs)
+        task_args = body[0]
+        input_doc, overrides = task_args
+
+        task_file_name = input_doc.original_file.name
+        user_id = overrides.owner_id if overrides else None
 
         PaperlessTask.objects.create(
-            task_id=task_id,
-            task_type=task_type,
-            trigger_source=trigger_source,
-            status=PaperlessTask.Status.PENDING,
-            input_data=input_data,
-            owner_id=owner_id,
+            type=PaperlessTask.TaskType.AUTO,
+            task_id=headers["id"],
+            status=states.PENDING,
+            task_file_name=task_file_name,
+            task_name=PaperlessTask.TaskName.CONSUME_FILE,
+            result=None,
+            date_created=timezone.now(),
+            date_started=None,
+            date_done=None,
+            owner_id=user_id,
         )
-    except Exception:
+    except Exception:  # pragma: no cover
+        # Don't let an exception in the signal handlers prevent
+        # a document from being consumed.
         logger.exception("Creating PaperlessTask failed")
 
 
 @task_prerun.connect
 def task_prerun_handler(sender=None, task_id=None, task=None, **kwargs) -> None:
     """
-    Marks the task STARTED when execution begins on a worker.
+
+    Updates the PaperlessTask to be started.  Sent before the task begins execution
+    on a worker.
 
     https://docs.celeryq.dev/en/stable/userguide/signals.html#task-prerun
     """
-    if task_id is None:
-        return
     try:
         close_old_connections()
-        PaperlessTask.objects.filter(task_id=task_id).update(
-            status=PaperlessTask.Status.STARTED,
-            date_started=timezone.now(),
-        )
-    except Exception:
+        task_instance = PaperlessTask.objects.filter(task_id=task_id).first()
+
+        if task_instance is not None:
+            task_instance.status = states.STARTED
+            task_instance.date_started = timezone.now()
+            task_instance.save()
+    except Exception:  # pragma: no cover
+        # Don't let an exception in the signal handlers prevent
+        # a document from being consumed.
         logger.exception("Setting PaperlessTask started failed")
 
 
@@ -1181,53 +1074,22 @@ def task_postrun_handler(
     **kwargs,
 ) -> None:
     """
-    Records task completion and result data.
+    Updates the result of the PaperlessTask.
 
     https://docs.celeryq.dev/en/stable/userguide/signals.html#task-postrun
     """
-    if task_id is None:
-        return
     try:
         close_old_connections()
-
-        status_map = {
-            "SUCCESS": PaperlessTask.Status.SUCCESS,
-            "FAILURE": PaperlessTask.Status.FAILURE,
-            "REVOKED": PaperlessTask.Status.REVOKED,
-        }
-        new_status = status_map.get(state, PaperlessTask.Status.FAILURE)
-
-        result_data: dict | None = None
-        result_message: str | None = None
-        if isinstance(retval, dict):
-            result_data = retval
-        elif isinstance(retval, str):
-            result_message = retval
-            result_data = _parse_legacy_result(retval)
-
-        now = timezone.now()
         task_instance = PaperlessTask.objects.filter(task_id=task_id).first()
-        if task_instance is None:
-            return
 
-        duration_seconds: float | None = None
-        wait_time_seconds: float | None = None
-        if task_instance.date_started:
-            duration_seconds = (now - task_instance.date_started).total_seconds()
-        if task_instance.date_started and task_instance.date_created:
-            wait_time_seconds = (
-                task_instance.date_started - task_instance.date_created
-            ).total_seconds()
-
-        PaperlessTask.objects.filter(task_id=task_id).update(
-            status=new_status,
-            result_data=result_data,
-            result_message=result_message,
-            date_done=now,
-            duration_seconds=duration_seconds,
-            wait_time_seconds=wait_time_seconds,
-        )
-    except Exception:
+        if task_instance is not None:
+            task_instance.status = state or states.FAILURE
+            task_instance.result = retval
+            task_instance.date_done = timezone.now()
+            task_instance.save()
+    except Exception:  # pragma: no cover
+        # Don't let an exception in the signal handlers prevent
+        # a document from being consumed.
         logger.exception("Updating PaperlessTask failed")
 
 
@@ -1241,33 +1103,21 @@ def task_failure_handler(
     **kwargs,
 ) -> None:
     """
-    Records failure details when a task raises an exception.
+    Updates the result of a failed PaperlessTask.
 
     https://docs.celeryq.dev/en/stable/userguide/signals.html#task-failure
     """
-    if task_id is None:
-        return
     try:
         close_old_connections()
+        task_instance = PaperlessTask.objects.filter(task_id=task_id).first()
 
-        result_data: dict = {
-            "error_type": type(exception).__name__ if exception else "Unknown",
-            "error_message": str(exception) if exception else "Unknown error",
-        }
-        if traceback:
-            import traceback as _tb
-
-            tb_str = "".join(_tb.format_tb(traceback))
-            result_data["traceback"] = tb_str[:5000]
-
-        PaperlessTask.objects.filter(task_id=task_id).update(
-            status=PaperlessTask.Status.FAILURE,
-            result_data=result_data,
-            result_message=str(exception) if exception else None,
-            date_done=timezone.now(),
-        )
-    except Exception:
-        logger.exception("Updating PaperlessTask on failure failed")
+        if task_instance is not None and task_instance.result is None:
+            task_instance.status = states.FAILURE
+            task_instance.result = traceback
+            task_instance.date_done = timezone.now()
+            task_instance.save()
+    except Exception:  # pragma: no cover
+        logger.exception("Updating PaperlessTask failed")
 
 
 @worker_process_init.connect

src/documents/tasks.py

@@ -10,6 +10,7 @@ from tempfile import mkstemp
 
 from celery import Task
 from celery import shared_task
+from celery import states
 from django.conf import settings
 from django.contrib.contenttypes.models import ContentType
 from django.db import models
@@ -40,6 +41,7 @@ from documents.models import Correspondent
 from documents.models import CustomFieldInstance
 from documents.models import Document
 from documents.models import DocumentType
+from documents.models import PaperlessTask
 from documents.models import ShareLink
 from documents.models import ShareLinkBundle
 from documents.models import StoragePath
@@ -82,8 +84,19 @@ def index_optimize() -> None:
 @shared_task
 def train_classifier(
     *,
+    scheduled=True,
     status_callback: Callable[[str], None] | None = None,
-) -> str:
+) -> None:
+    task = PaperlessTask.objects.create(
+        type=PaperlessTask.TaskType.SCHEDULED_TASK
+        if scheduled
+        else PaperlessTask.TaskType.MANUAL_TASK,
+        task_id=uuid.uuid4(),
+        task_name=PaperlessTask.TaskName.TRAIN_CLASSIFIER,
+        status=states.STARTED,
+        date_created=timezone.now(),
+        date_started=timezone.now(),
+    )
     if (
         not Tag.objects.filter(matching_algorithm=Tag.MATCH_AUTO).exists()
         and not DocumentType.objects.filter(matching_algorithm=Tag.MATCH_AUTO).exists()
@@ -97,22 +110,37 @@ def train_classifier(
         if settings.MODEL_FILE.exists():
             logger.info(f"Removing {settings.MODEL_FILE} so it won't be used")
             settings.MODEL_FILE.unlink()
-        return result
+        task.status = states.SUCCESS
+        task.result = result
+        task.date_done = timezone.now()
+        task.save()
+        return
 
     classifier = load_classifier()
 
     if not classifier:
         classifier = DocumentClassifier()
 
-    if classifier.train(status_callback=status_callback):
-        logger.info(
-            f"Saving updated classifier model to {settings.MODEL_FILE}...",
-        )
-        classifier.save()
-        return "Training completed successfully"
-    else:
-        logger.debug("Training data unchanged.")
-        return "Training data unchanged"
+    try:
+        if classifier.train(status_callback=status_callback):
+            logger.info(
+                f"Saving updated classifier model to {settings.MODEL_FILE}...",
+            )
+            classifier.save()
+            task.result = "Training completed successfully"
+        else:
+            logger.debug("Training data unchanged.")
+            task.result = "Training data unchanged"
+
+        task.status = states.SUCCESS
+
+    except Exception as e:
+        logger.warning("Classifier error: " + str(e))
+        task.status = states.FAILURE
+        task.result = str(e)
+
+    task.date_done = timezone.now()
+    task.save(update_fields=["status", "result", "date_done"])
 
 
 @shared_task(bind=True)
@@ -203,8 +231,8 @@ def consume_file(
 
 
 @shared_task
-def sanity_check(*, raise_on_error: bool = True) -> str:
-    messages = sanity_checker.check_sanity()
+def sanity_check(*, scheduled=True, raise_on_error=True):
+    messages = sanity_checker.check_sanity(scheduled=scheduled)
     messages.log_messages()
 
     if not messages.has_error and not messages.has_warning and not messages.has_info:
@@ -607,19 +635,42 @@ def update_document_parent_tags(tag: Tag, new_parent: Tag) -> None:
 def llmindex_index(
     *,
     iter_wrapper: IterWrapper[Document] = identity,
-    rebuild: bool = False,
-) -> str | None:
+    rebuild=False,
+    scheduled=True,
+    auto=False,
+) -> None:
     ai_config = AIConfig()
-    if not ai_config.llm_index_enabled:
+    if ai_config.llm_index_enabled:
+        task = PaperlessTask.objects.create(
+            type=PaperlessTask.TaskType.SCHEDULED_TASK
+            if scheduled
+            else PaperlessTask.TaskType.AUTO
+            if auto
+            else PaperlessTask.TaskType.MANUAL_TASK,
+            task_id=uuid.uuid4(),
+            task_name=PaperlessTask.TaskName.LLMINDEX_UPDATE,
+            status=states.STARTED,
+            date_created=timezone.now(),
+            date_started=timezone.now(),
+        )
+        from paperless_ai.indexing import update_llm_index
+
+        try:
+            result = update_llm_index(
+                iter_wrapper=iter_wrapper,
+                rebuild=rebuild,
+            )
+            task.status = states.SUCCESS
+            task.result = result
+        except Exception as e:
+            logger.error("LLM index error: " + str(e))
+            task.status = states.FAILURE
+            task.result = str(e)
+
+        task.date_done = timezone.now()
+        task.save(update_fields=["status", "result", "date_done"])
+    else:
         logger.info("LLM index is disabled, skipping update.")
-        return None
-
-    from paperless_ai.indexing import update_llm_index
-
-    return update_llm_index(
-        iter_wrapper=iter_wrapper,
-        rebuild=rebuild,
-    )
 
 
 @shared_task

src/documents/tests/conftest.py

@@ -13,8 +13,6 @@ from rest_framework.test import APIClient
 
 from documents.tests.factories import DocumentFactory
 
-UserModelT = get_user_model()
-
 if TYPE_CHECKING:
     from documents.models import Document
 
@@ -128,34 +126,15 @@ def rest_api_client():
     yield APIClient()
 
 
-@pytest.fixture()
-def regular_user(django_user_model: type[UserModelT]) -> UserModelT:
-    """Unprivileged authenticated user for permission boundary tests."""
-    return django_user_model.objects.create_user(username="regular", password="regular")
-
-
-@pytest.fixture()
-def admin_client(rest_api_client: APIClient, admin_user: UserModelT) -> APIClient:
-    """Admin client pre-authenticated and sending the v10 Accept header."""
-    rest_api_client.force_authenticate(user=admin_user)
-    rest_api_client.credentials(HTTP_ACCEPT="application/json; version=10")
-    return rest_api_client
-
-
-@pytest.fixture()
-def v9_client(rest_api_client: APIClient, admin_user: UserModelT) -> APIClient:
-    """Admin client pre-authenticated and sending the v9 Accept header."""
-    rest_api_client.force_authenticate(user=admin_user)
-    rest_api_client.credentials(HTTP_ACCEPT="application/json; version=9")
-    return rest_api_client
-
-
-@pytest.fixture()
-def user_client(rest_api_client: APIClient, regular_user: UserModelT) -> APIClient:
-    """Regular-user client pre-authenticated and sending the v10 Accept header."""
-    rest_api_client.force_authenticate(user=regular_user)
-    rest_api_client.credentials(HTTP_ACCEPT="application/json; version=10")
-    return rest_api_client
+@pytest.fixture
+def authenticated_rest_api_client(rest_api_client: APIClient):
+    """
+    The basic DRF ApiClient which has been authenticated
+    """
+    UserModel = get_user_model()
+    user = UserModel.objects.create_user(username="testuser", password="password")
+    rest_api_client.force_authenticate(user=user)
+    yield rest_api_client
 
 
 @pytest.fixture(scope="session", autouse=True)

src/documents/tests/factories.py

@@ -11,7 +11,6 @@ from documents.models import Correspondent
 from documents.models import Document
 from documents.models import DocumentType
 from documents.models import MatchingModel
-from documents.models import PaperlessTask
 from documents.models import StoragePath
 from documents.models import Tag
 
@@ -66,17 +65,3 @@ class DocumentFactory(DjangoModelFactory):
     correspondent = None
     document_type = None
     storage_path = None
-
-
-class PaperlessTaskFactory(DjangoModelFactory):
-    class Meta:
-        model = PaperlessTask
-
-    task_id = factory.LazyFunction(lambda: str(__import__("uuid").uuid4()))
-    task_type = PaperlessTask.TaskType.CONSUME_FILE
-    trigger_source = PaperlessTask.TriggerSource.WEB_UI
-    status = PaperlessTask.Status.PENDING
-    input_data = factory.LazyFunction(dict)
-    result_data = None
-    result_message = None
-    acknowledged = False

src/documents/tests/test_api_app_config.py

@@ -831,7 +831,7 @@ class TestApiAppConfig(DirectoriesMixin, APITestCase):
         config.save()
 
         with (
-            patch("documents.tasks.llmindex_index.apply_async") as mock_update,
+            patch("documents.tasks.llmindex_index.delay") as mock_update,
             patch("paperless_ai.indexing.vector_store_file_exists") as mock_exists,
         ):
             mock_exists.return_value = False

src/documents/tests/test_api_status.py

@@ -4,6 +4,7 @@ import tempfile
 from pathlib import Path
 from unittest import mock
 
+from celery import states
 from django.contrib.auth.models import Permission
 from django.contrib.auth.models import User
 from django.test import override_settings
@@ -12,7 +13,6 @@ from rest_framework.test import APITestCase
 
 from documents.models import PaperlessTask
 from documents.permissions import has_system_status_permission
-from documents.tests.factories import PaperlessTaskFactory
 from paperless import version
 
 
@@ -258,10 +258,10 @@ class TestSystemStatus(APITestCase):
         THEN:
             - The response contains an OK classifier status
         """
-        PaperlessTaskFactory(
-            task_type=PaperlessTask.TaskType.TRAIN_CLASSIFIER,
-            trigger_source=PaperlessTask.TriggerSource.SCHEDULED,
-            status=PaperlessTask.Status.SUCCESS,
+        PaperlessTask.objects.create(
+            type=PaperlessTask.TaskType.SCHEDULED_TASK,
+            status=states.SUCCESS,
+            task_name=PaperlessTask.TaskName.TRAIN_CLASSIFIER,
         )
         self.client.force_login(self.user)
         response = self.client.get(self.ENDPOINT)
@@ -295,11 +295,11 @@ class TestSystemStatus(APITestCase):
         THEN:
             - The response contains an ERROR classifier status
         """
-        PaperlessTaskFactory(
-            task_type=PaperlessTask.TaskType.TRAIN_CLASSIFIER,
-            trigger_source=PaperlessTask.TriggerSource.SCHEDULED,
-            status=PaperlessTask.Status.FAILURE,
-            result_message="Classifier training failed",
+        PaperlessTask.objects.create(
+            type=PaperlessTask.TaskType.SCHEDULED_TASK,
+            status=states.FAILURE,
+            task_name=PaperlessTask.TaskName.TRAIN_CLASSIFIER,
+            result="Classifier training failed",
         )
         self.client.force_login(self.user)
         response = self.client.get(self.ENDPOINT)
@@ -319,10 +319,10 @@ class TestSystemStatus(APITestCase):
         THEN:
             - The response contains an OK sanity check status
         """
-        PaperlessTaskFactory(
-            task_type=PaperlessTask.TaskType.SANITY_CHECK,
-            trigger_source=PaperlessTask.TriggerSource.SCHEDULED,
-            status=PaperlessTask.Status.SUCCESS,
+        PaperlessTask.objects.create(
+            type=PaperlessTask.TaskType.SCHEDULED_TASK,
+            status=states.SUCCESS,
+            task_name=PaperlessTask.TaskName.CHECK_SANITY,
         )
         self.client.force_login(self.user)
         response = self.client.get(self.ENDPOINT)
@@ -356,11 +356,11 @@ class TestSystemStatus(APITestCase):
         THEN:
             - The response contains an ERROR sanity check status
         """
-        PaperlessTaskFactory(
-            task_type=PaperlessTask.TaskType.SANITY_CHECK,
-            trigger_source=PaperlessTask.TriggerSource.SCHEDULED,
-            status=PaperlessTask.Status.FAILURE,
-            result_message="5 issues found.",
+        PaperlessTask.objects.create(
+            type=PaperlessTask.TaskType.SCHEDULED_TASK,
+            status=states.FAILURE,
+            task_name=PaperlessTask.TaskName.CHECK_SANITY,
+            result="5 issues found.",
         )
         self.client.force_login(self.user)
         response = self.client.get(self.ENDPOINT)
@@ -405,10 +405,10 @@ class TestSystemStatus(APITestCase):
             self.assertEqual(response.status_code, status.HTTP_200_OK)
             self.assertEqual(response.data["tasks"]["llmindex_status"], "WARNING")
 
-            PaperlessTaskFactory(
-                task_type=PaperlessTask.TaskType.LLM_INDEX,
-                trigger_source=PaperlessTask.TriggerSource.SCHEDULED,
-                status=PaperlessTask.Status.SUCCESS,
+            PaperlessTask.objects.create(
+                type=PaperlessTask.TaskType.SCHEDULED_TASK,
+                status=states.SUCCESS,
+                task_name=PaperlessTask.TaskName.LLMINDEX_UPDATE,
             )
             response = self.client.get(self.ENDPOINT)
             self.assertEqual(response.status_code, status.HTTP_200_OK)
@@ -425,11 +425,11 @@ class TestSystemStatus(APITestCase):
             - The response contains the correct AI status
         """
         with override_settings(AI_ENABLED=True, LLM_EMBEDDING_BACKEND="openai"):
-            PaperlessTaskFactory(
-                task_type=PaperlessTask.TaskType.LLM_INDEX,
-                trigger_source=PaperlessTask.TriggerSource.SCHEDULED,
-                status=PaperlessTask.Status.FAILURE,
-                result_message="AI index update failed",
+            PaperlessTask.objects.create(
+                type=PaperlessTask.TaskType.SCHEDULED_TASK,
+                status=states.FAILURE,
+                task_name=PaperlessTask.TaskName.LLMINDEX_UPDATE,
+                result="AI index update failed",
             )
             self.client.force_login(self.user)
             response = self.client.get(self.ENDPOINT)

src/documents/tests/test_api_tasks.py

@@ -1,586 +1,425 @@
-"""Tests for the /api/tasks/ endpoint.
-
-Covers:
-- v10 serializer (new field names)
-- v9 serializer (backwards-compatible field names)
-- Filtering, ordering, acknowledge, acknowledge_all, summary, active, run
-"""
-
 import uuid
-from datetime import timedelta
 from unittest import mock
 
-import pytest
+import celery
 from django.contrib.auth.models import Permission
 from django.contrib.auth.models import User
-from django.utils import timezone
 from rest_framework import status
-from rest_framework.test import APIClient
+from rest_framework.test import APITestCase
 
+from documents.models import Document
 from documents.models import PaperlessTask
-from documents.tests.factories import PaperlessTaskFactory
-
-pytestmark = pytest.mark.api
-
-ENDPOINT = "/api/tasks/"
-ACCEPT_V10 = "application/json; version=10"
-ACCEPT_V9 = "application/json; version=9"
+from documents.tests.utils import DirectoriesMixin
+from documents.views import TasksViewSet
 
 
-@pytest.mark.django_db()
-class TestGetTasksV10:
-    def test_list_returns_tasks(self, admin_client: APIClient) -> None:
-        """GET /api/tasks/ returns all tasks visible to the admin."""
-        PaperlessTaskFactory.create_batch(2)
+class TestTasks(DirectoriesMixin, APITestCase):
+    ENDPOINT = "/api/tasks/"
 
-        response = admin_client.get(ENDPOINT)
+    def setUp(self) -> None:
+        super().setUp()
 
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 2
+        self.user = User.objects.create_superuser(username="temp_admin")
+        self.client.force_authenticate(user=self.user)
 
-    def test_related_document_ids_populated_from_result_data(
-        self,
-        admin_client: APIClient,
-    ) -> None:
-        """related_document_ids includes the consumed document_id from result_data."""
-        PaperlessTaskFactory(
-            status=PaperlessTask.Status.SUCCESS,
-            result_data={"document_id": 7},
+    def test_get_tasks(self) -> None:
+        """
+        GIVEN:
+            - Attempted celery tasks
+        WHEN:
+            - API call is made to get tasks
+        THEN:
+            - Attempting and pending tasks are serialized and provided
+        """
+
+        task1 = PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_one.pdf",
         )
 
-        response = admin_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["related_document_ids"] == [7]
-
-    def test_related_document_ids_includes_duplicate_of(
-        self,
-        admin_client: APIClient,
-    ) -> None:
-        """related_document_ids includes duplicate_of when the file was already archived."""
-        PaperlessTaskFactory(
-            status=PaperlessTask.Status.SUCCESS,
-            result_data={"duplicate_of": 12},
+        task2 = PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_two.pdf",
         )
 
-        response = admin_client.get(ENDPOINT)
+        response = self.client.get(self.ENDPOINT)
 
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["related_document_ids"] == [12]
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(len(response.data), 2)
+        returned_task1 = response.data[1]
+        returned_task2 = response.data[0]
 
-    def test_filter_by_task_type(self, admin_client: APIClient) -> None:
-        """?task_type= filters results to tasks of that type only."""
-        PaperlessTaskFactory(task_type=PaperlessTask.TaskType.CONSUME_FILE)
-        PaperlessTaskFactory(task_type=PaperlessTask.TaskType.TRAIN_CLASSIFIER)
+        self.assertEqual(returned_task1["task_id"], task1.task_id)
+        self.assertEqual(returned_task1["status"], celery.states.PENDING)
+        self.assertEqual(returned_task1["task_file_name"], task1.task_file_name)
 
-        response = admin_client.get(
-            ENDPOINT,
-            {"task_type": PaperlessTask.TaskType.TRAIN_CLASSIFIER},
+        self.assertEqual(returned_task2["task_id"], task2.task_id)
+        self.assertEqual(returned_task2["status"], celery.states.PENDING)
+        self.assertEqual(returned_task2["task_file_name"], task2.task_file_name)
+
+    def test_get_single_task_status(self) -> None:
+        """
+        GIVEN
+            - Query parameter for a valid task ID
+        WHEN:
+            - API call is made to get task status
+        THEN:
+            - Single task data is returned
+        """
+
+        id1 = str(uuid.uuid4())
+        task1 = PaperlessTask.objects.create(
+            task_id=id1,
+            task_file_name="task_one.pdf",
         )
 
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 1
-        assert response.data[0]["task_type"] == PaperlessTask.TaskType.TRAIN_CLASSIFIER
-
-    def test_filter_by_status(self, admin_client: APIClient) -> None:
-        """?status= filters results to tasks with that status only."""
-        PaperlessTaskFactory(status=PaperlessTask.Status.PENDING)
-        PaperlessTaskFactory(status=PaperlessTask.Status.SUCCESS)
-
-        response = admin_client.get(
-            ENDPOINT,
-            {"status": PaperlessTask.Status.SUCCESS},
+        _ = PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_two.pdf",
         )
 
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 1
-        assert response.data[0]["status"] == PaperlessTask.Status.SUCCESS
+        response = self.client.get(self.ENDPOINT + f"?task_id={id1}")
 
-    def test_filter_by_task_id(self, admin_client: APIClient) -> None:
-        """?task_id= returns only the task with that UUID."""
-        task = PaperlessTaskFactory()
-        PaperlessTaskFactory()  # unrelated task that should not appear
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(len(response.data), 1)
+        returned_task1 = response.data[0]
 
-        response = admin_client.get(ENDPOINT, {"task_id": task.task_id})
+        self.assertEqual(returned_task1["task_id"], task1.task_id)
 
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 1
-        assert response.data[0]["task_id"] == task.task_id
-
-    def test_filter_by_acknowledged(self, admin_client: APIClient) -> None:
-        """?acknowledged=false returns only tasks that have not been acknowledged."""
-        PaperlessTaskFactory(acknowledged=False)
-        PaperlessTaskFactory(acknowledged=True)
-
-        response = admin_client.get(ENDPOINT, {"acknowledged": "false"})
-
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 1
-        assert response.data[0]["acknowledged"] is False
-
-    def test_filter_is_complete_true(self, admin_client: APIClient) -> None:
-        """?is_complete=true returns only SUCCESS and FAILURE tasks."""
-        PaperlessTaskFactory(status=PaperlessTask.Status.PENDING)
-        PaperlessTaskFactory(status=PaperlessTask.Status.SUCCESS)
-        PaperlessTaskFactory(status=PaperlessTask.Status.FAILURE)
-
-        response = admin_client.get(ENDPOINT, {"is_complete": "true"})
-
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 2
-        returned_statuses = {t["status"] for t in response.data}
-        assert returned_statuses == {
-            PaperlessTask.Status.SUCCESS,
-            PaperlessTask.Status.FAILURE,
-        }
-
-    def test_filter_is_complete_false(self, admin_client: APIClient) -> None:
-        """?is_complete=false returns only PENDING and STARTED tasks."""
-        PaperlessTaskFactory(status=PaperlessTask.Status.PENDING)
-        PaperlessTaskFactory(status=PaperlessTask.Status.STARTED)
-        PaperlessTaskFactory(status=PaperlessTask.Status.SUCCESS)
-
-        response = admin_client.get(ENDPOINT, {"is_complete": "false"})
-
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 2
-        returned_statuses = {t["status"] for t in response.data}
-        assert returned_statuses == {
-            PaperlessTask.Status.PENDING,
-            PaperlessTask.Status.STARTED,
-        }
-
-    def test_default_ordering_is_newest_first(self, admin_client: APIClient) -> None:
-        """Tasks are returned in descending date_created order (newest first)."""
-        base = timezone.now()
-        t1 = PaperlessTaskFactory(date_created=base)
-        t2 = PaperlessTaskFactory(date_created=base + timedelta(seconds=1))
-        t3 = PaperlessTaskFactory(date_created=base + timedelta(seconds=2))
-
-        response = admin_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        ids = [t["task_id"] for t in response.data]
-        assert ids == [t3.task_id, t2.task_id, t1.task_id]
-
-    def test_list_is_owner_aware(
-        self,
-        admin_user: User,
-        regular_user: User,
-    ) -> None:
-        """The task list only shows tasks the user owns or that are unowned."""
-        regular_user.user_permissions.add(
-            Permission.objects.get(codename="view_paperlesstask"),
+    def test_get_single_task_status_not_valid(self) -> None:
+        """
+        GIVEN
+            - Query parameter for a non-existent task ID
+        WHEN:
+            - API call is made to get task status
+        THEN:
+            - No task data is returned
+        """
+        PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_one.pdf",
         )
 
-        client = APIClient()
-        client.force_authenticate(user=regular_user)
-        client.credentials(HTTP_ACCEPT=ACCEPT_V10)
-
-        PaperlessTaskFactory(owner=admin_user)
-        shared_task = PaperlessTaskFactory()
-        own_task = PaperlessTaskFactory(owner=regular_user)
-
-        response = client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 2
-        returned_task_ids = {t["task_id"] for t in response.data}
-        assert shared_task.task_id in returned_task_ids
-        assert own_task.task_id in returned_task_ids
-
-
-@pytest.mark.django_db()
-class TestGetTasksV9:
-    def test_task_name_equals_task_type_value(self, v9_client: APIClient) -> None:
-        """task_name mirrors the task_type value for v9 backwards compatibility."""
-        PaperlessTaskFactory(task_type=PaperlessTask.TaskType.CONSUME_FILE)
-
-        response = v9_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["task_name"] == "consume_file"
-
-    def test_task_file_name_from_input_data(self, v9_client: APIClient) -> None:
-        """task_file_name is read from input_data['filename']."""
-        PaperlessTaskFactory(input_data={"filename": "report.pdf"})
-
-        response = v9_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["task_file_name"] == "report.pdf"
-
-    def test_task_file_name_none_when_no_filename_key(
-        self,
-        v9_client: APIClient,
-    ) -> None:
-        """task_file_name is None when filename is absent from input_data."""
-        PaperlessTaskFactory(input_data={})
-
-        response = v9_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["task_file_name"] is None
-
-    def test_type_scheduled_maps_to_scheduled_task(self, v9_client: APIClient) -> None:
-        """trigger_source=scheduled maps to type='SCHEDULED_TASK' in v9."""
-        PaperlessTaskFactory(trigger_source=PaperlessTask.TriggerSource.SCHEDULED)
-
-        response = v9_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["type"] == "SCHEDULED_TASK"
-
-    def test_type_system_maps_to_auto_task(self, v9_client: APIClient) -> None:
-        """trigger_source=system maps to type='AUTO_TASK' in v9."""
-        PaperlessTaskFactory(trigger_source=PaperlessTask.TriggerSource.SYSTEM)
-
-        response = v9_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["type"] == "AUTO_TASK"
-
-    def test_type_web_ui_maps_to_manual_task(self, v9_client: APIClient) -> None:
-        """trigger_source=web_ui maps to type='MANUAL_TASK' in v9."""
-        PaperlessTaskFactory(trigger_source=PaperlessTask.TriggerSource.WEB_UI)
-
-        response = v9_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["type"] == "MANUAL_TASK"
-
-    def test_type_manual_maps_to_manual_task(self, v9_client: APIClient) -> None:
-        """trigger_source=manual maps to type='MANUAL_TASK' in v9."""
-        PaperlessTaskFactory(trigger_source=PaperlessTask.TriggerSource.MANUAL)
-
-        response = v9_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["type"] == "MANUAL_TASK"
-
-    def test_related_document_from_result_data_document_id(
-        self,
-        v9_client: APIClient,
-    ) -> None:
-        """related_document is taken from result_data['document_id'] in v9."""
-        PaperlessTaskFactory(
-            status=PaperlessTask.Status.SUCCESS,
-            result_data={"document_id": 99},
+        _ = PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_two.pdf",
         )
 
