Update advanced-search-query.ts

The inverse of the serializer, deliberately narrow: it reads the forms
the serializer writes and nothing else, so a query it cannot show stays
text instead of being converted into something subtly different.

The guarantee is enforced rather than argued: a tree is only returned
when writing it out again reproduces the query character for character.
That turns every near miss into a plain "can't show this one" - a
boost, an implicit AND, a field alias, or AND and OR mixed at one level.
Aliases are refused for the same reason, since reading type: would mean
rewriting the query as it was read.

Per-word field prefixes are put back together into the single condition
they came from, but only when adjacent and on the same field, so the
order of what the user typed is never rearranged.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

.dockerignore

@@ -15,8 +15,6 @@
 # Test related
 **/.pytest_cache
 **/tests
-src/paperless_testing
-src/conftest.py
 **/*.spec.ts
 **/htmlcov
 # Local folders

.github/actions/apt-install/action.yml

@@ -1,53 +0,0 @@
-name: Install apt packages
-description: >
-  Installs apt packages with a per-attempt timeout and retries. The hosted runner's Azure apt mirror intermittently serves archives at a crawl, which apt's own timeout does not catch. A stuck attempt is killed and retried, which usually lands on a healthy mirror node.
-
-inputs:
-  packages:
-    description: Space separated list of packages to install
-    required: true
-  attempts:
-    description: Maximum number of attempts
-    default: "3"
-  attempt-timeout:
-    description: Seconds before a single attempt is killed
-    default: "180"
-runs:
-  using: composite
-  steps:
-    - name: Install ${{ inputs.packages }}
-      shell: bash
-      env:
-        PACKAGES: ${{ inputs.packages }}
-        ATTEMPTS: ${{ inputs.attempts }}
-        ATTEMPT_TIMEOUT: ${{ inputs.attempt-timeout }}
-      run: |
-        SUDO=""
-        if [ "$(id -u)" -ne 0 ]; then
-          SUDO="sudo"
-        fi
-
-        attempt_install() {
-          $SUDO apt-get update -q
-          # Intentionally unquoted, this is a list of package names
-          $SUDO apt-get install -q --yes --no-install-recommends $PACKAGES
-        }
-        export -f attempt_install
-        export SUDO PACKAGES
-
-        for attempt in $(seq 1 "${ATTEMPTS}"); do
-          echo "::group::apt install attempt ${attempt}/${ATTEMPTS}"
-          start=$(date +%s)
-          if timeout --kill-after=10 "${ATTEMPT_TIMEOUT}" bash -c attempt_install; then
-            echo "::endgroup::"
-            echo "Installed in $(( $(date +%s) - start ))s"
-            exit 0
-          fi
-          echo "::endgroup::"
-          echo "::warning::apt install attempt ${attempt}/${ATTEMPTS} failed or exceeded ${ATTEMPT_TIMEOUT}s"
-          # A killed attempt can leave dpkg half configured
-          $SUDO dpkg --configure -a || true
-        done
-
-        echo "::error::apt install failed after ${ATTEMPTS} attempts"
-        exit 1

.github/dependabot.yml

@@ -59,6 +59,7 @@ updates:
           - "drf-*"
           - "djangorestframework"
           - "whitenoise"
+          - "bleach"
           - "jinja2"
       # Async, Task Queuing & Caching
       async-tasks:

.github/workflows/ci-backend.yml

@@ -102,16 +102,17 @@ jobs:
         with:
           python-version: "${{ matrix.python-version }}"
       - name: Install uv
-        uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
+        uses: astral-sh/setup-uv@bec219d24cd3e171d82865faccec33120bb574f4 # v10.1.0
         with:
           version: ${{ env.DEFAULT_UV_VERSION }}
           enable-cache: true
           python-version: ${{ steps.setup-python.outputs.python-version }}
       - name: Install system dependencies
-        timeout-minutes: 12
+        timeout-minutes: 10
-        uses: $/.github/actions/apt-install
+        run: |
-        with:
+          sudo apt-get update -qq
-          packages: unpaper tesseract-ocr imagemagick ghostscript poppler-utils
+          sudo apt-get install -qq --no-install-recommends \
+            unpaper tesseract-ocr imagemagick ghostscript poppler-utils
       - name: Configure ImageMagick
         run: |
           sudo cp docker/rootfs/etc/ImageMagick-6/paperless-policy.xml /etc/ImageMagick-6/policy.xml
@@ -138,13 +139,13 @@ jobs:
             pytest
       - name: Upload test results to Codecov
         if: always()
-        uses: codecov/codecov-action@303a32d7a59b442fa8d48b6a1cc6825c09c847a5 # v7.1.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f # v7.0.0
         with:
           flags: backend-python-${{ matrix.python-version }}
           files: junit.xml
           report_type: test_results
       - name: Upload coverage to Codecov
-        uses: codecov/codecov-action@303a32d7a59b442fa8d48b6a1cc6825c09c847a5 # v7.1.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f # v7.0.0
         with:
           flags: backend-python-${{ matrix.python-version }}
           files: coverage.xml
@@ -175,7 +176,7 @@ jobs:
         with:
           python-version: "${{ env.DEFAULT_PYTHON }}"
       - name: Install uv
-        uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
+        uses: astral-sh/setup-uv@bec219d24cd3e171d82865faccec33120bb574f4 # v10.1.0
         with:
           version: ${{ env.DEFAULT_UV_VERSION }}
           enable-cache: true

.github/workflows/ci-docker.yml

@@ -106,7 +106,7 @@ jobs:
           echo "repository=${repo_name}"
           echo "name=${repo_name}" >> $GITHUB_OUTPUT
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@f87e5991a6d7451dcb8d9637bfbc97413f497069 # v4.4.1
+        uses: docker/setup-buildx-action@37fe631027851001ddb9b187196cc803df7f5f0e # v4.3.0
       - name: Login to GitHub Container Registry
         uses: docker/login-action@dbcb813823bdd20940b903addbd779551569679f # v4.6.0
         with:
@@ -132,7 +132,7 @@ jobs:
             type=semver,pattern={{major}}.{{minor}}
       - name: Build and push by digest
         id: build
-        uses: docker/build-push-action@c3c9e263c25d99ce0380d002d59b67737d91b0dc # v7.4.0
+        uses: docker/build-push-action@53b7df96c91f9c12dcc8a07bcb9ccacbed38856a # v7.3.0
         with:
           context: .
           file: ./Dockerfile
@@ -182,7 +182,7 @@ jobs:
           echo "Downloaded digests:"
           ls -la /tmp/digests/
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@f87e5991a6d7451dcb8d9637bfbc97413f497069 # v4.4.1
+        uses: docker/setup-buildx-action@37fe631027851001ddb9b187196cc803df7f5f0e # v4.3.0
       - name: Login to GitHub Container Registry
         uses: docker/login-action@dbcb813823bdd20940b903addbd779551569679f # v4.6.0
         with:

.github/workflows/ci-docs.yml

@@ -78,7 +78,7 @@ jobs:
         with:
           python-version: ${{ env.DEFAULT_PYTHON_VERSION }}
       - name: Install uv
-        uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
+        uses: astral-sh/setup-uv@bec219d24cd3e171d82865faccec33120bb574f4 # v10.1.0
         with:
           version: ${{ env.DEFAULT_UV_VERSION }}
           enable-cache: true

.github/workflows/ci-frontend.yml

@@ -174,13 +174,13 @@ jobs:
         run: cd src-ui && pnpm run test --max-workers=2 --shard=${{ matrix.shard-index }}/${{ matrix.shard-count }}
       - name: Upload test results to Codecov
         if: always()
-        uses: codecov/codecov-action@303a32d7a59b442fa8d48b6a1cc6825c09c847a5 # v7.1.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f # v7.0.0
         with:
           flags: frontend-node-${{ matrix.node-version }}
           directory: src-ui/
           report_type: test_results
       - name: Upload coverage to Codecov
-        uses: codecov/codecov-action@303a32d7a59b442fa8d48b6a1cc6825c09c847a5 # v7.1.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f # v7.0.0
         with:
           flags: frontend-node-${{ matrix.node-version }}
           directory: src-ui/coverage/
@@ -216,7 +216,7 @@ jobs:
         with:
           python-version: '3.12'
       - name: Install uv
-        uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
+        uses: astral-sh/setup-uv@bec219d24cd3e171d82865faccec33120bb574f4 # v10.1.0
         with:
           version: '0.12.x'
           enable-cache: false
@@ -231,10 +231,9 @@ jobs:
       - name: Install dependencies
         run: cd src-ui && pnpm install --frozen-lockfile
       - name: Install backend system dependencies
-        timeout-minutes: 12
+        run: |
-        uses: $/.github/actions/apt-install
+          apt-get update
-        with:
+          apt-get install --yes --quiet --no-install-recommends libmagic1
-          packages: libmagic1
       - name: Install backend dependencies
         env:
           PYTHON_VERSION: ${{ steps.setup-python.outputs.python-version }}

.github/workflows/ci-release.yml

@@ -59,7 +59,7 @@ jobs:
         with:
           python-version: ${{ env.DEFAULT_PYTHON_VERSION }}
       - name: Install uv
-        uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
+        uses: astral-sh/setup-uv@bec219d24cd3e171d82865faccec33120bb574f4 # v10.1.0
         with:
           version: ${{ env.DEFAULT_UV_VERSION }}
           enable-cache: false
@@ -70,10 +70,9 @@ jobs:
         run: |
           uv sync --python "${PYTHON_VERSION}" --dev --frozen
       - name: Install system dependencies
-        timeout-minutes: 12
+        run: |
-        uses: $/.github/actions/apt-install
+          sudo apt-get update -qq
-        with:
+          sudo apt-get install -qq --no-install-recommends gettext liblept5
-          packages: gettext liblept5
       # ---- Build Documentation ----
       - name: Build documentation
         env:
@@ -213,7 +212,7 @@ jobs:
         with:
           python-version: ${{ env.DEFAULT_PYTHON_VERSION }}
       - name: Install uv
-        uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
+        uses: astral-sh/setup-uv@bec219d24cd3e171d82865faccec33120bb574f4 # v10.1.0
         with:
           version: ${{ env.DEFAULT_UV_VERSION }}
           enable-cache: false

.github/workflows/ci-static-analysis.yml

@@ -44,7 +44,7 @@ jobs:
       - name: Run Semgrep
         run: semgrep scan --config auto --sarif-output results.sarif
       - name: Upload results to GitHub code scanning
-        uses: github/codeql-action/upload-sarif@2892aa5e19bbd11bc0cff5427e3b750a04d9e3c2 # v4.38.2
+        uses: github/codeql-action/upload-sarif@b96794f015dfd88f77b49b1c93e0fa7110f94c63 # v4.38.0
         if: always()
         with:
           sarif_file: results.sarif

.github/workflows/cleanup-tags.yml

@@ -29,7 +29,7 @@ jobs:
     steps:
       - name: Clean temporary images
         if: "${{ env.TOKEN != '' }}"
-        uses: stumpylog/image-cleaner-action/ephemeral@21f875bab2376314e0525c614e923c669434f0e4 # v0.13.0
+        uses: stumpylog/image-cleaner-action/ephemeral@4fe057d991d63b8f6d5d22c40f17c1bca2226537 # v0.12.0
         with:
           token: "${{ env.TOKEN }}"
           owner: "${{ github.repository_owner }}"
@@ -56,7 +56,7 @@ jobs:
     steps:
       - name: Clean untagged images
         if: "${{ env.TOKEN != '' }}"
-        uses: stumpylog/image-cleaner-action/untagged@21f875bab2376314e0525c614e923c669434f0e4 # v0.13.0
+        uses: stumpylog/image-cleaner-action/untagged@4fe057d991d63b8f6d5d22c40f17c1bca2226537 # v0.12.0
         with:
           token: "${{ env.TOKEN }}"
           owner: "${{ github.repository_owner }}"

.github/workflows/codeql-analysis.yml

@@ -39,7 +39,7 @@ jobs:
           persist-credentials: false
       # Initializes the CodeQL tools for scanning.
       - name: Initialize CodeQL
-        uses: github/codeql-action/init@2892aa5e19bbd11bc0cff5427e3b750a04d9e3c2 # v4.38.2
+        uses: github/codeql-action/init@b96794f015dfd88f77b49b1c93e0fa7110f94c63 # v4.38.0
         with:
           languages: ${{ matrix.language }}
           # If you wish to specify custom queries, you can do so here or in a config file.
@@ -47,4 +47,4 @@ jobs:
           # Prefix the list here with "+" to use these queries and those in the config file.
           # queries: ./path/to/local/query, your-org/your-repo/queries@main
       - name: Perform CodeQL Analysis
-        uses: github/codeql-action/analyze@2892aa5e19bbd11bc0cff5427e3b750a04d9e3c2 # v4.38.2
+        uses: github/codeql-action/analyze@b96794f015dfd88f77b49b1c93e0fa7110f94c63 # v4.38.0

.github/workflows/crowdin.yml

@@ -22,7 +22,7 @@ jobs:
           token: ${{ secrets.PNGX_BOT_PAT }}
           persist-credentials: false
       - name: crowdin action
-        uses: crowdin/github-action@9c23991700c0ec5256fd41089b9d9d7d540e424e # v3.3.0
+        uses: crowdin/github-action@0d5670f539973aea2f01abce61a8989934df0025 # v3.0.2
         with:
           upload_translations: false
           download_translations: true

.github/workflows/translate-strings.yml

@@ -25,12 +25,11 @@ jobs:
         id: setup-python
         uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
       - name: Install system dependencies
-        timeout-minutes: 12
+        run: |
-        uses: $/.github/actions/apt-install
+          sudo apt-get update -qq
-        with:
+          sudo apt-get install -qq --no-install-recommends gettext
-          packages: gettext
       - name: Install uv
-        uses: astral-sh/setup-uv@c18668ad3cf93ea998bef934396af7bb5c839dc7 # v10.2.0
+        uses: astral-sh/setup-uv@bec219d24cd3e171d82865faccec33120bb574f4 # v10.1.0
         with:
           version: ${{ env.DEFAULT_UV_VERSION }}
           enable-cache: true

.pre-commit-config.yaml

@@ -38,7 +38,7 @@ repos:
           - json
   # See https://github.com/prettier/prettier/issues/15742 for the fork reason
   - repo: https://github.com/rbubley/mirrors-prettier
-    rev: 'v3.9.9'
+    rev: 'v3.9.6'
     hooks:
       - id: prettier
         types_or:
@@ -46,16 +46,16 @@ repos:
           - ts
           - markdown
         additional_dependencies:
-          - prettier@3.9.9
+          - prettier@3.9.6
           - 'prettier-plugin-organize-imports@4.3.0'
   # Python hooks
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.16.9
+    rev: v0.16.5
     hooks:
       - id: ruff-check
       - id: ruff-format
   - repo: https://github.com/tox-dev/pyproject-fmt
-    rev: "v2.29.4"
+    rev: "v2.28.1"
     hooks:
       - id: pyproject-fmt
         additional_dependencies: [tomli]

Dockerfile

@@ -30,7 +30,7 @@ RUN set -eux \
 # Purpose: Installs s6-overlay and rootfs
 # Comments:
 #  - Don't leave anything extra in here either
-FROM ghcr.io/astral-sh/uv:0.12.23-python3.14-trixie-slim AS s6-overlay-base
+FROM ghcr.io/astral-sh/uv:0.12.16-python3.14-trixie-slim AS s6-overlay-base
 
 WORKDIR /usr/src/s6
 
@@ -171,9 +171,7 @@ RUN set -eux \
         && cp /etc/ImageMagick-6/paperless-policy.xml /etc/ImageMagick-6/policy.xml \
       && echo "Cleaning up image layer" \
         && rm --force --verbose *.deb \
-    && rm --recursive --force --verbose /var/lib/apt/lists/* \
+    && rm --recursive --force --verbose /var/lib/apt/lists/*
-    && echo "Configuring interactive shells to source the s6 container environment" \
-      && echo '. /etc/profile.d/contenv.sh' >> /etc/bash.bashrc
 
 WORKDIR /usr/src/paperless/src/
 

docker/compose/docker-compose.ci-test.yml

@@ -24,7 +24,7 @@ services:
     network_mode: host
     restart: unless-stopped
   greenmail:
-    image: docker.io/greenmail/standalone:2.1.14
+    image: docker.io/greenmail/standalone:2.1.13
     hostname: greenmail
     container_name: greenmail
     environment:

docker/rootfs/etc/profile.d/contenv.sh

@@ -1,18 +0,0 @@
-#!/bin/sh
-# Source s6 container environment for interactive shells.
-# Ensures variables resolved from *_FILE secret injection are visible
-# when using 'docker exec bash'. Does not affect s6 services (those
-# use with-contenv directly). Has no effect in non-container contexts
-# because the directory will not exist.
-# Note: sh/dash shells opened via 'docker exec sh' are not covered;
-# only bash-based sessions benefit from this file.
-_pngx_contenv="/run/s6/container_environment"
-if [ -d "${_pngx_contenv}" ]; then
-	for _pngx_f in "${_pngx_contenv}"/*; do
-		[ -f "${_pngx_f}" ] || continue
-		_pngx_name=$(basename "${_pngx_f}")
-		_pngx_val=$(cat "${_pngx_f}")
-		export "${_pngx_name}=${_pngx_val}"
-	done
-fi
-unset _pngx_contenv _pngx_f _pngx_name _pngx_val

docker/rootfs/etc/s6-overlay/s6-rc.d/svc-flower/run

@@ -2,7 +2,6 @@
 # shellcheck shell=bash
 
 declare -r log_prefix="[svc-flower]"
-declare -r flower_config="${PAPERLESS_SRC_DIR}/paperless/flowerconfig.py"
 
 echo "${log_prefix} Checking if we should start flower..."
 
