A ChatGPT script for testing

.github/workflows/ci.yml

@@ -275,12 +275,8 @@ jobs:
   tests-frontend-e2e:
     name: "Frontend E2E Tests (Node ${{ matrix.node-version }} - ${{ matrix.shard-index }}/${{ matrix.shard-count }})"
     runs-on: ubuntu-24.04
-    container: mcr.microsoft.com/playwright:v1.57.0-noble
     needs:
       - install-frontend-dependencies
-    env:
-      PLAYWRIGHT_BROWSERS_PATH: /ms-playwright
-      PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD: 1
     strategy:
       fail-fast: false
       matrix:
@@ -309,8 +305,19 @@ jobs:
           key: ${{ runner.os }}-frontenddeps-${{ hashFiles('src-ui/pnpm-lock.yaml') }}
       - name: Re-link Angular cli
         run: cd src-ui && pnpm link @angular/cli
+      - name: Cache Playwright browsers
+        uses: actions/cache@v4
+        with:
+          path: ~/.cache/ms-playwright
+          key: ${{ runner.os }}-playwright-${{ hashFiles('src-ui/pnpm-lock.yaml') }}
+          restore-keys: |
+            ${{ runner.os }}-playwright-
+      - name: Install Playwright system dependencies
+        run: npx playwright install-deps
       - name: Install dependencies
         run: cd src-ui && pnpm install --no-frozen-lockfile
+      - name: Install Playwright
+        run: cd src-ui && pnpm exec playwright install
       - name: Run Playwright e2e tests
         run: cd src-ui && pnpm exec playwright test --shard ${{ matrix.shard-index }}/${{ matrix.shard-count }}
   frontend-bundle-analysis:

scripts/tag_perf_probe.py

@@ -0,0 +1,139 @@
+# noqa: INP001
+
+"""
+Ad-hoc script to gauge Tag + treenode performance locally.
+
+It bootstraps a fresh SQLite DB in a temp folder (or PAPERLESS_DATA_DIR),
+uses locmem cache/redis to avoid external services, creates synthetic tags,
+and measures:
+ - creation time
+ - query count and wall time for the Tag list view
+
+Usage:
+    PAPERLESS_DEBUG=1 PAPERLESS_REDIS=locmem:// PYTHONPATH=src \
+    PAPERLESS_DATA_DIR=/tmp/paperless-tags-probe \
+    .venv/bin/python scripts/tag_perf_probe.py
+"""
+
+import os
+import sys
+import time
+from collections.abc import Iterable
+from contextlib import contextmanager
+
+os.environ.setdefault("DJANGO_SETTINGS_MODULE", "paperless.settings")
+os.environ.setdefault("PAPERLESS_DEBUG", "1")
+os.environ.setdefault("PAPERLESS_REDIS", "locmem://")
+os.environ.setdefault("PYTHONPATH", "src")
+
+import django
+
+django.setup()
+
+from django.contrib.auth import get_user_model  # noqa: E402
+from django.core.management import call_command  # noqa: E402
+from django.db import connection  # noqa: E402
+from django.test.client import RequestFactory  # noqa: E402
+from rest_framework.test import force_authenticate  # noqa: E402
+from treenode.signals import no_signals  # noqa: E402
+
+from documents.models import Tag  # noqa: E402
+from documents.views import TagViewSet  # noqa: E402
+
+User = get_user_model()
+
+
+@contextmanager
+def count_queries():
+    total = 0
+
+    def wrapper(execute, sql, params, many, context):
+        nonlocal total
+        total += 1
+        return execute(sql, params, many, context)
+
+    with connection.execute_wrapper(wrapper):
+        yield lambda: total
+
+
+def measure_list(tag_count: int, user) -> tuple[int, float]:
+    """Render Tag list with page_size=tag_count and return (queries, seconds)."""
+    rf = RequestFactory()
+    view = TagViewSet.as_view({"get": "list"})
+    request = rf.get("/api/tags/", {"page_size": tag_count})
+    force_authenticate(request, user=user)
+
+    with count_queries() as get_count:
+        start = time.perf_counter()
+        response = view(request)
+        response.render()
+        elapsed = time.perf_counter() - start
+        total_queries = get_count()
+
+    return total_queries, elapsed
+
+
+def bulk_create_tags(count: int, parents: Iterable[Tag] | None = None) -> None:
+    """Create tags; when parents provided, create one child per parent."""
+    if parents is None:
+        Tag.objects.bulk_create([Tag(name=f"Flat {i}") for i in range(count)])
+        return
+
+    children = []
+    for p in parents:
+        children.append(Tag(name=f"Child {p.id}", tn_parent=p))
+    Tag.objects.bulk_create(children)
+
+
+def run():
+    # Ensure tables exist when pointing at a fresh DATA_DIR.
+    call_command("migrate", interactive=False, verbosity=0)
+
+    user, _ = User.objects.get_or_create(
+        username="admin",
+        defaults={"is_superuser": True, "is_staff": True},
+    )
+
+    # Flat scenario
+    Tag.objects.all().delete()
+    start = time.perf_counter()
+    bulk_create_tags(200)
+    flat_create = time.perf_counter() - start
+    q, t = measure_list(tag_count=200, user=user)
+    print(f"Flat create 200 -> {flat_create:.2f}s; list -> {q} queries, {t:.2f}s")  # noqa: T201
+
+    # Nested scenario (parents + 2 children each => 600 total)
+    Tag.objects.all().delete()
+    start = time.perf_counter()
+    with no_signals():  # avoid per-save tree rebuild; rebuild once
+        parents = Tag.objects.bulk_create([Tag(name=f"Parent {i}") for i in range(200)])
+        children = []
+        for p in parents:
+            children.extend(
+                Tag(name=f"Child {p.id}-{j}", tn_parent=p) for j in range(2)
+            )
+        Tag.objects.bulk_create(children)
+    Tag.update_tree()
+    nested_create = time.perf_counter() - start
+    q, t = measure_list(tag_count=600, user=user)
+    print(f"Nested create 600 -> {nested_create:.2f}s; list -> {q} queries, {t:.2f}s")  # noqa: T201
+
+    # Larger nested scenario (1 child per parent, 3000 total)
+    Tag.objects.all().delete()
+    start = time.perf_counter()
+    with no_signals():
+        parents = Tag.objects.bulk_create(
+            [Tag(name=f"Parent {i}") for i in range(1500)],
+        )
+        bulk_create_tags(0, parents=parents)
+    Tag.update_tree()
+    big_create = time.perf_counter() - start
+    q, t = measure_list(tag_count=3000, user=user)
+    print(f"Nested create 3000 -> {big_create:.2f}s; list -> {q} queries, {t:.2f}s")  # noqa: T201
+
+
+if __name__ == "__main__":
+    if "runserver" in sys.argv:
+        print("Run directly: .venv/bin/python scripts/tag_perf_probe.py")  # noqa: T201
+        sys.exit(1)
+    run()