-        response = v9_client.get(ENDPOINT)
+        response = self.client.get(self.ENDPOINT + "?task_id=bad-task-id")
 
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["related_document"] == 99
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(len(response.data), 0)
 
-    def test_related_document_none_when_no_result_data(
-        self,
-        v9_client: APIClient,
-    ) -> None:
-        """related_document is None when result_data is absent in v9."""
-        PaperlessTaskFactory(result_data=None)
-
-        response = v9_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["related_document"] is None
-
-    def test_duplicate_documents_from_result_data(self, v9_client: APIClient) -> None:
-        """duplicate_documents includes duplicate_of from result_data in v9."""
-        PaperlessTaskFactory(
-            status=PaperlessTask.Status.SUCCESS,
-            result_data={"duplicate_of": 55},
+    def test_acknowledge_tasks(self) -> None:
+        """
+        GIVEN:
+            - Attempted celery tasks
+        WHEN:
+            - API call is made to get mark task as acknowledged
+        THEN:
+            - Task is marked as acknowledged
+        """
+        task = PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_one.pdf",
         )
 
-        response = v9_client.get(ENDPOINT)
+        response = self.client.get(self.ENDPOINT)
+        self.assertEqual(len(response.data), 1)
 
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["duplicate_documents"] == [55]
-
-    def test_duplicate_documents_empty_when_no_result_data(
-        self,
-        v9_client: APIClient,
-    ) -> None:
-        """duplicate_documents is an empty list when result_data is absent in v9."""
-        PaperlessTaskFactory(result_data=None)
-
-        response = v9_client.get(ENDPOINT)
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data[0]["duplicate_documents"] == []
-
-    def test_filter_by_task_name_maps_to_task_type(self, v9_client: APIClient) -> None:
-        """?task_name=consume_file filter maps to the task_type field for v9 compatibility."""
-        PaperlessTaskFactory(task_type=PaperlessTask.TaskType.CONSUME_FILE)
-        PaperlessTaskFactory(task_type=PaperlessTask.TaskType.TRAIN_CLASSIFIER)
-
-        response = v9_client.get(ENDPOINT, {"task_name": "consume_file"})
-
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 1
-        assert response.data[0]["task_name"] == "consume_file"
-
-    def test_filter_by_type_maps_to_trigger_source(self, v9_client: APIClient) -> None:
-        """?type=SCHEDULED_TASK filter maps to trigger_source=scheduled for v9 compatibility."""
-        PaperlessTaskFactory(trigger_source=PaperlessTask.TriggerSource.SCHEDULED)
-        PaperlessTaskFactory(trigger_source=PaperlessTask.TriggerSource.WEB_UI)
-
-        response = v9_client.get(ENDPOINT, {"type": "SCHEDULED_TASK"})
-
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 1
-        assert response.data[0]["type"] == "SCHEDULED_TASK"
-
-
-@pytest.mark.django_db()
-class TestAcknowledge:
-    def test_returns_count(self, admin_client: APIClient) -> None:
-        """POST acknowledge/ returns the count of tasks that were acknowledged."""
-        task1 = PaperlessTaskFactory()
-        task2 = PaperlessTaskFactory()
-
-        response = admin_client.post(
-            ENDPOINT + "acknowledge/",
-            {"tasks": [task1.id, task2.id]},
-            format="json",
-        )
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data == {"result": 2}
-
-    def test_acknowledged_tasks_excluded_from_unacked_filter(
-        self,
-        admin_client: APIClient,
-    ) -> None:
-        """Acknowledged tasks no longer appear when filtering with ?acknowledged=false."""
-        task = PaperlessTaskFactory()
-        admin_client.post(
-            ENDPOINT + "acknowledge/",
+        response = self.client.post(
+            self.ENDPOINT + "acknowledge/",
             {"tasks": [task.id]},
-            format="json",
+        )
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+
+        response = self.client.get(self.ENDPOINT + "?acknowledged=false")
+        self.assertEqual(len(response.data), 0)
+
+    def test_acknowledge_tasks_requires_change_permission(self) -> None:
+        """
+        GIVEN:
+            - A regular user initially without change permissions
+            - A regular user with change permissions
+        WHEN:
+            - API call is made to acknowledge tasks
+        THEN:
+            - The first user is forbidden from acknowledging tasks
+            - The second user is allowed to acknowledge tasks
+        """
+        regular_user = User.objects.create_user(username="test")
+        self.client.force_authenticate(user=regular_user)
+
+        task = PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_one.pdf",
         )
 
-        response = admin_client.get(ENDPOINT, {"acknowledged": "false"})
-
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 0
-
-    def test_requires_change_permission(self, user_client: APIClient) -> None:
-        """Regular users without change_paperlesstask permission receive 403."""
-        task = PaperlessTaskFactory()
-
-        response = user_client.post(
-            ENDPOINT + "acknowledge/",
+        response = self.client.post(
+            self.ENDPOINT + "acknowledge/",
             {"tasks": [task.id]},
-            format="json",
         )
+        self.assertEqual(response.status_code, status.HTTP_403_FORBIDDEN)
 
-        assert response.status_code == status.HTTP_403_FORBIDDEN
-
-    def test_succeeds_with_change_permission(self, regular_user: User) -> None:
-        """Users granted change_paperlesstask permission can acknowledge tasks."""
-        regular_user.user_permissions.add(
+        regular_user2 = User.objects.create_user(username="test2")
+        regular_user2.user_permissions.add(
             Permission.objects.get(codename="change_paperlesstask"),
         )
-        regular_user.save()
+        regular_user2.save()
+        self.client.force_authenticate(user=regular_user2)
 
-        client = APIClient()
-        client.force_authenticate(user=regular_user)
-        client.credentials(HTTP_ACCEPT=ACCEPT_V10)
-
-        task = PaperlessTaskFactory()
-        response = client.post(
-            ENDPOINT + "acknowledge/",
+        response = self.client.post(
+            self.ENDPOINT + "acknowledge/",
             {"tasks": [task.id]},
-            format="json",
+        )
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+
+    def test_tasks_owner_aware(self) -> None:
+        """
+        GIVEN:
+            - Existing PaperlessTasks with owner and with no owner
+        WHEN:
+            - API call is made to get tasks
+        THEN:
+            - Only tasks with no owner or request user are returned
+        """
+
+        regular_user = User.objects.create_user(username="test")
+        regular_user.user_permissions.add(*Permission.objects.all())
+        self.client.logout()
+        self.client.force_authenticate(user=regular_user)
+
+        task1 = PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_one.pdf",
+            owner=self.user,
         )
 
-        assert response.status_code == status.HTTP_200_OK
-
-
-@pytest.mark.django_db()
-class TestAcknowledgeAll:
-    def test_marks_only_completed_tasks(self, admin_client: APIClient) -> None:
-        """acknowledge_all/ marks only SUCCESS and FAILURE tasks as acknowledged."""
-        PaperlessTaskFactory(status=PaperlessTask.Status.SUCCESS, acknowledged=False)
-        PaperlessTaskFactory(status=PaperlessTask.Status.FAILURE, acknowledged=False)
-        PaperlessTaskFactory(status=PaperlessTask.Status.PENDING, acknowledged=False)
-
-        response = admin_client.post(ENDPOINT + "acknowledge_all/")
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data == {"result": 2}
-
-    def test_skips_already_acknowledged(self, admin_client: APIClient) -> None:
-        """acknowledge_all/ does not re-acknowledge tasks that are already acknowledged."""
-        PaperlessTaskFactory(status=PaperlessTask.Status.SUCCESS, acknowledged=True)
-        PaperlessTaskFactory(status=PaperlessTask.Status.SUCCESS, acknowledged=False)
-
-        response = admin_client.post(ENDPOINT + "acknowledge_all/")
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data == {"result": 1}
-
-    def test_skips_pending_and_started(self, admin_client: APIClient) -> None:
-        """acknowledge_all/ does not touch PENDING or STARTED tasks."""
-        PaperlessTaskFactory(status=PaperlessTask.Status.PENDING)
-        PaperlessTaskFactory(status=PaperlessTask.Status.STARTED)
-
-        response = admin_client.post(ENDPOINT + "acknowledge_all/")
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data == {"result": 0}
-
-    def test_includes_revoked(self, admin_client: APIClient) -> None:
-        """acknowledge_all/ marks REVOKED tasks as acknowledged."""
-        PaperlessTaskFactory(status=PaperlessTask.Status.REVOKED, acknowledged=False)
-
-        response = admin_client.post(ENDPOINT + "acknowledge_all/")
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data == {"result": 1}
-
-
-@pytest.mark.django_db()
-class TestSummary:
-    def test_returns_per_type_totals(self, admin_client: APIClient) -> None:
-        """summary/ returns per-type counts of total, success, and failure tasks."""
-        PaperlessTaskFactory(
-            task_type=PaperlessTask.TaskType.CONSUME_FILE,
-            status=PaperlessTask.Status.SUCCESS,
-        )
-        PaperlessTaskFactory(
-            task_type=PaperlessTask.TaskType.CONSUME_FILE,
-            status=PaperlessTask.Status.FAILURE,
-        )
-        PaperlessTaskFactory(
-            task_type=PaperlessTask.TaskType.TRAIN_CLASSIFIER,
-            status=PaperlessTask.Status.SUCCESS,
+        task2 = PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_two.pdf",
         )
 
-        response = admin_client.get(ENDPOINT + "summary/")
+        task3 = PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_three.pdf",
+            owner=regular_user,
+        )
 
-        assert response.status_code == status.HTTP_200_OK
-        by_type = {item["task_type"]: item for item in response.data}
-        assert by_type["consume_file"]["total_count"] == 2
-        assert by_type["consume_file"]["success_count"] == 1
-        assert by_type["consume_file"]["failure_count"] == 1
-        assert by_type["train_classifier"]["total_count"] == 1
+        response = self.client.get(self.ENDPOINT)
 
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(len(response.data), 2)
+        self.assertEqual(response.data[0]["task_id"], task3.task_id)
+        self.assertEqual(response.data[1]["task_id"], task2.task_id)
 
-@pytest.mark.django_db()
-class TestActive:
-    def test_returns_pending_and_started_only(self, admin_client: APIClient) -> None:
-        """active/ returns only tasks in PENDING or STARTED status."""
-        PaperlessTaskFactory(status=PaperlessTask.Status.PENDING)
-        PaperlessTaskFactory(status=PaperlessTask.Status.STARTED)
-        PaperlessTaskFactory(status=PaperlessTask.Status.SUCCESS)
-        PaperlessTaskFactory(status=PaperlessTask.Status.FAILURE)
+        acknowledge_response = self.client.post(
+            self.ENDPOINT + "acknowledge/",
+            {"tasks": [task1.id, task2.id, task3.id]},
+        )
+        self.assertEqual(acknowledge_response.status_code, status.HTTP_200_OK)
+        self.assertEqual(acknowledge_response.data, {"result": 2})
 
-        response = admin_client.get(ENDPOINT + "active/")
+    def test_task_result_no_error(self) -> None:
+        """
+        GIVEN:
+            - A celery task completed without error
+        WHEN:
+            - API call is made to get tasks
+        THEN:
+            - The returned data includes the task result
+        """
+        PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_one.pdf",
+            status=celery.states.SUCCESS,
+            result="Success. New document id 1 created",
+        )
 
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 2
-        active_statuses = {t["status"] for t in response.data}
-        assert active_statuses == {
-            PaperlessTask.Status.PENDING,
-            PaperlessTask.Status.STARTED,
+        response = self.client.get(self.ENDPOINT)
+
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(len(response.data), 1)
+
+        returned_data = response.data[0]
+
+        self.assertEqual(returned_data["result"], "Success. New document id 1 created")
+        self.assertEqual(returned_data["related_document"], "1")
+
+    def test_task_result_with_error(self) -> None:
+        """
+        GIVEN:
+            - A celery task completed with an exception
+        WHEN:
+            - API call is made to get tasks
+        THEN:
+            - The returned result is the exception info
+        """
+        PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_one.pdf",
+            status=celery.states.FAILURE,
+            result="test.pdf: Unexpected error during ingestion.",
+        )
+
+        response = self.client.get(self.ENDPOINT)
+
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(len(response.data), 1)
+
+        returned_data = response.data[0]
+
+        self.assertEqual(
+            returned_data["result"],
+            "test.pdf: Unexpected error during ingestion.",
+        )
+
+    def test_task_name_webui(self) -> None:
+        """
+        GIVEN:
+            - Attempted celery task
+            - Task was created through the webui
+        WHEN:
+            - API call is made to get tasks
+        THEN:
+            - Returned data include the filename
+        """
+        PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="test.pdf",
+            task_name=PaperlessTask.TaskName.CONSUME_FILE,
+            status=celery.states.SUCCESS,
+        )
+
+        response = self.client.get(self.ENDPOINT)
+
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(len(response.data), 1)
+
+        returned_data = response.data[0]
+
+        self.assertEqual(returned_data["task_file_name"], "test.pdf")
+
+    def test_task_name_consume_folder(self) -> None:
+        """
+        GIVEN:
+            - Attempted celery task
+            - Task was created through the consume folder
+        WHEN:
+            - API call is made to get tasks
+        THEN:
+            - Returned data include the filename
+        """
+        PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="anothertest.pdf",
+            task_name=PaperlessTask.TaskName.CONSUME_FILE,
+            status=celery.states.SUCCESS,
+        )
+
+        response = self.client.get(self.ENDPOINT)
+
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(len(response.data), 1)
+
+        returned_data = response.data[0]
+
+        self.assertEqual(returned_data["task_file_name"], "anothertest.pdf")
+
+    def test_task_result_duplicate_warning_includes_count(self) -> None:
+        """
+        GIVEN:
+            - A celery task succeeds, but a duplicate exists
+        WHEN:
+            - API call is made to get tasks
+        THEN:
+            - The returned data includes duplicate warning metadata
+        """
+        checksum = "duplicate-checksum"
+        Document.objects.create(
+            title="Existing",
+            content="",
+            mime_type="application/pdf",
+            checksum=checksum,
+        )
+        created_doc = Document.objects.create(
+            title="Created",
+            content="",
+            mime_type="application/pdf",
+            checksum=checksum,
+            archive_checksum="another-checksum",
+        )
+        PaperlessTask.objects.create(
+            task_id=str(uuid.uuid4()),
+            task_file_name="task_one.pdf",
+            status=celery.states.SUCCESS,
+            result=f"Success. New document id {created_doc.pk} created",
+        )
+
+        response = self.client.get(self.ENDPOINT)
+
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(len(response.data), 1)
+
+        returned_data = response.data[0]
+
+        self.assertEqual(returned_data["related_document"], str(created_doc.pk))
+
+    def test_run_train_classifier_task(self) -> None:
+        """
+        GIVEN:
+            - A superuser
+        WHEN:
+            - API call is made to run the train classifier task
+        THEN:
+            - The task is run
+        """
+        mock_train_classifier = mock.Mock(return_value="Task started")
+        TasksViewSet.TASK_AND_ARGS_BY_NAME = {
+            PaperlessTask.TaskName.TRAIN_CLASSIFIER: (
+                mock_train_classifier,
+                {"scheduled": False},
+            ),
         }
-
-    def test_excludes_revoked_tasks_from_active(self, admin_client: APIClient) -> None:
-        """active/ excludes REVOKED tasks."""
-        PaperlessTaskFactory(status=PaperlessTask.Status.REVOKED)
-
-        response = admin_client.get(ENDPOINT + "active/")
-
-        assert response.status_code == status.HTTP_200_OK
-        assert len(response.data) == 0
-
-
-@pytest.mark.django_db()
-class TestRun:
-    def test_forbidden_for_regular_user(self, user_client: APIClient) -> None:
-        """Regular users without add_paperlesstask permission receive 403 from run/."""
-        response = user_client.post(
-            ENDPOINT + "run/",
-            {"task_type": PaperlessTask.TaskType.TRAIN_CLASSIFIER},
-            format="json",
+        response = self.client.post(
+            self.ENDPOINT + "run/",
+            {"task_name": PaperlessTask.TaskName.TRAIN_CLASSIFIER},
         )
 
-        assert response.status_code == status.HTTP_403_FORBIDDEN
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertEqual(response.data, {"result": "Task started"})
+        mock_train_classifier.assert_called_once_with(scheduled=False)
 
-    def test_dispatches_via_apply_async_with_manual_trigger_header(
-        self,
-        admin_client: APIClient,
-    ) -> None:
-        """run/ dispatches the task via apply_async with trigger_source=manual in headers."""
-        fake_task_id = str(uuid.uuid4())
-        mock_async_result = mock.Mock()
-        mock_async_result.id = fake_task_id
-
-        mock_apply_async = mock.Mock(return_value=mock_async_result)
-
-        with mock.patch(
-            "documents.views.train_classifier.apply_async",
-            mock_apply_async,
-        ):
-            response = admin_client.post(
-                ENDPOINT + "run/",
-                {"task_type": PaperlessTask.TaskType.TRAIN_CLASSIFIER},
-                format="json",
-            )
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data == {"task_id": fake_task_id}
-        mock_apply_async.assert_called_once_with(
-            kwargs={},
-            headers={"trigger_source": "manual"},
+        # mock error
+        mock_train_classifier.reset_mock()
+        mock_train_classifier.side_effect = Exception("Error")
+        response = self.client.post(
+            self.ENDPOINT + "run/",
+            {"task_name": PaperlessTask.TaskName.TRAIN_CLASSIFIER},
         )
 
-    def test_returns_400_for_consume_file(self, admin_client: APIClient) -> None:
-        """consume_file cannot be manually triggered via the run endpoint."""
-        response = admin_client.post(
-            ENDPOINT + "run/",
-            {"task_type": PaperlessTask.TaskType.CONSUME_FILE},
-            format="json",
+        self.assertEqual(response.status_code, status.HTTP_500_INTERNAL_SERVER_ERROR)
+        mock_train_classifier.assert_called_once_with(scheduled=False)
+
+    @mock.patch("documents.tasks.sanity_check")
+    def test_run_task_requires_superuser(self, mock_check_sanity) -> None:
+        """
+        GIVEN:
+            - A regular user
+        WHEN:
+            - API call is made to run a task
+        THEN:
+            - The task is not run
+        """
+        regular_user = User.objects.create_user(username="test")
+        regular_user.user_permissions.add(*Permission.objects.all())
+        self.client.logout()
+        self.client.force_authenticate(user=regular_user)
+
+        response = self.client.post(
+            self.ENDPOINT + "run/",
+            {"task_name": PaperlessTask.TaskName.CHECK_SANITY},
         )
 
-        assert response.status_code == status.HTTP_400_BAD_REQUEST
-
-    def test_returns_400_for_invalid_task_type(self, admin_client: APIClient) -> None:
-        """run/ returns 400 for an unrecognized task_type value."""
-        response = admin_client.post(
-            ENDPOINT + "run/",
-            {"task_type": "not_a_real_type"},
-            format="json",
-        )
-
-        assert response.status_code == status.HTTP_400_BAD_REQUEST
-
-    def test_sanity_check_dispatched_with_correct_kwargs(
-        self,
-        admin_client: APIClient,
-    ) -> None:
-        """run/ dispatches sanity_check with raise_on_error=False and manual trigger header."""
-        fake_task_id = str(uuid.uuid4())
-        mock_async_result = mock.Mock()
-        mock_async_result.id = fake_task_id
-
-        mock_apply_async = mock.Mock(return_value=mock_async_result)
-
-        with mock.patch(
-            "documents.views.sanity_check.apply_async",
-            mock_apply_async,
-        ):
-            response = admin_client.post(
-                ENDPOINT + "run/",
-                {"task_type": PaperlessTask.TaskType.SANITY_CHECK},
-                format="json",
-            )
-
-        assert response.status_code == status.HTTP_200_OK
-        assert response.data == {"task_id": fake_task_id}
-        mock_apply_async.assert_called_once_with(
-            kwargs={"raise_on_error": False},
-            headers={"trigger_source": "manual"},
-        )
+        self.assertEqual(response.status_code, status.HTTP_403_FORBIDDEN)
+        mock_check_sanity.assert_not_called()

src/documents/tests/test_consumer.py

@@ -1328,7 +1328,7 @@ class PreConsumeTestCase(DirectoriesMixin, GetConsumerMixin, TestCase):
                     environment = args[1]
 
                     self.assertEqual(command[0], script.name)
-                    self.assertEqual(len(command), 1)
+                    self.assertEqual(command[1], str(self.test_file))
 
                     subset = {
                         "DOCUMENT_SOURCE_PATH": str(c.input_doc.original_file),
@@ -1478,7 +1478,11 @@ class PostConsumeTestCase(DirectoriesMixin, GetConsumerMixin, TestCase):
                 environment = args[1]
 
                 self.assertEqual(command[0], script.name)
-                self.assertEqual(len(command), 1)
+                self.assertEqual(command[1], str(doc.pk))
+                self.assertEqual(command[5], f"/api/documents/{doc.pk}/download/")
+                self.assertEqual(command[6], f"/api/documents/{doc.pk}/thumb/")
+                self.assertEqual(command[7], "my_bank")
+                self.assertCountEqual(command[8].split(","), ["a", "b"])
 
                 subset = {
                     "DOCUMENT_ID": str(doc.pk),

src/documents/tests/test_management.py

@@ -211,7 +211,7 @@ class TestCreateClassifier:
 
         call_command("document_create_classifier", skip_checks=True)
 
-        m.assert_called_once_with(status_callback=mocker.ANY)
+        m.assert_called_once_with(scheduled=False, status_callback=mocker.ANY)
         assert callable(m.call_args.kwargs["status_callback"])
 
     def test_create_classifier_callback_output(self, mocker: MockerFixture) -> None:

src/documents/tests/test_sanity_check.py

@@ -1,7 +1,7 @@
 """Tests for the sanity checker module.
 
 Tests exercise ``check_sanity`` as a whole, verifying document validation,
-orphan detection, and the iter_wrapper contract.
+orphan detection, task recording, and the iter_wrapper contract.
 """
 
 from __future__ import annotations
@@ -12,12 +12,13 @@ from typing import TYPE_CHECKING
 
 import pytest
 
+from documents.models import Document
+from documents.models import PaperlessTask
 from documents.sanity_checker import check_sanity
 
 if TYPE_CHECKING:
     from collections.abc import Iterable
 
-    from documents.models import Document
     from documents.tests.conftest import PaperlessDirs
 
 
@@ -228,6 +229,35 @@ class TestCheckSanityIterWrapper:
         assert not messages.has_error
 
 
+@pytest.mark.django_db
+class TestCheckSanityTaskRecording:
+    @pytest.mark.parametrize(
+        ("expected_type", "scheduled"),
+        [
+            pytest.param(PaperlessTask.TaskType.SCHEDULED_TASK, True, id="scheduled"),
+            pytest.param(PaperlessTask.TaskType.MANUAL_TASK, False, id="manual"),
+        ],
+    )
+    @pytest.mark.usefixtures("_media_settings")
+    def test_task_type(self, expected_type: str, *, scheduled: bool) -> None:
+        check_sanity(scheduled=scheduled)
+        task = PaperlessTask.objects.latest("date_created")
+        assert task.task_name == PaperlessTask.TaskName.CHECK_SANITY
+        assert task.type == expected_type
+
+    def test_success_status(self, sample_doc: Document) -> None:
+        check_sanity()
+        task = PaperlessTask.objects.latest("date_created")
+        assert task.status == "SUCCESS"
+
+    def test_failure_status(self, sample_doc: Document) -> None:
+        Path(sample_doc.source_path).unlink()
+        check_sanity()
+        task = PaperlessTask.objects.latest("date_created")
+        assert task.status == "FAILURE"
+        assert "Check logs for details" in task.result
+
+
 @pytest.mark.django_db
 class TestCheckSanityLogMessages:
     def test_logs_doc_issues(

src/documents/tests/test_task_signals.py

@@ -1,302 +1,250 @@
 import uuid
 from unittest import mock
 
-import pytest
+import celery
+from django.contrib.auth import get_user_model
+from django.test import TestCase
 
 from documents.data_models import ConsumableDocument
 from documents.data_models import DocumentMetadataOverrides
 from documents.data_models import DocumentSource
+from documents.models import Document
 from documents.models import PaperlessTask
+from documents.signals.handlers import add_to_index
+from documents.signals.handlers import before_task_publish_handler
+from documents.signals.handlers import task_failure_handler
+from documents.signals.handlers import task_postrun_handler
+from documents.signals.handlers import task_prerun_handler
+from documents.tests.test_consumer import fake_magic_from_file
+from documents.tests.utils import DirectoriesMixin
 
 
-@pytest.fixture
-def consume_input_doc():
-    doc = mock.MagicMock(spec=ConsumableDocument)
-    # original_file is a Path; configure the nested mock so .name works
-    doc.original_file = mock.MagicMock()
-    doc.original_file.name = "invoice.pdf"
-    doc.original_path = None
-    doc.mime_type = "application/pdf"
-    doc.mailrule_id = None
-    doc.source = DocumentSource.WebUI
-    return doc
+@mock.patch("documents.consumer.magic.from_file", fake_magic_from_file)
+class TestTaskSignalHandler(DirectoriesMixin, TestCase):
+    @classmethod
+    def setUpTestData(cls) -> None:
+        super().setUpTestData()
+        cls.user = get_user_model().objects.create_user(username="testuser")
 
-
-@pytest.fixture
-def consume_overrides(django_user_model):
-    user = django_user_model.objects.create_user(username="testuser")
-    overrides = mock.MagicMock(spec=DocumentMetadataOverrides)
-    overrides.owner_id = user.id
-    return overrides
-
-
-def send_publish(
-    task_name: str,
-    args: tuple,
-    kwargs: dict,
-    headers: dict | None = None,
-) -> str:
-    from documents.signals.handlers import before_task_publish_handler
-
-    task_id = str(uuid.uuid4())
-    hdrs = {"task": task_name, "id": task_id, **(headers or {})}
-    before_task_publish_handler(sender=task_name, headers=hdrs, body=(args, kwargs, {}))
-    return task_id
-
-
-@pytest.mark.django_db
-class TestBeforeTaskPublishHandler:
-    def test_creates_task_for_consume_file(self, consume_input_doc, consume_overrides):
-        task_id = send_publish(
-            "documents.tasks.consume_file",
-            (consume_input_doc, consume_overrides),
-            {},
-        )
-        task = PaperlessTask.objects.get(task_id=task_id)
-        assert task.task_type == PaperlessTask.TaskType.CONSUME_FILE
-        assert task.status == PaperlessTask.Status.PENDING
-        assert task.trigger_source == PaperlessTask.TriggerSource.WEB_UI
-        assert task.input_data["filename"] == "invoice.pdf"
-        assert task.owner_id == consume_overrides.owner_id
-
-    def test_creates_task_for_train_classifier(self):
-        task_id = send_publish("documents.tasks.train_classifier", (), {})
-        task = PaperlessTask.objects.get(task_id=task_id)
-        assert task.task_type == PaperlessTask.TaskType.TRAIN_CLASSIFIER
-        assert task.trigger_source == PaperlessTask.TriggerSource.MANUAL
-
-    def test_creates_task_for_sanity_check(self):
-        task_id = send_publish("documents.tasks.sanity_check", (), {})
-        task = PaperlessTask.objects.get(task_id=task_id)
-        assert task.task_type == PaperlessTask.TaskType.SANITY_CHECK
-
-    def test_creates_task_for_process_mail_accounts(self):
-        task_id = send_publish(
-            "paperless_mail.tasks.process_mail_accounts",
-            (),
-            {"account_ids": [1, 2]},
-        )
-        task = PaperlessTask.objects.get(task_id=task_id)
-        assert task.task_type == PaperlessTask.TaskType.MAIL_FETCH
-        assert task.input_data["account_ids"] == [1, 2]
-
-    def test_scheduled_header_sets_trigger_source(self):
-        task_id = send_publish(
-            "documents.tasks.train_classifier",
-            (),
-            {},
-            headers={"trigger_source": "scheduled"},
-        )
-        task = PaperlessTask.objects.get(task_id=task_id)
-        assert task.trigger_source == PaperlessTask.TriggerSource.SCHEDULED
-
-    def test_system_header_sets_trigger_source(self):
-        task_id = send_publish(
-            "documents.tasks.llmindex_index",
-            (),
-            {"rebuild": True},
-            headers={"trigger_source": "system"},
-        )
-        task = PaperlessTask.objects.get(task_id=task_id)
-        assert task.trigger_source == PaperlessTask.TriggerSource.SYSTEM
-
-    def test_ignores_untracked_task(self):
-        send_publish("documents.tasks.bulk_update_documents", ([1, 2],), {})
-        assert PaperlessTask.objects.count() == 0
-
-    def test_ignores_none_headers(self):
-        from documents.signals.handlers import before_task_publish_handler
-
-        before_task_publish_handler(sender=None, headers=None, body=None)
-        assert PaperlessTask.objects.count() == 0
-
-    def test_consume_folder_source_maps_correctly(
+    def util_call_before_task_publish_handler(
         self,
-        consume_input_doc,
-        consume_overrides,
-    ):
-        consume_input_doc.source = DocumentSource.ConsumeFolder
-        task_id = send_publish(
-            "documents.tasks.consume_file",
-            (consume_input_doc, consume_overrides),
+        headers_to_use,
+        body_to_use,
+    ) -> None:
+        """
+        Simple utility to call the pre-run handle and ensure it created a single task
+        instance
+        """
+        self.assertEqual(PaperlessTask.objects.all().count(), 0)
+
+        before_task_publish_handler(headers=headers_to_use, body=body_to_use)
+
+        self.assertEqual(PaperlessTask.objects.all().count(), 1)
+
+    def test_before_task_publish_handler_consume(self) -> None:
+        """
+        GIVEN:
+            - A celery task is started via the consume folder
+        WHEN:
+            - Task before publish handler is called
+        THEN:
+            - The task is created and marked as pending
+        """
+        headers = {
+            "id": str(uuid.uuid4()),
+            "task": "documents.tasks.consume_file",
+        }
+        body = (
+            # args
+            (
+                ConsumableDocument(
+                    source=DocumentSource.ConsumeFolder,
+                    original_file="/consume/hello-999.pdf",
+                ),
+                DocumentMetadataOverrides(
+                    title="Hello world",
+                    owner_id=self.user.id,
+                ),
+            ),
+            # kwargs
             {},
+            # celery stuff
+            {"callbacks": None, "errbacks": None, "chain": None, "chord": None},
+        )
+        self.util_call_before_task_publish_handler(
+            headers_to_use=headers,
+            body_to_use=body,
         )
-        task = PaperlessTask.objects.get(task_id=task_id)
-        assert task.trigger_source == PaperlessTask.TriggerSource.FOLDER_CONSUME
 