@@ -10,20 +9,12 @@ if [[ -n "${PAPERLESS_ENABLE_FLOWER}" ]]; then
 	# Small delay to allow celery to be up first
 	echo "${log_prefix} Starting flower in 5s"
 	sleep 5
-	cd "${PAPERLESS_SRC_DIR}" || exit 1
+	cd ${PAPERLESS_SRC_DIR}
-
-	# Only pass --conf if the file is actually there. The image does not ship one, and
-	# flower >= 2.1.0 exits with FileNotFoundError when an explicitly given --conf path
-	# does not exist (mher/flower#1391). Earlier versions silently ignored it.
-	declare -a conf_args=()
-	if [[ -f "${flower_config}" ]]; then
-		conf_args=(--conf="${flower_config}")
-	fi
 
 	if [[ -n "${USER_IS_NON_ROOT}" ]]; then
-		exec /usr/local/bin/celery --app paperless flower "${conf_args[@]}"
+		exec /usr/local/bin/celery --app paperless flower --conf=${PAPERLESS_SRC_DIR}/paperless/flowerconfig.py
 	else
-		exec s6-setuidgid paperless /usr/local/bin/celery --app paperless flower "${conf_args[@]}"
+		exec s6-setuidgid paperless /usr/local/bin/celery --app paperless flower --conf=${PAPERLESS_SRC_DIR}/paperless/flowerconfig.py
 	fi
 
 else

docs/advanced_usage.md

@@ -136,15 +136,13 @@ for suggested generation and embedding models.
 ### AI-assisted suggestions
 
 With AI enabled, Paperless-ngx can suggest a title, tags, correspondent, document type,
-storage path and dates by sending the document to the LLM using "Suggest" button on the document
+storage path and dates by sending the document to the LLM. This is **opt-in per request**
-detail page. You can choose which type of suggestions are requested by default under Settings >
+and surfaces through the "Suggest" control on the document detail page, alongside the
-Documents, either ML (classifier-based) suggestions, AI suggestions, or both. When both are requested
+classic classifier-based suggestions — it does not disable them. Suggestions are requested
-the results are combined.
+automatically when you open a document that carries an inbox tag unless "Automatically request
-
+suggestions for inbox documents" under Settings > Documents is disabled. Suggestion output
-Suggestions are requested automatically when you open a document that carries an inbox tag
+language can be steered with
-unless "Automatically request suggestions for inbox documents" under Settings > Documents is disabled.
+[`PAPERLESS_AI_LLM_OUTPUT_LANGUAGE`](configuration.md#PAPERLESS_AI_LLM_OUTPUT_LANGUAGE)
-
-Suggestion output language can be steered with [`PAPERLESS_AI_LLM_OUTPUT_LANGUAGE`](configuration.md#PAPERLESS_AI_LLM_OUTPUT_LANGUAGE)
 (otherwise it follows the user's UI language).
 
 ### The LLM index (RAG) and similar documents
@@ -155,11 +153,8 @@ in similar existing documents, and the document chat can retrieve relevant conte
 
 Enable it by setting
 [`PAPERLESS_AI_LLM_EMBEDDING_BACKEND`](configuration.md#PAPERLESS_AI_LLM_EMBEDDING_BACKEND)
-(`huggingface` for fully-local embeddings, or `ollama` / `openai-like`). By default, the main
+(`huggingface` for fully-local embeddings, or `ollama` / `openai-like`). The index is only
-LLM API key and endpoint are used, but an optional embedding-specific[API key](configuration.md#PAPERLESS_AI_LLM_EMBEDDING_API_KEY)
+built when AI is enabled **and** an embedding backend is set.
-and [endpoint](configuration.md#PAPERLESS_AI_LLM_EMBEDDING_ENDPOINT) can be configured.
-
-The index is only built when AI is enabled **and** an embedding backend is set.
 
 The index is updated automatically on a schedule controlled by
 [`PAPERLESS_LLM_INDEX_TASK_CRON`](configuration.md#PAPERLESS_LLM_INDEX_TASK_CRON) (daily by

docs/changelog.md

@@ -1,30 +1,5 @@
 # Changelog
 
-## paperless-ngx 3.2.1
-
-### Bug Fixes
-
-- Fix: only pass --conf to flower when flowerconfig.py exists [@bitfoo1](https://github.com/bitfoo1) ([#14182](https://github.com/paperless-ngx/paperless-ngx/pull/14182))
-- Fix: replace stale mail-fetch overlap check with a self-expiring lock [@stumpylog](https://github.com/stumpylog) ([#14189](https://github.com/paperless-ngx/paperless-ngx/pull/14189))
-- Fix: bump ocrmypdf to 17.12 to pick up the ligature text-layer fix [@stumpylog](https://github.com/stumpylog) ([#14190](https://github.com/paperless-ngx/paperless-ngx/pull/14190))
-- Fix: rebuild the search index automatically when it is missing Tantivy files [@stumpylog](https://github.com/stumpylog) ([#14180](https://github.com/paperless-ngx/paperless-ngx/pull/14180))
-
-### Dependencies
-
-- Chore(deps): Bump anyio from 4.12.1 to 4.14.2 in the uv group across 1 directory @[dependabot[bot]](https://github.com/apps/dependabot) ([#14175](https://github.com/paperless-ngx/paperless-ngx/pull/14175))
-
-### All App Changes
-
-<details>
-<summary>4 changes</summary>
-
-- Chore(deps): Bump anyio from 4.12.1 to 4.14.2 in the uv group across 1 directory @[dependabot[bot]](https://github.com/apps/dependabot) ([#14175](https://github.com/paperless-ngx/paperless-ngx/pull/14175))
-- Fix: replace stale mail-fetch overlap check with a self-expiring lock [@stumpylog](https://github.com/stumpylog) ([#14189](https://github.com/paperless-ngx/paperless-ngx/pull/14189))
-- Fix: bump ocrmypdf to 17.12 to pick up the ligature text-layer fix [@stumpylog](https://github.com/stumpylog) ([#14190](https://github.com/paperless-ngx/paperless-ngx/pull/14190))
-- Fix: rebuild the search index automatically when it is missing Tantivy files [@stumpylog](https://github.com/stumpylog) ([#14180](https://github.com/paperless-ngx/paperless-ngx/pull/14180))
-
-</details>
-
 ## paperless-ngx 3.2.0
 
 ### Features / Enhancements

docs/configuration.md

@@ -2133,13 +2133,6 @@ for language and resource considerations.
 
     Defaults to None.
 
-#### [`PAPERLESS_AI_LLM_EMBEDDING_API_KEY=<str>`](#PAPERLESS_AI_LLM_EMBEDDING_API_KEY) {#PAPERLESS_AI_LLM_EMBEDDING_API_KEY}
-
-: The API key to use for the embedding backend. If not supplied, embeddings use
-`PAPERLESS_AI_LLM_API_KEY`.
-
-    Defaults to None.
-
 #### [`PAPERLESS_AI_LLM_EMBEDDING_ENDPOINT=<str>`](#PAPERLESS_AI_LLM_EMBEDDING_ENDPOINT) {#PAPERLESS_AI_LLM_EMBEDDING_ENDPOINT}
 
 : The endpoint / url to use for the embedding backend. If not supplied, embeddings use
@@ -2224,19 +2217,6 @@ used with the OpenAI-compatible backend to target a custom provider or local gat
 
     Defaults to true, which allows internal endpoints.
 
-#### [`PAPERLESS_AI_LLM_EXTRA_PARAMS=<json>`](#PAPERLESS_AI_LLM_EXTRA_PARAMS) {#PAPERLESS_AI_LLM_EXTRA_PARAMS}
-
-: A JSON object of extra parameters sent with every LLM request, for providers that require a parameter Paperless does not
-set itself. Values here override Paperless' own, and no validation is performed. Whatever you put here is passed to the
-backend as-is, so an invalid parameter will simply be rejected by your provider. For example, current OpenAI reasoning
-models refuse tool calls on the chat completions API unless reasoning is off:
-
-    ```
-    PAPERLESS_AI_LLM_EXTRA_PARAMS={"reasoning_effort": "none"}
-    ```
-
-    Defaults to empty, which adds nothing to requests.
-
 #### [`PAPERLESS_LLM_INDEX_TASK_CRON=<cron expression>`](#PAPERLESS_LLM_INDEX_TASK_CRON) {#PAPERLESS_LLM_INDEX_TASK_CRON}
 
 : Configures the schedule to update the AI embeddings of text content and metadata for all documents. Only performed if

docs/development.md

@@ -150,7 +150,6 @@ pnpm ng build --configuration production
   is loaded as well. However, the tests rely on the default
   configuration. This is not ideal. But for now, make sure no settings
   except for DEBUG are overridden when testing.
-- Tests run in a random order each session, so that one test cannot quietly depend on another having run first. The seed is printed at the top of the run; pass `--randomly-seed=<seed>` to replay that exact order, or `--randomly-seed=last` to repeat the previous run.
 
 !!! note
 

pyproject.toml

@@ -1,9 +1,7 @@
 [project]
 name = "paperless-ngx"
-version = "3.2.1"
+version = "3.2.0"
-description = """\
+description = "A community-supported supercharged document management system: scan, index and archive all your physical documents"
-  A community-supported supercharged document management system: scan, index and archive all your physical documents\
-  """
 readme = "README.md"
 requires-python = ">=3.11"
 classifiers = [
@@ -12,13 +10,13 @@ classifiers = [
   "Programming Language :: Python :: 3.12",
   "Programming Language :: Python :: 3.13",
   "Programming Language :: Python :: 3.14",
-  "Programming Language :: Python :: 3.15",
 ]
 # TODO: Move certain things to groups and then utilize that further
 # This will allow testing to not install a webserver, mysql, etc
 dependencies = [
   "azure-ai-documentintelligence>=1.0.2",
   "babel>=2.17",
+  "bleach~=6.4.0",
   "celery[redis]~=5.6.2",
   "channels~=4.2",
   "channels-redis~=4.2",
@@ -34,7 +32,7 @@ dependencies = [
   "django-cors-headers~=4.9.0",
   "django-extensions~=4.1",
   "django-filter~=25.1",
-  "django-guardian>=3.3.3,<3.6",
+  "django-guardian>=3.3.3,<3.5",
   "django-multiselectfield~=1.0.1",
   "django-rich~=2.2.0",
   "django-soft-delete~=1.0.18",
@@ -44,7 +42,7 @@ dependencies = [
   "drf-spectacular-sidecar>=2026.7.1,<2026.10",
   "drf-writable-nested~=0.7.1",
   "filelock~=3.32.0",
-  "flower>=2.0.1,<2.3",
+  "flower>=2.0.1,<2.2",
   "gotenberg-client[httpx]~=1.0",
   "httpx-oauth~=0.17",
   "ijson>=3.5.1",
@@ -76,7 +74,6 @@ dependencies = [
   "tantivy~=0.26.0",
   "tika-client[httpx]~=1.0",
   "torch>=2.13,<2.15",
-  "turbohtml~=1.10.0",
   "watchfiles>=1.2",
   "whitenoise~=6.11",
   "whoosh-compat[tantivy]==0.3",
@@ -84,13 +81,13 @@ dependencies = [
 ]
 [project.optional-dependencies]
 mariadb = [
-  "mysqlclient>=2.2.7,<2.4",
+  "mysqlclient~=2.2.7",
 ]
 postgres = [
   "psycopg[c,pool]==3.3.4",
   # Direct dependency for proper resolution of the pre-built wheels
   "psycopg-c==3.3.4",
-  "psycopg-pool==3.3.2",
+  "psycopg-pool==3.3.1",
 ]
 webserver = [
   "granian[uvloop]>=2.7,<2.9",
@@ -112,7 +109,7 @@ lint = [
 testing = [
   "daphne",
   "factory-boy~=3.3.1",
-  "faker>=40.36,<40.40",
+  "faker>=40.36,<40.39",
   "imagehash",
   "pytest~=9.1.1",
   "pytest-cov~=7.1.0",
@@ -120,7 +117,7 @@ testing = [
   "pytest-env~=1.7.0",
   "pytest-httpx",
   "pytest-mock~=3.15.1",
-  "pytest-randomly~=5.0.0",
+  # "pytest-randomly~=4.0.1",
   "pytest-rerunfailures~=16.4",
   "pytest-sugar",
   "pytest-xdist~=3.8.0",
@@ -136,6 +133,7 @@ typing = [
   "mypy",
   "mypy-baseline",
   "pyrefly",
+  "types-bleach",
   "types-channels",
   "types-colorama",
   "types-dateparser",
@@ -150,7 +148,6 @@ typing = [
 
 [tool.uv]
 required-version = ">=0.9.0"
-prerelease = "disallow"
 environments = [
   "sys_platform == 'darwin'",
   "sys_platform == 'linux'",
@@ -247,10 +244,6 @@ per-file-ignores."docker/wait-for-redis.py" = [
 per-file-ignores."src/documents/models.py" = [
   "SIM115",
 ]
-per-file-ignores."src/documents/tests/*.py" = [
-  "TID251",
-]
-flake8-tidy-imports.banned-api."documents.tests".msg = "Shared test infrastructure lives in src/paperless_testing/."
 isort.force-single-line = true
 
 [tool.codespell]
@@ -278,9 +271,9 @@ plugins = [
 ]
 
 [tool.pyrefly]
+baseline = ".pyrefly-baseline.json"
 python-platform = "linux"
 search-path = [ "src" ]
-baseline = ".pyrefly-baseline.json"
 
 [tool.django-stubs]
 django_settings_module = "paperless.settings"
@@ -333,8 +326,6 @@ PAPERLESS_CACHE_BACKEND = "django.core.cache.backends.locmem.LocMemCache"
 PAPERLESS_CHANNELS_BACKEND = "channels.layers.InMemoryChannelLayer"
 # I don't think anything hits this, but just in case, basically infinite
 PAPERLESS_TOKEN_THROTTLE_RATE = "1000/min"
-# The 0.1s production default trips on a stalled CI runner, the date parsing tests then find no dates
-PAPERLESS_MATCH_REGEX_TIMEOUT_SECONDS = "5"
 
 [tool.coverage.run]
 source = [
@@ -343,15 +334,13 @@ source = [
 omit = [
   "*/tests/*",
   "manage.py",
-  "paperless/auth.py",
   "paperless/wsgi.py",
-  "src/conftest.py",
+  "paperless/auth.py",
-  "src/paperless_testing/*",
 ]
 [tool.coverage.report]
 exclude_also = [
-  "if AUDIT_LOG_ENABLED:",
   "if settings.AUDIT_LOG_ENABLED:",
+  "if AUDIT_LOG_ENABLED:",
   "if TYPE_CHECKING:",
 ]
 

src-ui/angular.json

@@ -91,11 +91,6 @@
                 "glob": "**/*",
                 "input": "node_modules/pdfjs-dist/iccs/",
                 "output": "/assets/iccs/"
-              },
-              {
-                "glob": "**/*",
-                "input": "node_modules/pdfjs-dist/cmaps/",
-                "output": "/assets/cmaps/"
               }
             ],
             "styles": [

src-ui/package.json

@@ -1,6 +1,6 @@
 {
   "name": "paperless-ngx-ui",
-  "version": "3.2.1",
+  "version": "3.2.0",
   "scripts": {
     "preinstall": "npx only-allow pnpm",
     "ng": "ng",
@@ -15,16 +15,16 @@
   },
   "private": true,
   "dependencies": {
-    "@angular/cdk": "^22.2.0",
+    "@angular/cdk": "^22.1.6",
-    "@angular/common": "~22.2.0",
+    "@angular/common": "~22.1.6",
-    "@angular/compiler": "~22.2.0",
+    "@angular/compiler": "~22.1.6",
-    "@angular/core": "~22.2.0",
+    "@angular/core": "~22.1.6",
-    "@angular/forms": "~22.2.0",
+    "@angular/forms": "~22.1.6",
-    "@angular/localize": "~22.2.0",
+    "@angular/localize": "~22.1.6",
-    "@angular/platform-browser": "~22.2.0",
+    "@angular/platform-browser": "~22.1.6",
-    "@angular/router": "~22.2.0",
+    "@angular/router": "~22.1.6",
     "@ng-bootstrap/ng-bootstrap": "^21.0.0",
-    "@ng-select/ng-select": "~24.1.2",
+    "@ng-select/ng-select": "~24.1.1",
     "@ngneat/dirty-check-forms": "^3.0.3",
     "@popperjs/core": "^2.11.8",
     "bootstrap": "^5.3.8",
@@ -45,29 +45,29 @@
   },
   "devDependencies": {
     "@angular-builders/jest": "^22.0.1",
-    "@angular-devkit/core": "^22.2.0",
+    "@angular-devkit/core": "^22.1.8",
-    "@angular-devkit/schematics": "^22.2.0",
+    "@angular-devkit/schematics": "^22.1.8",
     "@angular-eslint/builder": "22.5.0",
     "@angular-eslint/eslint-plugin": "22.5.0",
     "@angular-eslint/eslint-plugin-template": "22.5.0",
     "@angular-eslint/schematics": "22.5.0",
     "@angular-eslint/template-parser": "22.5.0",
-    "@angular/build": "22.2.0",
+    "@angular/build": "22.1.8",
-    "@angular/cli": "22.2.0",
+    "@angular/cli": "22.1.8",
-    "@angular/compiler-cli": "~22.2.0",
+    "@angular/compiler-cli": "~22.1.6",
     "@playwright/test": "^1.62.1",
     "@types/jest": "^30.0.0",
-    "@types/node": "^26.6.2",
+    "@types/node": "^26.5.0",
-    "@typescript-eslint/eslint-plugin": "^8.70.1",
+    "@typescript-eslint/eslint-plugin": "^8.70.0",
-    "@typescript-eslint/parser": "^8.70.1",
+    "@typescript-eslint/parser": "^8.70.0",
-    "@typescript-eslint/utils": "^8.70.1",
+    "@typescript-eslint/utils": "^8.70.0",
-    "eslint": "^10.11.0",
+    "eslint": "^10.10.0",
-    "jest": "30.5.2",
+    "jest": "30.5.1",
-    "jest-environment-jsdom": "^30.5.2",
+    "jest-environment-jsdom": "^30.5.1",
     "jest-junit": "^17.0.0",
     "jest-preset-angular": "^17.0.0",
     "jest-websocket-mock": "^2.5.0",
-    "prettier": "^3.9.9",
+    "prettier": "^3.9.6",
     "prettier-plugin-organize-imports": "^4.3.0",
     "ts-node": "~10.9.2",
     "typescript": "^6.0.3"