src-ui/src/app/components/common/edit-dialog/workflow-edit-dialog/workflow-edit-dialog.component.spec.ts

@@ -416,9 +416,6 @@ describe('WorkflowEditDialogComponent', () => {
       return newFilter
     }
 
-    const correspondentAny = addFilterOfType(TriggerFilterType.CorrespondentAny)
-    correspondentAny.get('values').setValue([11])
-
     const correspondentIs = addFilterOfType(TriggerFilterType.CorrespondentIs)
     correspondentIs.get('values').setValue(1)
 
@@ -428,18 +425,12 @@ describe('WorkflowEditDialogComponent', () => {
     const documentTypeIs = addFilterOfType(TriggerFilterType.DocumentTypeIs)
     documentTypeIs.get('values').setValue(1)
 
-    const documentTypeAny = addFilterOfType(TriggerFilterType.DocumentTypeAny)
-    documentTypeAny.get('values').setValue([12])
-
     const documentTypeNot = addFilterOfType(TriggerFilterType.DocumentTypeNot)
     documentTypeNot.get('values').setValue([1])
 
     const storagePathIs = addFilterOfType(TriggerFilterType.StoragePathIs)
     storagePathIs.get('values').setValue(1)
 
-    const storagePathAny = addFilterOfType(TriggerFilterType.StoragePathAny)
-    storagePathAny.get('values').setValue([13])
-
     const storagePathNot = addFilterOfType(TriggerFilterType.StoragePathNot)
     storagePathNot.get('values').setValue([1])
 
@@ -454,13 +445,10 @@ describe('WorkflowEditDialogComponent', () => {
     expect(formValues.triggers[0].filter_has_tags).toEqual([1])
     expect(formValues.triggers[0].filter_has_all_tags).toEqual([2, 3])
     expect(formValues.triggers[0].filter_has_not_tags).toEqual([4])
-    expect(formValues.triggers[0].filter_has_any_correspondents).toEqual([11])
     expect(formValues.triggers[0].filter_has_correspondent).toEqual(1)
     expect(formValues.triggers[0].filter_has_not_correspondents).toEqual([1])
-    expect(formValues.triggers[0].filter_has_any_document_types).toEqual([12])
     expect(formValues.triggers[0].filter_has_document_type).toEqual(1)
     expect(formValues.triggers[0].filter_has_not_document_types).toEqual([1])
-    expect(formValues.triggers[0].filter_has_any_storage_paths).toEqual([13])
     expect(formValues.triggers[0].filter_has_storage_path).toEqual(1)
     expect(formValues.triggers[0].filter_has_not_storage_paths).toEqual([1])
     expect(formValues.triggers[0].filter_custom_field_query).toEqual(
@@ -523,22 +511,16 @@ describe('WorkflowEditDialogComponent', () => {
 
     setFilter(TriggerFilterType.TagsAll, 11)
     setFilter(TriggerFilterType.TagsNone, 12)
-    setFilter(TriggerFilterType.CorrespondentAny, 16)
     setFilter(TriggerFilterType.CorrespondentNot, 13)
-    setFilter(TriggerFilterType.DocumentTypeAny, 17)
     setFilter(TriggerFilterType.DocumentTypeNot, 14)
-    setFilter(TriggerFilterType.StoragePathAny, 18)
     setFilter(TriggerFilterType.StoragePathNot, 15)
 
     const formValues = component['getFormValues']()
 
     expect(formValues.triggers[0].filter_has_all_tags).toEqual([11])
     expect(formValues.triggers[0].filter_has_not_tags).toEqual([12])
-    expect(formValues.triggers[0].filter_has_any_correspondents).toEqual([16])
     expect(formValues.triggers[0].filter_has_not_correspondents).toEqual([13])
-    expect(formValues.triggers[0].filter_has_any_document_types).toEqual([17])
     expect(formValues.triggers[0].filter_has_not_document_types).toEqual([14])
-    expect(formValues.triggers[0].filter_has_any_storage_paths).toEqual([18])
     expect(formValues.triggers[0].filter_has_not_storage_paths).toEqual([15])
   })
 
@@ -662,11 +644,8 @@ describe('WorkflowEditDialogComponent', () => {
       filter_has_tags: [],
       filter_has_all_tags: [],
       filter_has_not_tags: [],
-      filter_has_any_correspondents: [],
       filter_has_not_correspondents: [],
-      filter_has_any_document_types: [],
       filter_has_not_document_types: [],
-      filter_has_any_storage_paths: [],
       filter_has_not_storage_paths: [],
       filter_has_correspondent: null,
       filter_has_document_type: null,
@@ -724,14 +703,11 @@ describe('WorkflowEditDialogComponent', () => {
     trigger.filter_has_tags = [1]
     trigger.filter_has_all_tags = [2, 3]
     trigger.filter_has_not_tags = [4]
-    trigger.filter_has_any_correspondents = [10] as any
     trigger.filter_has_correspondent = 5 as any
     trigger.filter_has_not_correspondents = [6] as any
     trigger.filter_has_document_type = 7 as any
-    trigger.filter_has_any_document_types = [11] as any
     trigger.filter_has_not_document_types = [8] as any
     trigger.filter_has_storage_path = 9 as any
-    trigger.filter_has_any_storage_paths = [12] as any
     trigger.filter_has_not_storage_paths = [10] as any
     trigger.filter_custom_field_query = JSON.stringify([
       'AND',
@@ -742,8 +718,8 @@ describe('WorkflowEditDialogComponent', () => {
     component.ngOnInit()
     const triggerGroup = component.triggerFields.at(0) as FormGroup
     const filters = component.getFiltersFormArray(triggerGroup)
-    expect(filters.length).toBe(13)
-    const customFieldFilter = filters.at(12) as FormGroup
+    expect(filters.length).toBe(10)
+    const customFieldFilter = filters.at(9) as FormGroup
     expect(customFieldFilter.get('type').value).toBe(
       TriggerFilterType.CustomFieldQuery
     )
@@ -752,27 +728,12 @@ describe('WorkflowEditDialogComponent', () => {
   })
 