-    def test_email_source_maps_correctly(self, consume_input_doc, consume_overrides):
-        consume_input_doc.source = DocumentSource.MailFetch
-        task_id = send_publish(
-            "documents.tasks.consume_file",
-            (consume_input_doc, consume_overrides),
+        task = PaperlessTask.objects.get()
+        self.assertIsNotNone(task)
+        self.assertEqual(headers["id"], task.task_id)
+        self.assertEqual("hello-999.pdf", task.task_file_name)
+        self.assertEqual(PaperlessTask.TaskName.CONSUME_FILE, task.task_name)
+        self.assertEqual(self.user.id, task.owner_id)
+        self.assertEqual(celery.states.PENDING, task.status)
+
+    def test_task_prerun_handler(self) -> None:
+        """
+        GIVEN:
+            - A celery task is started via the consume folder
+        WHEN:
+            - Task starts execution
+        THEN:
+            - The task is marked as started
+        """
+
+        headers = {
+            "id": str(uuid.uuid4()),
+            "task": "documents.tasks.consume_file",
+        }
+        body = (
+            # args
+            (
+                ConsumableDocument(
+                    source=DocumentSource.ConsumeFolder,
+                    original_file="/consume/hello-99.pdf",
+                ),
+                None,
+            ),
+            # kwargs
             {},
-        )
-        task = PaperlessTask.objects.get(task_id=task_id)
-        assert task.trigger_source == PaperlessTask.TriggerSource.EMAIL_CONSUME
-
-
-@pytest.mark.django_db
-class TestTaskPrerunHandler:
-    def test_marks_task_started(self):
-        task = PaperlessTask.objects.create(
-            task_id=str(uuid.uuid4()),
-            task_type=PaperlessTask.TaskType.CONSUME_FILE,
-            trigger_source=PaperlessTask.TriggerSource.MANUAL,
-            status=PaperlessTask.Status.PENDING,
-        )
-        from documents.signals.handlers import task_prerun_handler
-
-        task_prerun_handler(task_id=task.task_id)
-        task.refresh_from_db()
-        assert task.status == PaperlessTask.Status.STARTED
-        assert task.date_started is not None
-
-    def test_ignores_unknown_task_id(self):
-        from documents.signals.handlers import task_prerun_handler
-
-        task_prerun_handler(task_id="nonexistent-id")  # must not raise
-
-    def test_ignores_none_task_id(self):
-        from documents.signals.handlers import task_prerun_handler
-
-        task_prerun_handler(task_id=None)  # must not raise
-
-
-@pytest.mark.django_db
-class TestTaskPostrunHandler:
-    def _started_task(self) -> PaperlessTask:
-        from django.utils import timezone
-
-        return PaperlessTask.objects.create(
-            task_id=str(uuid.uuid4()),
-            task_type=PaperlessTask.TaskType.TRAIN_CLASSIFIER,
-            trigger_source=PaperlessTask.TriggerSource.MANUAL,
-            status=PaperlessTask.Status.STARTED,
-            date_started=timezone.now(),
+            # celery stuff
+            {"callbacks": None, "errbacks": None, "chain": None, "chord": None},
         )
 
-    def test_records_success_with_dict_result(self):
-        task = self._started_task()
-        from documents.signals.handlers import task_postrun_handler
+        self.util_call_before_task_publish_handler(
+            headers_to_use=headers,
+            body_to_use=body,
+        )
+
+        task_prerun_handler(task_id=headers["id"])
+
+        task = PaperlessTask.objects.get()
+
+        self.assertEqual(celery.states.STARTED, task.status)
+
+    def test_task_postrun_handler(self) -> None:
+        """
+        GIVEN:
+            - A celery task is started via the consume folder
+        WHEN:
+            - Task finished execution
+        THEN:
+            - The task is marked as started
+        """
+        headers = {
+            "id": str(uuid.uuid4()),
+            "task": "documents.tasks.consume_file",
+        }
+        body = (
+            # args
+            (
+                ConsumableDocument(
+                    source=DocumentSource.ConsumeFolder,
+                    original_file="/consume/hello-9.pdf",
+                ),
+                None,
+            ),
+            # kwargs
+            {},
+            # celery stuff
+            {"callbacks": None, "errbacks": None, "chain": None, "chord": None},
+        )
+        self.util_call_before_task_publish_handler(
+            headers_to_use=headers,
+            body_to_use=body,
+        )
 
         task_postrun_handler(
-            task_id=task.task_id,
-            retval={"document_id": 42},
-            state="SUCCESS",
+            task_id=headers["id"],
+            retval="Success. New document id 1 created",
+            state=celery.states.SUCCESS,
         )
-        task.refresh_from_db()
-        assert task.status == PaperlessTask.Status.SUCCESS
-        assert task.result_data == {"document_id": 42}
-        assert task.date_done is not None
-        assert task.duration_seconds is not None
-        assert task.wait_time_seconds is not None
 
-    def test_records_failure_state(self):
-        task = self._started_task()
-        from documents.signals.handlers import task_postrun_handler
+        task = PaperlessTask.objects.get()
 
-        task_postrun_handler(task_id=task.task_id, retval="some error", state="FAILURE")
-        task.refresh_from_db()
-        assert task.status == PaperlessTask.Status.FAILURE
+        self.assertEqual(celery.states.SUCCESS, task.status)
 
-    def test_parses_legacy_new_document_string(self):
-        task = self._started_task()
-        from documents.signals.handlers import task_postrun_handler
-
-        task_postrun_handler(
-            task_id=task.task_id,
-            retval="New document id 42 created",
-            state="SUCCESS",
+    def test_task_failure_handler(self) -> None:
+        """
+        GIVEN:
+            - A celery task is started via the consume folder
+        WHEN:
+            - Task failed execution
+        THEN:
+            - The task is marked as failed
+        """
+        headers = {
+            "id": str(uuid.uuid4()),
+            "task": "documents.tasks.consume_file",
+        }
+        body = (
+            # args
+            (
+                ConsumableDocument(
+                    source=DocumentSource.ConsumeFolder,
+                    original_file="/consume/hello-9.pdf",
+                ),
+                None,
+            ),
+            # kwargs
+            {},
+            # celery stuff
+            {"callbacks": None, "errbacks": None, "chain": None, "chord": None},
         )
-        task.refresh_from_db()
-        assert task.result_data["document_id"] == 42
-        assert task.result_message == "New document id 42 created"
-
-    def test_parses_legacy_duplicate_string(self):
-        task = self._started_task()
-        from documents.signals.handlers import task_postrun_handler
-
-        task_postrun_handler(
-            task_id=task.task_id,
-            retval="It is a duplicate of some document (#99).",
-            state="FAILURE",
+        self.util_call_before_task_publish_handler(
+            headers_to_use=headers,
+            body_to_use=body,
         )
-        task.refresh_from_db()
-        assert task.result_data["duplicate_of"] == 99
-        assert task.result_data["duplicate_in_trash"] is False
-
-    def test_ignores_unknown_task_id(self):
-        from documents.signals.handlers import task_postrun_handler
-
-        task_postrun_handler(
-            task_id="nonexistent",
-            retval=None,
-            state="SUCCESS",
-        )  # must not raise
-
-    def test_records_revoked_state(self):
-        task = self._started_task()
-        from documents.signals.handlers import task_postrun_handler
-
-        task_postrun_handler(task_id=task.task_id, retval=None, state="REVOKED")
-        task.refresh_from_db()
-        assert task.status == PaperlessTask.Status.REVOKED
-
-
-@pytest.mark.django_db
-class TestTaskFailureHandler:
-    def test_records_failure_with_exception(self):
-        from django.utils import timezone
-
-        task = PaperlessTask.objects.create(
-            task_id=str(uuid.uuid4()),
-            task_type=PaperlessTask.TaskType.CONSUME_FILE,
-            trigger_source=PaperlessTask.TriggerSource.WEB_UI,
-            status=PaperlessTask.Status.STARTED,
-            date_started=timezone.now(),
-        )
-        from documents.signals.handlers import task_failure_handler
 
         task_failure_handler(
-            task_id=task.task_id,
-            exception=ValueError("PDF parse failed"),
-            traceback=None,
+            task_id=headers["id"],
+            exception="Example failure",
         )
-        task.refresh_from_db()
-        assert task.status == PaperlessTask.Status.FAILURE
-        assert task.result_data["error_type"] == "ValueError"
-        assert task.result_data["error_message"] == "PDF parse failed"
-        assert task.date_done is not None
 
-    def test_records_traceback_when_provided(self):
-        import sys
+        task = PaperlessTask.objects.get()
 
-        from django.utils import timezone
+        self.assertEqual(celery.states.FAILURE, task.status)
 
-        task = PaperlessTask.objects.create(
-            task_id=str(uuid.uuid4()),
-            task_type=PaperlessTask.TaskType.CONSUME_FILE,
-            trigger_source=PaperlessTask.TriggerSource.WEB_UI,
-            status=PaperlessTask.Status.STARTED,
-            date_started=timezone.now(),
+    def test_add_to_index_indexes_root_once_for_root_documents(self) -> None:
+        root = Document.objects.create(
+            title="root",
+            checksum="root",
+            mime_type="application/pdf",
         )
-        try:
-            raise ValueError("test error")
-        except ValueError:
-            tb = sys.exc_info()[2]
 
-        from documents.signals.handlers import task_failure_handler
+        with mock.patch("documents.search.get_backend") as mock_get_backend:
+            mock_backend = mock.MagicMock()
+            mock_get_backend.return_value = mock_backend
+            add_to_index(sender=None, document=root)
 
-        task_failure_handler(
-            task_id=task.task_id,
-            exception=ValueError("test error"),
-            traceback=tb,
+        mock_backend.add_or_update.assert_called_once_with(root, effective_content="")
+
+    def test_add_to_index_reindexes_root_for_version_documents(self) -> None:
+        root = Document.objects.create(
+            title="root",
+            checksum="root",
+            mime_type="application/pdf",
+        )
+        version = Document.objects.create(
+            title="version",
+            checksum="version",
+            mime_type="application/pdf",
+            root_document=root,
         )
-        task.refresh_from_db()
-        assert "traceback" in task.result_data
-        assert len(task.result_data["traceback"]) <= 5000
 
-    def test_ignores_none_task_id(self):
-        from documents.signals.handlers import task_failure_handler
+        with mock.patch("documents.search.get_backend") as mock_get_backend:
+            mock_backend = mock.MagicMock()
+            mock_get_backend.return_value = mock_backend
+            add_to_index(sender=None, document=version)
 
-        task_failure_handler(task_id=None, exception=ValueError("x"), traceback=None)
+        self.assertEqual(mock_backend.add_or_update.call_count, 1)
+        self.assertEqual(
+            mock_backend.add_or_update.call_args_list[0].args[0].id,
+            version.id,
+        )
+        self.assertEqual(
+            mock_backend.add_or_update.call_args_list[0].kwargs,
+            {"effective_content": version.content},
+        )

src/documents/tests/test_tasks.py

@@ -4,6 +4,7 @@ from pathlib import Path
 from unittest import mock
 
 import pytest
+from celery import states
 from django.conf import settings
 from django.test import TestCase
 from django.test import override_settings
@@ -13,6 +14,7 @@ from documents import tasks
 from documents.models import Correspondent
 from documents.models import Document
 from documents.models import DocumentType
+from documents.models import PaperlessTask
 from documents.models import Tag
 from documents.sanity_checker import SanityCheckFailedException
 from documents.sanity_checker import SanityCheckMessages
@@ -38,8 +40,7 @@ class TestClassifier(DirectoriesMixin, FileSystemAssertsMixin, TestCase):
     def test_train_classifier_with_auto_tag(self, load_classifier) -> None:
         load_classifier.return_value = None
         Tag.objects.create(matching_algorithm=Tag.MATCH_AUTO, name="test")
-        with self.assertRaises(ValueError):
-            tasks.train_classifier()
+        tasks.train_classifier()
         load_classifier.assert_called_once()
         self.assertIsNotFile(settings.MODEL_FILE)
 
@@ -47,8 +48,7 @@ class TestClassifier(DirectoriesMixin, FileSystemAssertsMixin, TestCase):
     def test_train_classifier_with_auto_type(self, load_classifier) -> None:
         load_classifier.return_value = None
         DocumentType.objects.create(matching_algorithm=Tag.MATCH_AUTO, name="test")
-        with self.assertRaises(ValueError):
-            tasks.train_classifier()
+        tasks.train_classifier()
         load_classifier.assert_called_once()
         self.assertIsNotFile(settings.MODEL_FILE)
 
@@ -56,8 +56,7 @@ class TestClassifier(DirectoriesMixin, FileSystemAssertsMixin, TestCase):
     def test_train_classifier_with_auto_correspondent(self, load_classifier) -> None:
         load_classifier.return_value = None
         Correspondent.objects.create(matching_algorithm=Tag.MATCH_AUTO, name="test")
-        with self.assertRaises(ValueError):
-            tasks.train_classifier()
+        tasks.train_classifier()
         load_classifier.assert_called_once()
         self.assertIsNotFile(settings.MODEL_FILE)
 
@@ -299,7 +298,7 @@ class TestAIIndex(DirectoriesMixin, TestCase):
         WHEN:
             - llmindex_index task is called
         THEN:
-            - update_llm_index is called and its result is returned
+            - update_llm_index is called, and the task is marked as success
         """
         Document.objects.create(
             title="test",
@@ -309,9 +308,13 @@ class TestAIIndex(DirectoriesMixin, TestCase):
         # lazy-loaded so mock the actual function
         with mock.patch("paperless_ai.indexing.update_llm_index") as update_llm_index:
             update_llm_index.return_value = "LLM index updated successfully."
-            result = tasks.llmindex_index()
+            tasks.llmindex_index()
             update_llm_index.assert_called_once()
-            self.assertEqual(result, "LLM index updated successfully.")
+            task = PaperlessTask.objects.get(
+                task_name=PaperlessTask.TaskName.LLMINDEX_UPDATE,
+            )
+            self.assertEqual(task.status, states.SUCCESS)
+            self.assertEqual(task.result, "LLM index updated successfully.")
 
     @override_settings(
         AI_ENABLED=True,
@@ -322,9 +325,9 @@ class TestAIIndex(DirectoriesMixin, TestCase):
         GIVEN:
             - Document exists, AI is enabled, llm index backend is set
         WHEN:
-            - llmindex_index task is called and update_llm_index raises an exception
+            - llmindex_index task is called
         THEN:
-            - the exception propagates to the caller
+            - update_llm_index raises an exception, and the task is marked as failure
         """
         Document.objects.create(
             title="test",
@@ -334,9 +337,13 @@ class TestAIIndex(DirectoriesMixin, TestCase):
         # lazy-loaded so mock the actual function
         with mock.patch("paperless_ai.indexing.update_llm_index") as update_llm_index:
             update_llm_index.side_effect = Exception("LLM index update failed.")
-            with self.assertRaises(Exception, msg="LLM index update failed."):
-                tasks.llmindex_index()
+            tasks.llmindex_index()
             update_llm_index.assert_called_once()
+            task = PaperlessTask.objects.get(
+                task_name=PaperlessTask.TaskName.LLMINDEX_UPDATE,
+            )
+            self.assertEqual(task.status, states.FAILURE)
+            self.assertIn("LLM index update failed.", task.result)
 
     def test_update_document_in_llm_index(self) -> None:
         """

src/documents/views.py

@@ -8,7 +8,6 @@ import zipfile
 from collections import defaultdict
 from collections import deque
 from datetime import datetime
-from datetime import timedelta
 from pathlib import Path
 from time import mktime
 from typing import TYPE_CHECKING
@@ -21,6 +20,7 @@ from urllib.parse import urlparse
 import httpx
 import magic
 import pathvalidate
+from celery import states
 from django.conf import settings
 from django.contrib.auth.models import Group
 from django.contrib.auth.models import User
@@ -91,7 +91,6 @@ from rest_framework.mixins import DestroyModelMixin
 from rest_framework.mixins import ListModelMixin
 from rest_framework.mixins import RetrieveModelMixin
 from rest_framework.mixins import UpdateModelMixin
-from rest_framework.permissions import IsAdminUser
 from rest_framework.permissions import IsAuthenticated
 from rest_framework.request import Request
 from rest_framework.response import Response
@@ -192,7 +191,7 @@ from documents.serialisers import PostDocumentSerializer
 from documents.serialisers import RemovePasswordDocumentsSerializer
 from documents.serialisers import ReprocessDocumentsSerializer
 from documents.serialisers import RotateDocumentsSerializer
-from documents.serialisers import RunTaskSerializer
+from documents.serialisers import RunTaskViewSerializer
 from documents.serialisers import SavedViewSerializer
 from documents.serialisers import SearchResultSerializer
 from documents.serialisers import SerializerWithPerms
@@ -201,9 +200,7 @@ from documents.serialisers import ShareLinkSerializer
 from documents.serialisers import StoragePathSerializer
 from documents.serialisers import StoragePathTestSerializer
 from documents.serialisers import TagSerializer
-from documents.serialisers import TaskSerializerV9
-from documents.serialisers import TaskSerializerV10
-from documents.serialisers import TaskSummarySerializer
+from documents.serialisers import TasksViewSerializer
 from documents.serialisers import TrashSerializer
 from documents.serialisers import UiSettingsViewSerializer
 from documents.serialisers import WorkflowActionSerializer
@@ -294,7 +291,7 @@ class IndexView(TemplateView):
         return context
 
 
-class PassUserMixin(GenericAPIView[Any]):
+class PassUserMixin(GenericAPIView):
     """
     Pass a user object to serializer
     """
@@ -460,10 +457,7 @@ class PermissionsAwareDocumentCountMixin(BulkPermissionMixin, PassUserMixin):
 
 
 @extend_schema_view(**generate_object_with_permissions_schema(CorrespondentSerializer))
-class CorrespondentViewSet(
-    PermissionsAwareDocumentCountMixin,
-    ModelViewSet[Correspondent],
-):
+class CorrespondentViewSet(PermissionsAwareDocumentCountMixin, ModelViewSet):
     model = Correspondent
 
     queryset = Correspondent.objects.select_related("owner").order_by(Lower("name"))
@@ -500,7 +494,7 @@ class CorrespondentViewSet(
 
 
 @extend_schema_view(**generate_object_with_permissions_schema(TagSerializer))
-class TagViewSet(PermissionsAwareDocumentCountMixin, ModelViewSet[Tag]):
+class TagViewSet(PermissionsAwareDocumentCountMixin, ModelViewSet):
     model = Tag
     serializer_class = TagSerializer
     document_count_through = Document.tags.through
@@ -579,10 +573,7 @@ class TagViewSet(PermissionsAwareDocumentCountMixin, ModelViewSet[Tag]):
 
 
 @extend_schema_view(**generate_object_with_permissions_schema(DocumentTypeSerializer))
-class DocumentTypeViewSet(
-    PermissionsAwareDocumentCountMixin,
-    ModelViewSet[DocumentType],
-):
+class DocumentTypeViewSet(PermissionsAwareDocumentCountMixin, ModelViewSet):
     model = DocumentType
 
     queryset = DocumentType.objects.select_related("owner").order_by(Lower("name"))
@@ -817,7 +808,7 @@ class DocumentViewSet(
     UpdateModelMixin,
     DestroyModelMixin,
     ListModelMixin,
-    GenericViewSet[Document],
+    GenericViewSet,
 ):
     model = Document
     queryset = Document.objects.all()
@@ -1257,10 +1248,7 @@ class DocumentViewSet(
         ),
     )
     def suggestions(self, request, pk=None):
-        doc = get_object_or_404(
-            Document.objects.select_related("owner").prefetch_related("versions"),
-            pk=pk,
-        )
+        doc = get_object_or_404(Document.objects.select_related("owner"), pk=pk)
         if request.user is not None and not has_perms_owner_aware(
             request.user,
             "view_document",
@@ -1964,7 +1952,7 @@ class ChatStreamingSerializer(serializers.Serializer):
     ],
     name="dispatch",
 )
-class ChatStreamingView(GenericAPIView[Any]):
+class ChatStreamingView(GenericAPIView):
     permission_classes = (IsAuthenticated,)
     serializer_class = ChatStreamingSerializer
 
@@ -2290,7 +2278,7 @@ class LogViewSet(ViewSet):
 
 
 @extend_schema_view(**generate_object_with_permissions_schema(SavedViewSerializer))
-class SavedViewViewSet(BulkPermissionMixin, PassUserMixin, ModelViewSet[SavedView]):
+class SavedViewViewSet(BulkPermissionMixin, PassUserMixin, ModelViewSet):
     model = SavedView
 
     queryset = SavedView.objects.select_related("owner").prefetch_related(
@@ -2768,7 +2756,7 @@ class RemovePasswordDocumentsView(DocumentOperationPermissionMixin):
         },
     ),
 )
-class PostDocumentView(GenericAPIView[Any]):
+class PostDocumentView(GenericAPIView):
     permission_classes = (IsAuthenticated,)
     serializer_class = PostDocumentSerializer
     parser_classes = (parsers.MultiPartParser,)
@@ -2889,7 +2877,7 @@ class PostDocumentView(GenericAPIView[Any]):
         },
     ),
 )
-class SelectionDataView(GenericAPIView[Any]):
+class SelectionDataView(GenericAPIView):
     permission_classes = (IsAuthenticated,)
     serializer_class = DocumentListSerializer
     parser_classes = (parsers.MultiPartParser, parsers.JSONParser)
@@ -2993,7 +2981,7 @@ class SelectionDataView(GenericAPIView[Any]):
         },
     ),
 )
-class SearchAutoCompleteView(GenericAPIView[Any]):
+class SearchAutoCompleteView(GenericAPIView):
     permission_classes = (IsAuthenticated,)
 
     def get(self, request, format=None):
@@ -3274,7 +3262,7 @@ class GlobalSearchView(PassUserMixin):
         },
     ),
 )
-class StatisticsView(GenericAPIView[Any]):
+class StatisticsView(GenericAPIView):
     permission_classes = (IsAuthenticated,)
 
     def get(self, request, format=None):
@@ -3376,7 +3364,7 @@ class StatisticsView(GenericAPIView[Any]):
         )
 
 
-class BulkDownloadView(DocumentSelectionMixin, GenericAPIView[Any]):
+class BulkDownloadView(DocumentSelectionMixin, GenericAPIView):
     permission_classes = (IsAuthenticated,)
     serializer_class = BulkDownloadSerializer
     parser_classes = (parsers.JSONParser,)
@@ -3429,7 +3417,7 @@ class BulkDownloadView(DocumentSelectionMixin, GenericAPIView[Any]):
 
 
 @extend_schema_view(**generate_object_with_permissions_schema(StoragePathSerializer))
-class StoragePathViewSet(PermissionsAwareDocumentCountMixin, ModelViewSet[StoragePath]):
+class StoragePathViewSet(PermissionsAwareDocumentCountMixin, ModelViewSet):
     model = StoragePath
 
     queryset = StoragePath.objects.select_related("owner").order_by(
@@ -3493,7 +3481,7 @@ class StoragePathViewSet(PermissionsAwareDocumentCountMixin, ModelViewSet[Storag
         return Response(result)
 
 
-class UiSettingsView(GenericAPIView[Any]):
+class UiSettingsView(GenericAPIView):
     queryset = UiSettings.objects.all()
     permission_classes = (IsAuthenticated, PaperlessObjectPermissions)
     serializer_class = UiSettingsViewSerializer
@@ -3591,7 +3579,7 @@ class UiSettingsView(GenericAPIView[Any]):
         },
     ),
 )
-class RemoteVersionView(GenericAPIView[Any]):
+class RemoteVersionView(GenericAPIView):
     cache_key = "remote_version_view_latest_release"
 
     def get(self, request, format=None):
@@ -3668,52 +3656,37 @@ class RemoteVersionView(GenericAPIView[Any]):
         ),
     ],
 )
-class TasksViewSet(ReadOnlyModelViewSet[PaperlessTask]):
+class TasksViewSet(ReadOnlyModelViewSet):
     permission_classes = (IsAuthenticated, PaperlessObjectPermissions)
+    serializer_class = TasksViewSerializer
     filter_backends = (
         DjangoFilterBackend,
         OrderingFilter,
         ObjectOwnedOrGrantedPermissionsFilter,
     )
     filterset_class = PaperlessTaskFilterSet
-    ordering_fields = [
-        "date_created",
-        "date_done",
-        "status",
-        "task_type",
-        "duration_seconds",
-        "wait_time_seconds",
-    ]
-    ordering = ["-date_created"]
 
-    def get_serializer_class(self):
-        # v9: use backwards-compatible serializer with old field names
-        if self.request.version and int(self.request.version) < 10:
-            return TaskSerializerV9
-        return TaskSerializerV10
+    TASK_AND_ARGS_BY_NAME = {
+        PaperlessTask.TaskName.INDEX_OPTIMIZE: (index_optimize, {}),
+        PaperlessTask.TaskName.TRAIN_CLASSIFIER: (
+            train_classifier,
+            {"scheduled": False},
+        ),
+        PaperlessTask.TaskName.CHECK_SANITY: (
+            sanity_check,
+            {"scheduled": False, "raise_on_error": False},
+        ),
+        PaperlessTask.TaskName.LLMINDEX_UPDATE: (
+            llmindex_index,
+            {"scheduled": False, "rebuild": False},
+        ),
+    }
 
     def get_queryset(self):
-        queryset = PaperlessTask.objects.all()
-        # v9 backwards compat: map old query params to new field names
-        if self.request.version and int(self.request.version) < 10:
-            task_name = self.request.query_params.get("task_name")
-            if task_name is not None:
-                queryset = queryset.filter(task_type=task_name)
-            task_type_old = self.request.query_params.get("type")
-            if task_type_old is not None:
-                # Old type values: AUTO_TASK -> SYSTEM, SCHEDULED_TASK -> SCHEDULED, MANUAL_TASK -> MANUAL
-                old_to_new = {
-                    "AUTO_TASK": PaperlessTask.TriggerSource.SYSTEM,
-                    "SCHEDULED_TASK": PaperlessTask.TriggerSource.SCHEDULED,
-                    "MANUAL_TASK": PaperlessTask.TriggerSource.MANUAL,
-                }
-                new_source = old_to_new.get(task_type_old)
-                if new_source:
-                    queryset = queryset.filter(trigger_source=new_source)
-        # v10+: direct task_id param for backwards compat
+        queryset = PaperlessTask.objects.all().order_by("-date_created")
         task_id = self.request.query_params.get("task_id")
         if task_id is not None:
-            queryset = queryset.filter(task_id=task_id)
+            queryset = PaperlessTask.objects.filter(task_id=task_id)
         return queryset
 
     @action(
@@ -3725,123 +3698,39 @@ class TasksViewSet(ReadOnlyModelViewSet[PaperlessTask]):
         serializer = AcknowledgeTasksViewSerializer(data=request.data)
         serializer.is_valid(raise_exception=True)
         task_ids = serializer.validated_data.get("tasks")
-        tasks = self.get_queryset().filter(id__in=task_ids)
-        count = tasks.update(acknowledged=True)
-        return Response({"result": count})
-
-    @action(
-        methods=["post"],
-        detail=False,
-        permission_classes=[IsAuthenticated, AcknowledgeTasksPermissions],
-    )
-    def acknowledge_all(self, request):
-        """Acknowledge all completed tasks visible to the requesting user."""
-        count = (
-            self.get_queryset()
-            .filter(
-                acknowledged=False,
-                status__in=[
-                    PaperlessTask.Status.SUCCESS,
-                    PaperlessTask.Status.FAILURE,
-                    PaperlessTask.Status.REVOKED,
-                ],
-            )
-            .update(acknowledged=True)
-        )
-        return Response({"result": count})
-
-    @action(methods=["get"], detail=False)
-    def summary(self, request):
-        """Aggregated task statistics per task_type over the last N days (default 30)."""
-        from django.db.models import Avg
-        from django.db.models import Count
-        from django.db.models import Max
-        from django.db.models import Q
-
-        days = int(request.query_params.get("days", 30))
-        cutoff = timezone.now() - timedelta(days=days)
-        queryset = self.get_queryset().filter(date_created__gte=cutoff)
-
-        data = queryset.values("task_type").annotate(
-            total_count=Count("id"),
-            pending_count=Count("id", filter=Q(status=PaperlessTask.Status.PENDING)),
-            success_count=Count("id", filter=Q(status=PaperlessTask.Status.SUCCESS)),
-            failure_count=Count("id", filter=Q(status=PaperlessTask.Status.FAILURE)),
-            avg_duration_seconds=Avg(
-                "duration_seconds",
-                filter=Q(duration_seconds__isnull=False),
-            ),
-            avg_wait_time_seconds=Avg(
-                "wait_time_seconds",
-                filter=Q(wait_time_seconds__isnull=False),
-            ),
-            last_run=Max("date_created"),
-            last_success=Max(
-                "date_done",
-                filter=Q(status=PaperlessTask.Status.SUCCESS),
-            ),
-            last_failure=Max(
-                "date_done",
-                filter=Q(status=PaperlessTask.Status.FAILURE),
-            ),
-        )
-        serializer = TaskSummarySerializer(data, many=True)
-        return Response(serializer.data)
-
-    @action(methods=["get"], detail=False)
-    def active(self, request):
-        """Currently pending and running tasks (capped at 50)."""
-        queryset = (
-            self.get_queryset()
-            .filter(
-                status__in=[PaperlessTask.Status.PENDING, PaperlessTask.Status.STARTED],
-            )
-            .order_by("-date_created")[:50]
-        )
-        serializer = self.get_serializer(queryset, many=True)
-        return Response(serializer.data)
-
-    @action(methods=["post"], detail=False, permission_classes=[IsAdminUser])
-    def run(self, request):
-        """Manually dispatch a background task. Superuser (admin) only."""
-        serializer = RunTaskSerializer(data=request.data)
-        serializer.is_valid(raise_exception=True)
-        task_type = serializer.validated_data.get("task_type")
-
-        task_func_map = {
-            PaperlessTask.TaskType.INDEX_OPTIMIZE: (index_optimize, {}),
-            PaperlessTask.TaskType.TRAIN_CLASSIFIER: (train_classifier, {}),
-            PaperlessTask.TaskType.SANITY_CHECK: (
-                sanity_check,
-                {"raise_on_error": False},
-            ),
-            PaperlessTask.TaskType.LLM_INDEX: (
-                llmindex_index,
-                {"rebuild": False},
-            ),
-        }
-
-        if task_type not in task_func_map:
-            return Response(
-                {"error": f"Task type '{task_type}' cannot be manually triggered"},
-                status=status.HTTP_400_BAD_REQUEST,
-            )
 
         try:
-            task_func, task_kwargs = task_func_map[task_type]
-            async_result = task_func.apply_async(
-                kwargs=task_kwargs,
-                headers={"trigger_source": "manual"},
+            tasks = PaperlessTask.objects.filter(id__in=task_ids)
+            if request.user is not None and not request.user.is_superuser:
+                tasks = tasks.filter(owner=request.user) | tasks.filter(owner=None)
+            result = tasks.update(
+                acknowledged=True,
             )
-            return Response({"task_id": async_result.id})
+            return Response({"result": result})
+        except Exception:
+            return HttpResponseBadRequest()
+
+    @action(methods=["post"], detail=False)
+    def run(self, request):
+        serializer = RunTaskViewSerializer(data=request.data)
+        serializer.is_valid(raise_exception=True)
+        task_name = serializer.validated_data.get("task_name")
+
+        if not request.user.is_superuser:
+            return HttpResponseForbidden("Insufficient permissions")
+
+        try:
+            task_func, task_args = self.TASK_AND_ARGS_BY_NAME[task_name]
+            result = task_func(**task_args)
+            return Response({"result": result})
         except Exception as e:
-            logger.warning(f"Error running task: {e!s}")
+            logger.warning(f"An error occurred running task: {e!s}")
             return HttpResponseServerError(
                 "Error running task, check logs for more detail.",
             )
 