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

@@ -253,24 +253,6 @@
               </div>
             </div>
 
-            @if (aiEnabled) {
-              <div class="row mb-3">
-                <div class="col-md-3 col-form-label pt-0">
-                  <span i18n>Suggestions default to</span>
-                </div>
-                <div class="col">
-                  <fieldset class="btn-group btn-group-sm">
-                    <input type="radio" class="btn-check" id="suggestionSourceBoth" [value]="SuggestionSource.Both" formControlName="documentEditingSuggestionSource">
-                    <label class="btn btn-outline-primary" for="suggestionSourceBoth"><ng-container i18n>Both</ng-container></label>
-                    <input type="radio" class="btn-check" id="suggestionSourceML" [value]="SuggestionSource.ML" formControlName="documentEditingSuggestionSource">
-                    <label class="btn btn-outline-primary" for="suggestionSourceML"><i-bs class="me-1" name="cpu"></i-bs><ng-container i18n>ML only</ng-container></label>
-                    <input type="radio" class="btn-check" id="suggestionSourceAI" [value]="SuggestionSource.AI" formControlName="documentEditingSuggestionSource">
-                    <label class="btn btn-outline-primary" for="suggestionSourceAI"><i-bs class="me-1" name="stars"></i-bs><ng-container i18n>AI only</ng-container></label>
-                  </fieldset>
-                </div>
-              </div>
-            }
-
             <div class="row">
               <div class="col">
                 <pngx-input-check i18n-title title="Automatically request suggestions for inbox documents" i18n-hint hint="If un-checked, suggestions must be requested via the Suggest button." formControlName="documentEditingAutoSuggest"></pngx-input-check>

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

@@ -307,7 +307,7 @@ describe('SettingsComponent', () => {
     expect(toastErrorSpy).toHaveBeenCalled()
     expect(storeSpy).toHaveBeenCalled()
     expect(appearanceSettingsSpy).not.toHaveBeenCalled()
-    expect(setSpy).toHaveBeenCalledTimes(35)
+    expect(setSpy).toHaveBeenCalledTimes(34)
     expect(setSpy).toHaveBeenCalledWith(SETTINGS_KEYS.SIDEBAR_HIDDEN_ITEMS, [
       HideableSidebarItemID.Workflows,
     ])

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

@@ -44,7 +44,6 @@ import {
   HIDEABLE_SIDEBAR_ITEM_IDS,
   HideableSidebarItemID,
   SETTINGS_KEYS,
-  SuggestionSource,
 } from 'src/app/data/ui-settings'
 import { User } from 'src/app/data/user'
 import { IfPermissionsDirective } from 'src/app/directives/if-permissions.directive'
@@ -185,7 +184,6 @@ export class SettingsComponent
     documentEditingRemoveInboxTags: new FormControl(null),
     documentEditingOverlayThumbnail: new FormControl(null),
     documentEditingAutoSuggest: new FormControl(null),
-    documentEditingSuggestionSource: new FormControl(null),
     documentDetailsHiddenFields: new FormControl([]),
     searchDbOnly: new FormControl(null),
     searchLink: new FormControl(null),
@@ -219,11 +217,6 @@ export class SettingsComponent
   public readonly PdfZoomScale = PdfZoomScale
 
   public readonly PdfEditorEditMode = PdfEditorEditMode
-  public readonly SuggestionSource = SuggestionSource
-
-  get aiEnabled(): boolean {
-    return this.settings.get(SETTINGS_KEYS.AI_ENABLED)
-  }
 
   public readonly documentDetailFieldOptions = documentDetailFieldOptions
   public readonly sidebarItemOptions = HIDEABLE_SIDEBAR_ITEM_IDS.map((id) => ({
@@ -411,9 +404,6 @@ export class SettingsComponent
       documentEditingAutoSuggest: this.settings.get(
         SETTINGS_KEYS.DOCUMENT_EDITING_AUTO_SUGGEST
       ),
-      documentEditingSuggestionSource: this.settings.get(
-        SETTINGS_KEYS.DOCUMENT_EDITING_SUGGESTION_SOURCE
-      ),
       documentDetailsHiddenFields: this.settings.get(
         SETTINGS_KEYS.DOCUMENT_DETAILS_HIDDEN_FIELDS
       ),
@@ -635,10 +625,6 @@ export class SettingsComponent
       SETTINGS_KEYS.DOCUMENT_EDITING_AUTO_SUGGEST,
       this.settingsForm.value.documentEditingAutoSuggest
     )
-    this.settings.set(
-      SETTINGS_KEYS.DOCUMENT_EDITING_SUGGESTION_SOURCE,
-      this.settingsForm.value.documentEditingSuggestionSource
-    )
     this.settings.set(
       SETTINGS_KEYS.DOCUMENT_DETAILS_HIDDEN_FIELDS,
       this.settingsForm.value.documentDetailsHiddenFields

src-ui/src/app/components/common/advanced-search-dialog/advanced-search-dialog.component.html

@@ -0,0 +1,347 @@
+<div class="modal-header">
+  <h4 class="modal-title" id="advanced-search-dialog-title" i18n>
+    Advanced search
+  </h4>
+  <button
+    type="button"
+    class="btn-close"
+    aria-label="Close"
+    i18n-aria-label
+    (click)="cancel()"
+  ></button>
+</div>
+<div class="modal-body">
+  @if (unreadable) {
+    <div class="alert alert-warning d-flex flex-wrap gap-2 align-items-center">
+      <div class="flex-grow-1">
+        <span i18n
+          >The current query uses syntax this editor can't show. It still works
+          as typed.</span
+        >
+      </div>
+      <button
+        type="button"
+        class="btn btn-sm btn-outline-secondary"
+        (click)="startOver()"
+        i18n
+      >
+        Start a new query
+      </button>
+    </div>
+  }
+
+  <ng-container
+    *ngTemplateOutlet="
+      groupTemplate;
+      context: { group: root, parent: null, depth: 0 }
+    "
+  ></ng-container>
+
+  <div class="mt-4">
+    <label
+      class="form-label small text-muted"
+      for="advanced-search-preview"
+      i18n
+      >Query</label
+    >
+    <pre
+      id="advanced-search-preview"
+      class="query-preview mb-0 p-2 rounded border"
+    >@if (generatedQuery) {{{ generatedQuery }}} @else {<span class="text-muted fst-italic" i18n>Nothing to search for yet</span>}</pre>
+  </div>
+</div>
+<div class="modal-footer">
+  <button
+    type="button"
+    class="btn btn-outline-secondary"
+    (click)="cancel()"
+    i18n
+  >
+    Cancel
+  </button>
+  <button
+    type="button"
+    class="btn btn-primary"
+    (click)="apply()"
+    [disabled]="!generatedQuery"
+    i18n
+  >
+    Apply
+  </button>
+</div>
+
+<ng-template
+  #groupTemplate
+  let-group="group"
+  let-parent="parent"
+  let-depth="depth"
+>
+  <div class="d-flex w-100 gap-2">
+    <div class="d-flex flex-grow-1 flex-column">
+      <div class="d-flex align-items-center flex-wrap">
+        <div
+          class="btn-group btn-group-xs"
+          role="group"
+          aria-label="Match"
+          i18n-aria-label
+        >
+          <input
+            type="radio"
+            class="btn-check"
+            [(ngModel)]="group.operator"
+            [ngModelOptions]="{ standalone: true }"
+            [value]="LogicalOperator.Or"
+            id="advancedSearchAny_{{ idFor(group) }}"
+            name="advancedSearchAny_{{ idFor(group) }}"
+          />
+          <label
+            class="btn btn-outline-primary"
+            for="advancedSearchAny_{{ idFor(group) }}"
+            i18n
+            >Any</label
+          >
+          <input
+            type="radio"
+            class="btn-check"
+            [(ngModel)]="group.operator"
+            [ngModelOptions]="{ standalone: true }"
+            [value]="LogicalOperator.And"
+            id="advancedSearchAll_{{ idFor(group) }}"
+            name="advancedSearchAll_{{ idFor(group) }}"
+          />
+          <label
+            class="btn btn-outline-primary"
+            for="advancedSearchAll_{{ idFor(group) }}"
+            i18n
+            >All</label
+          >
+          <input
+            type="radio"
+            class="btn-check"
+            [(ngModel)]="group.operator"
+            [ngModelOptions]="{ standalone: true }"
+            [value]="LogicalOperator.Not"
+            id="advancedSearchNot_{{ idFor(group) }}"
+            name="advancedSearchNot_{{ idFor(group) }}"
+          />
+          <label
+            class="btn btn-outline-secondary"
+            for="advancedSearchNot_{{ idFor(group) }}"
+            i18n
+            >Not</label
+          >
+        </div>
+        <span class="small text-muted ms-2">
+          @switch (group.operator) {
+            @case (LogicalOperator.And) {
+              <ng-container i18n>match all of these</ng-container>
+            }
+            @case (LogicalOperator.Or) {
+              <ng-container i18n>match any of these</ng-container>
+            }
+            @case (LogicalOperator.Not) {
+              <ng-container i18n>match none of these</ng-container>
+            }
+          }
+        </span>
+      </div>
+      <div class="list-group list-group-flush">
+        @for (element of group.children; track element) {
+          <div class="list-group-item px-0 d-flex flex-nowrap">
+            @if (element.type === ElementType.Group) {
+              <ng-container
+                *ngTemplateOutlet="
+                  groupTemplate;
+                  context: { group: element, parent: group, depth: depth + 1 }
+                "
+              ></ng-container>
+            } @else {
+              <ng-container
+                *ngTemplateOutlet="
+                  atomTemplate;
+                  context: { atom: element, parent: group }
+                "
+              ></ng-container>
+            }
+          </div>
+        }
+      </div>
+    </div>
+    <div
+      class="btn-group-vertical align-self-start ms-2 ps-2 border-start"
+      role="group"
+      aria-label="Group actions"
+      i18n-aria-label
+    >
+      <button
+        type="button"
+        class="btn btn-sm btn-outline-secondary text-primary"
+        title="Add condition"
+        i18n-title
+        (click)="addAtom(group)"
+        [disabled]="group.children.length >= maxAtoms"
+      >
+        <i-bs name="node-plus"></i-bs>
+      </button>
+      <button
+        type="button"
+        class="btn btn-sm btn-outline-secondary text-primary"
+        title="Add group"
+        i18n-title
+        (click)="addGroup(group)"
+        [disabled]="depth >= maxDepth"
+      >
+        <i-bs name="braces"></i-bs>
+      </button>
+      @if (parent) {
+        <button
+          type="button"
+          class="btn btn-sm btn-outline-secondary text-danger"
+          aria-label="Remove group"
+          i18n-aria-label
+          (click)="remove(parent, group)"
+        >
+          <i-bs name="x-circle"></i-bs>
+        </button>
+      }
+    </div>
+  </div>
+</ng-template>
+
+<ng-template #atomTemplate let-atom="atom" let-parent="parent">
+  <div class="d-flex align-items-center gap-1 w-100">
+    <div class="input-group input-group-sm flex-wrap">
+      <select
+        class="form-select flex-grow-0 w-auto"
+        [(ngModel)]="atom.field"
+        [ngModelOptions]="{ standalone: true }"
+        (ngModelChange)="onFieldChange(atom)"
+        aria-label="Field"
+        i18n-aria-label
+      >
+        @for (fieldGroup of fieldGroups; track fieldGroup.label) {
+          <optgroup [label]="fieldGroup.label">
+            @for (field of fieldGroup.fields; track field) {
+              <option [ngValue]="field">{{ fieldLabels[field] }}</option>
+            }
+          </optgroup>
+        }
+      </select>
+      <select
+        class="form-select flex-grow-0 w-auto"
+        [(ngModel)]="atom.operator"
+        [ngModelOptions]="{ standalone: true }"
+        (ngModelChange)="onOperatorChange(atom)"
+        aria-label="Condition"
+        i18n-aria-label
+      >
+        @for (operator of operatorsFor(atom); track operator) {
+          <option [ngValue]="operator">
+            {{ operatorLabel(atom, operator) }}
+          </option>
+        }
+      </select>
+
+      @switch (atom.operator) {
+        @case (Operator.DateKeyword) {
+          <select
+            class="form-select"
+            [(ngModel)]="atom.value"
+            [ngModelOptions]="{ standalone: true }"
+            aria-label="Period"
+            i18n-aria-label
+          >
+            <option [ngValue]="''" disabled i18n>Choose a period</option>
+            @for (keyword of dateKeywords; track keyword) {
+              <option [ngValue]="keyword">
+                {{ dateKeywordLabels[keyword] }}
+              </option>
+            }
+          </select>
+        }
+        @case (Operator.WithinLast) {
+          <input
+            class="form-control amount"
+            type="number"
+            min="1"
+            [(ngModel)]="atom.value"
+            [ngModelOptions]="{ standalone: true }"
+            aria-label="Amount"
+            i18n-aria-label
+          />
+          <select
+            class="form-select"
+            [(ngModel)]="atom.unit"
+            [ngModelOptions]="{ standalone: true }"
+            aria-label="Unit"
+            i18n-aria-label
+          >
+            @for (unit of dateUnits; track unit) {
+              <option [ngValue]="unit">{{ dateUnitLabels[unit] }}</option>
+            }
+          </select>
+        }
+        @case (Operator.Between) {
+          <input
+            class="form-control"
+            [type]="kindOf(atom) === FieldKind.Date ? 'date' : 'number'"
+            [(ngModel)]="atom.value"
+            [ngModelOptions]="{ standalone: true }"
+            aria-label="From"
+            i18n-aria-label
+          />
+          <span class="input-group-text" i18n>and</span>
+          <input
+            class="form-control"
+            [type]="kindOf(atom) === FieldKind.Date ? 'date' : 'number'"
+            [(ngModel)]="atom.valueTo"
+            [ngModelOptions]="{ standalone: true }"
+            aria-label="To"
+            i18n-aria-label
+          />
+        }
+        @default {
+          @if (kindOf(atom) === FieldKind.Date) {
+            <input
+              class="form-control"
+              type="date"
+              [(ngModel)]="atom.value"
+              [ngModelOptions]="{ standalone: true }"
+              aria-label="Date"
+              i18n-aria-label
+            />
+          } @else if (kindOf(atom) === FieldKind.Number) {
+            <input
+              class="form-control"
+              type="number"
+              min="0"
+              [(ngModel)]="atom.value"
+              [ngModelOptions]="{ standalone: true }"
+              aria-label="Number"
+              i18n-aria-label
+            />
+          } @else {
+            <input
+              class="form-control"
+              type="text"
+              [(ngModel)]="atom.value"
+              [ngModelOptions]="{ standalone: true }"
+              [placeholder]="placeholderFor(atom)"
+              aria-label="Value"
+              i18n-aria-label
+            />
+          }
+        }
+      }
+    </div>
+    <button
+      class="btn btn-link btn-sm text-danger px-1"
+      type="button"
+      (click)="remove(parent, atom)"
+      aria-label="Remove condition"
+      i18n-aria-label
+    >
+      <i-bs name="x-circle"></i-bs>
+    </button>
+  </div>
+</ng-template>

src-ui/src/app/components/common/advanced-search-dialog/advanced-search-dialog.component.scss

@@ -0,0 +1,11 @@
+.query-preview {
+  font-size: 0.8125rem;
+  white-space: pre-wrap;
+  word-break: break-word;
+  background-color: var(--pngx-bg-darker);
+  border-color: var(--bs-border-color) !important;
+}
+
+.input-group .amount {
+  flex: 0 1 5rem;
+}

src-ui/src/app/components/common/advanced-search-dialog/advanced-search-dialog.component.spec.ts