   it('should expose select metadata helpers', () => {
-    expect(component.isSelectMultiple(TriggerFilterType.CorrespondentAny)).toBe(
-      true
-    )
     expect(component.isSelectMultiple(TriggerFilterType.CorrespondentNot)).toBe(
       true
     )
     expect(component.isSelectMultiple(TriggerFilterType.CorrespondentIs)).toBe(
       false
     )
-    expect(component.isSelectMultiple(TriggerFilterType.DocumentTypeAny)).toBe(
-      true
-    )
-    expect(component.isSelectMultiple(TriggerFilterType.DocumentTypeIs)).toBe(
-      false
-    )
-    expect(component.isSelectMultiple(TriggerFilterType.StoragePathAny)).toBe(
-      true
-    )
-    expect(component.isSelectMultiple(TriggerFilterType.StoragePathIs)).toBe(
-      false
-    )
 
     component.correspondents = [{ id: 1, name: 'C1' } as any]
     component.documentTypes = [{ id: 2, name: 'DT' } as any]
@@ -784,15 +745,9 @@ describe('WorkflowEditDialogComponent', () => {
     expect(
       component.getFilterSelectItems(TriggerFilterType.DocumentTypeIs)
     ).toEqual(component.documentTypes)
-    expect(
-      component.getFilterSelectItems(TriggerFilterType.DocumentTypeAny)
-    ).toEqual(component.documentTypes)
     expect(
       component.getFilterSelectItems(TriggerFilterType.StoragePathIs)
     ).toEqual(component.storagePaths)
-    expect(
-      component.getFilterSelectItems(TriggerFilterType.StoragePathAny)
-    ).toEqual(component.storagePaths)
     expect(component.getFilterSelectItems(TriggerFilterType.TagsAll)).toEqual(
       []
     )

src-ui/src/app/components/common/edit-dialog/workflow-edit-dialog/workflow-edit-dialog.component.ts