 
-class ShareLinkViewSet(PassUserMixin, ModelViewSet[ShareLink]):
+class ShareLinkViewSet(ModelViewSet, PassUserMixin):
     model = ShareLink
 
     queryset = ShareLink.objects.all()
@@ -3858,7 +3747,7 @@ class ShareLinkViewSet(PassUserMixin, ModelViewSet[ShareLink]):
     ordering_fields = ("created", "expiration", "document")
 
 
-class ShareLinkBundleViewSet(PassUserMixin, ModelViewSet[ShareLinkBundle]):
+class ShareLinkBundleViewSet(ModelViewSet, PassUserMixin):
     model = ShareLinkBundle
 
     queryset = ShareLinkBundle.objects.all()
@@ -4215,7 +4104,7 @@ class BulkEditObjectsView(PassUserMixin):
         return Response({"result": "OK"})
 
 
-class WorkflowTriggerViewSet(ModelViewSet[WorkflowTrigger]):
+class WorkflowTriggerViewSet(ModelViewSet):
     permission_classes = (IsAuthenticated, PaperlessObjectPermissions)
 
     serializer_class = WorkflowTriggerSerializer
@@ -4233,7 +4122,7 @@ class WorkflowTriggerViewSet(ModelViewSet[WorkflowTrigger]):
         return super().partial_update(request, *args, **kwargs)
 
 
-class WorkflowActionViewSet(ModelViewSet[WorkflowAction]):
+class WorkflowActionViewSet(ModelViewSet):
     permission_classes = (IsAuthenticated, PaperlessObjectPermissions)
 
     serializer_class = WorkflowActionSerializer
@@ -4258,7 +4147,7 @@ class WorkflowActionViewSet(ModelViewSet[WorkflowAction]):
         return super().partial_update(request, *args, **kwargs)
 
 
-class WorkflowViewSet(ModelViewSet[Workflow]):
+class WorkflowViewSet(ModelViewSet):
     permission_classes = (IsAuthenticated, PaperlessObjectPermissions)
 
     serializer_class = WorkflowSerializer
@@ -4276,7 +4165,7 @@ class WorkflowViewSet(ModelViewSet[Workflow]):
     )
 
 
-class CustomFieldViewSet(PermissionsAwareDocumentCountMixin, ModelViewSet[CustomField]):
+class CustomFieldViewSet(PermissionsAwareDocumentCountMixin, ModelViewSet):
     permission_classes = (IsAuthenticated, PaperlessObjectPermissions)
 
     serializer_class = CustomFieldSerializer
@@ -4460,11 +4349,11 @@ class SystemStatusView(PassUserMixin):
 
         last_trained_task = (
             PaperlessTask.objects.filter(
-                task_type=PaperlessTask.TaskType.TRAIN_CLASSIFIER,
+                task_name=PaperlessTask.TaskName.TRAIN_CLASSIFIER,
                 status__in=[
-                    PaperlessTask.Status.SUCCESS,
-                    PaperlessTask.Status.FAILURE,
-                    PaperlessTask.Status.REVOKED,
+                    states.SUCCESS,
+                    states.FAILURE,
+                    states.REVOKED,
                 ],  # ignore running tasks
             )
             .order_by("-date_done")
@@ -4475,23 +4364,20 @@ class SystemStatusView(PassUserMixin):
         if last_trained_task is None:
             classifier_status = "WARNING"
             classifier_error = "No classifier training tasks found"
-        elif (
-            last_trained_task
-            and last_trained_task.status != PaperlessTask.Status.SUCCESS
-        ):
+        elif last_trained_task and last_trained_task.status != states.SUCCESS:
             classifier_status = "ERROR"
-            classifier_error = last_trained_task.result_message
+            classifier_error = last_trained_task.result
         classifier_last_trained = (
             last_trained_task.date_done if last_trained_task else None
         )
 
         last_sanity_check = (
             PaperlessTask.objects.filter(
-                task_type=PaperlessTask.TaskType.SANITY_CHECK,
+                task_name=PaperlessTask.TaskName.CHECK_SANITY,
                 status__in=[
-                    PaperlessTask.Status.SUCCESS,
-                    PaperlessTask.Status.FAILURE,
-                    PaperlessTask.Status.REVOKED,
+                    states.SUCCESS,
+                    states.FAILURE,
+                    states.REVOKED,
                 ],  # ignore running tasks
             )
             .order_by("-date_done")
@@ -4502,12 +4388,9 @@ class SystemStatusView(PassUserMixin):
         if last_sanity_check is None:
             sanity_check_status = "WARNING"
             sanity_check_error = "No sanity check tasks found"
-        elif (
-            last_sanity_check
-            and last_sanity_check.status != PaperlessTask.Status.SUCCESS
-        ):
+        elif last_sanity_check and last_sanity_check.status != states.SUCCESS:
             sanity_check_status = "ERROR"
-            sanity_check_error = last_sanity_check.result_message
+            sanity_check_error = last_sanity_check.result
         sanity_check_last_run = (
             last_sanity_check.date_done if last_sanity_check else None
         )
@@ -4520,7 +4403,7 @@ class SystemStatusView(PassUserMixin):
         else:
             last_llmindex_update = (
                 PaperlessTask.objects.filter(
-                    task_type=PaperlessTask.TaskType.LLM_INDEX,
+                    task_name=PaperlessTask.TaskName.LLMINDEX_UPDATE,
                 )
                 .order_by("-date_done")
                 .first()
@@ -4530,12 +4413,9 @@ class SystemStatusView(PassUserMixin):
             if last_llmindex_update is None:
                 llmindex_status = "WARNING"
                 llmindex_error = "No LLM index update tasks found"
-            elif (
-                last_llmindex_update
-                and last_llmindex_update.status == PaperlessTask.Status.FAILURE
-            ):
+            elif last_llmindex_update and last_llmindex_update.status == states.FAILURE:
                 llmindex_status = "ERROR"
-                llmindex_error = last_llmindex_update.result_message
+                llmindex_error = last_llmindex_update.result
             llmindex_last_modified = (
                 last_llmindex_update.date_done if last_llmindex_update else None
             )

src/paperless/serialisers.py

@@ -74,7 +74,7 @@ class PaperlessAuthTokenSerializer(AuthTokenSerializer):
         return attrs
 
 
-class UserSerializer(PasswordValidationMixin, serializers.ModelSerializer[User]):
+class UserSerializer(PasswordValidationMixin, serializers.ModelSerializer):
     password = ObfuscatedPasswordField(required=False)
     user_permissions = serializers.SlugRelatedField(
         many=True,
@@ -142,7 +142,7 @@ class UserSerializer(PasswordValidationMixin, serializers.ModelSerializer[User])
         return user
 
 
-class GroupSerializer(serializers.ModelSerializer[Group]):
+class GroupSerializer(serializers.ModelSerializer):
     permissions = serializers.SlugRelatedField(
         many=True,
         queryset=Permission.objects.exclude(content_type__app_label="admin"),
@@ -158,7 +158,7 @@ class GroupSerializer(serializers.ModelSerializer[Group]):
         )
 
 
-class SocialAccountSerializer(serializers.ModelSerializer[SocialAccount]):
+class SocialAccountSerializer(serializers.ModelSerializer):
     name = serializers.SerializerMethodField()
 
     class Meta:
@@ -176,7 +176,7 @@ class SocialAccountSerializer(serializers.ModelSerializer[SocialAccount]):
             return "Unknown App"
 
 
-class ProfileSerializer(PasswordValidationMixin, serializers.ModelSerializer[User]):
+class ProfileSerializer(PasswordValidationMixin, serializers.ModelSerializer):
     email = serializers.EmailField(allow_blank=True, required=False)
     password = ObfuscatedPasswordField(required=False, allow_null=False)
     auth_token = serializers.SlugRelatedField(read_only=True, slug_field="key")
@@ -209,9 +209,7 @@ class ProfileSerializer(PasswordValidationMixin, serializers.ModelSerializer[Use
         )
 
 
-class ApplicationConfigurationSerializer(
-    serializers.ModelSerializer[ApplicationConfiguration],
-):
+class ApplicationConfigurationSerializer(serializers.ModelSerializer):
     user_args = serializers.JSONField(binary=True, allow_null=True)
     barcode_tag_mapping = serializers.JSONField(binary=True, allow_null=True)
     llm_api_key = ObfuscatedPasswordField(

src/paperless/settings/__init__.py

@@ -133,6 +133,7 @@ INSTALLED_APPS = [
     "rest_framework",
     "rest_framework.authtoken",
     "django_filters",
+    "django_celery_results",
     "guardian",
     "allauth",
     "allauth.account",
@@ -668,6 +669,8 @@ CELERY_BROKER_TRANSPORT_OPTIONS = {
 CELERY_TASK_TRACK_STARTED = True
 CELERY_TASK_TIME_LIMIT: Final[int] = get_int_from_env("PAPERLESS_WORKER_TIMEOUT", 1800)
 
+CELERY_RESULT_EXTENDED = True
+CELERY_RESULT_BACKEND = "django-db"
 CELERY_CACHE_BACKEND = "default"
 
 # https://docs.celeryq.dev/en/stable/userguide/configuration.html#task-serializer

src/paperless/settings/custom.py

@@ -181,10 +181,7 @@ def parse_beat_schedule() -> dict:
         schedule[task["name"]] = {
             "task": task["task"],
             "schedule": crontab(minute, hour, day_week, day_month, month),
-            "options": {
-                **task["options"],
-                "headers": {"trigger_source": "scheduled"},
-            },
+            "options": task["options"],
         }
 
     return schedule

src/paperless/tests/settings/test_custom_parsers.py

@@ -186,66 +186,42 @@ def make_expected_schedule(
         "Check all e-mail accounts": {
             "task": "paperless_mail.tasks.process_mail_accounts",
             "schedule": crontab(minute="*/10"),
-            "options": {
-                "expires": mail_expire,
-                "headers": {"trigger_source": "scheduled"},
-            },
+            "options": {"expires": mail_expire},
         },
         "Train the classifier": {
             "task": "documents.tasks.train_classifier",
             "schedule": crontab(minute="5", hour="*/1"),
-            "options": {
-                "expires": classifier_expire,
-                "headers": {"trigger_source": "scheduled"},
-            },
+            "options": {"expires": classifier_expire},
         },
         "Optimize the index": {
             "task": "documents.tasks.index_optimize",
             "schedule": crontab(minute=0, hour=0),
-            "options": {
-                "expires": index_expire,
-                "headers": {"trigger_source": "scheduled"},
-            },
+            "options": {"expires": index_expire},
         },
         "Perform sanity check": {
             "task": "documents.tasks.sanity_check",
             "schedule": crontab(minute=30, hour=0, day_of_week="sun"),
-            "options": {
-                "expires": sanity_expire,
-                "headers": {"trigger_source": "scheduled"},
-            },
+            "options": {"expires": sanity_expire},
         },
         "Empty trash": {
             "task": "documents.tasks.empty_trash",
             "schedule": crontab(minute=0, hour="1"),
-            "options": {
-                "expires": empty_trash_expire,
-                "headers": {"trigger_source": "scheduled"},
-            },
+            "options": {"expires": empty_trash_expire},
         },
         "Check and run scheduled workflows": {
             "task": "documents.tasks.check_scheduled_workflows",
             "schedule": crontab(minute="5", hour="*/1"),
-            "options": {
-                "expires": workflow_expire,
-                "headers": {"trigger_source": "scheduled"},
-            },
+            "options": {"expires": workflow_expire},
         },
         "Rebuild LLM index": {
             "task": "documents.tasks.llmindex_index",
             "schedule": crontab(minute="10", hour="2"),
-            "options": {
-                "expires": llm_index_expire,
-                "headers": {"trigger_source": "scheduled"},
-            },
+            "options": {"expires": llm_index_expire},
         },
         "Cleanup expired share link bundles": {
             "task": "documents.tasks.cleanup_expired_share_link_bundles",
             "schedule": crontab(minute=0, hour="2"),
-            "options": {
-                "expires": share_link_cleanup_expire,
-                "headers": {"trigger_source": "scheduled"},
-            },
+            "options": {"expires": share_link_cleanup_expire},
         },
     }
 
@@ -308,16 +284,6 @@ class TestParseBeatSchedule:
         schedule = parse_beat_schedule()
         assert schedule == expected
 
-    def test_parse_beat_schedule_all_entries_have_trigger_source_header(self) -> None:
-        """Every beat entry must carry trigger_source=scheduled so the task signal
-        handler can identify scheduler-originated tasks."""
-        schedule = parse_beat_schedule()
-        for name, entry in schedule.items():
-            headers = entry.get("options", {}).get("headers", {})
-            assert headers.get("trigger_source") == "scheduled", (
-                f"Beat entry '{name}' is missing trigger_source header"
-            )
-
 
 class TestParseDbSettings:
     """Test suite for parse_db_settings function."""

src/paperless/views.py

@@ -1,6 +1,5 @@
 from collections import OrderedDict
 from pathlib import Path
-from typing import Any
 
 from allauth.mfa import signals
 from allauth.mfa.adapter import get_adapter as get_mfa_adapter
@@ -115,7 +114,7 @@ class FaviconView(View):
             return HttpResponseNotFound("favicon.ico not found")
 
 
-class UserViewSet(ModelViewSet[User]):
+class UserViewSet(ModelViewSet):
     _BOOL_NOT_PROVIDED = object()
     model = User
 
@@ -217,7 +216,7 @@ class UserViewSet(ModelViewSet[User]):
             return HttpResponseNotFound("TOTP not found")
 
 
-class GroupViewSet(ModelViewSet[Group]):
+class GroupViewSet(ModelViewSet):
     model = Group
 
     queryset = Group.objects.order_by(Lower("name"))
@@ -230,7 +229,7 @@ class GroupViewSet(ModelViewSet[Group]):
     ordering_fields = ("name",)
 
 
-class ProfileView(GenericAPIView[Any]):
+class ProfileView(GenericAPIView):
     """
     User profile view, only available when logged in
     """
@@ -289,7 +288,7 @@ class ProfileView(GenericAPIView[Any]):
         },
     ),
 )
-class TOTPView(GenericAPIView[Any]):
+class TOTPView(GenericAPIView):
     """
     TOTP views
     """
@@ -369,7 +368,7 @@ class TOTPView(GenericAPIView[Any]):
         },
     ),
 )
-class GenerateAuthTokenView(GenericAPIView[Any]):
+class GenerateAuthTokenView(GenericAPIView):
     """
     Generates (or re-generates) an auth token, requires a logged in user
     unlike the default DRF endpoint
@@ -398,7 +397,7 @@ class GenerateAuthTokenView(GenericAPIView[Any]):
         },
     ),
 )
-class ApplicationConfigurationViewSet(ModelViewSet[ApplicationConfiguration]):
+class ApplicationConfigurationViewSet(ModelViewSet):
     model = ApplicationConfiguration
 
     queryset = ApplicationConfiguration.objects
@@ -427,9 +426,10 @@ class ApplicationConfigurationViewSet(ModelViewSet[ApplicationConfiguration]):
             and not vector_store_file_exists()
         ):
             # AI index was just enabled and vector store file does not exist
-            llmindex_index.apply_async(
-                kwargs={"rebuild": True},
-                headers={"trigger_source": "system"},
+            llmindex_index.delay(
+                rebuild=True,
+                scheduled=False,
+                auto=True,
             )
 
 
@@ -450,7 +450,7 @@ class ApplicationConfigurationViewSet(ModelViewSet[ApplicationConfiguration]):
         },
     ),
 )
-class DisconnectSocialAccountView(GenericAPIView[Any]):
+class DisconnectSocialAccountView(GenericAPIView):
     """
     Disconnects a social account provider from the user account
     """
@@ -476,7 +476,7 @@ class DisconnectSocialAccountView(GenericAPIView[Any]):
         },
     ),
 )
-class SocialAccountProvidersView(GenericAPIView[Any]):
+class SocialAccountProvidersView(GenericAPIView):
     """
     List of social account providers
     """

src/paperless_ai/indexing.py

@@ -4,6 +4,7 @@ from datetime import timedelta
 from pathlib import Path
 from typing import TYPE_CHECKING
 
+from celery import states
 from django.conf import settings
 from django.utils import timezone
 
@@ -27,20 +28,17 @@ def queue_llm_index_update_if_needed(*, rebuild: bool, reason: str) -> bool:
     from documents.tasks import llmindex_index
 
     has_running = PaperlessTask.objects.filter(
-        task_type=PaperlessTask.TaskType.LLM_INDEX,
-        status__in=[PaperlessTask.Status.PENDING, PaperlessTask.Status.STARTED],
+        task_name=PaperlessTask.TaskName.LLMINDEX_UPDATE,
+        status__in=[states.PENDING, states.STARTED],
     ).exists()
     has_recent = PaperlessTask.objects.filter(
-        task_type=PaperlessTask.TaskType.LLM_INDEX,
+        task_name=PaperlessTask.TaskName.LLMINDEX_UPDATE,
         date_created__gte=(timezone.now() - timedelta(minutes=5)),
     ).exists()
     if has_running or has_recent:
         return False
 
-    llmindex_index.apply_async(
-        kwargs={"rebuild": rebuild},
-        headers={"trigger_source": "system"},
-    )
+    llmindex_index.delay(rebuild=rebuild, scheduled=False, auto=True)
     logger.warning(
         "Queued LLM index update%s: %s",
         " (rebuild)" if rebuild else "",

src/paperless_ai/tests/test_ai_indexing.py

@@ -3,13 +3,13 @@ from unittest.mock import MagicMock
 from unittest.mock import patch
 
 import pytest
+from celery import states
 from django.test import override_settings
 from django.utils import timezone
 from llama_index.core.base.embeddings.base import BaseEmbedding
 
 from documents.models import Document
 from documents.models import PaperlessTask
-from documents.tests.factories import PaperlessTaskFactory
 from paperless_ai import indexing
 
 
@@ -292,15 +292,13 @@ def test_queue_llm_index_update_if_needed_enqueues_when_idle_or_skips_recent() -
         )
 
     assert result is True
-    mock_task.apply_async.assert_called_once_with(
-        kwargs={"rebuild": True},
-        headers={"trigger_source": "system"},
-    )
+    mock_task.delay.assert_called_once_with(rebuild=True, scheduled=False, auto=True)
 
-    PaperlessTaskFactory(
-        task_type=PaperlessTask.TaskType.LLM_INDEX,
-        trigger_source=PaperlessTask.TriggerSource.SYSTEM,
-        status=PaperlessTask.Status.STARTED,
+    PaperlessTask.objects.create(
+        task_id="task-1",
+        task_name=PaperlessTask.TaskName.LLMINDEX_UPDATE,
+        status=states.STARTED,
+        date_created=timezone.now(),
     )
 
     # Existing running task
@@ -311,7 +309,7 @@ def test_queue_llm_index_update_if_needed_enqueues_when_idle_or_skips_recent() -
         )
 
     assert result is False
-    mock_task.apply_async.assert_not_called()
+    mock_task.delay.assert_not_called()
 
 
 @override_settings(

src/paperless_mail/serialisers.py

@@ -57,7 +57,7 @@ class MailAccountSerializer(OwnedObjectSerializer):
         return instance
 
 
-class AccountField(serializers.PrimaryKeyRelatedField[MailAccount]):
+class AccountField(serializers.PrimaryKeyRelatedField):
     def get_queryset(self):
         return MailAccount.objects.all().order_by("-id")
 

src/paperless_mail/views.py

@@ -1,7 +1,6 @@
 import datetime
 import logging
 from datetime import timedelta
-from typing import Any
 
 from django.http import HttpResponseBadRequest
 from django.http import HttpResponseForbidden
@@ -66,7 +65,7 @@ from paperless_mail.tasks import process_mail_accounts
         },
     ),
 )
-class MailAccountViewSet(PassUserMixin, ModelViewSet[MailAccount]):
+class MailAccountViewSet(ModelViewSet, PassUserMixin):
     model = MailAccount
 
     queryset = MailAccount.objects.all().order_by("pk")
@@ -160,7 +159,7 @@ class MailAccountViewSet(PassUserMixin, ModelViewSet[MailAccount]):
         return Response({"result": "OK"})
 
 
-class ProcessedMailViewSet(PassUserMixin, ReadOnlyModelViewSet[ProcessedMail]):
+class ProcessedMailViewSet(ReadOnlyModelViewSet, PassUserMixin):
     permission_classes = (IsAuthenticated, PaperlessObjectPermissions)
     serializer_class = ProcessedMailSerializer
     pagination_class = StandardPagination
@@ -188,7 +187,7 @@ class ProcessedMailViewSet(PassUserMixin, ReadOnlyModelViewSet[ProcessedMail]):
         return Response({"result": "OK", "deleted_mail_ids": mail_ids})
 
 
-class MailRuleViewSet(PassUserMixin, ModelViewSet[MailRule]):
+class MailRuleViewSet(ModelViewSet, PassUserMixin):
     model = MailRule
 
     queryset = MailRule.objects.all().order_by("order")
@@ -204,7 +203,7 @@ class MailRuleViewSet(PassUserMixin, ModelViewSet[MailRule]):
         responses={200: None},
     ),
 )
-class OauthCallbackView(GenericAPIView[Any]):
+class OauthCallbackView(GenericAPIView):
     permission_classes = (IsAuthenticated,)
 
     def get(self, request, format=None):

test_backend_profile.py

@@ -0,0 +1,346 @@
+# ruff: noqa: T201
+"""
+cProfile-based search pipeline profiling with a 20k-document dataset.
+
+Run with:
+    uv run pytest ../test_backend_profile.py \
+        -m profiling --override-ini="addopts=" -s -v
+
+Each scenario prints:
+  - Wall time for the operation
+  - cProfile stats sorted by cumulative time (top 25 callers)
+
+This is a developer tool, not a correctness test.  Nothing here should
+fail unless the code is broken.
+"""
+
+from __future__ import annotations
+
+import random
+import time
+from typing import TYPE_CHECKING
+
+import pytest
+from profiling import profile_cpu
+
+from documents.models import Document
+from documents.search._backend import TantivyBackend
+from documents.search._backend import reset_backend
+
+if TYPE_CHECKING:
+    from pathlib import Path
+
+# transaction=False (default): tests roll back, but the module-scoped fixture
+# commits its data outside the test transaction so it remains visible throughout.
+pytestmark = [pytest.mark.profiling, pytest.mark.django_db]
+
+# ---------------------------------------------------------------------------
+# Dataset constants
+# ---------------------------------------------------------------------------
+NUM_DOCS = 20_000
+SEED = 42
+
+# Terms and their approximate match rates across the corpus.
+# "rechnung"    -> ~70% of docs  (~14 000)
+# "mahnung"     -> ~20% of docs  (~4 000)
+# "kontonummer" -> ~5%  of docs  (~1 000)
+# "rarewort"    -> ~1%  of docs  (~200)
+COMMON_TERM = "rechnung"
+MEDIUM_TERM = "mahnung"
+RARE_TERM = "kontonummer"
+VERY_RARE_TERM = "rarewort"
+
+PAGE_SIZE = 25
+
+
+# ---------------------------------------------------------------------------
+# Helpers
+# ---------------------------------------------------------------------------
+
+_FILLER_WORDS = [
+    "dokument",  # codespell:ignore
+    "seite",
+    "datum",
+    "betrag",
+    "nummer",
+    "konto",
+    "firma",
+    "vertrag",
+    "lieferant",
+    "bestellung",
+    "steuer",
+    "mwst",
+    "leistung",
+    "auftrag",
+    "zahlung",
+]
+
+
+def _build_content(rng: random.Random) -> str:
+    """Return a short paragraph with terms embedded at the desired rates."""
+    words = rng.choices(_FILLER_WORDS, k=15)
+    if rng.random() < 0.70:
+        words.append(COMMON_TERM)
+    if rng.random() < 0.20:
+        words.append(MEDIUM_TERM)
+    if rng.random() < 0.05:
+        words.append(RARE_TERM)
+    if rng.random() < 0.01:
+        words.append(VERY_RARE_TERM)
+    rng.shuffle(words)
+    return " ".join(words)
+
+
+def _time(fn, *, label: str, runs: int = 3):
+    """Run *fn()* several times and report min/avg/max wall time (no cProfile)."""
+    times = []
+    result = None
+    for _ in range(runs):
+        t0 = time.perf_counter()
+        result = fn()
+        times.append(time.perf_counter() - t0)
+    mn, avg, mx = min(times), sum(times) / len(times), max(times)
+    print(
+        f"  {label}: min={mn * 1000:.1f}ms  avg={avg * 1000:.1f}ms  max={mx * 1000:.1f}ms  (n={runs})",
+    )
+    return result
+
+
+# ---------------------------------------------------------------------------
+# Fixtures
+# ---------------------------------------------------------------------------
+
+
+@pytest.fixture(scope="module")
+def module_db(django_db_setup, django_db_blocker):
+    """Unlock the DB for the whole module (module-scoped)."""
+    with django_db_blocker.unblock():
+        yield
+
+
+@pytest.fixture(scope="module")
+def large_backend(tmp_path_factory, module_db) -> TantivyBackend:
+    """
+    Build a 20 000-document DB + on-disk Tantivy index, shared across all
+    profiling scenarios in this module.  Teardown deletes all documents.
+    """
+    index_path: Path = tmp_path_factory.mktemp("tantivy_profile")
+
+    # ---- 1. Bulk-create Document rows ----------------------------------------
+    rng = random.Random(SEED)
+    docs = [
+        Document(
+            title=f"Document {i:05d}",
+            content=_build_content(rng),
+            checksum=f"{i:064x}",
+            pk=i + 1,
+        )
+        for i in range(NUM_DOCS)
+    ]
+    t0 = time.perf_counter()
+    Document.objects.bulk_create(docs, batch_size=1_000)
+    db_time = time.perf_counter() - t0
+    print(f"\n[setup] bulk_create {NUM_DOCS} docs: {db_time:.2f}s")
+
+    # ---- 2. Build Tantivy index -----------------------------------------------
+    backend = TantivyBackend(path=index_path)
+    backend.open()
+
+    t0 = time.perf_counter()
+    with backend.batch_update() as batch:
+        for doc in Document.objects.iterator(chunk_size=500):
+            batch.add_or_update(doc)
+    idx_time = time.perf_counter() - t0
+    print(f"[setup] index {NUM_DOCS} docs: {idx_time:.2f}s")
+
+    # ---- 3. Report corpus stats -----------------------------------------------
+    for term in (COMMON_TERM, MEDIUM_TERM, RARE_TERM, VERY_RARE_TERM):
+        count = len(backend.search_ids(term, user=None))
+        print(f"[setup]   '{term}' -> {count} hits")
+
+    yield backend
+
+    # ---- Teardown ------------------------------------------------------------
+    backend.close()
+    reset_backend()
+    Document.objects.all().delete()
+
+
+# ---------------------------------------------------------------------------
+# Profiling tests — each scenario is a separate function so pytest can run
+# them individually or all together with -m profiling.
+# ---------------------------------------------------------------------------
+
+
+class TestSearchIdsProfile:
+    """Profile backend.search_ids() — pure Tantivy, no DB."""
+
+    def test_search_ids_large(self, large_backend: TantivyBackend):
+        """~14 000 hits: how long does Tantivy take to collect all IDs?"""
+        profile_cpu(
+            lambda: large_backend.search_ids(COMMON_TERM, user=None),
+            label=f"search_ids('{COMMON_TERM}')  [large result set ~14k]",
+        )
+
+    def test_search_ids_medium(self, large_backend: TantivyBackend):
+        """~4 000 hits."""
+        profile_cpu(
+            lambda: large_backend.search_ids(MEDIUM_TERM, user=None),
+            label=f"search_ids('{MEDIUM_TERM}')  [medium result set ~4k]",
+        )
+
+    def test_search_ids_rare(self, large_backend: TantivyBackend):
+        """~1 000 hits."""
+        profile_cpu(
+            lambda: large_backend.search_ids(RARE_TERM, user=None),
+            label=f"search_ids('{RARE_TERM}')  [rare result set ~1k]",
+        )
+
+
+class TestIntersectAndOrderProfile:
+    """
+    Profile the DB intersection step: filter(pk__in=search_ids).
+    This is the 'intersect_and_order' logic from views.py.
+    """
+
+    def test_intersect_large(self, large_backend: TantivyBackend):
+        """Intersect 14k Tantivy IDs with all 20k ORM-visible docs."""
+        all_ids = large_backend.search_ids(COMMON_TERM, user=None)
+        qs = Document.objects.all()
+
+        print(f"\n  Tantivy returned {len(all_ids)} IDs")
+
+        profile_cpu(
+            lambda: list(qs.filter(pk__in=all_ids).values_list("pk", flat=True)),
+            label=f"filter(pk__in={len(all_ids)} ids)  [large, use_tantivy_sort=True path]",
+        )
+
+        # Also time it a few times to get stable numbers
+        print()
+        _time(
+            lambda: list(qs.filter(pk__in=all_ids).values_list("pk", flat=True)),
+            label=f"filter(pk__in={len(all_ids)}) repeated",
+        )
+
+    def test_intersect_rare(self, large_backend: TantivyBackend):
+        """Intersect ~1k Tantivy IDs — the happy path."""
+        all_ids = large_backend.search_ids(RARE_TERM, user=None)
+        qs = Document.objects.all()
+
+        print(f"\n  Tantivy returned {len(all_ids)} IDs")
+
+        profile_cpu(
+            lambda: list(qs.filter(pk__in=all_ids).values_list("pk", flat=True)),
+            label=f"filter(pk__in={len(all_ids)} ids)  [rare, use_tantivy_sort=True path]",
+        )
+
+
+class TestHighlightHitsProfile:
+    """Profile backend.highlight_hits() — per-doc Tantivy lookups with BM25 scoring."""
+
+    def test_highlight_page1(self, large_backend: TantivyBackend):
+        """25-doc highlight for page 1 (rank_start=1)."""
+        all_ids = large_backend.search_ids(COMMON_TERM, user=None)
+        page_ids = all_ids[:PAGE_SIZE]
+
+        profile_cpu(
+            lambda: large_backend.highlight_hits(
+                COMMON_TERM,
+                page_ids,
+                rank_start=1,
+            ),
+            label=f"highlight_hits page 1  (ids {all_ids[0]}..{all_ids[PAGE_SIZE - 1]})",
+        )
+
+    def test_highlight_page_middle(self, large_backend: TantivyBackend):
+        """25-doc highlight for a mid-corpus page (rank_start=page_offset+1)."""
+        all_ids = large_backend.search_ids(COMMON_TERM, user=None)
+        mid = len(all_ids) // 2
+        page_ids = all_ids[mid : mid + PAGE_SIZE]
+        page_offset = mid
+
+        profile_cpu(
+            lambda: large_backend.highlight_hits(
+                COMMON_TERM,
+                page_ids,
+                rank_start=page_offset + 1,
+            ),
+            label=f"highlight_hits page ~{mid // PAGE_SIZE}  (offset {page_offset})",
+        )
+
+    def test_highlight_repeated(self, large_backend: TantivyBackend):
+        """Multiple runs of page-1 highlight to see variance."""
+        all_ids = large_backend.search_ids(COMMON_TERM, user=None)
+        page_ids = all_ids[:PAGE_SIZE]
+
+        print()
+        _time(
+            lambda: large_backend.highlight_hits(COMMON_TERM, page_ids, rank_start=1),
+            label="highlight_hits page 1",
+            runs=5,
+        )
+
+
+class TestFullPipelineProfile:
+    """
+    Profile the combined pipeline as it runs in views.py:
+      search_ids -> filter(pk__in) -> highlight_hits
+    """
+
+    def _run_pipeline(
+        self,
+        backend: TantivyBackend,
+        term: str,
+        page: int = 1,
+    ):
+        all_ids = backend.search_ids(term, user=None)
+        qs = Document.objects.all()
+        visible_ids = set(qs.filter(pk__in=all_ids).values_list("pk", flat=True))
+        ordered_ids = [i for i in all_ids if i in visible_ids]
+
+        page_offset = (page - 1) * PAGE_SIZE
+        page_ids = ordered_ids[page_offset : page_offset + PAGE_SIZE]
+        hits = backend.highlight_hits(
+            term,
+            page_ids,
+            rank_start=page_offset + 1,
+        )
+        return ordered_ids, hits
+
+    def test_pipeline_large_page1(self, large_backend: TantivyBackend):
+        """Full pipeline: large result set, page 1."""
+        ordered_ids, hits = profile_cpu(
+            lambda: self._run_pipeline(large_backend, COMMON_TERM, page=1),
+            label=f"full pipeline  '{COMMON_TERM}'  page 1",
+        )[0]
+        print(f"  -> {len(ordered_ids)} total results, {len(hits)} hits on page")
+
+    def test_pipeline_large_page5(self, large_backend: TantivyBackend):
+        """Full pipeline: large result set, page 5."""
+        ordered_ids, hits = profile_cpu(
+            lambda: self._run_pipeline(large_backend, COMMON_TERM, page=5),
+            label=f"full pipeline  '{COMMON_TERM}'  page 5",
+        )[0]
+        print(f"  -> {len(ordered_ids)} total results, {len(hits)} hits on page")
+
+    def test_pipeline_rare(self, large_backend: TantivyBackend):
+        """Full pipeline: rare term, page 1 (fast path)."""
+        ordered_ids, hits = profile_cpu(
+            lambda: self._run_pipeline(large_backend, RARE_TERM, page=1),
+            label=f"full pipeline  '{RARE_TERM}'  page 1",
+        )[0]
+        print(f"  -> {len(ordered_ids)} total results, {len(hits)} hits on page")
+
+    def test_pipeline_repeated(self, large_backend: TantivyBackend):
+        """Repeated runs to get stable timing (no cProfile overhead)."""
+        print()
+        for term, label in [
+            (COMMON_TERM, f"'{COMMON_TERM}' (large)"),
+            (MEDIUM_TERM, f"'{MEDIUM_TERM}' (medium)"),
+            (RARE_TERM, f"'{RARE_TERM}'  (rare)"),
+        ]:
+            _time(
+                lambda t=term: self._run_pipeline(large_backend, t, page=1),
+                label=f"full pipeline {label} page 1",
+                runs=3,
+            )