@@ -0,0 +1,186 @@
+import { ComponentFixture, TestBed } from '@angular/core/testing'
+
+import { NgbActiveModal } from '@ng-bootstrap/ng-bootstrap'
+import { allIcons, NgxBootstrapIconsModule } from 'ngx-bootstrap-icons'
+import {
+  AdvancedSearchDateUnit,
+  AdvancedSearchField,
+  AdvancedSearchLogicalOperator,
+  AdvancedSearchOperator,
+  AdvancedSearchQueryAtom,
+  AdvancedSearchQueryElementType,
+  AdvancedSearchQueryGroup,
+} from 'src/app/data/advanced-search-query'
+import { AdvancedSearchDialogComponent } from './advanced-search-dialog.component'
+
+describe('AdvancedSearchDialogComponent', () => {
+  let component: AdvancedSearchDialogComponent
+  let fixture: ComponentFixture<AdvancedSearchDialogComponent>
+  let activeModal: NgbActiveModal
+
+  const firstAtom = (): AdvancedSearchQueryAtom =>
+    component.root.children[0] as AdvancedSearchQueryAtom
+
+  beforeEach(async () => {
+    await TestBed.configureTestingModule({
+      imports: [
+        AdvancedSearchDialogComponent,
+        NgxBootstrapIconsModule.pick(allIcons),
+      ],
+      providers: [NgbActiveModal],
+    }).compileComponents()
+
+    fixture = TestBed.createComponent(AdvancedSearchDialogComponent)
+    activeModal = TestBed.inject(NgbActiveModal)
+    component = fixture.componentInstance
+    fixture.detectChanges()
+  })
+
+  it('should start with one empty condition and nothing to search for', () => {
+    expect(component.root.children).toHaveLength(1)
+    expect(component.generatedQuery).toBe('')
+    expect(component.unreadable).toBeFalsy()
+  })
+
+  it('should show an existing query as conditions', () => {
+    component.query = 'title:invoice AND NOT tag:paid'
+    expect(component.unreadable).toBeFalsy()
+    expect(component.root.children).toHaveLength(2)
+    expect(component.generatedQuery).toBe('title:invoice AND NOT tag:paid')
+  })
+
+  it('should flag a query it cannot show and start empty', () => {
+    component.query = 'title:invoice^2'
+    expect(component.unreadable).toBeTruthy()
+    expect(component.generatedQuery).toBe('')
+  })
+
+  it('should clear the warning when starting a new query', () => {
+    component.query = 'title:invoice^2'
+    component.startOver()
+    expect(component.unreadable).toBeFalsy()
+    expect(component.root.children).toHaveLength(1)
+  })
+
+  it('should treat an empty query as a fresh start', () => {
+    component.query = '   '
+    expect(component.unreadable).toBeFalsy()
+    expect(component.root.children).toHaveLength(1)
+  })
+
+  it('should write the query as conditions are filled in', () => {
+    const atom = firstAtom()
+    atom.field = AdvancedSearchField.Title
+    atom.value = 'unpaid invoice'
+    expect(component.generatedQuery).toBe('title:unpaid AND title:invoice')
+  })
+
+  it('should offer the conditions of the chosen field', () => {
+    const atom = firstAtom()
+    atom.field = AdvancedSearchField.Added
+    component.onFieldChange(atom)
+    expect(component.operatorsFor(atom)).toContain(
+      AdvancedSearchOperator.WithinLast
+    )
+    expect(component.operatorsFor(atom)).not.toContain(
+      AdvancedSearchOperator.Phrase
+    )
+  })
+
+  it('should keep a condition the new field still offers', () => {
+    const atom = firstAtom()
+    atom.operator = AdvancedSearchOperator.Phrase
+    atom.field = AdvancedSearchField.Correspondent
+    component.onFieldChange(atom)
+    expect(atom.operator).toBe(AdvancedSearchOperator.Phrase)
+  })
+
+  it('should replace a condition the new field does not offer, and clear the value', () => {
+    const atom = firstAtom()
+    atom.operator = AdvancedSearchOperator.Phrase
+    atom.value = 'invoice'
+    atom.field = AdvancedSearchField.ASN
+    component.onFieldChange(atom)
+    expect(atom.operator).toBe(AdvancedSearchOperator.Equals)
+    expect(atom.value).toBe('')
+  })
+
+  it('should give a within-the-last condition a unit to start from', () => {
+    const atom = firstAtom()
+    atom.field = AdvancedSearchField.Added
+    atom.operator = AdvancedSearchOperator.WithinLast
+    component.onOperatorChange(atom)
+    expect(atom.unit).toBe(AdvancedSearchDateUnit.Day)
+    atom.value = '3'
+    expect(component.generatedQuery).toBe('added:[-3 days to now]')
+  })
+
+  it('should label date comparisons as dates', () => {
+    const atom = firstAtom()
+    atom.field = AdvancedSearchField.Created
+    expect(
+      component.operatorLabel(atom, AdvancedSearchOperator.AtLeast)
+    ).toEqual('is on or after')
+    atom.field = AdvancedSearchField.ASN
+    expect(
+      component.operatorLabel(atom, AdvancedSearchOperator.AtLeast)
+    ).toEqual('is at least')
+  })
+
+  it('should add and remove conditions', () => {
+    component.addAtom(component.root)
+    expect(component.root.children).toHaveLength(2)
+    component.remove(component.root, component.root.children[1])
+    expect(component.root.children).toHaveLength(1)
+  })
+
+  it('should add a group, which starts as Any', () => {
+    component.addGroup(component.root)
+    const group = component.root.children[1] as AdvancedSearchQueryGroup
+    expect(group.type).toBe(AdvancedSearchQueryElementType.Group)
+    expect(group.operator).toBe(AdvancedSearchLogicalOperator.Or)
+    expect(group.children).toHaveLength(1)
+  })
+
+  it('should give every group its own id, once', () => {
+    component.addGroup(component.root)
+    const group = component.root.children[1] as AdvancedSearchQueryGroup
+    expect(component.idFor(component.root)).not.toEqual(component.idFor(group))
+    expect(component.idFor(group)).toEqual(component.idFor(group))
+  })
+
+  it('should apply the query and close', () => {
+    const emitSpy = jest.spyOn(component.queryApplied, 'emit')
+    const closeSpy = jest.spyOn(activeModal, 'close')
+    const atom = firstAtom()
+    atom.value = 'invoice'
+    component.apply()
+    expect(emitSpy).toHaveBeenCalledWith('content:invoice')
+    expect(closeSpy).toHaveBeenCalled()
+  })
+
+  it('should close without applying on cancel', () => {
+    const emitSpy = jest.spyOn(component.queryApplied, 'emit')
+    const closeSpy = jest.spyOn(activeModal, 'close')
+    component.cancel()
+    expect(emitSpy).not.toHaveBeenCalled()
+    expect(closeSpy).toHaveBeenCalled()
+  })
+
+  it('should show the query it will apply', () => {
+    component.query = 'content:invoice OR content:receipt'
+    fixture.detectChanges()
+    const preview = fixture.nativeElement.querySelector(
+      '#advanced-search-preview'
+    )
+    expect(preview.textContent).toContain('content:invoice OR content:receipt')
+  })
+
+  it('should not offer to apply an empty query', () => {
+    fixture.detectChanges()
+    const apply = Array.from(
+      fixture.nativeElement.querySelectorAll('.modal-footer button')
+    ).pop() as HTMLButtonElement
+    expect(apply.disabled).toBeTruthy()
+  })
+})

src-ui/src/app/components/common/advanced-search-dialog/advanced-search-dialog.component.ts

@@ -0,0 +1,197 @@
+import { NgTemplateOutlet } from '@angular/common'
+import { Component, EventEmitter, inject, Input, Output } from '@angular/core'
+import { FormsModule } from '@angular/forms'
+import { NgbActiveModal } from '@ng-bootstrap/ng-bootstrap'
+import { NgxBootstrapIconsModule } from 'ngx-bootstrap-icons'
+import {
+  ADVANCED_SEARCH_DATE_KEYWORD_LABELS,
+  ADVANCED_SEARCH_DATE_KEYWORDS,
+  ADVANCED_SEARCH_DATE_OPERATOR_LABELS,
+  ADVANCED_SEARCH_DATE_UNIT_LABELS,
+  ADVANCED_SEARCH_FIELD_GROUPS,
+  ADVANCED_SEARCH_FIELD_KINDS,
+  ADVANCED_SEARCH_FIELD_LABELS,
+  ADVANCED_SEARCH_MAX_ATOMS,
+  ADVANCED_SEARCH_MAX_DEPTH,
+  ADVANCED_SEARCH_OPERATOR_LABELS,
+  ADVANCED_SEARCH_OPERATORS_BY_KIND,
+  AdvancedSearchDateUnit,
+  AdvancedSearchField,
+  AdvancedSearchFieldKind,
+  AdvancedSearchLogicalOperator,
+  AdvancedSearchOperator,
+  AdvancedSearchQueryAtom,
+  AdvancedSearchQueryElement,
+  AdvancedSearchQueryElementType,
+  AdvancedSearchQueryGroup,
+} from 'src/app/data/advanced-search-query'
+import {
+  parseAdvancedSearchQuery,
+  serializeAdvancedSearchQuery,
+} from 'src/app/utils/advanced-search-query'
+import { LoadingComponentWithPermissions } from '../../loading-component/loading.component'
+
+@Component({
+  selector: 'pngx-advanced-search-dialog',
+  templateUrl: './advanced-search-dialog.component.html',
+  styleUrl: './advanced-search-dialog.component.scss',
+  imports: [FormsModule, NgTemplateOutlet, NgxBootstrapIconsModule],
+})
+export class AdvancedSearchDialogComponent extends LoadingComponentWithPermissions {
+  private activeModal = inject(NgbActiveModal)
+
+  public readonly ElementType = AdvancedSearchQueryElementType
+  public readonly LogicalOperator = AdvancedSearchLogicalOperator
+  public readonly Operator = AdvancedSearchOperator
+  public readonly FieldKind = AdvancedSearchFieldKind
+  public readonly fieldGroups = ADVANCED_SEARCH_FIELD_GROUPS
+  public readonly fieldLabels = ADVANCED_SEARCH_FIELD_LABELS
+  public readonly dateKeywords = ADVANCED_SEARCH_DATE_KEYWORDS
+  public readonly dateKeywordLabels = ADVANCED_SEARCH_DATE_KEYWORD_LABELS
+  public readonly dateUnits = Object.values(AdvancedSearchDateUnit)
+  public readonly dateUnitLabels = ADVANCED_SEARCH_DATE_UNIT_LABELS
+  public readonly maxDepth = ADVANCED_SEARCH_MAX_DEPTH
+  public readonly maxAtoms = ADVANCED_SEARCH_MAX_ATOMS
+
+  @Output()
+  public queryApplied = new EventEmitter<string>()
+
+  public root: AdvancedSearchQueryGroup = this.emptyRoot()
+
+  // True when the query in the search box uses syntax the editor cannot show
+  public unreadable: boolean = false
+
+  private _query: string = ''
+
+  @Input()
+  set query(query: string) {
+    this._query = query ?? ''
+    const parsed = parseAdvancedSearchQuery(this._query)
+    this.unreadable = !!this._query.trim() && !parsed
+    this.root = parsed ?? this.emptyRoot()
+  }
+
+  get query(): string {
+    return this._query
+  }
+
+  constructor() {
+    super()
+    this.loading.set(false)
+  }
+
+  // Stable ids for the radio groups, without putting them in the query model
+  private ids = new WeakMap<object, number>()
+  private nextId = 0
+
+  public idFor(element: AdvancedSearchQueryElement): number {
+    if (!this.ids.has(element)) {
+      this.ids.set(element, this.nextId++)
+    }
+    return this.ids.get(element)
+  }
+
+  private emptyRoot(): AdvancedSearchQueryGroup {
+    return {
+      type: AdvancedSearchQueryElementType.Group,
+      operator: AdvancedSearchLogicalOperator.And,
+      children: [this.newAtom()],
+    }
+  }
+
+  private newAtom(): AdvancedSearchQueryAtom {
+    return {
+      type: AdvancedSearchQueryElementType.Atom,
+      field: AdvancedSearchField.Content,
+      operator: AdvancedSearchOperator.AllWords,
+      value: '',
+    }
+  }
+
+  public get generatedQuery(): string {
+    return serializeAdvancedSearchQuery(this.root)
+  }
+
+  public kindOf(atom: AdvancedSearchQueryAtom): AdvancedSearchFieldKind {
+    return ADVANCED_SEARCH_FIELD_KINDS[atom.field]
+  }
+
+  public operatorsFor(atom: AdvancedSearchQueryAtom): AdvancedSearchOperator[] {
+    return ADVANCED_SEARCH_OPERATORS_BY_KIND[this.kindOf(atom)]
+  }
+
+  public operatorLabel(
+    atom: AdvancedSearchQueryAtom,
+    operator: AdvancedSearchOperator
+  ): string {
+    return this.kindOf(atom) === AdvancedSearchFieldKind.Date
+      ? (ADVANCED_SEARCH_DATE_OPERATOR_LABELS[operator] ??
+          ADVANCED_SEARCH_OPERATOR_LABELS[operator])
+      : ADVANCED_SEARCH_OPERATOR_LABELS[operator]
+  }
+
+  public placeholderFor(atom: AdvancedSearchQueryAtom): string {
+    switch (atom.operator) {
+      case AdvancedSearchOperator.Phrase:
+        return $localize`phrase`
+      case AdvancedSearchOperator.StartsWith:
+        return $localize`beginning of a word`
+      default:
+        return $localize`words`
+    }
+  }
+
+  public onFieldChange(atom: AdvancedSearchQueryAtom) {
+    // Keep the condition only if the new field still offers it
+    if (!this.operatorsFor(atom).includes(atom.operator)) {
+      atom.operator = this.operatorsFor(atom)[0]
+    }
+    this.clearValues(atom)
+  }
+
+  public onOperatorChange(atom: AdvancedSearchQueryAtom) {
+    this.clearValues(atom)
+  }
+
+  private clearValues(atom: AdvancedSearchQueryAtom) {
+    atom.value = ''
+    atom.valueTo = undefined
+    atom.unit =
+      atom.operator === AdvancedSearchOperator.WithinLast
+        ? AdvancedSearchDateUnit.Day
+        : undefined
+  }
+
+  public addAtom(group: AdvancedSearchQueryGroup) {
+    group.children.push(this.newAtom())
+  }
+
+  public addGroup(group: AdvancedSearchQueryGroup) {
+    group.children.push({
+      type: AdvancedSearchQueryElementType.Group,
+      operator: AdvancedSearchLogicalOperator.Or,
+      children: [this.newAtom()],
+    })
+  }
+
+  public remove(
+    parent: AdvancedSearchQueryGroup,
+    element: AdvancedSearchQueryElement
+  ) {
+    parent.children = parent.children.filter((child) => child !== element)
+  }
+
+  public startOver() {
+    this.unreadable = false
+    this.root = this.emptyRoot()
+  }
+
+  public apply() {
+    this.queryApplied.emit(this.generatedQuery)
+    this.activeModal.close()
+  }
+
+  public cancel() {
+    this.activeModal.close()
+  }
+}

src-ui/src/app/components/common/pdf-viewer/pdf-viewer.component.scss

@@ -154,29 +154,11 @@
   & section {
     position: absolute;
     text-align: initial;
-    pointer-events: auto;
     box-sizing: border-box;
     transform-origin: 0 0;
   }
 
-  & :is(.linkAnnotation, .buttonWidgetAnnotation.pushButton) > a {
-    position: absolute;
-    inset: 0;
-    font-size: 1em;
-    transition: none;
-  }
-
-  & :is(.linkAnnotation, .buttonWidgetAnnotation.pushButton):not(.hasBorder)
-    > a:hover {
-    opacity: 0.2;
-    background-color: rgb(255 255 0);
-  }
-
   & .annotationTextContent {
     opacity: 0;
   }
 }
-
-:host ::ng-deep .textLayer.selecting ~ .annotationLayer section {
-  pointer-events: none;
-}

src-ui/src/app/components/common/pdf-viewer/pdf-viewer.component.spec.ts

@@ -1,11 +1,7 @@
 import { SimpleChange } from '@angular/core'
 import { ComponentFixture, TestBed } from '@angular/core/testing'
 import * as pdfjs from 'pdfjs-dist/legacy/build/pdf.mjs'
-import {
+import { PDFSinglePageViewer, PDFViewer } from 'pdfjs-dist/web/pdf_viewer.mjs'
-  LinkTarget,
-  PDFSinglePageViewer,
-  PDFViewer,
-} from 'pdfjs-dist/web/pdf_viewer.mjs'
 import { PngxPdfViewerComponent } from './pdf-viewer.component'
 import { PdfRenderMode, PdfZoomLevel, PdfZoomScale } from './pdf-viewer.types'
 
@@ -62,16 +58,6 @@ describe('PngxPdfViewerComponent', () => {
     expect((component as any).pdfViewer).toBeInstanceOf(PDFViewer)
   })
 
-  it('opens external links in a new tab', () => {
-    const linkService = (component as any).linkService
-    expect(linkService.options).toEqual(
-      expect.objectContaining({
-        externalLinkTarget: LinkTarget.BLANK,
-        externalLinkRel: 'noopener noreferrer nofollow',
-      })
-    )
-  })
-
   it('resolves the worker source relative to the document base URI', async () => {
     setBaseHref('/paperless/')
     const getDocumentSpy = jest.spyOn(pdfjs, 'getDocument')
@@ -90,7 +76,6 @@ describe('PngxPdfViewerComponent', () => {
       withCredentials: true,
       wasmUrl: expect.stringContaining('/paperless/assets/wasm/'),
       iccUrl: expect.stringContaining('/paperless/assets/iccs/'),
-      cMapUrl: expect.stringContaining('/paperless/assets/cmaps/'),
     })
   })
 

src-ui/src/app/components/common/pdf-viewer/pdf-viewer.component.ts

@@ -21,7 +21,6 @@ import {
 } from 'pdfjs-dist/legacy/build/pdf.mjs'
 import {
   EventBus,
-  LinkTarget,
   PDFFindController,
   PDFLinkService,
   PDFSinglePageViewer,
@@ -76,11 +75,7 @@ export class PngxPdfViewerComponent
   private lastViewerPage?: number
 
   private readonly eventBus = new EventBus()
-  private readonly linkService = new PDFLinkService({
+  private readonly linkService = new PDFLinkService({ eventBus: this.eventBus })
-    eventBus: this.eventBus,
-    externalLinkTarget: LinkTarget.BLANK,
-    externalLinkRel: 'noopener noreferrer nofollow',
-  })
   private readonly findController = new PDFFindController({
     eventBus: this.eventBus,
     linkService: this.linkService,
@@ -190,7 +185,6 @@ export class PngxPdfViewerComponent
       withCredentials: true,
       wasmUrl: new URL('assets/wasm/', this.document.baseURI).toString(),
       iccUrl: new URL('assets/iccs/', this.document.baseURI).toString(),
-      cMapUrl: new URL('assets/cmaps/', this.document.baseURI).toString(),
     }
     this.loadingTask = getDocument(initOptions)
     try {

src-ui/src/app/components/common/suggestions-dropdown/suggestions-dropdown.component.html

@@ -1,84 +1,58 @@
-<div class="d-flex align-items-center">
+<div class="btn-group">
-  <div class="btn-group">
+  <button type="button" class="btn btn-sm btn-outline-primary" (click)="clickSuggest()" [disabled]="disabled() || loading() || (suggestions() && !aiEnabled())" [aria-label]="noSuggestions ? 'No suggestions' : 'Suggest'" i18n-aria-label>
-    <button type="button" class="btn btn-sm btn-outline-primary" (click)="clickSuggest()" [disabled]="disabled() || loading() || (suggestions() && !aiEnabled())" [aria-label]="noSuggestions ? 'No suggestions' : 'Suggest'" i18n-aria-label>
+    @if (loading()) {
-      @if (loading()) {
+      <div class="spinner-border spinner-border-sm" role="status"></div>
-        <div class="spinner-border spinner-border-sm" role="status"></div>
+    } @else if (noSuggestions) {
-      } @else if (noSuggestions) {
+      <i-bs width="1.2em" height="1.2em" name="check-circle"></i-bs>
-        <i-bs width="1.2em" height="1.2em" name="check-circle"></i-bs>
+    } @else {
-      } @else {
+      <i-bs width="1.2em" height="1.2em" name="stars"></i-bs>
-        <i-bs width="1.2em" height="1.2em" name="lightbulb"></i-bs>
-      }
-      @if (noSuggestions) {
-        <span class="d-none d-lg-inline ps-1" i18n>No suggestions</span>
-      } @else {
-        <span class="d-none d-lg-inline ps-1" i18n>Suggest</span>
-      }
-      @if (totalSuggestions > 0) {
-        <span class="badge bg-primary ms-2">{{ totalSuggestions }}</span>
-      }
-    </button>
-
-    @if (aiEnabled()) {
-      <div class="btn-group" ngbDropdown #dropdown="ngbDropdown" [popperOptions]="popperOptions">
-        <button type="button" class="btn btn-sm btn-outline-primary" ngbDropdownToggle [disabled]="disabled() || loading() || !suggestions()" aria-expanded="false" aria-controls="suggestionsDropdown" aria-label="Suggestions dropdown">
-          <span class="visually-hidden" i18n>Show suggestions</span>
-        </button>
-
-        <div ngbDropdownMenu aria-labelledby="suggestionsDropdown" class="shadow suggestions-dropdown">
-          <div class="list-group list-group-flush small pb-0">
-            @if (novelSuggestions === 0 && reusableSuggestions === 0) {
-              <div class="list-group-item text-muted fst-italic">
-                <small class="text-muted small fst-italic" i18n>No novel suggestions</small>
-              </div>
-            }
-            @if (suggestions()?.suggested_tags?.length > 0) {
-              <small class="list-group-item text-uppercase text-muted small"><i-bs class="me-2" name="tags"></i-bs><ng-container i18n>Tags</ng-container></small>
-              @for (tag of suggestions().suggested_tags; track tag) {
-                <button type="button" class="list-group-item list-group-item-action bg-light" (click)="addTag.emit(tag)">{{ tag }}</button>
-              }
-            }
-            @if (suggestions()?.suggested_document_types?.length > 0) {
-              <div class="list-group-item text-uppercase text-muted small"><i-bs class="me-2" name="hash"></i-bs><ng-container i18n>Document Types</ng-container></div>
-              @for (type of suggestions().suggested_document_types; track type) {
-                <button type="button" class="list-group-item list-group-item-action bg-light" (click)="addDocumentType.emit(type)">{{ type }}</button>
-              }
-            }
-            @if (suggestions()?.suggested_correspondents?.length > 0) {
-              <div class="list-group-item text-uppercase text-muted small"><i-bs class="me-2" name="person"></i-bs><ng-container i18n>Correspondents</ng-container></div>
-              @for (correspondent of suggestions().suggested_correspondents; track correspondent) {
-                <button type="button" class="list-group-item list-group-item-action bg-light" (click)="addCorrespondent.emit(correspondent)">{{ correspondent }}</button>
-              }
-            }
-            @if (reusableSuggestions > 0) {
-              <div class="list-group-item text-muted fst-italic">
-                <small class="text-muted small fst-italic" i18n>{reusableSuggestions, plural, =1 {1 existing value suggested below} other {{{reusableSuggestions}} existing values suggested below}}</small>
-              </div>
-            }
-          </div>
-        </div>
-      </div>
     }
-  </div>
+    @if (noSuggestions) {
+      <span class="d-none d-lg-inline ps-1" i18n>No suggestions</span>
+    } @else {
+      <span class="d-none d-lg-inline ps-1" i18n>Suggest</span>
+    }
+    @if (totalSuggestions > 0) {
+      <span class="badge bg-primary ms-2">{{ totalSuggestions }}</span>
+    }
+  </button>
 