@@ -145,13 +145,10 @@ export enum TriggerFilterType {
   TagsAny = 'tags_any',
   TagsAll = 'tags_all',
   TagsNone = 'tags_none',
-  CorrespondentAny = 'correspondent_any',
   CorrespondentIs = 'correspondent_is',
   CorrespondentNot = 'correspondent_not',
-  DocumentTypeAny = 'document_type_any',
   DocumentTypeIs = 'document_type_is',
   DocumentTypeNot = 'document_type_not',
-  StoragePathAny = 'storage_path_any',
   StoragePathIs = 'storage_path_is',
   StoragePathNot = 'storage_path_not',
   CustomFieldQuery = 'custom_field_query',
@@ -175,11 +172,8 @@ type TriggerFilterAggregate = {
   filter_has_tags: number[]
   filter_has_all_tags: number[]
   filter_has_not_tags: number[]
-  filter_has_any_correspondents: number[]
   filter_has_not_correspondents: number[]
-  filter_has_any_document_types: number[]
   filter_has_not_document_types: number[]
-  filter_has_any_storage_paths: number[]
   filter_has_not_storage_paths: number[]
   filter_has_correspondent: number | null
   filter_has_document_type: number | null
@@ -225,14 +219,6 @@ const TRIGGER_FILTER_DEFINITIONS: TriggerFilterDefinition[] = [
     allowMultipleEntries: false,
     allowMultipleValues: true,
   },
-  {
-    id: TriggerFilterType.CorrespondentAny,
-    name: $localize`Has any of these correspondents`,
-    inputType: 'select',
-    allowMultipleEntries: false,
-    allowMultipleValues: true,
-    selectItems: 'correspondents',
-  },
   {
     id: TriggerFilterType.CorrespondentIs,
     name: $localize`Has correspondent`,
@@ -257,14 +243,6 @@ const TRIGGER_FILTER_DEFINITIONS: TriggerFilterDefinition[] = [
     allowMultipleValues: false,
     selectItems: 'documentTypes',
   },
-  {
-    id: TriggerFilterType.DocumentTypeAny,
-    name: $localize`Has any of these document types`,
-    inputType: 'select',
-    allowMultipleEntries: false,
-    allowMultipleValues: true,
-    selectItems: 'documentTypes',
-  },
   {
     id: TriggerFilterType.DocumentTypeNot,
     name: $localize`Does not have document types`,
@@ -281,14 +259,6 @@ const TRIGGER_FILTER_DEFINITIONS: TriggerFilterDefinition[] = [
     allowMultipleValues: false,
     selectItems: 'storagePaths',
   },
-  {
-    id: TriggerFilterType.StoragePathAny,
-    name: $localize`Has any of these storage paths`,
-    inputType: 'select',
-    allowMultipleEntries: false,
-    allowMultipleValues: true,
-    selectItems: 'storagePaths',
-  },
   {
     id: TriggerFilterType.StoragePathNot,
     name: $localize`Does not have storage paths`,
@@ -336,15 +306,6 @@ const FILTER_HANDLERS: Record<TriggerFilterType, FilterHandler> = {
     extract: (trigger) => trigger.filter_has_not_tags,
     hasValue: (value) => Array.isArray(value) && value.length > 0,
   },
-  [TriggerFilterType.CorrespondentAny]: {
-    apply: (aggregate, values) => {
-      aggregate.filter_has_any_correspondents = Array.isArray(values)
-        ? [...values]
-        : [values]
-    },
-    extract: (trigger) => trigger.filter_has_any_correspondents,
-    hasValue: (value) => Array.isArray(value) && value.length > 0,
-  },
   [TriggerFilterType.CorrespondentIs]: {
     apply: (aggregate, values) => {
       aggregate.filter_has_correspondent = Array.isArray(values)
@@ -372,15 +333,6 @@ const FILTER_HANDLERS: Record<TriggerFilterType, FilterHandler> = {
     extract: (trigger) => trigger.filter_has_document_type,
     hasValue: (value) => value !== null && value !== undefined,
   },
-  [TriggerFilterType.DocumentTypeAny]: {
-    apply: (aggregate, values) => {
-      aggregate.filter_has_any_document_types = Array.isArray(values)
-        ? [...values]
-        : [values]
-    },
-    extract: (trigger) => trigger.filter_has_any_document_types,
-    hasValue: (value) => Array.isArray(value) && value.length > 0,
-  },
   [TriggerFilterType.DocumentTypeNot]: {
     apply: (aggregate, values) => {
       aggregate.filter_has_not_document_types = Array.isArray(values)
@@ -399,15 +351,6 @@ const FILTER_HANDLERS: Record<TriggerFilterType, FilterHandler> = {
     extract: (trigger) => trigger.filter_has_storage_path,
     hasValue: (value) => value !== null && value !== undefined,
   },
-  [TriggerFilterType.StoragePathAny]: {
-    apply: (aggregate, values) => {
-      aggregate.filter_has_any_storage_paths = Array.isArray(values)
-        ? [...values]
-        : [values]
-    },
-    extract: (trigger) => trigger.filter_has_any_storage_paths,
-    hasValue: (value) => Array.isArray(value) && value.length > 0,
-  },
   [TriggerFilterType.StoragePathNot]: {
     apply: (aggregate, values) => {
       aggregate.filter_has_not_storage_paths = Array.isArray(values)
@@ -699,11 +642,8 @@ export class WorkflowEditDialogComponent
             filter_has_tags: [],
             filter_has_all_tags: [],
             filter_has_not_tags: [],
-            filter_has_any_correspondents: [],
             filter_has_not_correspondents: [],
-            filter_has_any_document_types: [],
             filter_has_not_document_types: [],
-            filter_has_any_storage_paths: [],
             filter_has_not_storage_paths: [],
             filter_has_correspondent: null,
             filter_has_document_type: null,
@@ -730,16 +670,10 @@ export class WorkflowEditDialogComponent
           trigger.filter_has_tags = aggregate.filter_has_tags
           trigger.filter_has_all_tags = aggregate.filter_has_all_tags
           trigger.filter_has_not_tags = aggregate.filter_has_not_tags
-          trigger.filter_has_any_correspondents =
-            aggregate.filter_has_any_correspondents
           trigger.filter_has_not_correspondents =
             aggregate.filter_has_not_correspondents
-          trigger.filter_has_any_document_types =
-            aggregate.filter_has_any_document_types
           trigger.filter_has_not_document_types =
             aggregate.filter_has_not_document_types
-          trigger.filter_has_any_storage_paths =
-            aggregate.filter_has_any_storage_paths
           trigger.filter_has_not_storage_paths =
             aggregate.filter_has_not_storage_paths
           trigger.filter_has_correspondent =
@@ -922,11 +856,8 @@ export class WorkflowEditDialogComponent
       case TriggerFilterType.TagsAny:
       case TriggerFilterType.TagsAll:
       case TriggerFilterType.TagsNone:
-      case TriggerFilterType.CorrespondentAny:
       case TriggerFilterType.CorrespondentNot:
-      case TriggerFilterType.DocumentTypeAny:
       case TriggerFilterType.DocumentTypeNot:
-      case TriggerFilterType.StoragePathAny:
       case TriggerFilterType.StoragePathNot:
         return true
       default:
@@ -1248,11 +1179,8 @@ export class WorkflowEditDialogComponent
       filter_has_tags: [],
       filter_has_all_tags: [],
       filter_has_not_tags: [],
-      filter_has_any_correspondents: [],
       filter_has_not_correspondents: [],
-      filter_has_any_document_types: [],
       filter_has_not_document_types: [],
-      filter_has_any_storage_paths: [],
       filter_has_not_storage_paths: [],
       filter_custom_field_query: null,
       filter_has_correspondent: null,

src-ui/src/app/data/workflow-trigger.ts

@@ -44,16 +44,10 @@ export interface WorkflowTrigger extends ObjectWithId {
 
   filter_has_not_tags?: number[] // Tag.id[]
 
-  filter_has_any_correspondents?: number[] // Correspondent.id[]
-
   filter_has_not_correspondents?: number[] // Correspondent.id[]
 
-  filter_has_any_document_types?: number[] // DocumentType.id[]
-
   filter_has_not_document_types?: number[] // DocumentType.id[]
 
-  filter_has_any_storage_paths?: number[] // StoragePath.id[]
-
   filter_has_not_storage_paths?: number[] // StoragePath.id[]
 
   filter_custom_field_query?: string

src/documents/apps.py

@@ -1,5 +1,9 @@
 from django.apps import AppConfig
+from django.db.models.signals import post_delete
+from django.db.models.signals import post_save
 from django.utils.translation import gettext_lazy as _
+from treenode.signals import post_delete_treenode
+from treenode.signals import post_save_treenode
 
 
 class DocumentsConfig(AppConfig):
@@ -8,12 +12,14 @@ class DocumentsConfig(AppConfig):
     verbose_name = _("Documents")
 
     def ready(self):
+        from documents.models import Tag
         from documents.signals import document_consumption_finished
         from documents.signals import document_updated
         from documents.signals.handlers import add_inbox_tags
         from documents.signals.handlers import add_to_index
         from documents.signals.handlers import run_workflows_added
         from documents.signals.handlers import run_workflows_updated
+        from documents.signals.handlers import schedule_tag_tree_update
         from documents.signals.handlers import set_correspondent
         from documents.signals.handlers import set_document_type
         from documents.signals.handlers import set_storage_path
@@ -28,6 +34,29 @@ class DocumentsConfig(AppConfig):
         document_consumption_finished.connect(run_workflows_added)
         document_updated.connect(run_workflows_updated)
 