test_classifier_profile.py

@@ -0,0 +1,605 @@
+# ruff: noqa: T201
+"""
+cProfile + tracemalloc classifier profiling test.
+
+Run with:
+    uv run pytest ../test_classifier_profile.py \
+        -m profiling --override-ini="addopts=" -s -v
+
+Corpus: 5 000 documents, 40 correspondents (25 AUTO), 25 doc types (15 AUTO),
+        50 tags (30 AUTO), 20 storage paths (12 AUTO).
+
+Document content is generated with Faker for realistic base text, with a
+per-label fingerprint injected so the MLP has a real learning signal.
+
+Scenarios:
+  - train()         full corpus — memory and CPU profiles
+  - second train()  no-op path — shows cost of the skip check
+  - save()/load()   round-trip — model file size and memory cost
+  - _update_data_vectorizer_hash()   isolated hash overhead
+  - predict_*()     four independent calls per document — the 4x redundant
+                    vectorization path used by the signal handlers
+  - _vectorize()    cache-miss vs cache-hit breakdown
+
+Memory: tracemalloc (delta + peak + top-20 allocation sites).
+CPU:    cProfile sorted by cumulative time (top 30).
+"""
+
+from __future__ import annotations
+
+import random
+import time
+from typing import TYPE_CHECKING
+
+import pytest
+from django.test import override_settings
+from faker import Faker
+from profiling import measure_memory
+from profiling import profile_cpu
+
+from documents.classifier import DocumentClassifier
+from documents.models import Correspondent
+from documents.models import Document
+from documents.models import DocumentType
+from documents.models import MatchingModel
+from documents.models import StoragePath
+from documents.models import Tag
+
+if TYPE_CHECKING:
+    from pathlib import Path
+
+pytestmark = [pytest.mark.profiling, pytest.mark.django_db]
+
+# ---------------------------------------------------------------------------
+# Corpus parameters
+# ---------------------------------------------------------------------------
+
+NUM_DOCS = 5_000
+NUM_CORRESPONDENTS = 40  # first 25 are MATCH_AUTO
+NUM_DOC_TYPES = 25  # first 15 are MATCH_AUTO
+NUM_TAGS = 50  # first 30 are MATCH_AUTO
+NUM_STORAGE_PATHS = 20  # first 12 are MATCH_AUTO
+
+NUM_AUTO_CORRESPONDENTS = 25
+NUM_AUTO_DOC_TYPES = 15
+NUM_AUTO_TAGS = 30
+NUM_AUTO_STORAGE_PATHS = 12
+
+SEED = 42
+
+
+# ---------------------------------------------------------------------------
+# Content generation
+# ---------------------------------------------------------------------------
+
+
+def _make_label_fingerprint(
+    fake: Faker,
+    label_seed: int,
+    n_words: int = 6,
+) -> list[str]:
+    """
+    Generate a small set of unique-looking words to use as the learning
+    fingerprint for a label.  Each label gets its own seeded Faker so the
+    fingerprints are distinct and reproducible.
+    """
+    per_label_fake = Faker()
+    per_label_fake.seed_instance(label_seed)
+    # Mix word() and last_name() to get varied, pronounceable tokens
+    words: list[str] = []
+    while len(words) < n_words:
+        w = per_label_fake.word().lower()
+        if w not in words:
+            words.append(w)
+    return words
+
+
+def _build_fingerprints(
+    num_correspondents: int,
+    num_doc_types: int,
+    num_tags: int,
+    num_paths: int,
+) -> tuple[list[list[str]], list[list[str]], list[list[str]], list[list[str]]]:
+    """Pre-generate per-label fingerprints.  Expensive once, free to reuse."""
+    fake = Faker()
+    # Use deterministic seeds offset by type so fingerprints don't collide
+    corr_fps = [
+        _make_label_fingerprint(fake, 1_000 + i) for i in range(num_correspondents)
+    ]
+    dtype_fps = [_make_label_fingerprint(fake, 2_000 + i) for i in range(num_doc_types)]
+    tag_fps = [_make_label_fingerprint(fake, 3_000 + i) for i in range(num_tags)]
+    path_fps = [_make_label_fingerprint(fake, 4_000 + i) for i in range(num_paths)]
+    return corr_fps, dtype_fps, tag_fps, path_fps
+
+
+def _build_content(
+    fake: Faker,
+    corr_fp: list[str] | None,
+    dtype_fp: list[str] | None,
+    tag_fps: list[list[str]],
+    path_fp: list[str] | None,
+) -> str:
+    """
+    Combine a Faker paragraph (realistic base text) with per-label
+    fingerprint words so the classifier has a genuine learning signal.
+    """
+    # 3-sentence paragraph provides realistic vocabulary
+    base = fake.paragraph(nb_sentences=3)
+
+    extras: list[str] = []
+    if corr_fp:
+        extras.extend(corr_fp)
+    if dtype_fp:
+        extras.extend(dtype_fp)
+    for fp in tag_fps:
+        extras.extend(fp)
+    if path_fp:
+        extras.extend(path_fp)
+
+    if extras:
+        return base + " " + " ".join(extras)
+    return base
+
+
+# ---------------------------------------------------------------------------
+# Module-scoped corpus fixture
+# ---------------------------------------------------------------------------
+
+
+@pytest.fixture(scope="module")
+def module_db(django_db_setup, django_db_blocker):
+    """Unlock the DB for the whole module (module-scoped)."""
+    with django_db_blocker.unblock():
+        yield
+
+
+@pytest.fixture(scope="module")
+def classifier_corpus(tmp_path_factory, module_db):
+    """
+    Build the full 5 000-document corpus once for all profiling tests.
+
+    Label objects are created individually (small number), documents are
+    bulk-inserted, and tag M2M rows go through the through-table.
+
+    Yields a dict with the model path and a sample content string for
+    prediction tests.  All rows are deleted on teardown.
+    """
+    model_path: Path = tmp_path_factory.mktemp("cls_profile") / "model.pickle"
+
+    with override_settings(MODEL_FILE=model_path):
+        fake = Faker()
+        Faker.seed(SEED)
+        rng = random.Random(SEED)
+
+        # Pre-generate fingerprints for all labels
+        print("\n[setup] Generating label fingerprints...")
+        corr_fps, dtype_fps, tag_fps, path_fps = _build_fingerprints(
+            NUM_CORRESPONDENTS,
+            NUM_DOC_TYPES,
+            NUM_TAGS,
+            NUM_STORAGE_PATHS,
+        )
+
+        # -----------------------------------------------------------------
+        # 1. Create label objects
+        # -----------------------------------------------------------------
+        print(f"[setup] Creating {NUM_CORRESPONDENTS} correspondents...")
+        correspondents: list[Correspondent] = []
+        for i in range(NUM_CORRESPONDENTS):
+            algo = (
+                MatchingModel.MATCH_AUTO
+                if i < NUM_AUTO_CORRESPONDENTS
+                else MatchingModel.MATCH_NONE
+            )
+            correspondents.append(
+                Correspondent.objects.create(
+                    name=fake.company(),
+                    matching_algorithm=algo,
+                ),
+            )
+
+        print(f"[setup] Creating {NUM_DOC_TYPES} document types...")
+        doc_types: list[DocumentType] = []
+        for i in range(NUM_DOC_TYPES):
+            algo = (
+                MatchingModel.MATCH_AUTO
+                if i < NUM_AUTO_DOC_TYPES
+                else MatchingModel.MATCH_NONE
+            )
+            doc_types.append(
+                DocumentType.objects.create(
+                    name=fake.bs()[:64],
+                    matching_algorithm=algo,
+                ),
+            )
+
+        print(f"[setup] Creating {NUM_TAGS} tags...")
+        tags: list[Tag] = []
+        for i in range(NUM_TAGS):
+            algo = (
+                MatchingModel.MATCH_AUTO
+                if i < NUM_AUTO_TAGS
+                else MatchingModel.MATCH_NONE
+            )
+            tags.append(
+                Tag.objects.create(
+                    name=f"{fake.word()} {i}",
+                    matching_algorithm=algo,
+                    is_inbox_tag=False,
+                ),
+            )
+
+        print(f"[setup] Creating {NUM_STORAGE_PATHS} storage paths...")
+        storage_paths: list[StoragePath] = []
+        for i in range(NUM_STORAGE_PATHS):
+            algo = (
+                MatchingModel.MATCH_AUTO
+                if i < NUM_AUTO_STORAGE_PATHS
+                else MatchingModel.MATCH_NONE
+            )
+            storage_paths.append(
+                StoragePath.objects.create(
+                    name=fake.word(),
+                    path=f"{fake.word()}/{fake.word()}/{{title}}",
+                    matching_algorithm=algo,
+                ),
+            )
+
+        # -----------------------------------------------------------------
+        # 2. Build document rows and M2M assignments
+        # -----------------------------------------------------------------
+        print(f"[setup] Building {NUM_DOCS} document rows...")
+        doc_rows: list[Document] = []
+        doc_tag_map: list[tuple[int, int]] = []  # (doc_position, tag_index)
+
+        for i in range(NUM_DOCS):
+            corr_idx = (
+                rng.randrange(NUM_CORRESPONDENTS) if rng.random() < 0.80 else None
+            )
+            dt_idx = rng.randrange(NUM_DOC_TYPES) if rng.random() < 0.80 else None
+            sp_idx = rng.randrange(NUM_STORAGE_PATHS) if rng.random() < 0.70 else None
+
+            # 1-4 tags; most documents get at least one
+            n_tags = rng.randint(1, 4) if rng.random() < 0.85 else 0
+            assigned_tag_indices = rng.sample(range(NUM_TAGS), min(n_tags, NUM_TAGS))
+
+            content = _build_content(
+                fake,
+                corr_fp=corr_fps[corr_idx] if corr_idx is not None else None,
+                dtype_fp=dtype_fps[dt_idx] if dt_idx is not None else None,
+                tag_fps=[tag_fps[ti] for ti in assigned_tag_indices],
+                path_fp=path_fps[sp_idx] if sp_idx is not None else None,
+            )
+
+            doc_rows.append(
+                Document(
+                    title=fake.sentence(nb_words=5),
+                    content=content,
+                    checksum=f"{i:064x}",
+                    correspondent=correspondents[corr_idx]
+                    if corr_idx is not None
+                    else None,
+                    document_type=doc_types[dt_idx] if dt_idx is not None else None,
+                    storage_path=storage_paths[sp_idx] if sp_idx is not None else None,
+                ),
+            )
+            for ti in assigned_tag_indices:
+                doc_tag_map.append((i, ti))
+
+        t0 = time.perf_counter()
+        Document.objects.bulk_create(doc_rows, batch_size=500)
+        print(
+            f"[setup] bulk_create {NUM_DOCS} documents: {time.perf_counter() - t0:.2f}s",
+        )
+
+        # -----------------------------------------------------------------
+        # 3. Bulk-create M2M through-table rows
+        # -----------------------------------------------------------------
+        created_docs = list(Document.objects.order_by("pk"))
+        through_rows = [
+            Document.tags.through(
+                document_id=created_docs[pos].pk,
+                tag_id=tags[ti].pk,
+            )
+            for pos, ti in doc_tag_map
+            if pos < len(created_docs)
+        ]
+        t0 = time.perf_counter()
+        Document.tags.through.objects.bulk_create(
+            through_rows,
+            batch_size=1_000,
+            ignore_conflicts=True,
+        )
+        print(
+            f"[setup] bulk_create {len(through_rows)} tag M2M rows: "
+            f"{time.perf_counter() - t0:.2f}s",
+        )
+
+        # Sample content for prediction tests
+        sample_content = _build_content(
+            fake,
+            corr_fp=corr_fps[0],
+            dtype_fp=dtype_fps[0],
+            tag_fps=[tag_fps[0], tag_fps[1], tag_fps[5]],
+            path_fp=path_fps[0],
+        )
+
+        yield {
+            "model_path": model_path,
+            "sample_content": sample_content,
+        }
+
+        # Teardown
+        print("\n[teardown] Removing corpus...")
+        Document.objects.all().delete()
+        Correspondent.objects.all().delete()
+        DocumentType.objects.all().delete()
+        Tag.objects.all().delete()
+        StoragePath.objects.all().delete()
+
+
+# ---------------------------------------------------------------------------
+# Training profiles
+# ---------------------------------------------------------------------------
+
+
+class TestClassifierTrainingProfile:
+    """Profile DocumentClassifier.train() on the full corpus."""
+
+    def test_train_memory(self, classifier_corpus, tmp_path):
+        """
+        Peak memory allocated during train().
+        tracemalloc reports the delta and top allocation sites.
+        """
+        model_path = tmp_path / "model.pickle"
+        with override_settings(MODEL_FILE=model_path):
+            classifier = DocumentClassifier()
+
+            result, _, _ = measure_memory(
+                classifier.train,
+                label=(
+                    f"train()  [{NUM_DOCS} docs | "
+                    f"{NUM_CORRESPONDENTS} correspondents ({NUM_AUTO_CORRESPONDENTS} AUTO) | "
+                    f"{NUM_DOC_TYPES} doc types ({NUM_AUTO_DOC_TYPES} AUTO) | "
+                    f"{NUM_TAGS} tags ({NUM_AUTO_TAGS} AUTO) | "
+                    f"{NUM_STORAGE_PATHS} paths ({NUM_AUTO_STORAGE_PATHS} AUTO)]"
+                ),
+            )
+            assert result is True, "train() must return True on first run"
+
+            print("\n  Classifiers trained:")
+            print(
+                f"    tags_classifier:           {classifier.tags_classifier is not None}",
+            )
+            print(
+                f"    correspondent_classifier:  {classifier.correspondent_classifier is not None}",
+            )
+            print(
+                f"    document_type_classifier:  {classifier.document_type_classifier is not None}",
+            )
+            print(
+                f"    storage_path_classifier:   {classifier.storage_path_classifier is not None}",
+            )
+            if classifier.data_vectorizer is not None:
+                vocab_size = len(classifier.data_vectorizer.vocabulary_)
+                print(f"    vocabulary size:           {vocab_size} terms")
+
+    def test_train_cpu(self, classifier_corpus, tmp_path):
+        """
+        CPU profile of train() — shows time spent in DB queries,
+        CountVectorizer.fit_transform(), and four MLPClassifier.fit() calls.
+        """
+        model_path = tmp_path / "model_cpu.pickle"
+        with override_settings(MODEL_FILE=model_path):
+            classifier = DocumentClassifier()
+            profile_cpu(
+                classifier.train,
+                label=f"train()  [{NUM_DOCS} docs]",
+                top=30,
+            )
+
+    def test_train_second_call_noop(self, classifier_corpus, tmp_path):
+        """
+        No-op path: second train() on unchanged data should return False.
+        Still queries the DB to build the hash — shown here as the remaining cost.
+        """
+        model_path = tmp_path / "model_noop.pickle"
+        with override_settings(MODEL_FILE=model_path):
+            classifier = DocumentClassifier()
+
+            t0 = time.perf_counter()
+            classifier.train()
+            first_ms = (time.perf_counter() - t0) * 1000
+
+            result, second_elapsed = profile_cpu(
+                classifier.train,
+                label="train() second call (no-op — same data unchanged)",
+                top=20,
+            )
+            assert result is False, "second train() should skip and return False"
+
+            print(f"\n  First train:  {first_ms:.1f} ms  (full fit)")
+            print(f"  Second train: {second_elapsed * 1000:.1f} ms  (skip)")
+            print(f"  Speedup:      {first_ms / (second_elapsed * 1000):.1f}x")
+
+    def test_vectorizer_hash_cost(self, classifier_corpus, tmp_path):
+        """
+        Isolate _update_data_vectorizer_hash() — pickles the entire
+        CountVectorizer just to SHA256 it.  Called at both save and load.
+        """
+        import pickle
+
+        model_path = tmp_path / "model_hash.pickle"
+        with override_settings(MODEL_FILE=model_path):
+            classifier = DocumentClassifier()
+            classifier.train()
+
+            profile_cpu(
+                classifier._update_data_vectorizer_hash,
+                label="_update_data_vectorizer_hash()  [pickle.dumps vectorizer + sha256]",
+                top=10,
+            )
+
+            pickled_size = len(pickle.dumps(classifier.data_vectorizer))
+            vocab_size = len(classifier.data_vectorizer.vocabulary_)
+            print(f"\n  Vocabulary size:       {vocab_size} terms")
+            print(f"  Pickled vectorizer:    {pickled_size / 1024:.1f} KiB")
+
+    def test_save_load_roundtrip(self, classifier_corpus, tmp_path):
+        """
+        Profile save() and load() — model file size directly reflects how
+        much memory the classifier occupies on disk (and roughly in RAM).
+        """
+        model_path = tmp_path / "model_saveload.pickle"
+        with override_settings(MODEL_FILE=model_path):
+            classifier = DocumentClassifier()
+            classifier.train()
+
+            _, save_peak, _ = measure_memory(
+                classifier.save,
+                label="save()  [pickle.dumps + HMAC + atomic rename]",
+            )
+
+            file_size_kib = model_path.stat().st_size / 1024
+            print(f"\n  Model file size: {file_size_kib:.1f} KiB")
+
+            classifier2 = DocumentClassifier()
+            _, load_peak, _ = measure_memory(
+                classifier2.load,
+                label="load()  [read file + verify HMAC + pickle.loads]",
+            )
+
+            print("\n  Summary:")
+            print(f"    Model file size:  {file_size_kib:.1f} KiB")
+            print(f"    Save peak memory: {save_peak:.1f} KiB")
+            print(f"    Load peak memory: {load_peak:.1f} KiB")
+
+
+# ---------------------------------------------------------------------------
+# Prediction profiles
+# ---------------------------------------------------------------------------
+
+
+class TestClassifierPredictionProfile:
+    """
+    Profile the four predict_*() methods — specifically the redundant
+    per-call vectorization overhead from the signal handler pattern.
+    """
+
+    @pytest.fixture(autouse=True)
+    def trained_classifier(self, classifier_corpus, tmp_path):
+        model_path = tmp_path / "model_pred.pickle"
+        self._ctx = override_settings(MODEL_FILE=model_path)
+        self._ctx.enable()
+        self.classifier = DocumentClassifier()
+        self.classifier.train()
+        self.content = classifier_corpus["sample_content"]
+        yield
+        self._ctx.disable()
+
+    def test_predict_all_four_separately_cpu(self):
+        """
+        Profile all four predict_*() calls in the order the signal handlers
+        fire them.  Call 1 is a cache miss; calls 2-4 hit the locmem cache
+        but still pay sha256 + pickle.loads each time.
+        """
+        from django.core.cache import caches
+
+        caches["read-cache"].clear()
+
+        content = self.content
+        print(f"\n  Content length: {len(content)} chars")
+
+        calls = [
+            ("predict_correspondent", self.classifier.predict_correspondent),
+            ("predict_document_type", self.classifier.predict_document_type),
+            ("predict_tags", self.classifier.predict_tags),
+            ("predict_storage_path", self.classifier.predict_storage_path),
+        ]
+
+        timings: list[tuple[str, float]] = []
+        for name, fn in calls:
+            _, elapsed = profile_cpu(
+                lambda f=fn: f(content),
+                label=f"{name}()  [call {len(timings) + 1}/4]",
+                top=15,
+            )
+            timings.append((name, elapsed * 1000))
+
+        print("\n  Per-call timings (sequential, locmem cache):")
+        for name, ms in timings:
+            print(f"    {name:<32s}  {ms:8.3f} ms")
+        print(f"    {'TOTAL':<32s}  {sum(t for _, t in timings):8.3f} ms")
+
+    def test_predict_all_four_memory(self):
+        """
+        Memory allocated for the full four-prediction sequence, both cold
+        and warm, to show pickle serialization allocation per call.
+        """
+        from django.core.cache import caches
+
+        content = self.content
+        calls = [
+            self.classifier.predict_correspondent,
+            self.classifier.predict_document_type,
+            self.classifier.predict_tags,
+            self.classifier.predict_storage_path,
+        ]
+
+        caches["read-cache"].clear()
+        measure_memory(
+            lambda: [fn(content) for fn in calls],
+            label="all four predict_*()  [cache COLD — first call misses]",
+        )
+
+        measure_memory(
+            lambda: [fn(content) for fn in calls],
+            label="all four predict_*()  [cache WARM — all calls hit]",
+        )
+
+    def test_vectorize_cache_miss_vs_hit(self):
+        """
+        Isolate the cost of a cache miss (sha256 + transform + pickle.dumps)
+        vs a cache hit (sha256 + pickle.loads).
+        """
+        from django.core.cache import caches
+
+        read_cache = caches["read-cache"]
+        content = self.content
+
+        read_cache.clear()
+        _, miss_elapsed = profile_cpu(
+            lambda: self.classifier._vectorize(content),
+            label="_vectorize()  [MISS: sha256 + transform + pickle.dumps]",
+            top=15,
+        )
+
+        _, hit_elapsed = profile_cpu(
+            lambda: self.classifier._vectorize(content),
+            label="_vectorize()  [HIT:  sha256 + pickle.loads]",
+            top=15,
+        )
+
+        print(f"\n  Cache miss: {miss_elapsed * 1000:.3f} ms")
+        print(f"  Cache hit:  {hit_elapsed * 1000:.3f} ms")
+        print(f"  Hit is {miss_elapsed / hit_elapsed:.1f}x faster than miss")
+
+    def test_content_hash_overhead(self):
+        """
+        Micro-benchmark the sha256 of the content string — paid on every
+        _vectorize() call regardless of cache state, including x4 per doc.
+        """
+        import hashlib
+
+        content = self.content
+        encoded = content.encode()
+        runs = 5_000
+
+        t0 = time.perf_counter()
+        for _ in range(runs):
+            hashlib.sha256(encoded).hexdigest()
+        us_per_call = (time.perf_counter() - t0) / runs * 1_000_000
+
+        print(f"\n  Content: {len(content)} chars / {len(encoded)} bytes")
+        print(f"  sha256 cost per call:   {us_per_call:.2f} us  (avg over {runs} runs)")
+        print(f"  x4 calls per document:  {us_per_call * 4:.2f} us  total overhead")