   @if (aiEnabled()) {
-    <div ngbDropdown autoClose="outside" placement="bottom-end" [popperOptions]="popperOptions">
+    <div class="btn-group" ngbDropdown #dropdown="ngbDropdown" [popperOptions]="popperOptions">
-      <button type="button" class="btn btn-sm btn-link position-relative" ngbDropdownToggle [disabled]="disabled() || loading()" i18n-title title="Suggestion options">
+      <button type="button" class="btn btn-sm btn-outline-primary" ngbDropdownToggle [disabled]="disabled() || loading() || !suggestions()" aria-expanded="false" aria-controls="suggestionsDropdown" aria-label="Suggestions dropdown">
-        <i-bs name="three-dots"></i-bs>
+        <span class="visually-hidden" i18n>Show suggestions</span>
-        @if (source() !== defaultSource()) {
-          <span class="position-absolute top-0 start-100 translate-middle p-1 bg-primary border border-light rounded-circle">
-            <span class="visually-hidden" i18n>Not using default</span>
-          </span>
-        }
       </button>
-      <div ngbDropdownMenu class="shadow p-3">
+
-        <div class="small text-muted mb-2" i18n>Suggest using:</div>
+      <div ngbDropdownMenu aria-labelledby="suggestionsDropdown" class="shadow suggestions-dropdown">
-        <div class="form-check small">
+        <div class="list-group list-group-flush small pb-0">
-          <input class="form-check-input" type="checkbox" id="suggestionSourceML" [checked]="useML" [disabled]="useML && !useAI" (change)="setSources($event.target.checked, useAI)">
+          @if (novelSuggestions === 0 && reusableSuggestions === 0) {
-          <label class="form-check-label d-inline-flex align-items-center gap-1" for="suggestionSourceML"><i-bs name="cpu"></i-bs><ng-container i18n>ML</ng-container></label>
+            <div class="list-group-item text-muted fst-italic">
-        </div>
+              <small class="text-muted small fst-italic" i18n>No novel suggestions</small>
-        <div class="form-check small">
+            </div>
-          <input class="form-check-input" type="checkbox" id="suggestionSourceAI" [checked]="useAI" [disabled]="useAI && !useML" (change)="setSources(useML, $event.target.checked)">
+          }
-          <label class="form-check-label d-inline-flex align-items-center gap-1" for="suggestionSourceAI"><i-bs name="stars"></i-bs><ng-container i18n>AI</ng-container></label>
+          @if (suggestions()?.suggested_tags?.length > 0) {
+            <small class="list-group-item text-uppercase text-muted small"><i-bs class="me-2" name="tags"></i-bs><ng-container i18n>Tags</ng-container></small>
+            @for (tag of suggestions().suggested_tags; track tag) {
+              <button type="button" class="list-group-item list-group-item-action bg-light" (click)="addTag.emit(tag)">{{ tag }}</button>
+            }
+          }
+          @if (suggestions()?.suggested_document_types?.length > 0) {
+            <div class="list-group-item text-uppercase text-muted small"><i-bs class="me-2" name="hash"></i-bs><ng-container i18n>Document Types</ng-container></div>
+            @for (type of suggestions().suggested_document_types; track type) {
+              <button type="button" class="list-group-item list-group-item-action bg-light" (click)="addDocumentType.emit(type)">{{ type }}</button>
+            }
+          }
+          @if (suggestions()?.suggested_correspondents?.length > 0) {
+            <div class="list-group-item text-uppercase text-muted small"><i-bs class="me-2" name="person"></i-bs><ng-container i18n>Correspondents</ng-container></div>
+            @for (correspondent of suggestions().suggested_correspondents; track correspondent) {
+              <button type="button" class="list-group-item list-group-item-action bg-light" (click)="addCorrespondent.emit(correspondent)">{{ correspondent }}</button>
+            }
+          }
+          @if (reusableSuggestions > 0) {
+            <div class="list-group-item text-muted fst-italic">
+              <small class="text-muted small fst-italic" i18n>{reusableSuggestions, plural, =1 {1 existing value suggested below} other {{{reusableSuggestions}} existing values suggested below}}</small>
+            </div>
+          }
         </div>
       </div>
     </div>

src-ui/src/app/components/common/suggestions-dropdown/suggestions-dropdown.component.scss

@@ -1,7 +1,3 @@
 .suggestions-dropdown {
   min-width: 250px;
 }
-
-.btn-link.dropdown-toggle::after {
-  display: none;
-}

src-ui/src/app/components/common/suggestions-dropdown/suggestions-dropdown.component.spec.ts

@@ -1,7 +1,6 @@
 import { ComponentFixture, TestBed } from '@angular/core/testing'
 import { NgbDropdownModule } from '@ng-bootstrap/ng-bootstrap'
 import { NgxBootstrapIconsModule, allIcons } from 'ngx-bootstrap-icons'
-import { SuggestionSource } from 'src/app/data/ui-settings'
 import { SuggestionsDropdownComponent } from './suggestions-dropdown.component'
 
 describe('SuggestionsDropdownComponent', () => {
@@ -180,71 +179,14 @@ describe('SuggestionsDropdownComponent', () => {
 
   it('should toggle dropdown when clickSuggest is called and suggestions are not null', () => {
     fixture.componentRef.setInput('aiEnabled', true)
-    fixture.componentRef.setInput('fetchedSources', [SuggestionSource.ML])
     fixture.detectChanges()
     fixture.componentRef.setInput('suggestions', {
       suggested_correspondents: [],
       suggested_tags: [],
       suggested_document_types: [],
     })
-    fixture.detectChanges()
     component.clickSuggest()
-    expect(component.dropdown.isOpen()).toBeTruthy()
+    expect(component.dropdown.open).toBeTruthy()
     expect(fixture.nativeElement.textContent).toContain('No novel suggestions')
   })
-
-  it('should fetch unfetched sources and show existing suggestions', () => {
-    jest.spyOn(component.getSuggestions, 'emit')
-    fixture.componentRef.setInput('aiEnabled', true)
-    fixture.componentRef.setInput('source', SuggestionSource.Both)
-    fixture.componentRef.setInput('fetchedSources', [SuggestionSource.ML])
-    fixture.componentRef.setInput('suggestions', { tags: [1] })
-    fixture.detectChanges()
-    component.clickSuggest()
-    expect(component.getSuggestions.emit).toHaveBeenCalledWith(
-      SuggestionSource.Both
-    )
-    expect(component.dropdown.isOpen()).toBeTruthy()
-  })
-
-  it('should only show source options when AI is enabled', () => {
-    expect(
-      fixture.nativeElement.querySelector('#suggestionSourceML')
-    ).toBeNull()
-    fixture.componentRef.setInput('aiEnabled', true)
-    fixture.detectChanges()
-    fixture.nativeElement
-      .querySelector('button[title="Suggestion options"]')
-      .click()
-    fixture.detectChanges()
-    expect(
-      fixture.nativeElement.querySelector('#suggestionSourceML')
-    ).not.toBeNull()
-  })
-
-  it('should emit source changes and never allow no source', () => {
-    const emitSpy = jest.spyOn(component.sourceChange, 'emit')
-    component.setSources(true, true)
-    expect(emitSpy).toHaveBeenCalledWith(SuggestionSource.Both)
-    component.setSources(true, false)
-    expect(emitSpy).toHaveBeenCalledWith(SuggestionSource.ML)
-    component.setSources(false, true)
-    expect(emitSpy).toHaveBeenCalledWith(SuggestionSource.AI)
-
-    emitSpy.mockClear()
-    component.setSources(false, false)
-    expect(emitSpy).not.toHaveBeenCalled()
-  })
-
-  it('should indicate a non-default source', () => {
-    fixture.componentRef.setInput('aiEnabled', true)
-    fixture.componentRef.setInput('source', SuggestionSource.AI)
-    fixture.componentRef.setInput('defaultSource', SuggestionSource.AI)
-    fixture.detectChanges()
-    expect(fixture.nativeElement.textContent).not.toContain('Not using default')
-
-    fixture.componentRef.setInput('source', SuggestionSource.Both)
-    fixture.detectChanges()
-    expect(fixture.nativeElement.textContent).toContain('Not using default')
-  })
 })

src-ui/src/app/components/common/suggestions-dropdown/suggestions-dropdown.component.ts

@@ -8,7 +8,6 @@ import {
 import { NgbDropdown, NgbDropdownModule } from '@ng-bootstrap/ng-bootstrap'
 import { NgxBootstrapIconsModule } from 'ngx-bootstrap-icons'
 import { DocumentSuggestions } from 'src/app/data/document-suggestions'
-import { SuggestionSource } from 'src/app/data/ui-settings'
 import { pngxPopperOptions } from 'src/app/utils/popper-options'
 
 @Component({
@@ -19,16 +18,12 @@ import { pngxPopperOptions } from 'src/app/utils/popper-options'
 })
 export class SuggestionsDropdownComponent {
   public popperOptions = pngxPopperOptions
-  public readonly SuggestionSource = SuggestionSource
 
   @ViewChild('dropdown') dropdown: NgbDropdown
   readonly suggestions = input<DocumentSuggestions>(null)
   readonly aiEnabled = input(false)
   readonly loading = input(false)
   readonly disabled = input(false)
-  readonly source = input<SuggestionSource>(SuggestionSource.ML)
-  readonly defaultSource = input<SuggestionSource>(SuggestionSource.ML)
-  readonly fetchedSources = input<SuggestionSource[]>([])
 
   readonly appliedTags = input<number[]>([])
   readonly appliedCorrespondent = input<number>(null)
@@ -36,10 +31,8 @@ export class SuggestionsDropdownComponent {
   readonly appliedStoragePath = input<number>(null)
 
   @Output()
-  getSuggestions: EventEmitter<SuggestionSource> = new EventEmitter()
+  getSuggestions: EventEmitter<SuggestionsDropdownComponent> =
-
+    new EventEmitter()
-  @Output()
-  sourceChange: EventEmitter<SuggestionSource> = new EventEmitter()
 
   @Output()
   addTag: EventEmitter<string> = new EventEmitter()
@@ -60,42 +53,12 @@ export class SuggestionsDropdownComponent {
     }
 
     if (!this.suggestions()) {
-      this.getSuggestions.emit(this.source())
+      this.getSuggestions.emit(this)
-    } else if (this.hasUnfetchedSources) {
-      // sources changed, fetch the rest and show what we have meanwhile
-      this.getSuggestions.emit(this.source())
-      this.dropdown?.open()
     } else {
       this.dropdown?.toggle()
     }
   }
 
-  get useML(): boolean {
-    return this.source() !== SuggestionSource.AI
-  }
-
-  get useAI(): boolean {
-    return this.source() !== SuggestionSource.ML
-  }
-
-  get hasUnfetchedSources(): boolean {
-    const fetched = this.fetchedSources()
-    return (
-      (this.useML && !fetched.includes(SuggestionSource.ML)) ||
-      (this.useAI && !fetched.includes(SuggestionSource.AI))
-    )
-  }
-
-  public setSources(ml: boolean, ai: boolean) {
-    if (ml && ai) {
-      this.sourceChange.emit(SuggestionSource.Both)
-    } else if (ml) {
-      this.sourceChange.emit(SuggestionSource.ML)
-    } else if (ai) {
-      this.sourceChange.emit(SuggestionSource.AI)
-    }
-  }
-
   get novelSuggestions(): number {
     return (
       (this.suggestions()?.suggested_correspondents?.length ?? 0) +

src-ui/src/app/components/document-detail/document-detail.component.html

@@ -134,15 +134,11 @@
                 [loading]="suggestionsLoading()"
                 [suggestions]="suggestions()"
                 [aiEnabled]="aiEnabled"
-                [source]="suggestionSource"
-                [defaultSource]="defaultSuggestionSource"
-                [fetchedSources]="fetchedSuggestionSources()"
                 [appliedTags]="documentForm.value.tags"
                 [appliedCorrespondent]="documentForm.value.correspondent"
                 [appliedDocumentType]="documentForm.value.document_type"
                 [appliedStoragePath]="documentForm.value.storage_path"
-                (getSuggestions)="getSuggestions($event)"
+                (getSuggestions)="getSuggestions()"
-                (sourceChange)="suggestionSourceOverride.set($event)"
                 (addTag)="createTag($event)"
                 (addDocumentType)="createDocumentType($event)"
                 (addCorrespondent)="createCorrespondent($event)">

src-ui/src/app/components/document-detail/document-detail.component.spec.ts

@@ -43,7 +43,7 @@ import {
 } from 'src/app/data/filter-rule-type'
 import { StoragePath } from 'src/app/data/storage-path'
 import { Tag } from 'src/app/data/tag'
-import { SETTINGS_KEYS, SuggestionSource } from 'src/app/data/ui-settings'
+import { SETTINGS_KEYS } from 'src/app/data/ui-settings'
 import { PermissionsGuard } from 'src/app/guards/permissions.guard'
 import { CustomDatePipe } from 'src/app/pipes/custom-date.pipe'
 import { DocumentTitlePipe } from 'src/app/pipes/document-title.pipe'
@@ -1528,113 +1528,6 @@ describe('DocumentDetailComponent', () => {
     expect(component.suggestionsLoading()).toBeFalsy()
   })
 
-  it('should get and merge ML and AI suggestions when source is both', () => {
-    settingsService.set(
-      SETTINGS_KEYS.DOCUMENT_EDITING_SUGGESTION_SOURCE,
-      SuggestionSource.Both
-    )
-    const getSetting = settingsService.get.bind(settingsService)
-    jest
-      .spyOn(settingsService, 'get')
-      .mockImplementation((key) =>
-        key === SETTINGS_KEYS.AI_ENABLED ? true : getSetting(key)
-      )
-    const suggestionsSpy = jest
-      .spyOn(documentService, 'getSuggestions')
-      .mockReturnValue(of({ tags: [42], dates: ['2024-01-01'] }))
-    const aiSuggestionsSpy = jest
-      .spyOn(documentService, 'getAiSuggestions')
-      .mockReturnValue(
-        of({ title: 'AI title', tags: [42, 43], suggested_tags: ['New'] })
-      )
-    initNormally()
-    expect(suggestionsSpy).toHaveBeenCalled()
-    expect(aiSuggestionsSpy).toHaveBeenCalled()
-    expect(component.suggestions().title).toEqual('AI title')
-    expect(component.suggestions().tags).toEqual([42, 43])
-    expect(component.suggestions().suggested_tags).toEqual(['New'])
-    expect(component.suggestions().dates).toEqual(['2024-01-01'])
-  })
-
-  it('should only fetch sources not yet fetched for the document', () => {
-    settingsService.set(SETTINGS_KEYS.DOCUMENT_EDITING_AUTO_SUGGEST, false)
-    settingsService.set(
-      SETTINGS_KEYS.DOCUMENT_EDITING_SUGGESTION_SOURCE,
-      SuggestionSource.ML
-    )
-    const getSetting = settingsService.get.bind(settingsService)
-    jest
-      .spyOn(settingsService, 'get')
-      .mockImplementation((key) =>
-        key === SETTINGS_KEYS.AI_ENABLED ? true : getSetting(key)
-      )
-    const suggestionsSpy = jest
-      .spyOn(documentService, 'getSuggestions')
-      .mockReturnValue(of({ tags: [42] }))
-    const aiSuggestionsSpy = jest
-      .spyOn(documentService, 'getAiSuggestions')
-      .mockReturnValue(of({ tags: [43] }))
-    initNormally()
-
-    component.getSuggestions()
-    expect(suggestionsSpy).toHaveBeenCalledTimes(1)
-    expect(aiSuggestionsSpy).not.toHaveBeenCalled()
-
-    component.getSuggestions(SuggestionSource.Both)
-    expect(suggestionsSpy).toHaveBeenCalledTimes(1)
-    expect(aiSuggestionsSpy).toHaveBeenCalledTimes(1)
-    expect(component.suggestions().tags).toEqual([42, 43])
-
-    component.getSuggestions(SuggestionSource.Both)
-    expect(suggestionsSpy).toHaveBeenCalledTimes(1)
-    expect(aiSuggestionsSpy).toHaveBeenCalledTimes(1)
-  })
-
-  it('should use the per-document source override and reset it on document change', () => {
-    settingsService.set(SETTINGS_KEYS.DOCUMENT_EDITING_AUTO_SUGGEST, false)
-    const getSetting = settingsService.get.bind(settingsService)
-    jest
-      .spyOn(settingsService, 'get')
-      .mockImplementation((key) =>
-        key === SETTINGS_KEYS.AI_ENABLED ? true : getSetting(key)
-      )
-    initNormally()
-    expect(component.suggestionSource).toEqual(SuggestionSource.AI)
-    component.suggestionSourceOverride.set(SuggestionSource.ML)
-    expect(component.suggestionSource).toEqual(SuggestionSource.ML)
-
-    jest
-      .spyOn(documentService, 'get')
-      .mockReturnValueOnce(of(Object.assign({}, doc)))
-    ;(component as any).loadDocument(doc.id, true)
-    expect(component.suggestionSourceOverride()).toBeNull()
-    expect(component.fetchedSuggestionSources()).toEqual([])
-  })
-
-  it('should keep suggestions from one source if the other fails', () => {
-    settingsService.set(
-      SETTINGS_KEYS.DOCUMENT_EDITING_SUGGESTION_SOURCE,
-      SuggestionSource.Both
-    )
-    const getSetting = settingsService.get.bind(settingsService)
-    jest
-      .spyOn(settingsService, 'get')
-      .mockImplementation((key) =>
-        key === SETTINGS_KEYS.AI_ENABLED ? true : getSetting(key)
-      )
-    const errorSpy = jest.spyOn(toastService, 'showError')
-    jest
-      .spyOn(documentService, 'getSuggestions')
-      .mockReturnValue(of({ tags: [42] }))
-    jest
-      .spyOn(documentService, 'getAiSuggestions')
-      .mockReturnValue(throwError(() => new Error('failed')))
-    initNormally()
-    expect(errorSpy).toHaveBeenCalled()
-    expect(component.suggestions().tags).toEqual([42])
-    expect(component.fetchedSuggestionSources()).toEqual([SuggestionSource.ML])
-  })
-
   it('should show error if needed for get suggestions', () => {
     const suggestionsSpy = jest.spyOn(documentService, 'getSuggestions')
     const errorSpy = jest.spyOn(toastService, 'showError')