+        # treenode updates the entire tree on every save/delete via hooks
+        # so disconnect for Tags and run once-per-transaction.
+        post_save.disconnect(
+            post_save_treenode,
+            sender=Tag,
+            dispatch_uid="post_save_treenode",
+        )
+        post_delete.disconnect(
+            post_delete_treenode,
+            sender=Tag,
+            dispatch_uid="post_delete_treenode",
+        )
+        post_save.connect(
+            schedule_tag_tree_update,
+            sender=Tag,
+            dispatch_uid="paperless_tag_mark_dirty_save",
+        )
+        post_delete.connect(
+            schedule_tag_tree_update,
+            sender=Tag,
+            dispatch_uid="paperless_tag_mark_dirty_delete",
+        )
+
         import documents.schema  # noqa: F401
 
         AppConfig.ready(self)

src/documents/matching.py

@@ -403,18 +403,6 @@ def existing_document_matches_workflow(
                 f"Document tags {list(document.tags.all())} include excluded tags {list(trigger_has_not_tags_qs)}",
             )
 
-    allowed_correspondent_ids = set(
-        trigger.filter_has_any_correspondents.values_list("id", flat=True),
-    )
-    if (
-        allowed_correspondent_ids
-        and document.correspondent_id not in allowed_correspondent_ids
-    ):
-        return (
-            False,
-            f"Document correspondent {document.correspondent} is not one of {list(trigger.filter_has_any_correspondents.all())}",
-        )
-
     # Document correspondent vs trigger has_correspondent
     if (
         trigger.filter_has_correspondent_id is not None
@@ -436,17 +424,6 @@ def existing_document_matches_workflow(
             f"Document correspondent {document.correspondent} is excluded by {list(trigger.filter_has_not_correspondents.all())}",
         )
 
-    allowed_document_type_ids = set(
-        trigger.filter_has_any_document_types.values_list("id", flat=True),
-    )
-    if allowed_document_type_ids and (
-        document.document_type_id not in allowed_document_type_ids
-    ):
-        return (
-            False,
-            f"Document doc type {document.document_type} is not one of {list(trigger.filter_has_any_document_types.all())}",
-        )
-
     # Document document_type vs trigger has_document_type
     if (
         trigger.filter_has_document_type_id is not None
@@ -468,17 +445,6 @@ def existing_document_matches_workflow(
             f"Document doc type {document.document_type} is excluded by {list(trigger.filter_has_not_document_types.all())}",
         )
 
-    allowed_storage_path_ids = set(
-        trigger.filter_has_any_storage_paths.values_list("id", flat=True),
-    )
-    if allowed_storage_path_ids and (
-        document.storage_path_id not in allowed_storage_path_ids
-    ):
-        return (
-            False,
-            f"Document storage path {document.storage_path} is not one of {list(trigger.filter_has_any_storage_paths.all())}",
-        )
-
     # Document storage_path vs trigger has_storage_path
     if (
         trigger.filter_has_storage_path_id is not None
@@ -566,10 +532,6 @@ def prefilter_documents_by_workflowtrigger(
 
     # Correspondent, DocumentType, etc. filtering
 
-    if trigger.filter_has_any_correspondents.exists():
-        documents = documents.filter(
-            correspondent__in=trigger.filter_has_any_correspondents.all(),
-        )
     if trigger.filter_has_correspondent is not None:
         documents = documents.filter(
             correspondent=trigger.filter_has_correspondent,
@@ -579,10 +541,6 @@ def prefilter_documents_by_workflowtrigger(
             correspondent__in=trigger.filter_has_not_correspondents.all(),
         )
 
-    if trigger.filter_has_any_document_types.exists():
-        documents = documents.filter(
-            document_type__in=trigger.filter_has_any_document_types.all(),
-        )
     if trigger.filter_has_document_type is not None:
         documents = documents.filter(
             document_type=trigger.filter_has_document_type,
@@ -592,10 +550,6 @@ def prefilter_documents_by_workflowtrigger(
             document_type__in=trigger.filter_has_not_document_types.all(),
         )
 
-    if trigger.filter_has_any_storage_paths.exists():
-        documents = documents.filter(
-            storage_path__in=trigger.filter_has_any_storage_paths.all(),
-        )
     if trigger.filter_has_storage_path is not None:
         documents = documents.filter(
             storage_path=trigger.filter_has_storage_path,
@@ -650,11 +604,8 @@ def document_matches_workflow(
             "filter_has_tags",
             "filter_has_all_tags",
             "filter_has_not_tags",
-            "filter_has_any_document_types",
             "filter_has_not_document_types",
-            "filter_has_any_correspondents",
             "filter_has_not_correspondents",
-            "filter_has_any_storage_paths",
             "filter_has_not_storage_paths",
         )
     )

src/documents/migrations/1075_workflowtrigger_filter_has_any_correspondents_and_more.py

@@ -1,43 +0,0 @@
-# Generated by Django 5.2.7 on 2025-12-17 22:25
-
-from django.db import migrations
-from django.db import models
-
-
-class Migration(migrations.Migration):
-    dependencies = [
-        ("documents", "1074_workflowrun_deleted_at_workflowrun_restored_at_and_more"),
-    ]
-
-    operations = [
-        migrations.AddField(
-            model_name="workflowtrigger",
-            name="filter_has_any_correspondents",
-            field=models.ManyToManyField(
-                blank=True,
-                related_name="workflowtriggers_has_any_correspondent",
-                to="documents.correspondent",
-                verbose_name="has one of these correspondents",
-            ),
-        ),
-        migrations.AddField(
-            model_name="workflowtrigger",
-            name="filter_has_any_document_types",
-            field=models.ManyToManyField(
-                blank=True,
-                related_name="workflowtriggers_has_any_document_type",
-                to="documents.documenttype",
-                verbose_name="has one of these document types",
-            ),
-        ),
-        migrations.AddField(
-            model_name="workflowtrigger",
-            name="filter_has_any_storage_paths",
-            field=models.ManyToManyField(
-                blank=True,
-                related_name="workflowtriggers_has_any_storage_path",
-                to="documents.storagepath",
-                verbose_name="has one of these storage paths",
-            ),
-        ),
-    ]

src/documents/models.py

@@ -1087,13 +1087,6 @@ class WorkflowTrigger(models.Model):
         verbose_name=_("has this document type"),
     )
 