test_doclist_profile.py

@@ -0,0 +1,293 @@
+"""
+Document list API profiling — no search, pure ORM path.
+
+Run with:
+    uv run pytest ../test_doclist_profile.py \
+        -m profiling --override-ini="addopts=" -s -v
+
+Corpus: 5 000 documents, 30 correspondents, 20 doc types, 80 tags,
+        ~500 notes (10 %), 10 custom fields with instances on ~50 % of docs.
+
+Scenarios
+---------
+TestDocListProfile
+  - test_list_default_ordering     GET /api/documents/ created desc, page 1, page_size=25
+  - test_list_title_ordering       same with ordering=title
+  - test_list_page_size_comparison page_size=10 / 25 / 100 in sequence
+  - test_list_detail_fields        GET /api/documents/{id}/ — single document serializer cost
+  - test_list_cpu_profile          cProfile of one list request
+
+TestSelectionDataProfile
+  - test_selection_data_unfiltered  _get_selection_data_for_queryset(all docs) in isolation
+  - test_selection_data_via_api     GET /api/documents/?include_selection_data=true
+  - test_selection_data_filtered    filtered vs unfiltered COUNT query comparison
+"""
+
+from __future__ import annotations
+
+import datetime
+import random
+import time
+
+import pytest
+from django.contrib.auth.models import User
+from faker import Faker
+from profiling import profile_block
+from profiling import profile_cpu
+from rest_framework.test import APIClient
+
+from documents.models import Correspondent
+from documents.models import CustomField
+from documents.models import CustomFieldInstance
+from documents.models import Document
+from documents.models import DocumentType
+from documents.models import Note
+from documents.models import Tag
+from documents.views import DocumentViewSet
+
+pytestmark = [pytest.mark.profiling, pytest.mark.django_db]
+
+# ---------------------------------------------------------------------------
+# Corpus parameters
+# ---------------------------------------------------------------------------
+
+NUM_DOCS = 5_000
+NUM_CORRESPONDENTS = 30
+NUM_DOC_TYPES = 20
+NUM_TAGS = 80
+NOTE_FRACTION = 0.10
+CUSTOM_FIELD_COUNT = 10
+CUSTOM_FIELD_FRACTION = 0.50
+PAGE_SIZE = 25
+SEED = 42
+
+
+# ---------------------------------------------------------------------------
+# Module-scoped corpus fixture
+# ---------------------------------------------------------------------------
+
+
+@pytest.fixture(scope="module")
+def module_db(django_db_setup, django_db_blocker):
+    """Unlock the DB for the whole module (module-scoped)."""
+    with django_db_blocker.unblock():
+        yield
+
+
+@pytest.fixture(scope="module")
+def doclist_corpus(module_db):
+    """
+    Build a 5 000-document corpus with tags, notes, custom fields, correspondents,
+    and doc types.  All objects are deleted on teardown.
+    """
+    fake = Faker()
+    Faker.seed(SEED)
+    rng = random.Random(SEED)
+
+    print(f"\n[setup] Creating {NUM_CORRESPONDENTS} correspondents...")  # noqa: T201
+    correspondents = [
+        Correspondent.objects.create(name=f"dlcorp-{i}-{fake.company()}"[:128])
+        for i in range(NUM_CORRESPONDENTS)
+    ]
+
+    print(f"[setup] Creating {NUM_DOC_TYPES} doc types...")  # noqa: T201
+    doc_types = [
+        DocumentType.objects.create(name=f"dltype-{i}-{fake.word()}"[:128])
+        for i in range(NUM_DOC_TYPES)
+    ]
+
+    print(f"[setup] Creating {NUM_TAGS} tags...")  # noqa: T201
+    tags = [
+        Tag.objects.create(name=f"dltag-{i}-{fake.word()}"[:100])
+        for i in range(NUM_TAGS)
+    ]
+
+    print(f"[setup] Creating {CUSTOM_FIELD_COUNT} custom fields...")  # noqa: T201
+    custom_fields = [
+        CustomField.objects.create(
+            name=f"Field {i}",
+            data_type=CustomField.FieldDataType.STRING,
+        )
+        for i in range(CUSTOM_FIELD_COUNT)
+    ]
+
+    note_user = User.objects.create_user(username="doclistnoteuser", password="x")
+    owner = User.objects.create_superuser(username="doclistowner", password="admin")
+
+    print(f"[setup] Building {NUM_DOCS} document rows...")  # noqa: T201
+    base_date = datetime.date(2018, 1, 1)
+    raw_docs = []
+    for i in range(NUM_DOCS):
+        day_offset = rng.randint(0, 6 * 365)
+        raw_docs.append(
+            Document(
+                title=fake.sentence(nb_words=rng.randint(3, 8)).rstrip("."),
+                content="\n\n".join(
+                    fake.paragraph(nb_sentences=rng.randint(2, 5))
+                    for _ in range(rng.randint(1, 3))
+                ),
+                checksum=f"DL{i:07d}",
+                correspondent=rng.choice(correspondents + [None] * 5),
+                document_type=rng.choice(doc_types + [None] * 4),
+                created=base_date + datetime.timedelta(days=day_offset),
+                owner=owner if rng.random() < 0.8 else None,
+            ),
+        )
+    t0 = time.perf_counter()
+    documents = Document.objects.bulk_create(raw_docs)
+    print(f"[setup] bulk_create {NUM_DOCS} docs: {time.perf_counter() - t0:.2f}s")  # noqa: T201
+
+    t0 = time.perf_counter()
+    for doc in documents:
+        k = rng.randint(0, 5)
+        if k:
+            doc.tags.add(*rng.sample(tags, k))
+    print(f"[setup] tag M2M assignments: {time.perf_counter() - t0:.2f}s")  # noqa: T201
+
+    note_docs = rng.sample(documents, int(NUM_DOCS * NOTE_FRACTION))
+    Note.objects.bulk_create(
+        [
+            Note(
+                document=doc,
+                note=fake.sentence(nb_words=rng.randint(4, 15)),
+                user=note_user,
+            )
+            for doc in note_docs
+        ],
+    )
+
+    cf_docs = rng.sample(documents, int(NUM_DOCS * CUSTOM_FIELD_FRACTION))
+    CustomFieldInstance.objects.bulk_create(
+        [
+            CustomFieldInstance(
+                document=doc,
+                field=rng.choice(custom_fields),
+                value_text=fake.word(),
+            )
+            for doc in cf_docs
+        ],
+    )
+
+    first_doc_pk = documents[0].pk
+
+    yield {"owner": owner, "first_doc_pk": first_doc_pk, "tags": tags}
+
+    print("\n[teardown] Removing doclist corpus...")  # noqa: T201
+    Document.objects.all().delete()
+    Correspondent.objects.all().delete()
+    DocumentType.objects.all().delete()
+    Tag.objects.all().delete()
+    CustomField.objects.all().delete()
+    User.objects.filter(username__in=["doclistnoteuser", "doclistowner"]).delete()
+
+
+# ---------------------------------------------------------------------------
+# TestDocListProfile
+# ---------------------------------------------------------------------------
+
+
+class TestDocListProfile:
+    """Profile GET /api/documents/ — pure ORM path, no Tantivy."""
+
+    @pytest.fixture(autouse=True)
+    def _client(self, doclist_corpus):
+        owner = doclist_corpus["owner"]
+        self.client = APIClient()
+        self.client.force_authenticate(user=owner)
+        self.first_doc_pk = doclist_corpus["first_doc_pk"]
+
+    def test_list_default_ordering(self):
+        """GET /api/documents/ default ordering (-created), page 1, page_size=25."""
+        with profile_block(
+            f"GET /api/documents/ default ordering  [page_size={PAGE_SIZE}]",
+        ):
+            response = self.client.get(
+                f"/api/documents/?page=1&page_size={PAGE_SIZE}",
+            )
+        assert response.status_code == 200
+
+    def test_list_title_ordering(self):
+        """GET /api/documents/ ordered by title — tests ORM sort path."""
+        with profile_block(
+            f"GET /api/documents/?ordering=title  [page_size={PAGE_SIZE}]",
+        ):
+            response = self.client.get(
+                f"/api/documents/?ordering=title&page=1&page_size={PAGE_SIZE}",
+            )
+        assert response.status_code == 200
+
+    def test_list_page_size_comparison(self):
+        """Compare serializer cost at page_size=10, 25, 100."""
+        for page_size in [10, 25, 100]:
+            with profile_block(f"GET /api/documents/  [page_size={page_size}]"):
+                response = self.client.get(
+                    f"/api/documents/?page=1&page_size={page_size}",
+                )
+            assert response.status_code == 200
+
+    def test_list_detail_fields(self):
+        """GET /api/documents/{id}/ — per-doc serializer cost with all relations."""
+        pk = self.first_doc_pk
+        with profile_block(f"GET /api/documents/{pk}/ — single doc serializer"):
+            response = self.client.get(f"/api/documents/{pk}/")
+        assert response.status_code == 200
+
+    def test_list_cpu_profile(self):
+        """cProfile of one list request — surfaces hot frames in serializer."""
+        profile_cpu(
+            lambda: self.client.get(
+                f"/api/documents/?page=1&page_size={PAGE_SIZE}",
+            ),
+            label=f"GET /api/documents/ cProfile  [page_size={PAGE_SIZE}]",
+            top=30,
+        )
+
+
+# ---------------------------------------------------------------------------
+# TestSelectionDataProfile
+# ---------------------------------------------------------------------------
+
+
+class TestSelectionDataProfile:
+    """Profile _get_selection_data_for_queryset — the 5+ COUNT queries per request."""
+
+    @pytest.fixture(autouse=True)
+    def _setup(self, doclist_corpus):
+        owner = doclist_corpus["owner"]
+        self.client = APIClient()
+        self.client.force_authenticate(user=owner)
+        self.tags = doclist_corpus["tags"]
+
+    def test_selection_data_unfiltered(self):
+        """Call _get_selection_data_for_queryset(all docs) directly — COUNT queries in isolation."""
+        viewset = DocumentViewSet()
+        qs = Document.objects.all()
+
+        with profile_block("_get_selection_data_for_queryset(all docs) — direct call"):
+            viewset._get_selection_data_for_queryset(qs)
+
+    def test_selection_data_via_api(self):
+        """Full API round-trip with include_selection_data=true."""
+        with profile_block(
+            f"GET /api/documents/?include_selection_data=true  [page_size={PAGE_SIZE}]",
+        ):
+            response = self.client.get(
+                f"/api/documents/?page=1&page_size={PAGE_SIZE}&include_selection_data=true",
+            )
+        assert response.status_code == 200
+        assert "selection_data" in response.data
+
+    def test_selection_data_filtered(self):
+        """selection_data on a tag-filtered queryset — filtered COUNT vs unfiltered."""
+        tag = self.tags[0]
+        viewset = DocumentViewSet()
+        filtered_qs = Document.objects.filter(tags=tag)
+        unfiltered_qs = Document.objects.all()
+
+        print(f"\n  Tag '{tag.name}' matches {filtered_qs.count()} docs")  # noqa: T201
+
+        with profile_block("_get_selection_data_for_queryset(unfiltered)"):
+            viewset._get_selection_data_for_queryset(unfiltered_qs)
+
+        with profile_block("_get_selection_data_for_queryset(filtered by tag)"):
+            viewset._get_selection_data_for_queryset(filtered_qs)

test_matching_profile.py

@@ -0,0 +1,284 @@
+"""
+Matching pipeline profiling.
+
+Run with:
+    uv run pytest ../test_matching_profile.py \
+        -m profiling --override-ini="addopts=" -s -v
+
+Corpus: 1 document + 50 correspondents, 100 tags, 25 doc types, 20 storage
+        paths. Labels are spread across all six matching algorithms
+        (NONE, ANY, ALL, LITERAL, REGEX, FUZZY, AUTO).
+
+Classifier is passed as None -- MATCH_AUTO models skip prediction gracefully,
+which is correct for isolating the ORM query and Python-side evaluation cost.
+
+Scenarios
+---------
+TestMatchingPipelineProfile
+  - test_match_correspondents   50 correspondents, algorithm mix
+  - test_match_tags             100 tags
+  - test_match_document_types   25 doc types
+  - test_match_storage_paths    20 storage paths
+  - test_full_match_sequence    all four in order (cumulative consumption cost)
+  - test_algorithm_breakdown    each MATCH_* algorithm in isolation
+"""
+
+from __future__ import annotations
+
+import random
+
+import pytest
+from faker import Faker
+from profiling import profile_block
+
+from documents.matching import match_correspondents
+from documents.matching import match_document_types
+from documents.matching import match_storage_paths
+from documents.matching import match_tags
+from documents.models import Correspondent
+from documents.models import Document
+from documents.models import DocumentType
+from documents.models import MatchingModel
+from documents.models import StoragePath
+from documents.models import Tag
+
+pytestmark = [pytest.mark.profiling, pytest.mark.django_db]
+
+NUM_CORRESPONDENTS = 50
+NUM_TAGS = 100
+NUM_DOC_TYPES = 25
+NUM_STORAGE_PATHS = 20
+SEED = 42
+
+# Algorithm distribution across labels (cycles through in order)
+_ALGORITHMS = [
+    MatchingModel.MATCH_NONE,
+    MatchingModel.MATCH_ANY,
+    MatchingModel.MATCH_ALL,
+    MatchingModel.MATCH_LITERAL,
+    MatchingModel.MATCH_REGEX,
+    MatchingModel.MATCH_FUZZY,
+    MatchingModel.MATCH_AUTO,
+]
+
+
+def _algo(i: int) -> int:
+    return _ALGORITHMS[i % len(_ALGORITHMS)]
+
+
+# ---------------------------------------------------------------------------
+# Module-scoped corpus fixture
+# ---------------------------------------------------------------------------
+
+
+@pytest.fixture(scope="module")
+def module_db(django_db_setup, django_db_blocker):
+    """Unlock the DB for the whole module (module-scoped)."""
+    with django_db_blocker.unblock():
+        yield
+
+
+@pytest.fixture(scope="module")
+def matching_corpus(module_db):
+    """
+    1 document with realistic content + dense matching model sets.
+    Classifier=None so MATCH_AUTO models are simply skipped.
+    """
+    fake = Faker()
+    Faker.seed(SEED)
+    random.seed(SEED)
+
+    # ---- matching models ---------------------------------------------------
+    print(f"\n[setup] Creating {NUM_CORRESPONDENTS} correspondents...")  # noqa: T201
+    correspondents = []
+    for i in range(NUM_CORRESPONDENTS):
+        algo = _algo(i)
+        match_text = (
+            fake.word()
+            if algo not in (MatchingModel.MATCH_NONE, MatchingModel.MATCH_AUTO)
+            else ""
+        )
+        if algo == MatchingModel.MATCH_REGEX:
+            match_text = r"\b" + fake.word() + r"\b"
+        correspondents.append(
+            Correspondent.objects.create(
+                name=f"mcorp-{i}-{fake.company()}"[:128],
+                matching_algorithm=algo,
+                match=match_text,
+            ),
+        )
+
+    print(f"[setup] Creating {NUM_TAGS} tags...")  # noqa: T201
+    tags = []
+    for i in range(NUM_TAGS):
+        algo = _algo(i)
+        match_text = (
+            fake.word()
+            if algo not in (MatchingModel.MATCH_NONE, MatchingModel.MATCH_AUTO)
+            else ""
+        )
+        if algo == MatchingModel.MATCH_REGEX:
+            match_text = r"\b" + fake.word() + r"\b"
+        tags.append(
+            Tag.objects.create(
+                name=f"mtag-{i}-{fake.word()}"[:100],
+                matching_algorithm=algo,
+                match=match_text,
+            ),
+        )
+
+    print(f"[setup] Creating {NUM_DOC_TYPES} doc types...")  # noqa: T201
+    doc_types = []
+    for i in range(NUM_DOC_TYPES):
+        algo = _algo(i)
+        match_text = (
+            fake.word()
+            if algo not in (MatchingModel.MATCH_NONE, MatchingModel.MATCH_AUTO)
+            else ""
+        )
+        if algo == MatchingModel.MATCH_REGEX:
+            match_text = r"\b" + fake.word() + r"\b"
+        doc_types.append(
+            DocumentType.objects.create(
+                name=f"mtype-{i}-{fake.word()}"[:128],
+                matching_algorithm=algo,
+                match=match_text,
+            ),
+        )
+
+    print(f"[setup] Creating {NUM_STORAGE_PATHS} storage paths...")  # noqa: T201
+    storage_paths = []
+    for i in range(NUM_STORAGE_PATHS):
+        algo = _algo(i)
+        match_text = (
+            fake.word()
+            if algo not in (MatchingModel.MATCH_NONE, MatchingModel.MATCH_AUTO)
+            else ""
+        )
+        if algo == MatchingModel.MATCH_REGEX:
+            match_text = r"\b" + fake.word() + r"\b"
+        storage_paths.append(
+            StoragePath.objects.create(
+                name=f"mpath-{i}-{fake.word()}",
+                path=f"{fake.word()}/{{title}}",
+                matching_algorithm=algo,
+                match=match_text,
+            ),
+        )
+
+    # ---- document with diverse content ------------------------------------
+    doc = Document.objects.create(
+        title="quarterly invoice payment tax financial statement",
+        content=" ".join(fake.paragraph(nb_sentences=5) for _ in range(3)),
+        checksum="MATCHPROF0001",
+    )
+
+    print(f"[setup] Document pk={doc.pk}, content length={len(doc.content)} chars")  # noqa: T201
+    print(  # noqa: T201
+        f"  Correspondents: {NUM_CORRESPONDENTS} "
+        f"({sum(1 for c in correspondents if c.matching_algorithm == MatchingModel.MATCH_AUTO)} AUTO)",
+    )
+    print(  # noqa: T201
+        f"  Tags: {NUM_TAGS} "
+        f"({sum(1 for t in tags if t.matching_algorithm == MatchingModel.MATCH_AUTO)} AUTO)",
+    )
+
+    yield {"doc": doc}
+
+    # Teardown
+    print("\n[teardown] Removing matching corpus...")  # noqa: T201
+    Document.objects.all().delete()
+    Correspondent.objects.all().delete()
+    Tag.objects.all().delete()
+    DocumentType.objects.all().delete()
+    StoragePath.objects.all().delete()
+
+
+# ---------------------------------------------------------------------------
+# TestMatchingPipelineProfile
+# ---------------------------------------------------------------------------
+
+
+class TestMatchingPipelineProfile:
+    """Profile the matching functions called per document during consumption."""
+
+    @pytest.fixture(autouse=True)
+    def _setup(self, matching_corpus):
+        self.doc = matching_corpus["doc"]
+
+    def test_match_correspondents(self):
+        """50 correspondents, algorithm mix. Query count + time."""
+        with profile_block(
+            f"match_correspondents()  [{NUM_CORRESPONDENTS} correspondents, mixed algorithms]",
+        ):
+            result = match_correspondents(self.doc, classifier=None)
+        print(f"  -> {len(result)} matched")  # noqa: T201
+
+    def test_match_tags(self):
+        """100 tags -- densest set in real installs."""
+        with profile_block(f"match_tags()  [{NUM_TAGS} tags, mixed algorithms]"):
+            result = match_tags(self.doc, classifier=None)
+        print(f"  -> {len(result)} matched")  # noqa: T201
+
+    def test_match_document_types(self):
+        """25 doc types."""
+        with profile_block(
+            f"match_document_types()  [{NUM_DOC_TYPES} types, mixed algorithms]",
+        ):
+            result = match_document_types(self.doc, classifier=None)
+        print(f"  -> {len(result)} matched")  # noqa: T201
+
+    def test_match_storage_paths(self):
+        """20 storage paths."""
+        with profile_block(
+            f"match_storage_paths()  [{NUM_STORAGE_PATHS} paths, mixed algorithms]",
+        ):
+            result = match_storage_paths(self.doc, classifier=None)
+        print(f"  -> {len(result)} matched")  # noqa: T201
+
+    def test_full_match_sequence(self):
+        """All four match_*() calls in order -- cumulative cost per document consumed."""
+        with profile_block(
+            "full match sequence: correspondents + doc_types + tags + storage_paths",
+        ):
+            match_correspondents(self.doc, classifier=None)
+            match_document_types(self.doc, classifier=None)
+            match_tags(self.doc, classifier=None)
+            match_storage_paths(self.doc, classifier=None)
+
+    def test_algorithm_breakdown(self):
+        """Create one correspondent per algorithm and time each independently."""
+        import time
+
+        from documents.matching import matches
+
+        fake = Faker()
+        algo_names = {
+            MatchingModel.MATCH_NONE: "MATCH_NONE",
+            MatchingModel.MATCH_ANY: "MATCH_ANY",
+            MatchingModel.MATCH_ALL: "MATCH_ALL",
+            MatchingModel.MATCH_LITERAL: "MATCH_LITERAL",
+            MatchingModel.MATCH_REGEX: "MATCH_REGEX",
+            MatchingModel.MATCH_FUZZY: "MATCH_FUZZY",
+        }
+        doc = self.doc
+        print()  # noqa: T201
+
+        for algo, name in algo_names.items():
+            match_text = fake.word() if algo != MatchingModel.MATCH_NONE else ""
+            if algo == MatchingModel.MATCH_REGEX:
+                match_text = r"\b" + fake.word() + r"\b"
+            model = Correspondent(
+                name=f"algo-test-{name}",
+                matching_algorithm=algo,
+                match=match_text,
+            )
+            # Time 1000 iterations to get stable microsecond readings
+            runs = 1_000
+            t0 = time.perf_counter()
+            for _ in range(runs):
+                matches(model, doc)
+            us_per_call = (time.perf_counter() - t0) / runs * 1_000_000
+            print(  # noqa: T201
+                f"  {name:<20s}  {us_per_call:8.2f} us/call  (match={match_text[:20]!r})",
+            )

test_sanity_profile.py

@@ -0,0 +1,154 @@
+"""
+Sanity checker profiling.
+
+Run with:
+    uv run pytest ../test_sanity_profile.py \
+        -m profiling --override-ini="addopts=" -s -v
+
+Corpus: 2 000 documents with stub files (original + archive + thumbnail)
+        created in a temp MEDIA_ROOT.
+
+Scenarios
+---------
+TestSanityCheckerProfile
+  - test_sanity_full_corpus    full check_sanity() -- cProfile + tracemalloc
+  - test_sanity_query_pattern  profile_block summary: query count + time
+"""
+
+from __future__ import annotations
+
+import hashlib
+import time
+
+import pytest
+from django.test import override_settings
+from profiling import measure_memory
+from profiling import profile_block
+from profiling import profile_cpu
+
+from documents.models import Document
+from documents.sanity_checker import check_sanity
+
+pytestmark = [pytest.mark.profiling, pytest.mark.django_db]
+
+NUM_DOCS = 2_000
+SEED = 42
+
+
+# ---------------------------------------------------------------------------
+# Module-scoped fixture: temp directories + corpus
+# ---------------------------------------------------------------------------
+
+
+@pytest.fixture(scope="module")
+def module_db(django_db_setup, django_db_blocker):
+    """Unlock the DB for the whole module (module-scoped)."""
+    with django_db_blocker.unblock():
+        yield
+
+
+@pytest.fixture(scope="module")
+def sanity_corpus(tmp_path_factory, module_db):
+    """
+    Build a 2 000-document corpus.  For each document create stub files
+    (1-byte placeholders) in ORIGINALS_DIR, ARCHIVE_DIR, and THUMBNAIL_DIR
+    so the sanity checker's file-existence and checksum checks have real targets.
+    """
+    media = tmp_path_factory.mktemp("sanity_media")
+    originals_dir = media / "documents" / "originals"
+    archive_dir = media / "documents" / "archive"
+    thumb_dir = media / "documents" / "thumbnails"
+    for d in (originals_dir, archive_dir, thumb_dir):
+        d.mkdir(parents=True)
+
+    # Use override_settings as a context manager for the whole fixture lifetime
+    settings_ctx = override_settings(
+        MEDIA_ROOT=media,
+        ORIGINALS_DIR=originals_dir,
+        ARCHIVE_DIR=archive_dir,
+        THUMBNAIL_DIR=thumb_dir,
+        MEDIA_LOCK=media / "media.lock",
+    )
+    settings_ctx.enable()
+
+    print(f"\n[setup] Creating {NUM_DOCS} documents with stub files...")  # noqa: T201
+    t0 = time.perf_counter()
+    docs = []
+    for i in range(NUM_DOCS):
+        content = f"document content for doc {i}"
+        checksum = hashlib.sha256(content.encode()).hexdigest()
+
+        orig_filename = f"{i:07d}.pdf"
+        arch_filename = f"{i:07d}.pdf"
+
+        orig_path = originals_dir / orig_filename
+        arch_path = archive_dir / arch_filename
+
+        orig_path.write_bytes(content.encode())
+        arch_path.write_bytes(content.encode())
+
+        docs.append(
+            Document(
+                title=f"Document {i:05d}",
+                content=content,
+                checksum=checksum,
+                archive_checksum=checksum,
+                filename=orig_filename,
+                archive_filename=arch_filename,
+                mime_type="application/pdf",
+            ),
+        )
+
+    created = Document.objects.bulk_create(docs, batch_size=500)
+
+    # Thumbnails use doc.pk, so create them after bulk_create assigns pks
+    for doc in created:
+        thumb_path = thumb_dir / f"{doc.pk:07d}.webp"
+        thumb_path.write_bytes(b"\x00")  # minimal thumbnail stub
+
+    print(  # noqa: T201
+        f"[setup] bulk_create + file creation: {time.perf_counter() - t0:.2f}s",
+    )
+
+    yield {"media": media}
+
+    # Teardown
+    print("\n[teardown] Removing sanity corpus...")  # noqa: T201
+    Document.objects.all().delete()
+    settings_ctx.disable()
+
+
+# ---------------------------------------------------------------------------
+# TestSanityCheckerProfile
+# ---------------------------------------------------------------------------
+
+
+class TestSanityCheckerProfile:
+    """Profile check_sanity() on a realistic corpus with real files."""
+
+    @pytest.fixture(autouse=True)
+    def _setup(self, sanity_corpus):
+        self.media = sanity_corpus["media"]
+
+    def test_sanity_full_corpus(self):
+        """Full check_sanity() -- cProfile surfaces hot frames, tracemalloc shows peak."""
+        _, elapsed = profile_cpu(
+            lambda: check_sanity(scheduled=False),
+            label=f"check_sanity()  [{NUM_DOCS} docs, real files]",
+            top=25,
+        )
+
+        _, peak_kib, delta_kib = measure_memory(
+            lambda: check_sanity(scheduled=False),
+            label=f"check_sanity()  [{NUM_DOCS} docs] -- memory",
+        )
+
+        print("\n  Summary:")  # noqa: T201
+        print(f"    Wall time (CPU profile run): {elapsed * 1000:.1f} ms")  # noqa: T201
+        print(f"    Peak memory (second run):    {peak_kib:.1f} KiB")  # noqa: T201
+        print(f"    Memory delta:                {delta_kib:+.1f} KiB")  # noqa: T201
+
+    def test_sanity_query_pattern(self):
+        """profile_block view: query count + query time + wall time in one summary."""
+        with profile_block(f"check_sanity()  [{NUM_DOCS} docs] -- query count"):
+            check_sanity(scheduled=False)

test_search_profiling.py

@@ -0,0 +1,273 @@
+"""
+Search performance profiling tests.
+
+Run explicitly — excluded from the normal test suite:
+
+    uv run pytest -m profiling -s -p no:xdist --override-ini="addopts=" -v
+
+The ``-s`` flag is required to see profile_block() output.
+The ``-p no:xdist`` flag disables parallel execution for accurate measurements.
+
+Corpus: 5 000 documents generated deterministically from a fixed Faker seed,
+with realistic variety: 30 correspondents, 15 document types, 50 tags, ~500
+notes spread across ~10 % of documents.
+"""
+
+from __future__ import annotations
+
+import random
+
+import pytest
+from django.contrib.auth.models import User
+from faker import Faker
+from profiling import profile_block
+from rest_framework.test import APIClient
+
+from documents.models import Correspondent
+from documents.models import Document
+from documents.models import DocumentType
+from documents.models import Note
+from documents.models import Tag
+from documents.search import get_backend
+from documents.search import reset_backend
+from documents.search._backend import SearchMode
+
+pytestmark = [pytest.mark.profiling, pytest.mark.search, pytest.mark.django_db]
+
+# ---------------------------------------------------------------------------
+# Corpus parameters
+# ---------------------------------------------------------------------------
+
+DOC_COUNT = 5_000
+SEED = 42
+NUM_CORRESPONDENTS = 30
+NUM_DOC_TYPES = 15
+NUM_TAGS = 50
+NOTE_FRACTION = 0.10  # ~500 documents get a note
+PAGE_SIZE = 25
+
+
+def _build_corpus(rng: random.Random, fake: Faker) -> None:
+    """
+    Insert the full corpus into the database and index it.
+
+    Uses bulk_create for the Document rows (fast) then handles the M2M tag
+    relationships and notes individually.  Indexes the full corpus with a
+    single backend.rebuild() call.
+    """
+    import datetime
+
+    # ---- lookup objects -------------------------------------------------
+    correspondents = [
+        Correspondent.objects.create(name=f"profcorp-{i}-{fake.company()}"[:128])
+        for i in range(NUM_CORRESPONDENTS)
+    ]
+    doc_types = [
+        DocumentType.objects.create(name=f"proftype-{i}-{fake.word()}"[:128])
+        for i in range(NUM_DOC_TYPES)
+    ]
+    tags = [
+        Tag.objects.create(name=f"proftag-{i}-{fake.word()}"[:100])
+        for i in range(NUM_TAGS)
+    ]
+    note_user = User.objects.create_user(username="profnoteuser", password="x")
+
+    # ---- bulk-create documents ------------------------------------------
+    base_date = datetime.date(2018, 1, 1)
+    raw_docs = []
+    for i in range(DOC_COUNT):
+        day_offset = rng.randint(0, 6 * 365)
+        created = base_date + datetime.timedelta(days=day_offset)
+        raw_docs.append(
+            Document(
+                title=fake.sentence(nb_words=rng.randint(3, 9)).rstrip("."),
+                content="\n\n".join(
+                    fake.paragraph(nb_sentences=rng.randint(3, 7))
+                    for _ in range(rng.randint(2, 5))
+                ),
+                checksum=f"PROF{i:07d}",
+                correspondent=rng.choice(correspondents + [None] * 8),
+                document_type=rng.choice(doc_types + [None] * 4),
+                created=created,
+            ),
+        )
+    documents = Document.objects.bulk_create(raw_docs)
+
+    # ---- tags (M2M, post-bulk) ------------------------------------------
+    for doc in documents:
+        k = rng.randint(0, 5)
+        if k:
+            doc.tags.add(*rng.sample(tags, k))
+
+    # ---- notes on ~10 % of docs -----------------------------------------
+    note_docs = rng.sample(documents, int(DOC_COUNT * NOTE_FRACTION))
+    for doc in note_docs:
+        Note.objects.create(
+            document=doc,
+            note=fake.sentence(nb_words=rng.randint(6, 20)),
+            user=note_user,
+        )
+
+    # ---- build Tantivy index --------------------------------------------
+    backend = get_backend()
+    qs = Document.objects.select_related(
+        "correspondent",
+        "document_type",
+        "storage_path",
+        "owner",
+    ).prefetch_related("tags", "notes__user", "custom_fields__field")
+    backend.rebuild(qs)
+
+
+class TestSearchProfiling:
+    """
+    Performance profiling for the Tantivy search backend and DRF API layer.
+
+    Each test builds a fresh 5 000-document corpus, exercises one hot path,
+    and prints profile_block() measurements to stdout.  No correctness
+    assertions — the goal is to surface hot spots and track regressions.
+    """
+
+    @pytest.fixture(autouse=True)
+    def _setup(self, tmp_path, settings):
+        index_dir = tmp_path / "index"
+        index_dir.mkdir()
+        settings.INDEX_DIR = index_dir
+
+        reset_backend()
+        rng = random.Random(SEED)
+        fake = Faker()
+        Faker.seed(SEED)
+
+        self.user = User.objects.create_superuser(
+            username="profiler",
+            password="admin",
+        )
+        self.client = APIClient()
+        self.client.force_authenticate(user=self.user)
+
+        _build_corpus(rng, fake)
+        yield
+        reset_backend()
+
+    # -- 1. Backend: search_ids relevance ---------------------------------
+
+    def test_profile_search_ids_relevance(self):
+        """Profile: search_ids() with relevance ordering across several queries."""
+        backend = get_backend()
+        queries = [
+            "invoice payment",
+            "annual report",
+            "bank statement",
+            "contract agreement",
+            "receipt",
+        ]
+        with profile_block(f"search_ids — relevance ({len(queries)} queries)"):
+            for q in queries:
+                backend.search_ids(q, user=None)
+
+    # -- 2. Backend: search_ids with Tantivy-native sort ------------------
+
+    def test_profile_search_ids_sorted(self):
+        """Profile: search_ids() sorted by a Tantivy fast field (created)."""
+        backend = get_backend()
+        with profile_block("search_ids — sorted by created (asc + desc)"):
+            backend.search_ids(
+                "the",
+                user=None,
+                sort_field="created",
+                sort_reverse=False,
+            )
+            backend.search_ids(
+                "the",
+                user=None,
+                sort_field="created",
+                sort_reverse=True,
+            )
+
+    # -- 3. Backend: highlight_hits for a page of 25 ----------------------
+
+    def test_profile_highlight_hits(self):
+        """Profile: highlight_hits() for a 25-document page."""
+        backend = get_backend()
+        all_ids = backend.search_ids("report", user=None)
+        page_ids = all_ids[:PAGE_SIZE]
+        with profile_block(f"highlight_hits — {len(page_ids)} docs"):
+            backend.highlight_hits("report", page_ids)
+
+    # -- 4. Backend: autocomplete -----------------------------------------
+
+    def test_profile_autocomplete(self):
+        """Profile: autocomplete() with eight common prefixes."""
+        backend = get_backend()
+        prefixes = ["inv", "pay", "con", "rep", "sta", "acc", "doc", "fin"]
+        with profile_block(f"autocomplete — {len(prefixes)} prefixes"):
+            for prefix in prefixes:
+                backend.autocomplete(prefix, limit=10)
+
+    # -- 5. Backend: simple-mode search (TEXT and TITLE) ------------------
+
+    def test_profile_search_ids_simple_modes(self):
+        """Profile: search_ids() in TEXT and TITLE simple-search modes."""
+        backend = get_backend()
+        queries = ["invoice 2023", "annual report", "bank statement"]
+        with profile_block(
+            f"search_ids — TEXT + TITLE modes ({len(queries)} queries each)",
+        ):
+            for q in queries:
+                backend.search_ids(q, user=None, search_mode=SearchMode.TEXT)
+                backend.search_ids(q, user=None, search_mode=SearchMode.TITLE)
+
+    # -- 6. API: full round-trip, relevance + page 1 ----------------------
+
+    def test_profile_api_relevance_search(self):
+        """Profile: full API search round-trip, relevance order, page 1."""
+        with profile_block(
+            f"API /documents/?query=… relevance (page 1, page_size={PAGE_SIZE})",
+        ):
+            response = self.client.get(
+                f"/api/documents/?query=invoice+payment&page=1&page_size={PAGE_SIZE}",
+            )
+        assert response.status_code == 200
+
+    # -- 7. API: full round-trip, ORM-ordered (title) ---------------------
+
+    def test_profile_api_orm_sorted_search(self):
+        """Profile: full API search round-trip with ORM-delegated sort (title)."""
+        with profile_block("API /documents/?query=…&ordering=title"):
+            response = self.client.get(
+                f"/api/documents/?query=report&ordering=title&page=1&page_size={PAGE_SIZE}",
+            )
+        assert response.status_code == 200
+
+    # -- 8. API: full round-trip, score sort ------------------------------
+
+    def test_profile_api_score_sort(self):
+        """Profile: full API search with ordering=-score (relevance, preserve order)."""
+        with profile_block("API /documents/?query=…&ordering=-score"):
+            response = self.client.get(
+                f"/api/documents/?query=statement&ordering=-score&page=1&page_size={PAGE_SIZE}",
+            )
+        assert response.status_code == 200
+
+    # -- 9. API: full round-trip, with selection_data ---------------------
+
+    def test_profile_api_with_selection_data(self):
+        """Profile: full API search including include_selection_data=true."""
+        with profile_block("API /documents/?query=…&include_selection_data=true"):
+            response = self.client.get(
+                f"/api/documents/?query=contract&page=1&page_size={PAGE_SIZE}"
+                "&include_selection_data=true",
+            )
+        assert response.status_code == 200
+        assert "selection_data" in response.data
+
+    # -- 10. API: paginated (page 2) --------------------------------------
+
+    def test_profile_api_page_2(self):
+        """Profile: full API search, page 2 — exercises page offset arithmetic."""
+        with profile_block(f"API /documents/?query=…&page=2&page_size={PAGE_SIZE}"):
+            response = self.client.get(
+                f"/api/documents/?query=the&page=2&page_size={PAGE_SIZE}",
+            )
+        assert response.status_code == 200