src-ui/src/app/components/document-detail/document-detail.component.ts

@@ -28,7 +28,7 @@ import {
 import { dirtyCheck, DirtyComponent } from '@ngneat/dirty-check-forms'
 import { NgxBootstrapIconsModule } from 'ngx-bootstrap-icons'
 import { DeviceDetectorService } from 'ngx-device-detector'
-import { BehaviorSubject, merge, Observable, of, Subject, timer } from 'rxjs'
+import { BehaviorSubject, Observable, of, Subject, timer } from 'rxjs'
 import {
   catchError,
   debounceTime,
@@ -48,10 +48,7 @@ import { DataType } from 'src/app/data/datatype'
 import { Document, DocumentVersionInfo } from 'src/app/data/document'
 import { DocumentMetadata } from 'src/app/data/document-metadata'
 import { DocumentNote } from 'src/app/data/document-note'
-import {
+import { DocumentSuggestions } from 'src/app/data/document-suggestions'
-  DocumentSuggestions,
-  mergeSuggestions,
-} from 'src/app/data/document-suggestions'
 import { DocumentType } from 'src/app/data/document-type'
 import { FilterRule } from 'src/app/data/filter-rule'
 import {
@@ -66,7 +63,7 @@ import {
 import { ObjectWithId } from 'src/app/data/object-with-id'
 import { StoragePath } from 'src/app/data/storage-path'
 import { Tag } from 'src/app/data/tag'
-import { SETTINGS_KEYS, SuggestionSource } from 'src/app/data/ui-settings'
+import { SETTINGS_KEYS } from 'src/app/data/ui-settings'
 import { User } from 'src/app/data/user'
 import { IfPermissionsDirective } from 'src/app/directives/if-permissions.directive'
 import { CustomDatePipe } from 'src/app/pipes/custom-date.pipe'
@@ -243,10 +240,6 @@ export class DocumentDetailComponent
   private readonly autoSuggestSetting = this.settings.getSignal<boolean>(
     SETTINGS_KEYS.DOCUMENT_EDITING_AUTO_SUGGEST
   )
-  private readonly suggestionSourceSetting =
-    this.settings.getSignal<SuggestionSource>(
-      SETTINGS_KEYS.DOCUMENT_EDITING_SUGGESTION_SOURCE
-    )
   private readonly hiddenFieldsSetting = this.settings.getSignal<
     DocumentDetailFieldID[]
   >(SETTINGS_KEYS.DOCUMENT_DETAILS_HIDDEN_FIELDS)
@@ -268,9 +261,6 @@ export class DocumentDetailComponent
   readonly metadata = signal<DocumentMetadata>(undefined)
   readonly suggestions = signal<DocumentSuggestions>(undefined)
   readonly suggestionsLoading = signal(false)
-  // per-document, resets on navigation
-  readonly suggestionSourceOverride = signal<SuggestionSource>(null)
-  readonly fetchedSuggestionSources = signal<SuggestionSource[]>([])
   readonly users = signal<User[]>(undefined)
 
   readonly title = signal<string>(undefined)
@@ -375,15 +365,6 @@ export class DocumentDetailComponent
     return this.autoSuggestSetting()
   }
 
-  get defaultSuggestionSource(): SuggestionSource {
-    return this.aiEnabled ? this.suggestionSourceSetting() : SuggestionSource.ML
-  }
-
-  get suggestionSource(): SuggestionSource {
-    if (!this.aiEnabled) return SuggestionSource.ML
-    return this.suggestionSourceOverride() ?? this.defaultSuggestionSource
-  }
-
   get archiveContentRenderType(): ContentRenderType {
     const hasArchiveVersion =
       this.metadata()?.has_archive_version ??
@@ -609,8 +590,6 @@ export class DocumentDetailComponent
           }
           this.documentId.set(doc.id)
           this.suggestions.set(null)
-          this.suggestionSourceOverride.set(null)
-          this.fetchedSuggestionSources.set([])
           const openDocument = this.openDocumentService.getOpenDocument(
             this.documentId()
           )
@@ -1098,44 +1077,29 @@ export class DocumentDetailComponent
     return this.documentForm.get('custom_fields') as FormArray
   }
 
-  getSuggestions(source: SuggestionSource = this.suggestionSource) {
+  getSuggestions() {
-    const sources = (
-      source === SuggestionSource.Both
-        ? [SuggestionSource.ML, SuggestionSource.AI]
-        : [source]
-    ).filter((s) => !this.fetchedSuggestionSources().includes(s))
-    if (!sources.length) return
-
     this.suggestionsLoading.set(true)
-    merge(
+    const suggestionsObservable = this.aiEnabled
-      ...sources.map((s) =>
+      ? this.documentsService.getAiSuggestions(this.documentId())
-        (s === SuggestionSource.AI
+      : this.documentsService.getSuggestions(this.documentId())
-          ? this.documentsService.getAiSuggestions(this.documentId())
+    suggestionsObservable
-          : this.documentsService.getSuggestions(this.documentId())
-        ).pipe(
-          first(),
-          map((result) => ({ source: s, result })),
-          catchError((error) => {
-            this.toastService.showError(
-              $localize`Error retrieving suggestions.`,
-              error
-            )
-            return of(null)
-          })
-        )
-      )
-    )
       .pipe(
+        first(),
         takeUntil(this.unsubscribeNotifier),
         takeUntil(this.docChangeNotifier),
         finalize(() => this.suggestionsLoading.set(false))
       )
-      .subscribe((response) => {
+      .subscribe({
-        if (!response) return
+        next: (result) => {
-        this.fetchedSuggestionSources.update((f) => [...f, response.source])
+          this.suggestions.set(result)
-        this.suggestions.set(
+        },
-          mergeSuggestions(this.suggestions(), response.result)
+        error: (error) => {
-        )
+          this.suggestions.set(null)
+          this.toastService.showError(
+            $localize`Error retrieving suggestions.`,
+            error
+          )
+        },
       })
   }
 

src-ui/src/app/components/document-list/document-list.component.spec.ts

@@ -146,19 +146,6 @@ describe('DocumentListComponent', () => {
     expect(reloadSpy).toHaveBeenCalled()
   })
 
-  it('should stop reloading on document deleted after destroy', () => {
-    const reloadSpy = jest.spyOn(documentListService, 'reload')
-    const documentDeletedSubject = new Subject<boolean>()
-    jest
-      .spyOn(websocketStatusService, 'onDocumentDeleted')
-      .mockReturnValue(documentDeletedSubject)
-    fixture.detectChanges()
-    fixture.destroy()
-    reloadSpy.mockClear()
-    documentDeletedSubject.next(true)
-    expect(reloadSpy).not.toHaveBeenCalled()
-  })
-
   it('should show score sort fields on fulltext queries', () => {
     documentListService.setFilterRules([
       {

src-ui/src/app/components/document-list/document-list.component.ts

@@ -270,12 +270,9 @@ export class DocumentListComponent
         this.list.reload()
       })
 
-    this.websocketStatusService
+    this.websocketStatusService.onDocumentDeleted().subscribe(() => {
-      .onDocumentDeleted()
+      this.list.reload()
-      .pipe(takeUntil(this.unsubscribeNotifier))
+    })
-      .subscribe(() => {
-        this.list.reload()
-      })
 
     this.route.paramMap
       .pipe(

src-ui/src/app/components/document-list/filter-editor/filter-editor.component.html

@@ -17,6 +17,12 @@
             }
           </select>
         }
+        @if (advancedSearchEditorAvailable) {
+          <button class="btn btn-sm btn-outline-primary" type="button" (click)="openAdvancedSearchEditor()"
+            title="Edit query" i18n-title [disabled]="disabled">
+            <i-bs name="sliders"></i-bs>
+          </button>
+        }
         @if (_textFilter) {
           <button class="btn btn-link btn-sm px-2 position-absolute top-0 end-0 z-10" (click)="resetTextField()" aria-label="Clear search" i18n-aria-label>
             <i-bs width="1em" height="1em" name="x"></i-bs>

src-ui/src/app/components/document-list/filter-editor/filter-editor.component.spec.ts

@@ -12,6 +12,8 @@ import {
   NgbDatepickerModule,
   NgbDropdownItem,
   NgbDropdownModule,
+  NgbModal,
+  NgbModalRef,
   NgbTypeaheadModule,
 } from '@ng-bootstrap/ng-bootstrap'
 import { NgSelectComponent, NgSelectModule } from '@ng-select/ng-select'
@@ -2515,4 +2517,45 @@ describe('FilterEditorComponent', () => {
 
     expect(component.textFilter).toEqual('help ')
   })
+
+  it('should open the advanced search editor with the current query and apply the result', () => {
+    const modalService: NgbModal = TestBed.inject(NgbModal)
+    let modal: NgbModalRef
+    modalService.activeInstances.subscribe(
+      (instances) => (modal = instances[0])
+    )
+    component.textFilterTarget = 'fulltext-query'
+    component.updateTextFilter('title:invoice')
+    fixture.detectChanges()
+
+    const editorButton = fixture.debugElement.query(
+      By.css('button[title="Edit query"]')
+    )
+    expect(editorButton).not.toBeNull()
+    editorButton.triggerEventHandler('click')
+    fixture.detectChanges()
+
+    expect(modal.componentInstance.query).toEqual('title:invoice')
+
+    const rulesSpy = jest.spyOn(component.filterRulesChange, 'next')
+    modal.componentInstance.queryApplied.emit('title:invoice AND NOT tag:paid')
+    expect(component.textFilter).toEqual('title:invoice AND NOT tag:paid')
+    expect(documentService.searchQuery).toEqual(
+      'title:invoice AND NOT tag:paid'
+    )
+    expect(rulesSpy).toHaveBeenCalledWith([
+      {
+        rule_type: FILTER_FULLTEXT_QUERY,
+        value: 'title:invoice AND NOT tag:paid',
+      },
+    ])
+  })
+
+  it('should not offer the advanced search editor for other targets', () => {
+    component.textFilterTarget = 'title-content'
+    fixture.detectChanges()
+    expect(
+      fixture.debugElement.query(By.css('button[title="Edit query"]'))
+    ).toBeNull()
+  })
 })

src-ui/src/app/components/document-list/filter-editor/filter-editor.component.ts

@@ -15,12 +15,13 @@ import {
 import { FormsModule, ReactiveFormsModule } from '@angular/forms'
 import {
   NgbDropdownModule,
+  NgbModal,
   NgbTypeahead,
   NgbTypeaheadModule,
 } from '@ng-bootstrap/ng-bootstrap'
 import { NgxBootstrapIconsModule } from 'ngx-bootstrap-icons'
 import { TourNgBootstrap } from 'ngx-ui-tour-ng-bootstrap'
-import { Observable, Subject, from } from 'rxjs'
+import { first, Observable, Subject, from } from 'rxjs'
 import {
   catchError,
   debounceTime,
@@ -121,6 +122,7 @@ import {
   PermissionsFilterDropdownComponent,
   PermissionsSelectionModel,
 } from '../../common/permissions-filter-dropdown/permissions-filter-dropdown.component'
+import { AdvancedSearchDialogComponent } from '../../common/advanced-search-dialog/advanced-search-dialog.component'
 import { LoadingComponentWithPermissions } from '../../loading-component/loading.component'
 
 const TEXT_FILTER_TARGET_TITLE = 'title'
@@ -286,6 +288,7 @@ export class FilterEditorComponent
   permissionsService = inject(PermissionsService)
   private customFieldService = inject(CustomFieldsService)
   private searchService = inject(SearchService)
+  private modalService = inject(NgbModal)
 
   generateFilterName() {
     if (this.filterRules.length == 1) {
@@ -1372,6 +1375,23 @@ export class FilterEditorComponent
     }
   }
 
+  get advancedSearchEditorAvailable(): boolean {
+    return this.textFilterTarget === TEXT_FILTER_TARGET_FULLTEXT_QUERY
+  }
+
+  openAdvancedSearchEditor() {
+    const modal = this.modalService.open(AdvancedSearchDialogComponent, {
+      backdrop: 'static',
+      size: 'lg',
+    })
+    modal.componentInstance.query = this._textFilter ?? ''
+    modal.componentInstance.queryApplied
+      .pipe(first())
+      .subscribe((query: string) => {
+        this.updateTextFilter(query)
+      })
+  }
+
   textFilterKeydown(event: KeyboardEvent) {
     if (event.key == 'Enter') {
       if (event.defaultPrevented) {