-    filter_has_any_document_types = models.ManyToManyField(
-        DocumentType,
-        blank=True,
-        related_name="workflowtriggers_has_any_document_type",
-        verbose_name=_("has one of these document types"),
-    )
-
     filter_has_not_document_types = models.ManyToManyField(
         DocumentType,
         blank=True,
@@ -1116,13 +1109,6 @@ class WorkflowTrigger(models.Model):
         verbose_name=_("does not have these correspondent(s)"),
     )
 
-    filter_has_any_correspondents = models.ManyToManyField(
-        Correspondent,
-        blank=True,
-        related_name="workflowtriggers_has_any_correspondent",
-        verbose_name=_("has one of these correspondents"),
-    )
-
     filter_has_storage_path = models.ForeignKey(
         StoragePath,
         null=True,
@@ -1131,13 +1117,6 @@ class WorkflowTrigger(models.Model):
         verbose_name=_("has this storage path"),
     )
 
-    filter_has_any_storage_paths = models.ManyToManyField(
-        StoragePath,
-        blank=True,
-        related_name="workflowtriggers_has_any_storage_path",
-        verbose_name=_("has one of these storage paths"),
-    )
-
     filter_has_not_storage_paths = models.ManyToManyField(
         StoragePath,
         blank=True,

src/documents/serialisers.py

@@ -578,30 +578,34 @@ class TagSerializer(MatchingModelSerializer, OwnedObjectSerializer):
         ),
     )
     def get_children(self, obj):
-        filter_q = self.context.get("document_count_filter")
-        request = self.context.get("request")
-        if filter_q is None:
-            user = getattr(request, "user", None) if request else None
-            filter_q = get_document_count_filter_for_user(user)
-            self.context["document_count_filter"] = filter_q
+        children_map = self.context.get("children_map")
+        if children_map is not None:
+            children = children_map.get(obj.pk, [])
+        else:
+            filter_q = self.context.get("document_count_filter")
+            request = self.context.get("request")
+            if filter_q is None:
+                user = getattr(request, "user", None) if request else None
+                filter_q = get_document_count_filter_for_user(user)
+                self.context["document_count_filter"] = filter_q
 
-        children_queryset = (
-            obj.get_children_queryset()
-            .select_related("owner")
-            .annotate(document_count=Count("documents", filter=filter_q))
-        )
+            children = (
+                obj.get_children_queryset()
+                .select_related("owner")
+                .annotate(document_count=Count("documents", filter=filter_q))
+            )
 
-        view = self.context.get("view")
-        ordering = (
-            OrderingFilter().get_ordering(request, children_queryset, view)
-            if request and view
-            else None
-        )
-        ordering = ordering or (Lower("name"),)
-        children_queryset = children_queryset.order_by(*ordering)
+            view = self.context.get("view")
+            ordering = (
+                OrderingFilter().get_ordering(request, children, view)
+                if request and view
+                else None
+            )
+            ordering = ordering or (Lower("name"),)
+            children = children.order_by(*ordering)
 
         serializer = TagSerializer(
-            children_queryset,
+            children,
             many=True,
             user=self.user,
             full_perms=self.full_perms,
@@ -2275,11 +2279,8 @@ class WorkflowTriggerSerializer(serializers.ModelSerializer):
             "filter_has_all_tags",
             "filter_has_not_tags",
             "filter_custom_field_query",
-            "filter_has_any_correspondents",
             "filter_has_not_correspondents",
-            "filter_has_any_document_types",
             "filter_has_not_document_types",
-            "filter_has_any_storage_paths",
             "filter_has_not_storage_paths",
             "filter_has_correspondent",
             "filter_has_document_type",
@@ -2517,26 +2518,14 @@ class WorkflowSerializer(serializers.ModelSerializer):
                 filter_has_tags = trigger.pop("filter_has_tags", None)
                 filter_has_all_tags = trigger.pop("filter_has_all_tags", None)
                 filter_has_not_tags = trigger.pop("filter_has_not_tags", None)
-                filter_has_any_correspondents = trigger.pop(
-                    "filter_has_any_correspondents",
-                    None,
-                )
                 filter_has_not_correspondents = trigger.pop(
                     "filter_has_not_correspondents",
                     None,
                 )
-                filter_has_any_document_types = trigger.pop(
-                    "filter_has_any_document_types",
-                    None,
-                )
                 filter_has_not_document_types = trigger.pop(
                     "filter_has_not_document_types",
                     None,
                 )
-                filter_has_any_storage_paths = trigger.pop(
-                    "filter_has_any_storage_paths",
-                    None,
-                )
                 filter_has_not_storage_paths = trigger.pop(
                     "filter_has_not_storage_paths",
                     None,
@@ -2553,26 +2542,14 @@ class WorkflowSerializer(serializers.ModelSerializer):
                     trigger_instance.filter_has_all_tags.set(filter_has_all_tags)
                 if filter_has_not_tags is not None:
                     trigger_instance.filter_has_not_tags.set(filter_has_not_tags)
-                if filter_has_any_correspondents is not None:
-                    trigger_instance.filter_has_any_correspondents.set(
-                        filter_has_any_correspondents,
-                    )
                 if filter_has_not_correspondents is not None:
                     trigger_instance.filter_has_not_correspondents.set(
                         filter_has_not_correspondents,
                     )
-                if filter_has_any_document_types is not None:
-                    trigger_instance.filter_has_any_document_types.set(
-                        filter_has_any_document_types,
-                    )
                 if filter_has_not_document_types is not None:
                     trigger_instance.filter_has_not_document_types.set(
                         filter_has_not_document_types,
                     )
-                if filter_has_any_storage_paths is not None:
-                    trigger_instance.filter_has_any_storage_paths.set(
-                        filter_has_any_storage_paths,
-                    )
                 if filter_has_not_storage_paths is not None:
                     trigger_instance.filter_has_not_storage_paths.set(
                         filter_has_not_storage_paths,

src/documents/signals/handlers.py

@@ -19,6 +19,7 @@ from django.db import DatabaseError
 from django.db import close_old_connections
 from django.db import connections
 from django.db import models
+from django.db import transaction
 from django.db.models import Q
 from django.dispatch import receiver
 from django.utils import timezone
@@ -60,6 +61,8 @@ if TYPE_CHECKING:
 
 logger = logging.getLogger("paperless.handlers")
 