test_workflow_profile.py

@@ -0,0 +1,231 @@
+"""
+Workflow trigger matching profiling.
+
+Run with:
+    uv run pytest ../test_workflow_profile.py \
+        -m profiling --override-ini="addopts=" -s -v
+
+Corpus: 500 documents + correspondents + tags + sets of WorkflowTrigger
+        objects at 5 and 20 count to allow scaling comparisons.
+
+Scenarios
+---------
+TestWorkflowMatchingProfile
+  - test_existing_document_5_workflows    existing_document_matches_workflow x 5 triggers
+  - test_existing_document_20_workflows   same x 20 triggers
+  - test_workflow_prefilter               prefilter_documents_by_workflowtrigger on 500 docs
+  - test_trigger_type_comparison          compare DOCUMENT_ADDED vs DOCUMENT_UPDATED overhead
+"""
+
+from __future__ import annotations
+
+import random
+import time
+
+import pytest
+from faker import Faker
+from profiling import profile_block
+
+from documents.matching import existing_document_matches_workflow
+from documents.matching import prefilter_documents_by_workflowtrigger
+from documents.models import Correspondent
+from documents.models import Document
+from documents.models import Tag
+from documents.models import Workflow
+from documents.models import WorkflowAction
+from documents.models import WorkflowTrigger
+
+pytestmark = [pytest.mark.profiling, pytest.mark.django_db]
+
+NUM_DOCS = 500
+NUM_CORRESPONDENTS = 10
+NUM_TAGS = 20
+SEED = 42
+
+
+# ---------------------------------------------------------------------------
+# Module-scoped fixture
+# ---------------------------------------------------------------------------
+
+
+@pytest.fixture(scope="module")
+def module_db(django_db_setup, django_db_blocker):
+    """Unlock the DB for the whole module (module-scoped)."""
+    with django_db_blocker.unblock():
+        yield
+
+
+@pytest.fixture(scope="module")
+def workflow_corpus(module_db):
+    """
+    500 documents + correspondents + tags + sets of workflow triggers
+    at 5 and 20 count to allow scaling comparisons.
+    """
+    fake = Faker()
+    Faker.seed(SEED)
+    rng = random.Random(SEED)
+
+    # ---- lookup objects ---------------------------------------------------
+    print("\n[setup] Creating lookup objects...")  # noqa: T201
+    correspondents = [
+        Correspondent.objects.create(name=f"wfcorp-{i}-{fake.company()}"[:128])
+        for i in range(NUM_CORRESPONDENTS)
+    ]
+    tags = [
+        Tag.objects.create(name=f"wftag-{i}-{fake.word()}"[:100])
+        for i in range(NUM_TAGS)
+    ]
+
+    # ---- documents --------------------------------------------------------
+    print(f"[setup] Building {NUM_DOCS} documents...")  # noqa: T201
+    raw_docs = []
+    for i in range(NUM_DOCS):
+        raw_docs.append(
+            Document(
+                title=fake.sentence(nb_words=4).rstrip("."),
+                content=fake.paragraph(nb_sentences=3),
+                checksum=f"WF{i:07d}",
+                correspondent=rng.choice(correspondents + [None] * 3),
+            ),
+        )
+    documents = Document.objects.bulk_create(raw_docs, batch_size=500)
+    for doc in documents:
+        k = rng.randint(0, 3)
+        if k:
+            doc.tags.add(*rng.sample(tags, k))
+
+    sample_doc = documents[0]
+    print(f"[setup] Sample doc pk={sample_doc.pk}")  # noqa: T201
+
+    # ---- build triggers at scale 5 and 20 --------------------------------
+    _wf_counter = [0]
+
+    def _make_triggers(n: int, trigger_type: int) -> list[WorkflowTrigger]:
+        triggers = []
+        for i in range(n):
+            # Alternate between no filter and a correspondent filter
+            corr = correspondents[i % NUM_CORRESPONDENTS] if i % 3 == 0 else None
+            trigger = WorkflowTrigger.objects.create(
+                type=trigger_type,
+                filter_has_correspondent=corr,
+            )
+            action = WorkflowAction.objects.create(
+                type=WorkflowAction.WorkflowActionType.ASSIGNMENT,
+            )
+            idx = _wf_counter[0]
+            _wf_counter[0] += 1
+            wf = Workflow.objects.create(name=f"wf-profile-{idx}")
+            wf.triggers.add(trigger)
+            wf.actions.add(action)
+            triggers.append(trigger)
+        return triggers
+
+    print("[setup] Creating workflow triggers...")  # noqa: T201
+    triggers_5 = _make_triggers(5, WorkflowTrigger.WorkflowTriggerType.DOCUMENT_UPDATED)
+    triggers_20 = _make_triggers(
+        20,
+        WorkflowTrigger.WorkflowTriggerType.DOCUMENT_UPDATED,
+    )
+    triggers_added = _make_triggers(
+        5,
+        WorkflowTrigger.WorkflowTriggerType.DOCUMENT_ADDED,
+    )
+
+    yield {
+        "doc": sample_doc,
+        "triggers_5": triggers_5,
+        "triggers_20": triggers_20,
+        "triggers_added": triggers_added,
+    }
+
+    # Teardown
+    print("\n[teardown] Removing workflow corpus...")  # noqa: T201
+    Workflow.objects.all().delete()
+    WorkflowTrigger.objects.all().delete()
+    WorkflowAction.objects.all().delete()
+    Document.objects.all().delete()
+    Correspondent.objects.all().delete()
+    Tag.objects.all().delete()
+
+
+# ---------------------------------------------------------------------------
+# TestWorkflowMatchingProfile
+# ---------------------------------------------------------------------------
+
+
+class TestWorkflowMatchingProfile:
+    """Profile workflow trigger evaluation per document save."""
+
+    @pytest.fixture(autouse=True)
+    def _setup(self, workflow_corpus):
+        self.doc = workflow_corpus["doc"]
+        self.triggers_5 = workflow_corpus["triggers_5"]
+        self.triggers_20 = workflow_corpus["triggers_20"]
+        self.triggers_added = workflow_corpus["triggers_added"]
+
+    def test_existing_document_5_workflows(self):
+        """existing_document_matches_workflow x 5 DOCUMENT_UPDATED triggers."""
+        doc = self.doc
+        triggers = self.triggers_5
+
+        with profile_block(
+            f"existing_document_matches_workflow  [{len(triggers)} triggers]",
+        ):
+            for trigger in triggers:
+                existing_document_matches_workflow(doc, trigger)
+
+    def test_existing_document_20_workflows(self):
+        """existing_document_matches_workflow x 20 triggers -- shows linear scaling."""
+        doc = self.doc
+        triggers = self.triggers_20
+
+        with profile_block(
+            f"existing_document_matches_workflow  [{len(triggers)} triggers]",
+        ):
+            for trigger in triggers:
+                existing_document_matches_workflow(doc, trigger)
+
+        # Also time each call individually to show per-trigger overhead
+        timings = []
+        for trigger in triggers:
+            t0 = time.perf_counter()
+            existing_document_matches_workflow(doc, trigger)
+            timings.append((time.perf_counter() - t0) * 1_000_000)
+        avg_us = sum(timings) / len(timings)
+        print(f"\n  Per-trigger avg: {avg_us:.1f} us  (n={len(timings)})")  # noqa: T201
+
+    def test_workflow_prefilter(self):
+        """prefilter_documents_by_workflowtrigger on 500 docs -- tag + correspondent filters."""
+        qs = Document.objects.all()
+        print(f"\n  Corpus: {qs.count()} documents")  # noqa: T201
+
+        for trigger in self.triggers_20[:3]:
+            label = (
+                f"prefilter_documents_by_workflowtrigger  "
+                f"[corr={trigger.filter_has_correspondent_id}]"
+            )
+            with profile_block(label):
+                result = prefilter_documents_by_workflowtrigger(qs, trigger)
+                # Evaluate the queryset
+                count = result.count()
+            print(f"  -> {count} docs passed filter")  # noqa: T201
+
+    def test_trigger_type_comparison(self):
+        """Compare per-call overhead of DOCUMENT_UPDATED vs DOCUMENT_ADDED."""
+        doc = self.doc
+        runs = 200
+
+        for label, triggers in [
+            ("DOCUMENT_UPDATED", self.triggers_5),
+            ("DOCUMENT_ADDED", self.triggers_added),
+        ]:
+            t0 = time.perf_counter()
+            for _ in range(runs):
+                for trigger in triggers:
+                    existing_document_matches_workflow(doc, trigger)
+            total_calls = runs * len(triggers)
+            us_per_call = (time.perf_counter() - t0) / total_calls * 1_000_000
+            print(  # noqa: T201
+                f"  {label:<22s}  {us_per_call:.2f} us/call  "
+                f"({total_calls} calls, {len(triggers)} triggers)",
+            )

uv.lock

@@ -875,15 +875,15 @@ wheels = [
 
 [[package]]
 name = "django"
-version = "5.2.13"
+version = "5.2.12"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "asgiref", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "sqlparse", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/1f/c5/c69e338eb2959f641045802e5ea87ca4bf5ac90c5fd08953ca10742fad51/django-5.2.13.tar.gz", hash = "sha256:a31589db5188d074c63f0945c3888fad104627dfcc236fb2b97f71f89da33bc4", size = 10890368, upload-time = "2026-04-07T14:02:15.072Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/bd/55/b9445fc0695b03746f355c05b2eecc54c34e05198c686f4fc4406b722b52/django-5.2.12.tar.gz", hash = "sha256:6b809af7165c73eff5ce1c87fdae75d4da6520d6667f86401ecf55b681eb1eeb", size = 10860574, upload-time = "2026-03-03T13:56:05.509Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/59/b1/51ab36b2eefcf8cdb9338c7188668a157e29e30306bfc98a379704c9e10d/django-5.2.13-py3-none-any.whl", hash = "sha256:5788fce61da23788a8ce6f02583765ab060d396720924789f97fa42119d37f7a", size = 8310982, upload-time = "2026-04-07T14:02:08.883Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/32/4b144e125678efccf5d5b61581de1c4088d6b0286e46096e3b8de0d556c8/django-5.2.12-py3-none-any.whl", hash = "sha256:4853482f395c3a151937f6991272540fcbf531464f254a347bf7c89f53c8cff7", size = 8310245, upload-time = "2026-03-03T13:56:01.174Z" },
 ]
 
 [[package]]
@@ -935,6 +935,19 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/12/bf/af8ad2aa5a402f278b444ca70729fb12ee96ddb89c19c32a2d7c5189358f/django_cachalot-2.9.0-py3-none-any.whl", hash = "sha256:b80ac4930613a7849988ea772a53598d262a15eaf55e5ec8c78accae7fdd99ff", size = 57814, upload-time = "2026-01-28T05:23:28.741Z" },
 ]
 
+[[package]]
+name = "django-celery-results"
+version = "2.6.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "celery", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
+    { name = "django", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a6/b5/9966c28e31014c228305e09d48b19b35522a8f941fe5af5f81f40dc8fa80/django_celery_results-2.6.0.tar.gz", hash = "sha256:9abcd836ae6b61063779244d8887a88fe80bbfaba143df36d3cb07034671277c", size = 83985, upload-time = "2025-04-10T08:23:52.677Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/2c/da/70f0f3c5364735344c4bc89e53413bcaae95b4fc1de4e98a7a3b9fb70c88/django_celery_results-2.6.0-py3-none-any.whl", hash = "sha256:b9ccdca2695b98c7cbbb8dea742311ba9a92773d71d7b4944a676e69a7df1c73", size = 38351, upload-time = "2025-04-10T08:23:49.965Z" },
+]
+
 [[package]]
 name = "django-compression-middleware"
 version = "0.5.0"
@@ -2856,6 +2869,7 @@ dependencies = [
     { name = "django-allauth", extra = ["mfa", "socialaccount"], marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "django-auditlog", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "django-cachalot", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
+    { name = "django-celery-results", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "django-compression-middleware", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "django-cors-headers", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "django-extensions", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
@@ -3000,10 +3014,11 @@ requires-dist = [
     { name = "channels-redis", specifier = "~=4.2" },
     { name = "concurrent-log-handler", specifier = "~=0.9.25" },
     { name = "dateparser", specifier = "~=1.2" },
-    { name = "django", specifier = "~=5.2.13" },
+    { name = "django", specifier = "~=5.2.10" },
     { name = "django-allauth", extras = ["mfa", "socialaccount"], specifier = "~=65.15.0" },
     { name = "django-auditlog", specifier = "~=3.4.1" },
     { name = "django-cachalot", specifier = "~=2.9.0" },
+    { name = "django-celery-results", specifier = "~=2.6.0" },
     { name = "django-compression-middleware", specifier = "~=0.5.0" },
     { name = "django-cors-headers", specifier = "~=4.9.0" },
     { name = "django-extensions", specifier = "~=4.1" },
@@ -3072,7 +3087,7 @@ dev = [
     { name = "faker", specifier = "~=40.12.0" },
     { name = "imagehash" },
     { name = "prek", specifier = "~=0.3.0" },
-    { name = "pytest", specifier = "~=9.0.3" },
+    { name = "pytest", specifier = "~=9.0.0" },
     { name = "pytest-cov", specifier = "~=7.1.0" },
     { name = "pytest-django", specifier = "~=4.12.0" },
     { name = "pytest-env", specifier = "~=1.6.0" },
@@ -3095,7 +3110,7 @@ testing = [
     { name = "factory-boy", specifier = "~=3.3.1" },
     { name = "faker", specifier = "~=40.12.0" },
     { name = "imagehash" },
-    { name = "pytest", specifier = "~=9.0.3" },
+    { name = "pytest", specifier = "~=9.0.0" },
     { name = "pytest-cov", specifier = "~=7.1.0" },
     { name = "pytest-django", specifier = "~=4.12.0" },
     { name = "pytest-env", specifier = "~=1.6.0" },
@@ -3250,70 +3265,70 @@ wheels = [
 