src-ui/src/app/data/advanced-search-query.ts

@@ -0,0 +1,262 @@
+// Fields and forms documented in docs/usage.md > "Document searches"
+
+export enum AdvancedSearchField {
+  Any = '',
+  Title = 'title',
+  Content = 'content',
+  OriginalFilename = 'original_filename',
+  NoteText = 'notes.note',
+  NoteAuthor = 'notes.user',
+  CustomFieldName = 'custom_fields.name',
+  CustomFieldValue = 'custom_fields.value',
+  Correspondent = 'correspondent',
+  DocumentType = 'document_type',
+  StoragePath = 'storage_path',
+  Tag = 'tag',
+  ASN = 'asn',
+  PageCount = 'page_count',
+  NumNotes = 'num_notes',
+  Created = 'created',
+  Added = 'added',
+  Modified = 'modified',
+  Checksum = 'checksum',
+}
+
+export enum AdvancedSearchFieldKind {
+  Text = 'text',
+  Number = 'number',
+  Date = 'date',
+  Checksum = 'checksum',
+}
+
+export const ADVANCED_SEARCH_FIELD_KINDS: Record<
+  AdvancedSearchField,
+  AdvancedSearchFieldKind
+> = {
+  [AdvancedSearchField.Any]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.Title]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.Content]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.OriginalFilename]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.NoteText]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.NoteAuthor]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.CustomFieldName]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.CustomFieldValue]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.Correspondent]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.DocumentType]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.StoragePath]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.Tag]: AdvancedSearchFieldKind.Text,
+  [AdvancedSearchField.ASN]: AdvancedSearchFieldKind.Number,
+  [AdvancedSearchField.PageCount]: AdvancedSearchFieldKind.Number,
+  [AdvancedSearchField.NumNotes]: AdvancedSearchFieldKind.Number,
+  [AdvancedSearchField.Created]: AdvancedSearchFieldKind.Date,
+  [AdvancedSearchField.Added]: AdvancedSearchFieldKind.Date,
+  [AdvancedSearchField.Modified]: AdvancedSearchFieldKind.Date,
+  [AdvancedSearchField.Checksum]: AdvancedSearchFieldKind.Checksum,
+}
+
+export enum AdvancedSearchOperator {
+  AllWords = 'all',
+  AnyWord = 'any',
+  Phrase = 'phrase',
+  StartsWith = 'prefix',
+  Equals = 'eq',
+  AtLeast = 'gte',
+  AtMost = 'lte',
+  Between = 'between',
+  DateKeyword = 'keyword',
+  WithinLast = 'within',
+}
+
+export const ADVANCED_SEARCH_OPERATORS_BY_KIND: Record<
+  AdvancedSearchFieldKind,
+  AdvancedSearchOperator[]
+> = {
+  [AdvancedSearchFieldKind.Text]: [
+    AdvancedSearchOperator.AllWords,
+    AdvancedSearchOperator.AnyWord,
+    AdvancedSearchOperator.Phrase,
+    AdvancedSearchOperator.StartsWith,
+  ],
+  [AdvancedSearchFieldKind.Number]: [
+    AdvancedSearchOperator.Equals,
+    AdvancedSearchOperator.AtLeast,
+    AdvancedSearchOperator.AtMost,
+    AdvancedSearchOperator.Between,
+  ],
+  [AdvancedSearchFieldKind.Date]: [
+    AdvancedSearchOperator.DateKeyword,
+    AdvancedSearchOperator.WithinLast,
+    AdvancedSearchOperator.AtLeast,
+    AdvancedSearchOperator.AtMost,
+    AdvancedSearchOperator.Between,
+  ],
+  [AdvancedSearchFieldKind.Checksum]: [AdvancedSearchOperator.StartsWith],
+}
+
+export const ADVANCED_SEARCH_DATE_KEYWORDS = [
+  'today',
+  'yesterday',
+  'tomorrow',
+  'previous week',
+  'this month',
+  'previous month',
+  'previous quarter',
+  'this year',
+  'previous year',
+] as const
+
+export type AdvancedSearchDateKeyword =
+  (typeof ADVANCED_SEARCH_DATE_KEYWORDS)[number]
+
+export enum AdvancedSearchDateUnit {
+  Day = 'day',
+  Week = 'week',
+  Month = 'month',
+  Year = 'year',
+}
+
+export enum AdvancedSearchLogicalOperator {
+  And = 'AND',
+  Or = 'OR',
+  Not = 'NOT',
+}
+
+export enum AdvancedSearchQueryElementType {
+  Atom = 'atom',
+  Group = 'group',
+}
+
+export interface AdvancedSearchQueryAtom {
+  type: AdvancedSearchQueryElementType.Atom
+  field: AdvancedSearchField
+  operator: AdvancedSearchOperator
+  value?: string
+  valueTo?: string
+  unit?: AdvancedSearchDateUnit // for WithinLast
+}
+
+export interface AdvancedSearchQueryGroup {
+  type: AdvancedSearchQueryElementType.Group
+  operator: AdvancedSearchLogicalOperator
+  children: AdvancedSearchQueryElement[]
+}
+
+export type AdvancedSearchQueryElement =
+  AdvancedSearchQueryAtom | AdvancedSearchQueryGroup
+
+export const ADVANCED_SEARCH_MAX_DEPTH = 2
+export const ADVANCED_SEARCH_MAX_ATOMS = 10
+
+export const ADVANCED_SEARCH_FIELD_LABELS: Record<AdvancedSearchField, string> =
+  {
+    [AdvancedSearchField.Any]: $localize`Any field`,
+    [AdvancedSearchField.Title]: $localize`Title`,
+    [AdvancedSearchField.Content]: $localize`Content`,
+    [AdvancedSearchField.OriginalFilename]: $localize`File name`,
+    [AdvancedSearchField.NoteText]: $localize`Note text`,
+    [AdvancedSearchField.NoteAuthor]: $localize`Note author`,
+    [AdvancedSearchField.CustomFieldName]: $localize`Custom field name`,
+    [AdvancedSearchField.CustomFieldValue]: $localize`Custom field value`,
+    [AdvancedSearchField.Correspondent]: $localize`Correspondent name`,
+    [AdvancedSearchField.DocumentType]: $localize`Document type name`,
+    [AdvancedSearchField.StoragePath]: $localize`Storage path name`,
+    [AdvancedSearchField.Tag]: $localize`Tag name`,
+    [AdvancedSearchField.ASN]: $localize`ASN`,
+    [AdvancedSearchField.PageCount]: $localize`Pages`,
+    [AdvancedSearchField.NumNotes]: $localize`Number of notes`,
+    [AdvancedSearchField.Created]: $localize`Created`,
+    [AdvancedSearchField.Added]: $localize`Added`,
+    [AdvancedSearchField.Modified]: $localize`Modified`,
+    [AdvancedSearchField.Checksum]: $localize`Checksum`,
+  }
+
+export const ADVANCED_SEARCH_FIELD_GROUPS: {
+  label: string
+  fields: AdvancedSearchField[]
+}[] = [
+  {
+    label: $localize`Text`,
+    fields: [
+      AdvancedSearchField.Any,
+      AdvancedSearchField.Title,
+      AdvancedSearchField.Content,
+      AdvancedSearchField.OriginalFilename,
+      AdvancedSearchField.NoteText,
+      AdvancedSearchField.NoteAuthor,
+      AdvancedSearchField.CustomFieldName,
+      AdvancedSearchField.CustomFieldValue,
+    ],
+  },
+  {
+    label: $localize`Names`,
+    fields: [
+      AdvancedSearchField.Correspondent,
+      AdvancedSearchField.DocumentType,
+      AdvancedSearchField.StoragePath,
+      AdvancedSearchField.Tag,
+    ],
+  },
+  {
+    label: $localize`Numbers`,
+    fields: [
+      AdvancedSearchField.ASN,
+      AdvancedSearchField.PageCount,
+      AdvancedSearchField.NumNotes,
+    ],
+  },
+  {
+    label: $localize`Dates`,
+    fields: [
+      AdvancedSearchField.Created,
+      AdvancedSearchField.Added,
+      AdvancedSearchField.Modified,
+    ],
+  },
+  { label: $localize`Other`, fields: [AdvancedSearchField.Checksum] },
+]
+
+export const ADVANCED_SEARCH_OPERATOR_LABELS: Record<
+  AdvancedSearchOperator,
+  string
+> = {
+  [AdvancedSearchOperator.AllWords]: $localize`contains all words`,
+  [AdvancedSearchOperator.AnyWord]: $localize`contains any word`,
+  [AdvancedSearchOperator.Phrase]: $localize`contains the phrase`,
+  [AdvancedSearchOperator.StartsWith]: $localize`starts with`,
+  [AdvancedSearchOperator.Equals]: $localize`is`,
+  [AdvancedSearchOperator.AtLeast]: $localize`is at least`,
+  [AdvancedSearchOperator.AtMost]: $localize`is at most`,
+  [AdvancedSearchOperator.Between]: $localize`is between`,
+  [AdvancedSearchOperator.DateKeyword]: $localize`is`,
+  [AdvancedSearchOperator.WithinLast]: $localize`is within the last`,
+}
+
+// Comparing dates reads differently than comparing counts
+export const ADVANCED_SEARCH_DATE_OPERATOR_LABELS: Partial<
+  Record<AdvancedSearchOperator, string>
+> = {
+  [AdvancedSearchOperator.AtLeast]: $localize`is on or after`,
+  [AdvancedSearchOperator.AtMost]: $localize`is on or before`,
+}
+
+export const ADVANCED_SEARCH_DATE_KEYWORD_LABELS: Record<string, string> = {
+  today: $localize`today`,
+  yesterday: $localize`yesterday`,
+  tomorrow: $localize`tomorrow`,
+  'previous week': $localize`previous week`,
+  'this month': $localize`this month`,
+  'previous month': $localize`previous month`,
+  'previous quarter': $localize`previous quarter`,
+  'this year': $localize`this year`,
+  'previous year': $localize`previous year`,
+}
+
+export const ADVANCED_SEARCH_DATE_UNIT_LABELS: Record<
+  AdvancedSearchDateUnit,
+  string
+> = {
+  [AdvancedSearchDateUnit.Day]: $localize`days`,
+  [AdvancedSearchDateUnit.Week]: $localize`weeks`,
+  [AdvancedSearchDateUnit.Month]: $localize`months`,
+  [AdvancedSearchDateUnit.Year]: $localize`years`,
+}

src-ui/src/app/data/document-suggestions.ts

@@ -15,33 +15,3 @@ export interface DocumentSuggestions {
 
   dates?: string[] // ISO-formatted date string e.g. 2022-11-03
 }
-
-const union = <T>(a: T[] = [], b: T[] = []): T[] => [...new Set([...a, ...b])]
-
-export function mergeSuggestions(
-  a: DocumentSuggestions,
-  b: DocumentSuggestions
-): DocumentSuggestions {
-  if (!a) return b
-  return {
-    title: a.title || b.title,
-    tags: union(a.tags, b.tags),
-    suggested_tags: union(a.suggested_tags, b.suggested_tags),
-    correspondents: union(a.correspondents, b.correspondents),
-    suggested_correspondents: union(
-      a.suggested_correspondents,
-      b.suggested_correspondents
-    ),
-    document_types: union(a.document_types, b.document_types),
-    suggested_document_types: union(
-      a.suggested_document_types,
-      b.suggested_document_types
-    ),
-    storage_paths: union(a.storage_paths, b.storage_paths),
-    suggested_storage_paths: union(
-      a.suggested_storage_paths,
-      b.suggested_storage_paths
-    ),
-    dates: union(a.dates, b.dates),
-  }
-}

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

@@ -353,14 +353,6 @@ export const PaperlessConfigOptions: ConfigOption[] = [
     config_key: 'PAPERLESS_AI_LLM_EMBEDDING_MODEL',
     category: ConfigCategory.AI,
   },
-  {
-    key: 'llm_embedding_api_key',
-    title: $localize`LLM Embedding API Key`,
-    type: ConfigOptionType.Password,
-    note: $localize`Used for embeddings when set, otherwise LLM API key is used.`,
-    config_key: 'PAPERLESS_AI_LLM_EMBEDDING_API_KEY',
-    category: ConfigCategory.AI,
-  },
   {
     key: 'llm_embedding_endpoint',
     title: $localize`LLM Embedding Endpoint`,
@@ -465,7 +457,6 @@ export interface PaperlessConfig extends ObjectWithId {
   ai_enabled: boolean
   llm_embedding_backend: string
   llm_embedding_model: string
-  llm_embedding_api_key: string
   llm_embedding_endpoint: string
   llm_embedding_chunk_size: number
   llm_context_size: number

src-ui/src/app/data/ui-settings.ts

@@ -20,12 +20,6 @@ export enum GlobalSearchType {
   TITLE_CONTENT = 'title-content',
 }
 
-export enum SuggestionSource {
-  ML = 'ml',
-  AI = 'ai',
-  Both = 'both',
-}
-
 export enum CollapsibleSection {
   ATTRIBUTES = 'attributes',
 }
@@ -104,8 +98,6 @@ export const SETTINGS_KEYS = {
     'general-settings:document-editing:overlay-thumbnail',
   DOCUMENT_EDITING_AUTO_SUGGEST:
     'general-settings:document-editing:auto-suggest',
-  DOCUMENT_EDITING_SUGGESTION_SOURCE:
-    'general-settings:document-editing:suggestion-source',
   DOCUMENT_DETAILS_HIDDEN_FIELDS:
     'general-settings:document-details:hidden-fields',
   SEARCH_DB_ONLY: 'general-settings:search:db-only',
@@ -334,11 +326,6 @@ export const SETTINGS: UiSetting[] = [
     type: 'boolean',
     default: true,
   },
-  {
-    key: SETTINGS_KEYS.DOCUMENT_EDITING_SUGGESTION_SOURCE,
-    type: 'string',
-    default: SuggestionSource.AI,
-  },
   {
     key: SETTINGS_KEYS.DOCUMENT_DETAILS_HIDDEN_FIELDS,
     type: 'array',