+_tag_tree_update_scheduled = False
+
 
 def add_inbox_tags(sender, document: Document, logging_group=None, **kwargs):
     if document.owner is not None:
@@ -944,3 +947,26 @@ def close_connection_pool_on_worker_init(**kwargs):
     for conn in connections.all(initialized_only=True):
         if conn.alias == "default" and hasattr(conn, "pool") and conn.pool:
             conn.close_pool()
+
+
+def schedule_tag_tree_update(**_kwargs):
+    """
+    Schedule a single Tag.update_tree() at transaction commit.
+
+    Treenode's default post_save hooks rebuild the entire tree on every save,
+    which is very slow for large tag sets so collapse to one update per
+    transaction.
+    """
+    global _tag_tree_update_scheduled
+    if _tag_tree_update_scheduled:
+        return
+    _tag_tree_update_scheduled = True
+
+    def _run():
+        global _tag_tree_update_scheduled
+        try:
+            Tag.update_tree()
+        finally:
+            _tag_tree_update_scheduled = False
+
+    transaction.on_commit(_run)

src/documents/tests/test_api_workflows.py

@@ -186,11 +186,8 @@ class TestApiWorkflows(DirectoriesMixin, APITestCase):
                             "filter_has_tags": [self.t1.id],
                             "filter_has_all_tags": [self.t2.id],
                             "filter_has_not_tags": [self.t3.id],
-                            "filter_has_any_correspondents": [self.c.id],
                             "filter_has_not_correspondents": [self.c2.id],
-                            "filter_has_any_document_types": [self.dt.id],
                             "filter_has_not_document_types": [self.dt2.id],
-                            "filter_has_any_storage_paths": [self.sp.id],
                             "filter_has_not_storage_paths": [self.sp2.id],
                             "filter_custom_field_query": json.dumps(
                                 [
@@ -251,26 +248,14 @@ class TestApiWorkflows(DirectoriesMixin, APITestCase):
             set(trigger.filter_has_not_tags.values_list("id", flat=True)),
             {self.t3.id},
         )
-        self.assertSetEqual(
-            set(trigger.filter_has_any_correspondents.values_list("id", flat=True)),
-            {self.c.id},
-        )
         self.assertSetEqual(
             set(trigger.filter_has_not_correspondents.values_list("id", flat=True)),
             {self.c2.id},
         )
-        self.assertSetEqual(
-            set(trigger.filter_has_any_document_types.values_list("id", flat=True)),
-            {self.dt.id},
-        )
         self.assertSetEqual(
             set(trigger.filter_has_not_document_types.values_list("id", flat=True)),
             {self.dt2.id},
         )
-        self.assertSetEqual(
-            set(trigger.filter_has_any_storage_paths.values_list("id", flat=True)),
-            {self.sp.id},
-        )
         self.assertSetEqual(
             set(trigger.filter_has_not_storage_paths.values_list("id", flat=True)),
             {self.sp2.id},
@@ -434,11 +419,8 @@ class TestApiWorkflows(DirectoriesMixin, APITestCase):
                             "filter_has_tags": [self.t1.id],
                             "filter_has_all_tags": [self.t2.id],
                             "filter_has_not_tags": [self.t3.id],
-                            "filter_has_any_correspondents": [self.c.id],
                             "filter_has_not_correspondents": [self.c2.id],
-                            "filter_has_any_document_types": [self.dt.id],
                             "filter_has_not_document_types": [self.dt2.id],
-                            "filter_has_any_storage_paths": [self.sp.id],
                             "filter_has_not_storage_paths": [self.sp2.id],
                             "filter_custom_field_query": json.dumps(
                                 ["AND", [[self.cf1.id, "exact", "value"]]],
@@ -468,26 +450,14 @@ class TestApiWorkflows(DirectoriesMixin, APITestCase):
             workflow.triggers.first().filter_has_not_tags.first(),
             self.t3,
         )
-        self.assertEqual(
-            workflow.triggers.first().filter_has_any_correspondents.first(),
-            self.c,
-        )
         self.assertEqual(
             workflow.triggers.first().filter_has_not_correspondents.first(),
             self.c2,
         )
-        self.assertEqual(
-            workflow.triggers.first().filter_has_any_document_types.first(),
-            self.dt,
-        )
         self.assertEqual(
             workflow.triggers.first().filter_has_not_document_types.first(),
             self.dt2,
         )
-        self.assertEqual(
-            workflow.triggers.first().filter_has_any_storage_paths.first(),
-            self.sp,
-        )
         self.assertEqual(
             workflow.triggers.first().filter_has_not_storage_paths.first(),
             self.sp2,

src/documents/tests/test_tag_hierarchy.py

@@ -250,3 +250,16 @@ class TestTagHierarchy(APITestCase):
             row for row in response.data["results"] if row["id"] == self.parent.pk
         )
         assert any(child["id"] == self.child.pk for child in parent_entry["children"])
+
+    def test_tag_tree_deferred_update_runs_on_commit(self):
+        from django.db import transaction
+
+        # Create tags inside an explicit transaction and commit.
+        with transaction.atomic():
+            parent = Tag.objects.create(name="Parent 2")
+            child = Tag.objects.create(name="Child 2", tn_parent=parent)
+        # After commit, tn_* fields should be populated.
+        parent.refresh_from_db()
+        child.refresh_from_db()
+        assert parent.tn_children_count == 1
+        assert child.tn_ancestors_count == 1

src/documents/tests/test_workflows.py

@@ -1276,76 +1276,6 @@ class TestWorkflows(
             )
             self.assertIn(expected_str, cm.output[1])
 