 [[package]]
 name = "pillow"
-version = "12.2.0"
+version = "12.1.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/8c/21/c2bcdd5906101a30244eaffc1b6e6ce71a31bd0742a01eb89e660ebfac2d/pillow-12.2.0.tar.gz", hash = "sha256:a830b1a40919539d07806aa58e1b114df53ddd43213d9c8b75847eee6c0182b5", size = 46987819, upload-time = "2026-04-01T14:46:17.687Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/1f/42/5c74462b4fd957fcd7b13b04fb3205ff8349236ea74c7c375766d6c82288/pillow-12.1.1.tar.gz", hash = "sha256:9ad8fa5937ab05218e2b6a4cff30295ad35afd2f83ac592e68c0d871bb0fdbc4", size = 46980264, upload-time = "2026-02-11T04:23:07.146Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/68/e1/748f5663efe6edcfc4e74b2b93edfb9b8b99b67f21a854c3ae416500a2d9/pillow-12.2.0-cp311-cp311-macosx_10_10_x86_64.whl", hash = "sha256:8be29e59487a79f173507c30ddf57e733a357f67881430449bb32614075a40ab", size = 5354347, upload-time = "2026-04-01T14:42:44.255Z" },
-    { url = "https://files.pythonhosted.org/packages/47/a1/d5ff69e747374c33a3b53b9f98cca7889fce1fd03d79cdc4e1bccc6c5a87/pillow-12.2.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:71cde9a1e1551df7d34a25462fc60325e8a11a82cc2e2f54578e5e9a1e153d65", size = 4695873, upload-time = "2026-04-01T14:42:46.452Z" },
-    { url = "https://files.pythonhosted.org/packages/df/21/e3fbdf54408a973c7f7f89a23b2cb97a7ef30c61ab4142af31eee6aebc88/pillow-12.2.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:f490f9368b6fc026f021db16d7ec2fbf7d89e2edb42e8ec09d2c60505f5729c7", size = 6280168, upload-time = "2026-04-01T14:42:49.228Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/f1/00b7278c7dd52b17ad4329153748f87b6756ec195ff786c2bdf12518337d/pillow-12.2.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:8bd7903a5f2a4545f6fd5935c90058b89d30045568985a71c79f5fd6edf9b91e", size = 8088188, upload-time = "2026-04-01T14:42:51.735Z" },
-    { url = "https://files.pythonhosted.org/packages/ad/cf/220a5994ef1b10e70e85748b75649d77d506499352be135a4989c957b701/pillow-12.2.0-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3997232e10d2920a68d25191392e3a4487d8183039e1c74c2297f00ed1c50705", size = 6394401, upload-time = "2026-04-01T14:42:54.343Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/bd/e51a61b1054f09437acfbc2ff9106c30d1eb76bc1453d428399946781253/pillow-12.2.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e74473c875d78b8e9d5da2a70f7099549f9eb37ded4e2f6a463e60125bccd176", size = 7079655, upload-time = "2026-04-01T14:42:56.954Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/3d/45132c57d5fb4b5744567c3817026480ac7fc3ce5d4c47902bc0e7f6f853/pillow-12.2.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:56a3f9c60a13133a98ecff6197af34d7824de9b7b38c3654861a725c970c197b", size = 6503105, upload-time = "2026-04-01T14:42:59.847Z" },
-    { url = "https://files.pythonhosted.org/packages/7d/2e/9df2fc1e82097b1df3dce58dc43286aa01068e918c07574711fcc53e6fb4/pillow-12.2.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:90e6f81de50ad6b534cab6e5aef77ff6e37722b2f5d908686f4a5c9eba17a909", size = 7203402, upload-time = "2026-04-01T14:43:02.664Z" },
-    { url = "https://files.pythonhosted.org/packages/58/be/7482c8a5ebebbc6470b3eb791812fff7d5e0216c2be3827b30b8bb6603ed/pillow-12.2.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:2d192a155bbcec180f8564f693e6fd9bccff5a7af9b32e2e4bf8c9c69dbad6b5", size = 5308279, upload-time = "2026-04-01T14:43:13.246Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/95/0a351b9289c2b5cbde0bacd4a83ebc44023e835490a727b2a3bd60ddc0f4/pillow-12.2.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:f3f40b3c5a968281fd507d519e444c35f0ff171237f4fdde090dd60699458421", size = 4695490, upload-time = "2026-04-01T14:43:15.584Z" },
-    { url = "https://files.pythonhosted.org/packages/de/af/4e8e6869cbed569d43c416fad3dc4ecb944cb5d9492defaed89ddd6fe871/pillow-12.2.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:03e7e372d5240cc23e9f07deca4d775c0817bffc641b01e9c3af208dbd300987", size = 6284462, upload-time = "2026-04-01T14:43:18.268Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/9e/c05e19657fd57841e476be1ab46c4d501bffbadbafdc31a6d665f8b737b6/pillow-12.2.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:b86024e52a1b269467a802258c25521e6d742349d760728092e1bc2d135b4d76", size = 8094744, upload-time = "2026-04-01T14:43:20.716Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/54/1789c455ed10176066b6e7e6da1b01e50e36f94ba584dc68d9eebfe9156d/pillow-12.2.0-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:7371b48c4fa448d20d2714c9a1f775a81155050d383333e0a6c15b1123dda005", size = 6398371, upload-time = "2026-04-01T14:43:23.443Z" },
-    { url = "https://files.pythonhosted.org/packages/43/e3/fdc657359e919462369869f1c9f0e973f353f9a9ee295a39b1fea8ee1a77/pillow-12.2.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:62f5409336adb0663b7caa0da5c7d9e7bdbaae9ce761d34669420c2a801b2780", size = 7087215, upload-time = "2026-04-01T14:43:26.758Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/f8/2f6825e441d5b1959d2ca5adec984210f1ec086435b0ed5f52c19b3b8a6e/pillow-12.2.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:01afa7cf67f74f09523699b4e88c73fb55c13346d212a59a2db1f86b0a63e8c5", size = 6509783, upload-time = "2026-04-01T14:43:29.56Z" },
-    { url = "https://files.pythonhosted.org/packages/67/f9/029a27095ad20f854f9dba026b3ea6428548316e057e6fc3545409e86651/pillow-12.2.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:fc3d34d4a8fbec3e88a79b92e5465e0f9b842b628675850d860b8bd300b159f5", size = 7212112, upload-time = "2026-04-01T14:43:32.091Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/01/53d10cf0dbad820a8db274d259a37ba50b88b24768ddccec07355382d5ad/pillow-12.2.0-cp313-cp313-ios_13_0_arm64_iphoneos.whl", hash = "sha256:8297651f5b5679c19968abefd6bb84d95fe30ef712eb1b2d9b2d31ca61267f4c", size = 4100837, upload-time = "2026-04-01T14:43:41.506Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/98/f3a6657ecb698c937f6c76ee564882945f29b79bad496abcba0e84659ec5/pillow-12.2.0-cp313-cp313-ios_13_0_arm64_iphonesimulator.whl", hash = "sha256:50d8520da2a6ce0af445fa6d648c4273c3eeefbc32d7ce049f22e8b5c3daecc2", size = 4176528, upload-time = "2026-04-01T14:43:43.773Z" },
-    { url = "https://files.pythonhosted.org/packages/69/bc/8986948f05e3ea490b8442ea1c1d4d990b24a7e43d8a51b2c7d8b1dced36/pillow-12.2.0-cp313-cp313-ios_13_0_x86_64_iphonesimulator.whl", hash = "sha256:766cef22385fa1091258ad7e6216792b156dc16d8d3fa607e7545b2b72061f1c", size = 3640401, upload-time = "2026-04-01T14:43:45.87Z" },
-    { url = "https://files.pythonhosted.org/packages/34/46/6c717baadcd62bc8ed51d238d521ab651eaa74838291bda1f86fe1f864c9/pillow-12.2.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:5d2fd0fa6b5d9d1de415060363433f28da8b1526c1c129020435e186794b3795", size = 5308094, upload-time = "2026-04-01T14:43:48.438Z" },
-    { url = "https://files.pythonhosted.org/packages/71/43/905a14a8b17fdb1ccb58d282454490662d2cb89a6bfec26af6d3520da5ec/pillow-12.2.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:56b25336f502b6ed02e889f4ece894a72612fe885889a6e8c4c80239ff6e5f5f", size = 4695402, upload-time = "2026-04-01T14:43:51.292Z" },
-    { url = "https://files.pythonhosted.org/packages/73/dd/42107efcb777b16fa0393317eac58f5b5cf30e8392e266e76e51cff28c3d/pillow-12.2.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:f1c943e96e85df3d3478f7b691f229887e143f81fedab9b20205349ab04d73ed", size = 6280005, upload-time = "2026-04-01T14:43:54.242Z" },
-    { url = "https://files.pythonhosted.org/packages/a8/68/b93e09e5e8549019e61acf49f65b1a8530765a7f812c77a7461bca7e4494/pillow-12.2.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:03f6fab9219220f041c74aeaa2939ff0062bd5c364ba9ce037197f4c6d498cd9", size = 8090669, upload-time = "2026-04-01T14:43:57.335Z" },
-    { url = "https://files.pythonhosted.org/packages/4b/6e/3ccb54ce8ec4ddd1accd2d89004308b7b0b21c4ac3d20fa70af4760a4330/pillow-12.2.0-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5cdfebd752ec52bf5bb4e35d9c64b40826bc5b40a13df7c3cda20a2c03a0f5ed", size = 6395194, upload-time = "2026-04-01T14:43:59.864Z" },
-    { url = "https://files.pythonhosted.org/packages/67/ee/21d4e8536afd1a328f01b359b4d3997b291ffd35a237c877b331c1c3b71c/pillow-12.2.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:eedf4b74eda2b5a4b2b2fb4c006d6295df3bf29e459e198c90ea48e130dc75c3", size = 7082423, upload-time = "2026-04-01T14:44:02.74Z" },
-    { url = "https://files.pythonhosted.org/packages/78/5f/e9f86ab0146464e8c133fe85df987ed9e77e08b29d8d35f9f9f4d6f917ba/pillow-12.2.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:00a2865911330191c0b818c59103b58a5e697cae67042366970a6b6f1b20b7f9", size = 6505667, upload-time = "2026-04-01T14:44:05.381Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/1e/409007f56a2fdce61584fd3acbc2bbc259857d555196cedcadc68c015c82/pillow-12.2.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:1e1757442ed87f4912397c6d35a0db6a7b52592156014706f17658ff58bbf795", size = 7208580, upload-time = "2026-04-01T14:44:08.39Z" },
-    { url = "https://files.pythonhosted.org/packages/4d/a4/b342930964e3cb4dce5038ae34b0eab4653334995336cd486c5a8c25a00c/pillow-12.2.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:042db20a421b9bafecc4b84a8b6e444686bd9d836c7fd24542db3e7df7baad9b", size = 5309927, upload-time = "2026-04-01T14:44:18.89Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/de/23198e0a65a9cf06123f5435a5d95cea62a635697f8f03d134d3f3a96151/pillow-12.2.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:dd025009355c926a84a612fecf58bb315a3f6814b17ead51a8e48d3823d9087f", size = 4698624, upload-time = "2026-04-01T14:44:21.115Z" },
-    { url = "https://files.pythonhosted.org/packages/01/a6/1265e977f17d93ea37aa28aa81bad4fa597933879fac2520d24e021c8da3/pillow-12.2.0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:88ddbc66737e277852913bd1e07c150cc7bb124539f94c4e2df5344494e0a612", size = 6321252, upload-time = "2026-04-01T14:44:23.663Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/83/5982eb4a285967baa70340320be9f88e57665a387e3a53a7f0db8231a0cd/pillow-12.2.0-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:d362d1878f00c142b7e1a16e6e5e780f02be8195123f164edf7eddd911eefe7c", size = 8126550, upload-time = "2026-04-01T14:44:26.772Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/48/6ffc514adce69f6050d0753b1a18fd920fce8cac87620d5a31231b04bfc5/pillow-12.2.0-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:2c727a6d53cb0018aadd8018c2b938376af27914a68a492f59dfcaca650d5eea", size = 6433114, upload-time = "2026-04-01T14:44:29.615Z" },
-    { url = "https://files.pythonhosted.org/packages/36/a3/f9a77144231fb8d40ee27107b4463e205fa4677e2ca2548e14da5cf18dce/pillow-12.2.0-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:efd8c21c98c5cc60653bcb311bef2ce0401642b7ce9d09e03a7da87c878289d4", size = 7115667, upload-time = "2026-04-01T14:44:32.773Z" },
-    { url = "https://files.pythonhosted.org/packages/c1/fc/ac4ee3041e7d5a565e1c4fd72a113f03b6394cc72ab7089d27608f8aaccb/pillow-12.2.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:9f08483a632889536b8139663db60f6724bfcb443c96f1b18855860d7d5c0fd4", size = 6538966, upload-time = "2026-04-01T14:44:35.252Z" },
-    { url = "https://files.pythonhosted.org/packages/c0/a8/27fb307055087f3668f6d0a8ccb636e7431d56ed0750e07a60547b1e083e/pillow-12.2.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:dac8d77255a37e81a2efcbd1fc05f1c15ee82200e6c240d7e127e25e365c39ea", size = 7238241, upload-time = "2026-04-01T14:44:37.875Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/98/4595daa2365416a86cb0d495248a393dfc84e96d62ad080c8546256cb9c0/pillow-12.2.0-cp314-cp314-ios_13_0_arm64_iphoneos.whl", hash = "sha256:3adc9215e8be0448ed6e814966ecf3d9952f0ea40eb14e89a102b87f450660d8", size = 4100848, upload-time = "2026-04-01T14:44:48.48Z" },
-    { url = "https://files.pythonhosted.org/packages/0b/79/40184d464cf89f6663e18dfcf7ca21aae2491fff1a16127681bf1fa9b8cf/pillow-12.2.0-cp314-cp314-ios_13_0_arm64_iphonesimulator.whl", hash = "sha256:6a9adfc6d24b10f89588096364cc726174118c62130c817c2837c60cf08a392b", size = 4176515, upload-time = "2026-04-01T14:44:51.353Z" },
-    { url = "https://files.pythonhosted.org/packages/b0/63/703f86fd4c422a9cf722833670f4f71418fb116b2853ff7da722ea43f184/pillow-12.2.0-cp314-cp314-ios_13_0_x86_64_iphonesimulator.whl", hash = "sha256:6a6e67ea2e6feda684ed370f9a1c52e7a243631c025ba42149a2cc5934dec295", size = 3640159, upload-time = "2026-04-01T14:44:53.588Z" },
-    { url = "https://files.pythonhosted.org/packages/71/e0/fb22f797187d0be2270f83500aab851536101b254bfa1eae10795709d283/pillow-12.2.0-cp314-cp314-macosx_10_15_x86_64.whl", hash = "sha256:2bb4a8d594eacdfc59d9e5ad972aa8afdd48d584ffd5f13a937a664c3e7db0ed", size = 5312185, upload-time = "2026-04-01T14:44:56.039Z" },
-    { url = "https://files.pythonhosted.org/packages/ba/8c/1a9e46228571de18f8e28f16fabdfc20212a5d019f3e3303452b3f0a580d/pillow-12.2.0-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:80b2da48193b2f33ed0c32c38140f9d3186583ce7d516526d462645fd98660ae", size = 4695386, upload-time = "2026-04-01T14:44:58.663Z" },
-    { url = "https://files.pythonhosted.org/packages/70/62/98f6b7f0c88b9addd0e87c217ded307b36be024d4ff8869a812b241d1345/pillow-12.2.0-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:22db17c68434de69d8ecfc2fe821569195c0c373b25cccb9cbdacf2c6e53c601", size = 6280384, upload-time = "2026-04-01T14:45:01.5Z" },
-    { url = "https://files.pythonhosted.org/packages/5e/03/688747d2e91cfbe0e64f316cd2e8005698f76ada3130d0194664174fa5de/pillow-12.2.0-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:7b14cc0106cd9aecda615dd6903840a058b4700fcb817687d0ee4fc8b6e389be", size = 8091599, upload-time = "2026-04-01T14:45:04.5Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/35/577e22b936fcdd66537329b33af0b4ccfefaeabd8aec04b266528cddb33c/pillow-12.2.0-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8cbeb542b2ebc6fcdacabf8aca8c1a97c9b3ad3927d46b8723f9d4f033288a0f", size = 6396021, upload-time = "2026-04-01T14:45:07.117Z" },
-    { url = "https://files.pythonhosted.org/packages/11/8d/d2532ad2a603ca2b93ad9f5135732124e57811d0168155852f37fbce2458/pillow-12.2.0-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4bfd07bc812fbd20395212969e41931001fd59eb55a60658b0e5710872e95286", size = 7083360, upload-time = "2026-04-01T14:45:09.763Z" },
-    { url = "https://files.pythonhosted.org/packages/5e/26/d325f9f56c7e039034897e7380e9cc202b1e368bfd04d4cbe6a441f02885/pillow-12.2.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:9aba9a17b623ef750a4d11b742cbafffeb48a869821252b30ee21b5e91392c50", size = 6507628, upload-time = "2026-04-01T14:45:12.378Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/f7/769d5632ffb0988f1c5e7660b3e731e30f7f8ec4318e94d0a5d674eb65a4/pillow-12.2.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:deede7c263feb25dba4e82ea23058a235dcc2fe1f6021025dc71f2b618e26104", size = 7209321, upload-time = "2026-04-01T14:45:15.122Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/ab/1b426a3974cb0e7da5c29ccff4807871d48110933a57207b5a676cccc155/pillow-12.2.0-cp314-cp314t-macosx_10_15_x86_64.whl", hash = "sha256:57850958fe9c751670e49b2cecf6294acc99e562531f4bd317fa5ddee2068463", size = 5314225, upload-time = "2026-04-01T14:45:25.637Z" },
-    { url = "https://files.pythonhosted.org/packages/19/1e/dce46f371be2438eecfee2a1960ee2a243bbe5e961890146d2dee1ff0f12/pillow-12.2.0-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:d5d38f1411c0ed9f97bcb49b7bd59b6b7c314e0e27420e34d99d844b9ce3b6f3", size = 4698541, upload-time = "2026-04-01T14:45:28.355Z" },
-    { url = "https://files.pythonhosted.org/packages/55/c3/7fbecf70adb3a0c33b77a300dc52e424dc22ad8cdc06557a2e49523b703d/pillow-12.2.0-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5c0a9f29ca8e79f09de89293f82fc9b0270bb4af1d58bc98f540cc4aedf03166", size = 6322251, upload-time = "2026-04-01T14:45:30.924Z" },
-    { url = "https://files.pythonhosted.org/packages/1c/3c/7fbc17cfb7e4fe0ef1642e0abc17fc6c94c9f7a16be41498e12e2ba60408/pillow-12.2.0-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:1610dd6c61621ae1cf811bef44d77e149ce3f7b95afe66a4512f8c59f25d9ebe", size = 8127807, upload-time = "2026-04-01T14:45:33.908Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/c3/a8ae14d6defd2e448493ff512fae903b1e9bd40b72efb6ec55ce0048c8ce/pillow-12.2.0-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0a34329707af4f73cf1782a36cd2289c0368880654a2c11f027bcee9052d35dd", size = 6433935, upload-time = "2026-04-01T14:45:36.623Z" },
-    { url = "https://files.pythonhosted.org/packages/6e/32/2880fb3a074847ac159d8f902cb43278a61e85f681661e7419e6596803ed/pillow-12.2.0-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8e9c4f5b3c546fa3458a29ab22646c1c6c787ea8f5ef51300e5a60300736905e", size = 7116720, upload-time = "2026-04-01T14:45:39.258Z" },
-    { url = "https://files.pythonhosted.org/packages/46/87/495cc9c30e0129501643f24d320076f4cc54f718341df18cc70ec94c44e1/pillow-12.2.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:fb043ee2f06b41473269765c2feae53fc2e2fbf96e5e22ca94fb5ad677856f06", size = 6540498, upload-time = "2026-04-01T14:45:41.879Z" },
-    { url = "https://files.pythonhosted.org/packages/18/53/773f5edca692009d883a72211b60fdaf8871cbef075eaa9d577f0a2f989e/pillow-12.2.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:f278f034eb75b4e8a13a54a876cc4a5ab39173d2cdd93a638e1b467fc545ac43", size = 7239413, upload-time = "2026-04-01T14:45:44.705Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/b7/2437044fb910f499610356d1352e3423753c98e34f915252aafecc64889f/pillow-12.2.0-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:0538bd5e05efec03ae613fd89c4ce0368ecd2ba239cc25b9f9be7ed426b0af1f", size = 5273969, upload-time = "2026-04-01T14:45:55.538Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/f4/8316e31de11b780f4ac08ef3654a75555e624a98db1056ecb2122d008d5a/pillow-12.2.0-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:394167b21da716608eac917c60aa9b969421b5dcbbe02ae7f013e7b85811c69d", size = 4659674, upload-time = "2026-04-01T14:45:58.093Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/37/664fca7201f8bb2aa1d20e2c3d5564a62e6ae5111741966c8319ca802361/pillow-12.2.0-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5d04bfa02cc2d23b497d1e90a0f927070043f6cbf303e738300532379a4b4e0f", size = 5288479, upload-time = "2026-04-01T14:46:01.141Z" },
-    { url = "https://files.pythonhosted.org/packages/49/62/5b0ed78fce87346be7a5cfcfaaad91f6a1f98c26f86bdbafa2066c647ef6/pillow-12.2.0-pp311-pypy311_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:0c838a5125cee37e68edec915651521191cef1e6aa336b855f495766e77a366e", size = 7032230, upload-time = "2026-04-01T14:46:03.874Z" },
-    { url = "https://files.pythonhosted.org/packages/c3/28/ec0fc38107fc32536908034e990c47914c57cd7c5a3ece4d8d8f7ffd7e27/pillow-12.2.0-pp311-pypy311_pp73-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4a6c9fa44005fa37a91ebfc95d081e8079757d2e904b27103f4f5fa6f0bf78c0", size = 5355404, upload-time = "2026-04-01T14:46:06.33Z" },
-    { url = "https://files.pythonhosted.org/packages/5e/8b/51b0eddcfa2180d60e41f06bd6d0a62202b20b59c68f5a132e615b75aecf/pillow-12.2.0-pp311-pypy311_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:25373b66e0dd5905ed63fa3cae13c82fbddf3079f2c8bf15c6fb6a35586324c1", size = 6002215, upload-time = "2026-04-01T14:46:08.83Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/46/5da1ec4a5171ee7bf1a0efa064aba70ba3d6e0788ce3f5acd1375d23c8c0/pillow-12.1.1-cp311-cp311-macosx_10_10_x86_64.whl", hash = "sha256:e879bb6cd5c73848ef3b2b48b8af9ff08c5b71ecda8048b7dd22d8a33f60be32", size = 5304084, upload-time = "2026-02-11T04:20:27.501Z" },
+    { url = "https://files.pythonhosted.org/packages/78/93/a29e9bc02d1cf557a834da780ceccd54e02421627200696fcf805ebdc3fb/pillow-12.1.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:365b10bb9417dd4498c0e3b128018c4a624dc11c7b97d8cc54effe3b096f4c38", size = 4657866, upload-time = "2026-02-11T04:20:29.827Z" },
+    { url = "https://files.pythonhosted.org/packages/13/84/583a4558d492a179d31e4aae32eadce94b9acf49c0337c4ce0b70e0a01f2/pillow-12.1.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d4ce8e329c93845720cd2014659ca67eac35f6433fd3050393d85f3ecef0dad5", size = 6232148, upload-time = "2026-02-11T04:20:31.329Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/e2/53c43334bbbb2d3b938978532fbda8e62bb6e0b23a26ce8592f36bcc4987/pillow-12.1.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:fc354a04072b765eccf2204f588a7a532c9511e8b9c7f900e1b64e3e33487090", size = 8038007, upload-time = "2026-02-11T04:20:34.225Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/a6/3d0e79c8a9d58150dd98e199d7c1c56861027f3829a3a60b3c2784190180/pillow-12.1.1-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:7e7976bf1910a8116b523b9f9f58bf410f3e8aa330cd9a2bb2953f9266ab49af", size = 6345418, upload-time = "2026-02-11T04:20:35.858Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/c8/46dfeac5825e600579157eea177be43e2f7ff4a99da9d0d0a49533509ac5/pillow-12.1.1-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:597bd9c8419bc7c6af5604e55847789b69123bbe25d65cc6ad3012b4f3c98d8b", size = 7034590, upload-time = "2026-02-11T04:20:37.91Z" },
+    { url = "https://files.pythonhosted.org/packages/af/bf/e6f65d3db8a8bbfeaf9e13cc0417813f6319863a73de934f14b2229ada18/pillow-12.1.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:2c1fc0f2ca5f96a3c8407e41cca26a16e46b21060fe6d5b099d2cb01412222f5", size = 6458655, upload-time = "2026-02-11T04:20:39.496Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/c2/66091f3f34a25894ca129362e510b956ef26f8fb67a0e6417bc5744e56f1/pillow-12.1.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:578510d88c6229d735855e1f278aa305270438d36a05031dfaae5067cc8eb04d", size = 7159286, upload-time = "2026-02-11T04:20:41.139Z" },
+    { url = "https://files.pythonhosted.org/packages/07/d3/8df65da0d4df36b094351dce696f2989bec731d4f10e743b1c5f4da4d3bf/pillow-12.1.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:ab323b787d6e18b3d91a72fc99b1a2c28651e4358749842b8f8dfacd28ef2052", size = 5262803, upload-time = "2026-02-11T04:20:47.653Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/71/5026395b290ff404b836e636f51d7297e6c83beceaa87c592718747e670f/pillow-12.1.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:adebb5bee0f0af4909c30db0d890c773d1a92ffe83da908e2e9e720f8edf3984", size = 4657601, upload-time = "2026-02-11T04:20:49.328Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/2e/1001613d941c67442f745aff0f7cc66dd8df9a9c084eb497e6a543ee6f7e/pillow-12.1.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:bb66b7cc26f50977108790e2456b7921e773f23db5630261102233eb355a3b79", size = 6234995, upload-time = "2026-02-11T04:20:51.032Z" },
+    { url = "https://files.pythonhosted.org/packages/07/26/246ab11455b2549b9233dbd44d358d033a2f780fa9007b61a913c5b2d24e/pillow-12.1.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:aee2810642b2898bb187ced9b349e95d2a7272930796e022efaf12e99dccd293", size = 8045012, upload-time = "2026-02-11T04:20:52.882Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/8b/07587069c27be7535ac1fe33874e32de118fbd34e2a73b7f83436a88368c/pillow-12.1.1-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a0b1cd6232e2b618adcc54d9882e4e662a089d5768cd188f7c245b4c8c44a397", size = 6349638, upload-time = "2026-02-11T04:20:54.444Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/79/6df7b2ee763d619cda2fb4fea498e5f79d984dae304d45a8999b80d6cf5c/pillow-12.1.1-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:7aac39bcf8d4770d089588a2e1dd111cbaa42df5a94be3114222057d68336bd0", size = 7041540, upload-time = "2026-02-11T04:20:55.97Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/5e/2ba19e7e7236d7529f4d873bdaf317a318896bac289abebd4bb00ef247f0/pillow-12.1.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:ab174cd7d29a62dd139c44bf74b698039328f45cb03b4596c43473a46656b2f3", size = 6462613, upload-time = "2026-02-11T04:20:57.542Z" },
+    { url = "https://files.pythonhosted.org/packages/03/03/31216ec124bb5c3dacd74ce8efff4cc7f52643653bad4825f8f08c697743/pillow-12.1.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:339ffdcb7cbeaa08221cd401d517d4b1fe7a9ed5d400e4a8039719238620ca35", size = 7166745, upload-time = "2026-02-11T04:20:59.196Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/11/6db24d4bd7685583caeae54b7009584e38da3c3d4488ed4cd25b439de486/pillow-12.1.1-cp313-cp313-ios_13_0_arm64_iphoneos.whl", hash = "sha256:d242e8ac078781f1de88bf823d70c1a9b3c7950a44cdf4b7c012e22ccbcd8e4e", size = 4062689, upload-time = "2026-02-11T04:21:06.804Z" },
+    { url = "https://files.pythonhosted.org/packages/33/c0/ce6d3b1fe190f0021203e0d9b5b99e57843e345f15f9ef22fcd43842fd21/pillow-12.1.1-cp313-cp313-ios_13_0_arm64_iphonesimulator.whl", hash = "sha256:02f84dfad02693676692746df05b89cf25597560db2857363a208e393429f5e9", size = 4138535, upload-time = "2026-02-11T04:21:08.452Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/c6/d5eb6a4fb32a3f9c21a8c7613ec706534ea1cf9f4b3663e99f0d83f6fca8/pillow-12.1.1-cp313-cp313-ios_13_0_x86_64_iphonesimulator.whl", hash = "sha256:e65498daf4b583091ccbb2556c7000abf0f3349fcd57ef7adc9a84a394ed29f6", size = 3601364, upload-time = "2026-02-11T04:21:10.194Z" },
+    { url = "https://files.pythonhosted.org/packages/14/a1/16c4b823838ba4c9c52c0e6bbda903a3fe5a1bdbf1b8eb4fff7156f3e318/pillow-12.1.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:6c6db3b84c87d48d0088943bf33440e0c42370b99b1c2a7989216f7b42eede60", size = 5262561, upload-time = "2026-02-11T04:21:11.742Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/ad/ad9dc98ff24f485008aa5cdedaf1a219876f6f6c42a4626c08bc4e80b120/pillow-12.1.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:8b7e5304e34942bf62e15184219a7b5ad4ff7f3bb5cca4d984f37df1a0e1aee2", size = 4657460, upload-time = "2026-02-11T04:21:13.786Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/1b/f1a4ea9a895b5732152789326202a82464d5254759fbacae4deea3069334/pillow-12.1.1-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:18e5bddd742a44b7e6b1e773ab5db102bd7a94c32555ba656e76d319d19c3850", size = 6232698, upload-time = "2026-02-11T04:21:15.949Z" },
+    { url = "https://files.pythonhosted.org/packages/95/f4/86f51b8745070daf21fd2e5b1fe0eb35d4db9ca26e6d58366562fb56a743/pillow-12.1.1-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:fc44ef1f3de4f45b50ccf9136999d71abb99dca7706bc75d222ed350b9fd2289", size = 8041706, upload-time = "2026-02-11T04:21:17.723Z" },
+    { url = "https://files.pythonhosted.org/packages/29/9b/d6ecd956bb1266dd1045e995cce9b8d77759e740953a1c9aad9502a0461e/pillow-12.1.1-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5a8eb7ed8d4198bccbd07058416eeec51686b498e784eda166395a23eb99138e", size = 6346621, upload-time = "2026-02-11T04:21:19.547Z" },
+    { url = "https://files.pythonhosted.org/packages/71/24/538bff45bde96535d7d998c6fed1a751c75ac7c53c37c90dc2601b243893/pillow-12.1.1-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:47b94983da0c642de92ced1702c5b6c292a84bd3a8e1d1702ff923f183594717", size = 7038069, upload-time = "2026-02-11T04:21:21.378Z" },
+    { url = "https://files.pythonhosted.org/packages/94/0e/58cb1a6bc48f746bc4cb3adb8cabff73e2742c92b3bf7a220b7cf69b9177/pillow-12.1.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:518a48c2aab7ce596d3bf79d0e275661b846e86e4d0e7dec34712c30fe07f02a", size = 6460040, upload-time = "2026-02-11T04:21:23.148Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/57/9045cb3ff11eeb6c1adce3b2d60d7d299d7b273a2e6c8381a524abfdc474/pillow-12.1.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:a550ae29b95c6dc13cf69e2c9dc5747f814c54eeb2e32d683e5e93af56caa029", size = 7164523, upload-time = "2026-02-11T04:21:25.01Z" },
+    { url = "https://files.pythonhosted.org/packages/19/2a/b9d62794fc8a0dd14c1943df68347badbd5511103e0d04c035ffe5cf2255/pillow-12.1.1-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:0330d233c1a0ead844fc097a7d16c0abff4c12e856c0b325f231820fee1f39da", size = 5264880, upload-time = "2026-02-11T04:21:32.865Z" },
+    { url = "https://files.pythonhosted.org/packages/26/9d/e03d857d1347fa5ed9247e123fcd2a97b6220e15e9cb73ca0a8d91702c6e/pillow-12.1.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:5dae5f21afb91322f2ff791895ddd8889e5e947ff59f71b46041c8ce6db790bc", size = 4660616, upload-time = "2026-02-11T04:21:34.97Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/ec/8a6d22afd02570d30954e043f09c32772bfe143ba9285e2fdb11284952cd/pillow-12.1.1-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:2e0c664be47252947d870ac0d327fea7e63985a08794758aa8af5b6cb6ec0c9c", size = 6269008, upload-time = "2026-02-11T04:21:36.623Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/1d/6d875422c9f28a4a361f495a5f68d9de4a66941dc2c619103ca335fa6446/pillow-12.1.1-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:691ab2ac363b8217f7d31b3497108fb1f50faab2f75dfb03284ec2f217e87bf8", size = 8073226, upload-time = "2026-02-11T04:21:38.585Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/cd/134b0b6ee5eda6dc09e25e24b40fdafe11a520bc725c1d0bbaa5e00bf95b/pillow-12.1.1-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e9e8064fb1cc019296958595f6db671fba95209e3ceb0c4734c9baf97de04b20", size = 6380136, upload-time = "2026-02-11T04:21:40.562Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/a9/7628f013f18f001c1b98d8fffe3452f306a70dc6aba7d931019e0492f45e/pillow-12.1.1-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:472a8d7ded663e6162dafdf20015c486a7009483ca671cece7a9279b512fcb13", size = 7067129, upload-time = "2026-02-11T04:21:42.521Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/f8/66ab30a2193b277785601e82ee2d49f68ea575d9637e5e234faaa98efa4c/pillow-12.1.1-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:89b54027a766529136a06cfebeecb3a04900397a3590fd252160b888479517bf", size = 6491807, upload-time = "2026-02-11T04:21:44.22Z" },
+    { url = "https://files.pythonhosted.org/packages/da/0b/a877a6627dc8318fdb84e357c5e1a758c0941ab1ddffdafd231983788579/pillow-12.1.1-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:86172b0831b82ce4f7877f280055892b31179e1576aa00d0df3bb1bbf8c3e524", size = 7190954, upload-time = "2026-02-11T04:21:46.114Z" },
+    { url = "https://files.pythonhosted.org/packages/03/d0/bebb3ffbf31c5a8e97241476c4cf8b9828954693ce6744b4a2326af3e16b/pillow-12.1.1-cp314-cp314-ios_13_0_arm64_iphoneos.whl", hash = "sha256:417423db963cb4be8bac3fc1204fe61610f6abeed1580a7a2cbb2fbda20f12af", size = 4062652, upload-time = "2026-02-11T04:21:53.19Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/c0/0e16fb0addda4851445c28f8350d8c512f09de27bbb0d6d0bbf8b6709605/pillow-12.1.1-cp314-cp314-ios_13_0_arm64_iphonesimulator.whl", hash = "sha256:b957b71c6b2387610f556a7eb0828afbe40b4a98036fc0d2acfa5a44a0c2036f", size = 4138823, upload-time = "2026-02-11T04:22:03.088Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/fb/6170ec655d6f6bb6630a013dd7cf7bc218423d7b5fa9071bf63dc32175ae/pillow-12.1.1-cp314-cp314-ios_13_0_x86_64_iphonesimulator.whl", hash = "sha256:097690ba1f2efdeb165a20469d59d8bb03c55fb6621eb2041a060ae8ea3e9642", size = 3601143, upload-time = "2026-02-11T04:22:04.909Z" },
+    { url = "https://files.pythonhosted.org/packages/59/04/dc5c3f297510ba9a6837cbb318b87dd2b8f73eb41a43cc63767f65cb599c/pillow-12.1.1-cp314-cp314-macosx_10_15_x86_64.whl", hash = "sha256:2815a87ab27848db0321fb78c7f0b2c8649dee134b7f2b80c6a45c6831d75ccd", size = 5266254, upload-time = "2026-02-11T04:22:07.656Z" },
+    { url = "https://files.pythonhosted.org/packages/05/30/5db1236b0d6313f03ebf97f5e17cda9ca060f524b2fcc875149a8360b21c/pillow-12.1.1-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:f7ed2c6543bad5a7d5530eb9e78c53132f93dfa44a28492db88b41cdab885202", size = 4657499, upload-time = "2026-02-11T04:22:09.613Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/18/008d2ca0eb612e81968e8be0bbae5051efba24d52debf930126d7eaacbba/pillow-12.1.1-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:652a2c9ccfb556235b2b501a3a7cf3742148cd22e04b5625c5fe057ea3e3191f", size = 6232137, upload-time = "2026-02-11T04:22:11.434Z" },
+    { url = "https://files.pythonhosted.org/packages/70/f1/f14d5b8eeb4b2cd62b9f9f847eb6605f103df89ef619ac68f92f748614ea/pillow-12.1.1-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:d6e4571eedf43af33d0fc233a382a76e849badbccdf1ac438841308652a08e1f", size = 8042721, upload-time = "2026-02-11T04:22:13.321Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/d6/17824509146e4babbdabf04d8171491fa9d776f7061ff6e727522df9bd03/pillow-12.1.1-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b574c51cf7d5d62e9be37ba446224b59a2da26dc4c1bb2ecbe936a4fb1a7cb7f", size = 6347798, upload-time = "2026-02-11T04:22:15.449Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/ee/c85a38a9ab92037a75615aba572c85ea51e605265036e00c5b67dfafbfe2/pillow-12.1.1-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a37691702ed687799de29a518d63d4682d9016932db66d4e90c345831b02fb4e", size = 7039315, upload-time = "2026-02-11T04:22:17.24Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/f3/bc8ccc6e08a148290d7523bde4d9a0d6c981db34631390dc6e6ec34cacf6/pillow-12.1.1-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:f95c00d5d6700b2b890479664a06e754974848afaae5e21beb4d83c106923fd0", size = 6462360, upload-time = "2026-02-11T04:22:19.111Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/ab/69a42656adb1d0665ab051eec58a41f169ad295cf81ad45406963105408f/pillow-12.1.1-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:559b38da23606e68681337ad74622c4dbba02254fc9cb4488a305dd5975c7eeb", size = 7165438, upload-time = "2026-02-11T04:22:21.041Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/9d/efd18493f9de13b87ede7c47e69184b9e859e4427225ea962e32e56a49bc/pillow-12.1.1-cp314-cp314t-macosx_10_15_x86_64.whl", hash = "sha256:1f90cff8aa76835cba5769f0b3121a22bd4eb9e6884cfe338216e557a9a548b8", size = 5268612, upload-time = "2026-02-11T04:22:29.884Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/f1/4f42eb2b388eb2ffc660dcb7f7b556c1015c53ebd5f7f754965ef997585b/pillow-12.1.1-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:1f1be78ce9466a7ee64bfda57bdba0f7cc499d9794d518b854816c41bf0aa4e9", size = 4660567, upload-time = "2026-02-11T04:22:31.799Z" },
+    { url = "https://files.pythonhosted.org/packages/01/54/df6ef130fa43e4b82e32624a7b821a2be1c5653a5fdad8469687a7db4e00/pillow-12.1.1-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:42fc1f4677106188ad9a55562bbade416f8b55456f522430fadab3cef7cd4e60", size = 6269951, upload-time = "2026-02-11T04:22:33.921Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/48/618752d06cc44bb4aae8ce0cd4e6426871929ed7b46215638088270d9b34/pillow-12.1.1-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:98edb152429ab62a1818039744d8fbb3ccab98a7c29fc3d5fcef158f3f1f68b7", size = 8074769, upload-time = "2026-02-11T04:22:35.877Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/bd/f1d71eb39a72fa088d938655afba3e00b38018d052752f435838961127d8/pillow-12.1.1-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d470ab1178551dd17fdba0fef463359c41aaa613cdcd7ff8373f54be629f9f8f", size = 6381358, upload-time = "2026-02-11T04:22:37.698Z" },
+    { url = "https://files.pythonhosted.org/packages/64/ef/c784e20b96674ed36a5af839305f55616f8b4f8aa8eeccf8531a6e312243/pillow-12.1.1-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6408a7b064595afcab0a49393a413732a35788f2a5092fdc6266952ed67de586", size = 7068558, upload-time = "2026-02-11T04:22:39.597Z" },
+    { url = "https://files.pythonhosted.org/packages/73/cb/8059688b74422ae61278202c4e1ad992e8a2e7375227be0a21c6b87ca8d5/pillow-12.1.1-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:5d8c41325b382c07799a3682c1c258469ea2ff97103c53717b7893862d0c98ce", size = 6493028, upload-time = "2026-02-11T04:22:42.73Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/da/e3c008ed7d2dd1f905b15949325934510b9d1931e5df999bb15972756818/pillow-12.1.1-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:c7697918b5be27424e9ce568193efd13d925c4481dd364e43f5dff72d33e10f8", size = 7191940, upload-time = "2026-02-11T04:22:44.543Z" },
+    { url = "https://files.pythonhosted.org/packages/56/11/5d43209aa4cb58e0cc80127956ff1796a68b928e6324bbf06ef4db34367b/pillow-12.1.1-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:600fd103672b925fe62ed08e0d874ea34d692474df6f4bf7ebe148b30f89f39f", size = 5228606, upload-time = "2026-02-11T04:22:52.106Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/d5/3b005b4e4fda6698b371fa6c21b097d4707585d7db99e98d9b0b87ac612a/pillow-12.1.1-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:665e1b916b043cef294bc54d47bf02d87e13f769bc4bc5fa225a24b3a6c5aca9", size = 4622321, upload-time = "2026-02-11T04:22:53.827Z" },
+    { url = "https://files.pythonhosted.org/packages/df/36/ed3ea2d594356fd8037e5a01f6156c74bc8d92dbb0fa60746cc96cabb6e8/pillow-12.1.1-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:495c302af3aad1ca67420ddd5c7bd480c8867ad173528767d906428057a11f0e", size = 5247579, upload-time = "2026-02-11T04:22:56.094Z" },
+    { url = "https://files.pythonhosted.org/packages/54/9a/9cc3e029683cf6d20ae5085da0dafc63148e3252c2f13328e553aaa13cfb/pillow-12.1.1-pp311-pypy311_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:8fd420ef0c52c88b5a035a0886f367748c72147b2b8f384c9d12656678dfdfa9", size = 6989094, upload-time = "2026-02-11T04:22:58.288Z" },
+    { url = "https://files.pythonhosted.org/packages/00/98/fc53ab36da80b88df0967896b6c4b4cd948a0dc5aa40a754266aa3ae48b3/pillow-12.1.1-pp311-pypy311_pp73-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f975aa7ef9684ce7e2c18a3aa8f8e2106ce1e46b94ab713d156b2898811651d3", size = 5313850, upload-time = "2026-02-11T04:23:00.554Z" },
+    { url = "https://files.pythonhosted.org/packages/30/02/00fa585abfd9fe9d73e5f6e554dc36cc2b842898cbfc46d70353dae227f8/pillow-12.1.1-pp311-pypy311_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8089c852a56c2966cf18835db62d9b34fef7ba74c726ad943928d494fa7f4735", size = 5963343, upload-time = "2026-02-11T04:23:02.934Z" },
 ]
 
 [[package]]
@@ -3754,7 +3769,7 @@ wheels = [
 
 [[package]]
 name = "pytest"
-version = "9.0.3"
+version = "9.0.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "iniconfig", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
@@ -3762,9 +3777,9 @@ dependencies = [
     { name = "pluggy", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
     { name = "pygments", marker = "sys_platform == 'darwin' or sys_platform == 'linux'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/7d/0d/549bd94f1a0a402dc8cf64563a117c0f3765662e2e668477624baeec44d5/pytest-9.0.3.tar.gz", hash = "sha256:b86ada508af81d19edeb213c681b1d48246c1a91d304c6c81a427674c17eb91c", size = 1572165, upload-time = "2026-04-07T17:16:18.027Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/d1/db/7ef3487e0fb0049ddb5ce41d3a49c235bf9ad299b6a25d5780a89f19230f/pytest-9.0.2.tar.gz", hash = "sha256:75186651a92bd89611d1d9fc20f0b4345fd827c41ccd5c299a868a05d70edf11", size = 1568901, upload-time = "2025-12-06T21:30:51.014Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d4/24/a372aaf5c9b7208e7112038812994107bc65a84cd00e0354a88c2c77a617/pytest-9.0.3-py3-none-any.whl", hash = "sha256:2c5efc453d45394fdd706ade797c0a81091eccd1d6e4bccfcd476e2b8e0ab5d9", size = 375249, upload-time = "2026-04-07T17:16:16.13Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/ab/b3226f0bd7cdcf710fbede2b3548584366da3b19b5021e74f5bde2a8fa3f/pytest-9.0.2-py3-none-any.whl", hash = "sha256:711ffd45bf766d5264d487b917733b453d917afd2b0ad65223959f59089f875b", size = 374801, upload-time = "2025-12-06T21:30:49.154Z" },
 ]
 
 [[package]]