-    def test_document_added_any_filters(self):
-        trigger = WorkflowTrigger.objects.create(
-            type=WorkflowTrigger.WorkflowTriggerType.DOCUMENT_ADDED,
-        )
-        trigger.filter_has_any_correspondents.set([self.c])
-        trigger.filter_has_any_document_types.set([self.dt])
-        trigger.filter_has_any_storage_paths.set([self.sp])
-
-        matching_doc = Document.objects.create(
-            title="sample test",
-            correspondent=self.c,
-            document_type=self.dt,
-            storage_path=self.sp,
-            original_filename="sample.pdf",
-            checksum="checksum-any-match",
-        )
-
-        matched, reason = existing_document_matches_workflow(matching_doc, trigger)
-        self.assertTrue(matched)
-        self.assertIsNone(reason)
-
-        wrong_correspondent = Document.objects.create(
-            title="wrong correspondent",
-            correspondent=self.c2,
-            document_type=self.dt,
-            storage_path=self.sp,
-            original_filename="sample2.pdf",
-        )
-        matched, reason = existing_document_matches_workflow(
-            wrong_correspondent,
-            trigger,
-        )
-        self.assertFalse(matched)
-        self.assertIn("correspondent", reason)
-
-        other_document_type = DocumentType.objects.create(name="Other")
-        wrong_document_type = Document.objects.create(
-            title="wrong doc type",
-            correspondent=self.c,
-            document_type=other_document_type,
-            storage_path=self.sp,
-            original_filename="sample3.pdf",
-            checksum="checksum-wrong-doc-type",
-        )
-        matched, reason = existing_document_matches_workflow(
-            wrong_document_type,
-            trigger,
-        )
-        self.assertFalse(matched)
-        self.assertIn("doc type", reason)
-
-        other_storage_path = StoragePath.objects.create(
-            name="Other path",
-            path="/other/",
-        )
-        wrong_storage_path = Document.objects.create(
-            title="wrong storage",
-            correspondent=self.c,
-            document_type=self.dt,
-            storage_path=other_storage_path,
-            original_filename="sample4.pdf",
-            checksum="checksum-wrong-storage-path",
-        )
-        matched, reason = existing_document_matches_workflow(
-            wrong_storage_path,
-            trigger,
-        )
-        self.assertFalse(matched)
-        self.assertIn("storage path", reason)
-
     def test_document_added_custom_field_query_no_match(self):
         trigger = WorkflowTrigger.objects.create(
             type=WorkflowTrigger.WorkflowTriggerType.DOCUMENT_ADDED,
@@ -1454,39 +1384,6 @@ class TestWorkflows(
         self.assertIn(doc1, filtered)
         self.assertNotIn(doc2, filtered)
 
-    def test_prefilter_documents_any_filters(self):
-        trigger = WorkflowTrigger.objects.create(
-            type=WorkflowTrigger.WorkflowTriggerType.DOCUMENT_ADDED,
-        )
-        trigger.filter_has_any_correspondents.set([self.c])
-        trigger.filter_has_any_document_types.set([self.dt])
-        trigger.filter_has_any_storage_paths.set([self.sp])
-
-        allowed_document = Document.objects.create(
-            title="allowed",
-            correspondent=self.c,
-            document_type=self.dt,
-            storage_path=self.sp,
-            original_filename="doc-allowed.pdf",
-            checksum="checksum-any-allowed",
-        )
-        blocked_document = Document.objects.create(
-            title="blocked",
-            correspondent=self.c2,
-            document_type=self.dt,
-            storage_path=self.sp,
-            original_filename="doc-blocked.pdf",
-            checksum="checksum-any-blocked",
-        )
-
-        filtered = prefilter_documents_by_workflowtrigger(
-            Document.objects.all(),
-            trigger,
-        )
-
-        self.assertIn(allowed_document, filtered)
-        self.assertNotIn(blocked_document, filtered)
-
     def test_consumption_trigger_requires_filter_configuration(self):
         serializer = WorkflowTriggerSerializer(
             data={

src/documents/views.py

@@ -448,8 +448,43 @@ class TagViewSet(ModelViewSet, PermissionsAwareDocumentCountMixin):
     def get_serializer_context(self):
         context = super().get_serializer_context()
         context["document_count_filter"] = self.get_document_count_filter()
+        if hasattr(self, "_children_map"):
+            context["children_map"] = self._children_map
         return context
 
+    def list(self, request, *args, **kwargs):
+        """
+        Build a children map once to avoid per-parent queries in the serializer.
+        """
+        queryset = self.filter_queryset(self.get_queryset())
+        ordering = OrderingFilter().get_ordering(request, queryset, self) or (
+            Lower("name"),
+        )
+        queryset = queryset.order_by(*ordering)
+
+        all_tags = list(queryset)
+        descendant_pks = {pk for tag in all_tags for pk in tag.get_descendants_pks()}
+
+        if descendant_pks:
+            filter_q = self.get_document_count_filter()
+            children_source = (
+                Tag.objects.filter(pk__in=descendant_pks | {t.pk for t in all_tags})
+                .select_related("owner")
+                .annotate(document_count=Count("documents", filter=filter_q))
+                .order_by(*ordering)
+            )
+        else:
+            children_source = all_tags
+
+        children_map = {}
+        for tag in children_source:
+            children_map.setdefault(tag.tn_parent_id, []).append(tag)
+        self._children_map = children_map
+
+        page = self.paginate_queryset(queryset)
+        serializer = self.get_serializer(page, many=True)
+        return self.get_paginated_response(serializer.data)
+
     def perform_update(self, serializer):
         old_parent = self.get_object().get_parent()
         tag = serializer.save()