src-ui/src/app/utils/advanced-search-query.spec.ts

@@ -0,0 +1,515 @@
+import {
+  AdvancedSearchDateUnit,
+  AdvancedSearchField,
+  AdvancedSearchLogicalOperator,
+  AdvancedSearchOperator,
+  AdvancedSearchQueryAtom,
+  AdvancedSearchQueryElement,
+  AdvancedSearchQueryElementType,
+  AdvancedSearchQueryGroup,
+} from '../data/advanced-search-query'
+import {
+  parseAdvancedSearchQuery,
+  serializeAdvancedSearchQuery,
+} from './advanced-search-query'
+
+const atom = (
+  field: AdvancedSearchField,
+  operator: AdvancedSearchOperator,
+  value?: string,
+  extra: Partial<AdvancedSearchQueryAtom> = {}
+): AdvancedSearchQueryAtom => ({
+  type: AdvancedSearchQueryElementType.Atom,
+  field,
+  operator,
+  value,
+  ...extra,
+})
+
+const group = (
+  operator: AdvancedSearchLogicalOperator,
+  ...children: AdvancedSearchQueryElement[]
+): AdvancedSearchQueryGroup => ({
+  type: AdvancedSearchQueryElementType.Group,
+  operator,
+  children,
+})
+
+const { And, Or, Not } = AdvancedSearchLogicalOperator
+
+describe('serializeAdvancedSearchQuery', () => {
+  describe('text fields', () => {
+    it.each([
+      [AdvancedSearchOperator.AllWords, 'invoice', 'title:invoice'],
+      [
+        AdvancedSearchOperator.AllWords,
+        '  invoice   unpaid ',
+        'title:invoice AND title:unpaid',
+      ],
+      [
+        AdvancedSearchOperator.AnyWord,
+        'invoice unpaid',
+        'title:invoice OR title:unpaid',
+      ],
+      [
+        AdvancedSearchOperator.Phrase,
+        'quick brown fox',
+        'title:"quick brown fox"',
+      ],
+      [AdvancedSearchOperator.Phrase, 'say "hi"', 'title:"say hi"'],
+      [AdvancedSearchOperator.StartsWith, 'invoi', 'title:invoi*'],
+      [AdvancedSearchOperator.StartsWith, 'in*v?oi', 'title:invoi*'],
+    ])('%s %j writes %s', (operator, value, expected) => {
+      expect(
+        serializeAdvancedSearchQuery(
+          atom(AdvancedSearchField.Title, operator, value)
+        )
+      ).toBe(expected)
+    })
+
+    it('writes bare words for the Any field', () => {
+      expect(
+        serializeAdvancedSearchQuery(
+          atom(AdvancedSearchField.Any, AdvancedSearchOperator.AllWords, 'a b')
+        )
+      ).toBe('a AND b')
+    })
+
+    it.each([
+      ['A-1312/99', 'custom_fields.value:A-1312/99'],
+      ["O'Brien", "custom_fields.value:O'Brien"],
+      ["'quoted'", `custom_fields.value:"'quoted'"`],
+      ['foo:bar', 'custom_fields.value:"foo:bar"'],
+      ['(x)', 'custom_fields.value:"(x)"'],
+      ['2024*', 'custom_fields.value:"2024*"'],
+      ['a,b', 'custom_fields.value:"a,b"'],
+      ['OR', 'custom_fields.value:"OR"'],
+      ['or', 'custom_fields.value:or'],
+    ])(
+      'quotes %j only when the grammar would read it as syntax',
+      (value, expected) => {
+        expect(
+          serializeAdvancedSearchQuery(
+            atom(
+              AdvancedSearchField.CustomFieldValue,
+              AdvancedSearchOperator.AllWords,
+              value
+            )
+          )
+        ).toBe(expected)
+      }
+    )
+
+    it('uses the dotted names for custom fields', () => {
+      expect(
+        serializeAdvancedSearchQuery(
+          group(
+            And,
+            atom(
+              AdvancedSearchField.CustomFieldName,
+              AdvancedSearchOperator.Phrase,
+              'status'
+            ),
+            atom(
+              AdvancedSearchField.CustomFieldValue,
+              AdvancedSearchOperator.AllWords,
+              'paid'
+            )
+          )
+        )
+      ).toBe('custom_fields.name:"status" AND custom_fields.value:paid')
+    })
+
+    it('uses the dotted names for notes', () => {
+      expect(
+        serializeAdvancedSearchQuery(
+          group(
+            And,
+            atom(
+              AdvancedSearchField.NoteText,
+              AdvancedSearchOperator.AllWords,
+              'call'
+            ),
+            atom(
+              AdvancedSearchField.NoteAuthor,
+              AdvancedSearchOperator.AllWords,
+              'alice'
+            )
+          )
+        )
+      ).toBe('notes.note:call AND notes.user:alice')
+    })
+
+    it.each([
+      [AdvancedSearchOperator.AllWords, ''],
+      [AdvancedSearchOperator.AllWords, '  '],
+      [AdvancedSearchOperator.AllWords, '!! --'],
+      [AdvancedSearchOperator.Phrase, '""'],
+      [AdvancedSearchOperator.StartsWith, 'two words'],
+      [AdvancedSearchOperator.StartsWith, '***'],
+      [AdvancedSearchOperator.StartsWith, undefined],
+    ])('leaves out %s %j', (operator, value) => {
+      expect(
+        serializeAdvancedSearchQuery(
+          atom(AdvancedSearchField.Title, operator, value)
+        )
+      ).toBe('')
+    })
+  })
+
+  describe('checksum', () => {
+    it('lowercases the prefix', () => {
+      expect(
+        serializeAdvancedSearchQuery(
+          atom(
+            AdvancedSearchField.Checksum,
+            AdvancedSearchOperator.StartsWith,
+            '9F86D081'
+          )
+        )
+      ).toBe('checksum:9f86d081*')
+    })
+  })
+
+  describe('number fields', () => {
+    it.each([
+      [AdvancedSearchOperator.Equals, '42', undefined, 'asn:42'],
+      [AdvancedSearchOperator.AtLeast, '50', undefined, 'asn:[50 to]'],
+      [AdvancedSearchOperator.AtMost, '50', undefined, 'asn:[to 50]'],
+      [AdvancedSearchOperator.Between, '50', '150', 'asn:[50 to 150]'],
+      [AdvancedSearchOperator.Equals, '4.2', undefined, ''],
+      [AdvancedSearchOperator.Equals, '-1', undefined, ''],
+      [AdvancedSearchOperator.Equals, '2024-01-01', undefined, ''],
+      [AdvancedSearchOperator.Between, '50', '', ''],
+    ])('%s %j %j writes %j', (operator, value, valueTo, expected) => {
+      expect(
+        serializeAdvancedSearchQuery(
+          atom(AdvancedSearchField.ASN, operator, value, { valueTo })
+        )
+      ).toBe(expected)
+    })
+  })
+
+  describe('date fields', () => {
+    it.each([
+      ['today', 'added:today'],
+      ['previous month', 'added:"previous month"'],
+      ['last tuesday', ''],
+      ['', ''],
+    ])('keyword %j writes %j', (value, expected) => {
+      expect(
+        serializeAdvancedSearchQuery(
+          atom(
+            AdvancedSearchField.Added,
+            AdvancedSearchOperator.DateKeyword,
+            value
+          )
+        )
+      ).toBe(expected)
+    })
+
+    it.each([
+      ['1', AdvancedSearchDateUnit.Day, 'added:[-1 day to now]'],
+      ['3', AdvancedSearchDateUnit.Month, 'added:[-3 months to now]'],
+      ['2', AdvancedSearchDateUnit.Week, 'added:[-2 weeks to now]'],
+      ['0', AdvancedSearchDateUnit.Year, ''],
+      ['1.5', AdvancedSearchDateUnit.Year, ''],
+      ['3', undefined, ''],
+      ['3', 'fortnight' as AdvancedSearchDateUnit, ''],
+    ])('within the last %j %j writes %j', (value, unit, expected) => {
+      expect(
+        serializeAdvancedSearchQuery(
+          atom(
+            AdvancedSearchField.Added,
+            AdvancedSearchOperator.WithinLast,
+            value,
+            {
+              unit,
+            }
+          )
+        )
+      ).toBe(expected)
+    })
+
+    it.each([
+      [
+        AdvancedSearchOperator.AtLeast,
+        '2024-01-01',
+        undefined,
+        'created:[2024-01-01 to]',
+      ],
+      [
+        AdvancedSearchOperator.AtMost,
+        '2024-01-01',
+        undefined,
+        'created:[to 2024-01-01]',
+      ],
+      [
+        AdvancedSearchOperator.Between,
+        '2024-01-01',
+        '2024-03-31',
+        'created:[2024-01-01 to 2024-03-31]',
+      ],
+      [AdvancedSearchOperator.AtLeast, '2024', undefined, ''],
+      [AdvancedSearchOperator.Between, '2024-01-01', 'now', ''],
+    ])('%s %j %j writes %j', (operator, value, valueTo, expected) => {
+      expect(
+        serializeAdvancedSearchQuery(
+          atom(AdvancedSearchField.Created, operator, value, { valueTo })
+        )
+      ).toBe(expected)
+    })
+  })
+
+  describe('groups', () => {
+    const invoice = atom(
+      AdvancedSearchField.Content,
+      AdvancedSearchOperator.AllWords,
+      'invoice'
+    )
+    const letter = atom(
+      AdvancedSearchField.Title,
+      AdvancedSearchOperator.AllWords,
+      'letter'
+    )
+    const paid = atom(
+      AdvancedSearchField.Tag,
+      AdvancedSearchOperator.AllWords,
+      'paid'
+    )
+    const twoWords = atom(
+      AdvancedSearchField.Title,
+      AdvancedSearchOperator.AllWords,
+      'a b'
+    )
+    const anyWords = atom(
+      AdvancedSearchField.Title,
+      AdvancedSearchOperator.AnyWord,
+      'a b'
+    )
+    const empty = atom(
+      AdvancedSearchField.Title,
+      AdvancedSearchOperator.AllWords,
+      ''
+    )
+
+    it.each([
+      ['an empty group', group(And), ''],
+      ['a group of empty atoms', group(Or, empty, group(And, empty)), ''],
+      [
+        'a single child without parentheses',
+        group(Or, invoice),
+        'content:invoice',
+      ],
+      ['All', group(And, invoice, letter), 'content:invoice AND title:letter'],
+      ['Any', group(Or, invoice, letter), 'content:invoice OR title:letter'],
+      [
+        'skipped empty atoms',
+        group(And, empty, invoice, empty),
+        'content:invoice',
+      ],
+      ['Not with one child', group(Not, paid), 'NOT tag:paid'],
+      [
+        'Not as none of',
+        group(Not, paid, letter),
+        'NOT (tag:paid OR title:letter)',
+      ],
+      [
+        'Not with a compound child',
+        group(Not, twoWords),
+        'NOT (title:a AND title:b)',
+      ],
+      [
+        'Not inside All',
+        group(And, invoice, group(Not, paid)),
+        'content:invoice AND NOT tag:paid',
+      ],
+      [
+        'Not inside Any',
+        group(Or, invoice, group(Not, paid)),
+        'content:invoice OR NOT tag:paid',
+      ],
+      [
+        'Any inside All',
+        group(And, invoice, group(Or, letter, paid)),
+        'content:invoice AND (title:letter OR tag:paid)',
+      ],
+      [
+        'All inside Any',
+        group(Or, invoice, group(And, letter, paid)),
+        'content:invoice OR (title:letter AND tag:paid)',
+      ],
+      [
+        'All inside All flattened',
+        group(And, invoice, group(And, letter, paid)),
+        'content:invoice AND title:letter AND tag:paid',
+      ],
+      [
+        'an all-words atom inside Any',
+        group(Or, invoice, twoWords),
+        'content:invoice OR (title:a AND title:b)',
+      ],
+      [
+        'an any-word atom inside All',
+        group(And, invoice, anyWords),
+        'content:invoice AND (title:a OR title:b)',
+      ],
+      [
+        'an all-words atom inside Not with siblings',
+        group(Not, paid, twoWords),
+        'NOT (tag:paid OR (title:a AND title:b))',
+      ],
+    ])('writes %s', (_, tree, expected) => {
+      expect(serializeAdvancedSearchQuery(tree)).toBe(expected)
+    })
+
+    it('writes the mockup example', () => {
+      expect(
+        serializeAdvancedSearchQuery(
+          group(
+            And,
+            invoice,
+            group(
+              Or,
+              atom(
+                AdvancedSearchField.Correspondent,
+                AdvancedSearchOperator.Phrase,
+                'acme corp'
+              ),
+              atom(
+                AdvancedSearchField.Content,
+                AdvancedSearchOperator.Phrase,
+                'acme corporation'
+              )
+            ),
+            atom(
+              AdvancedSearchField.Added,
+              AdvancedSearchOperator.WithinLast,
+              '3',
+              {
+                unit: AdvancedSearchDateUnit.Month,
+              }
+            ),
+            group(Not, paid)
+          )
+        )
+      ).toBe(
+        'content:invoice AND (correspondent:"acme corp" OR content:"acme corporation") AND added:[-3 months to now] AND NOT tag:paid'
+      )
+    })
+  })
+})
+
+describe('parseAdvancedSearchQuery', () => {
+  const canonical = [
+    'title:invoice',
+    'title:invoice AND title:unpaid',
+    'content:invoice OR content:receipt',
+    'a AND b',
+    'title:"quick brown fox"',
+    'title:invoi*',
+    "custom_fields.value:O'Brien",
+    'custom_fields.value:"foo:bar"',
+    'custom_fields.name:"status" AND custom_fields.value:paid',
+    'notes.note:call AND notes.user:alice',
+    'checksum:9f86d081*',
+    'asn:42',
+    'asn:[50 to 150]',
+    'asn:[50 to]',
+    'asn:[to 50]',
+    'page_count:[10 to]',
+    'added:today',
+    'added:"previous month"',
+    'added:[-1 day to now]',
+    'added:[-3 months to now]',
+    'created:[2024-01-01 to 2024-03-31]',
+    'created:[2024-01-01 to]',
+    'created:[to 2024-01-01]',
+    'NOT tag:paid',
+    'NOT (tag:paid OR title:letter)',
+    'NOT (title:a AND title:b)',
+    'content:invoice OR NOT tag:paid',
+    'content:invoice AND (title:letter OR tag:paid)',
+    'content:invoice OR (title:letter AND tag:paid)',
+    'content:invoice AND (correspondent:"acme corp" OR content:"acme corporation") AND added:[-3 months to now] AND NOT tag:paid',
+  ]
+
+  it.each(canonical)('reads back %s unchanged', (query) => {
+    const tree = parseAdvancedSearchQuery(query)
+    expect(tree).not.toBeNull()
+    expect(serializeAdvancedSearchQuery(tree)).toBe(query)
+  })
+
+  it('reads surrounding whitespace', () => {
+    expect(
+      serializeAdvancedSearchQuery(
+        parseAdvancedSearchQuery('  title:invoice  ')
+      )
+    ).toBe('title:invoice')
+  })
+
+  it('puts the words of one condition back together', () => {
+    expect(parseAdvancedSearchQuery('title:invoice AND title:unpaid')).toEqual(
+      group(
+        And,
+        atom(
+          AdvancedSearchField.Title,
+          AdvancedSearchOperator.AllWords,
+          'invoice unpaid'
+        )
+      )
+    )
+  })
+
+  it('keeps words of different fields apart', () => {
+    expect(parseAdvancedSearchQuery('title:a OR content:b')).toEqual(
+      group(
+        Or,
+        atom(AdvancedSearchField.Title, AdvancedSearchOperator.AllWords, 'a'),
+        atom(AdvancedSearchField.Content, AdvancedSearchOperator.AllWords, 'b')
+      )
+    )
+  })
+
+  it('reads a range as its condition', () => {
+    expect(parseAdvancedSearchQuery('added:[-3 months to now]')).toEqual(
+      group(
+        And,
+        atom(
+          AdvancedSearchField.Added,
+          AdvancedSearchOperator.WithinLast,
+          '3',
+          {
+            unit: AdvancedSearchDateUnit.Month,
+          }
+        )
+      )
+    )
+  })
+
+  it.each([
+    ['', 'nothing'],
+    ['   ', 'whitespace'],
+    ['type:invoice', 'a field alias'],
+    ['notes:call', 'a bare notes prefix'],
+    ['unknown:x', 'an unknown field'],
+    ['title:a AND content:b OR title:c', 'AND and OR mixed at one level'],
+    ['title:a b', 'an implicit AND'],
+    ['title:"unterminated', 'an unterminated phrase'],
+    ['asn:[50 to', 'an unterminated range'],
+    ['title:a*b', 'a wildcard in the middle'],
+    ['title:invoice^2', 'a boost'],
+    ['asn:[50 to abc]', 'a non-numeric bound'],
+    ['asn:invoice', 'a word on a number field'],
+    ['created:[2024 to 2025]', 'a year-only date range'],
+    ['added:"last tuesday"', 'a date keyword paperless does not have'],
+    ['title:[a to b]', 'a range on a text field'],
+    ['checksum:9f86d081', 'a checksum without a wildcard'],
+    ['(title:invoice)', 'parentheses the editor would not write'],
+    ['title:invoice AND', 'a trailing operator'],
+    ['title:"a"b', 'a value running into the next'],
+    ['ADDED:today', 'an uppercase field name'],
+  ])('leaves %s alone, having %s', (query) => {
+    expect(parseAdvancedSearchQuery(query)).toBeNull()
+  })
+})

src-ui/src/app/utils/advanced-search-query.ts

@@ -0,0 +1,500 @@
+import {
+  ADVANCED_SEARCH_DATE_KEYWORDS,
+  ADVANCED_SEARCH_FIELD_KINDS,
+  AdvancedSearchDateUnit,
+  AdvancedSearchField,
+  AdvancedSearchFieldKind,
+  AdvancedSearchLogicalOperator,
+  AdvancedSearchOperator,
+  AdvancedSearchQueryAtom,
+  AdvancedSearchQueryElement,
+  AdvancedSearchQueryElementType,
+  AdvancedSearchQueryGroup,
+} from '../data/advanced-search-query'
+
+// Anything the query grammar would read as syntax rather than as a word
+const SYNTAX_CHARS = /[\s():"[\]*?,{}^~\\]/
+const SYNTAX_CHARS_GLOBAL = new RegExp(SYNTAX_CHARS, 'g')
+// A word wrapped in single quotes is also syntax, an apostrophe inside one is not
+const EDGE_SINGLE_QUOTE = /^'|'$/
+const RESERVED_WORDS = /^(AND|OR|NOT|TO)$/
+const HAS_WORD_CHAR = /[\p{L}\p{N}]/u
+const WHOLE_NUMBER = /^\d+$/
+const ISO_DATE = /^\d{4}-\d{2}-\d{2}$/
+
+interface Serialized {
+  text: string
+  // The operator joining the top level of `text`, null when it is self-delimiting
+  join:
+    AdvancedSearchLogicalOperator.And | AdvancedSearchLogicalOperator.Or | null
+}
+
+const prefix = (field: AdvancedSearchField) => (field ? `${field}:` : '')
+
+const quote = (text: string) => `"${text.replace(/"/g, '')}"`
+
+const words = (value: string) =>
+  value
+    .trim()
+    .split(/\s+/)
+    .filter((word) => HAS_WORD_CHAR.test(word))
+
+const word = (w: string) =>
+  SYNTAX_CHARS.test(w) || EDGE_SINGLE_QUOTE.test(w) || RESERVED_WORDS.test(w)
+    ? quote(w)
+    : w
+
+const atomic = (text: string): Serialized => ({ text, join: null })
+
+function serializeWords(
+  atom: AdvancedSearchQueryAtom,
+  join: AdvancedSearchLogicalOperator.And | AdvancedSearchLogicalOperator.Or
+): Serialized {
+  // A field applies only to the word right after it, so repeat it per word
+  const terms = words(atom.value ?? '').map(
+    (w) => `${prefix(atom.field)}${word(w)}`
+  )
+  if (terms.length === 0) return null
+  if (terms.length === 1) return atomic(terms[0])
+  return { text: terms.join(` ${join} `), join }
+}
+
+function serializeRange(
+  atom: AdvancedSearchQueryAtom,
+  isValid: (v: string) => boolean
+): Serialized {
+  const lo = atom.value?.trim() ?? ''
+  const hi = atom.valueTo?.trim() ?? ''
+  const field = prefix(atom.field)
+  switch (atom.operator) {
+    case AdvancedSearchOperator.Equals:
+      return isValid(lo) ? atomic(`${field}${lo}`) : null
+    case AdvancedSearchOperator.AtLeast:
+      return isValid(lo) ? atomic(`${field}[${lo} to]`) : null
+    case AdvancedSearchOperator.AtMost:
+      return isValid(lo) ? atomic(`${field}[to ${lo}]`) : null
+    case AdvancedSearchOperator.Between:
+      return isValid(lo) && isValid(hi)
+        ? atomic(`${field}[${lo} to ${hi}]`)
+        : null
+  }
+  return null
+}
+
+function serializeAtom(atom: AdvancedSearchQueryAtom): Serialized {
+  const field = prefix(atom.field)
+  const value = atom.value?.trim() ?? ''
+
+  switch (atom.operator) {
+    case AdvancedSearchOperator.AllWords:
+      return serializeWords(atom, AdvancedSearchLogicalOperator.And)
+    case AdvancedSearchOperator.AnyWord:
+      return serializeWords(atom, AdvancedSearchLogicalOperator.Or)
+    case AdvancedSearchOperator.Phrase:
+      return HAS_WORD_CHAR.test(value)
+        ? atomic(`${field}${quote(value)}`)
+        : null
+    case AdvancedSearchOperator.StartsWith: {
+      if (/\s/.test(value)) return null
+      let stem = value.replace(SYNTAX_CHARS_GLOBAL, '')
+      if (!HAS_WORD_CHAR.test(stem)) return null
+      // checksum is indexed as-is, in lowercase
+      if (atom.field === AdvancedSearchField.Checksum) {
+        stem = stem.toLowerCase()
+      }
+      return atomic(`${field}${stem}*`)
+    }
+    case AdvancedSearchOperator.DateKeyword:
+      return (ADVANCED_SEARCH_DATE_KEYWORDS as readonly string[]).includes(
+        value
+      )
+        ? atomic(`${field}${word(value)}`)
+        : null
+    case AdvancedSearchOperator.WithinLast: {
+      const amount = parseInt(value, 10)
+      const units = Object.values(AdvancedSearchDateUnit) as string[]
+      if (!WHOLE_NUMBER.test(value) || amount < 1) return null
+      if (!units.includes(atom.unit)) return null
+      const unit = amount === 1 ? atom.unit : `${atom.unit}s`
+      return atomic(`${field}[-${amount} ${unit} to now]`)
+    }
+    case AdvancedSearchOperator.Equals:
+    case AdvancedSearchOperator.AtLeast:
+    case AdvancedSearchOperator.AtMost:
+    case AdvancedSearchOperator.Between:
+      return serializeRange(
+        atom,
+        ADVANCED_SEARCH_FIELD_KINDS[atom.field] === AdvancedSearchFieldKind.Date
+          ? (v) => ISO_DATE.test(v)
+          : (v) => WHOLE_NUMBER.test(v)
+      )
+  }
+  return null
+}
+
+function wrap(
+  child: Serialized,
+  parentJoin: AdvancedSearchLogicalOperator
+): string {
+  return child.join === null || child.join === parentJoin
+    ? child.text
+    : `(${child.text})`
+}
+
+function serializeGroup(group: AdvancedSearchQueryGroup): Serialized {
+  const children = group.children.map(serializeElement).filter(Boolean)
+  if (children.length === 0) return null
+
+  if (group.operator === AdvancedSearchLogicalOperator.Not) {
+    // A Not group matches documents matching none of its children
+    if (children.length === 1 && children[0].join === null) {
+      return atomic(`NOT ${children[0].text}`)
+    }
+    const inner =
+      children.length === 1
+        ? children[0].text
+        : children
+            .map((c) => wrap(c, AdvancedSearchLogicalOperator.Or))
+            .join(' OR ')
+    return atomic(`NOT (${inner})`)
+  }
+
+  if (children.length === 1) return children[0]
+  return {
+    text: children
+      .map((c) => wrap(c, group.operator))
+      .join(` ${group.operator} `),
+    join: group.operator,
+  }
+}
+
+function serializeElement(element: AdvancedSearchQueryElement): Serialized {
+  return element.type === AdvancedSearchQueryElementType.Group
+    ? serializeGroup(element)
+    : serializeAtom(element)
+}
+
+/**
+ * Writes an editor tree as a full-text query. Atoms that are not filled in
+ * (or not valid) are left out, and empty groups with them.
+ */
+export function serializeAdvancedSearchQuery(
+  element: AdvancedSearchQueryElement
+): string {
+  return serializeElement(element)?.text ?? ''
+}
+
+// --- Reading a query back into the editor --------------------------------
+//
+// Deliberately narrow: this reads the forms serializeAdvancedSearchQuery
+// writes, and nothing else. A query it cannot read is not a failure, it just
+// stays text, so there is never a lossy or surprising conversion. The final
+// round-trip check below is what holds that promise: a tree is only returned
+// when writing it out again reproduces the query exactly.
+
+const RELATIVE_BOUND = /^-(\d+) (day|week|month|year)s?$/
+const FIELD_PREFIX = /^([a-z_]+(?:\.[a-z_]+)?):/
+const KEYWORD_TOKEN = /^(AND|OR|NOT)(?=[\s(]|$)/
+// Either bound may be missing: [50 to 150], [50 to], [to 50]
+const RANGE_BOUNDS = /^(?:(.+?) )?to(?: (.+))?$/
+const TRAILING_WILDCARD = /^([^*?]+)\*$/
+
+class UnreadableQuery extends Error {}
+
+interface Token {
+  type: 'term' | 'AND' | 'OR' | 'NOT' | '(' | ')'
+  field?: string
+  value?: string
+  quoted?: boolean
+  range?: boolean
+}
+
+// A parsed element, plus what it takes to merge the per-word terms the
+// serializer writes for "contains all words" back into a single condition
+interface Parsed {
+  element: AdvancedSearchQueryElement
+  word?: { field: AdvancedSearchField; text: string }
+}
+
+function tokenize(query: string): Token[] {
+  const tokens: Token[] = []
+  let i = 0
+  while (i < query.length) {
+    const rest = query.slice(i)
+    if (/^\s/.test(rest)) {
+      i++
+      continue
+    }
+    if (rest[0] === '(' || rest[0] === ')') {
+      tokens.push({ type: rest[0] as '(' | ')' })
+      i++
+      continue
+    }
+    const keyword = KEYWORD_TOKEN.exec(rest)
+    if (keyword) {
+      tokens.push({ type: keyword[1] as 'AND' | 'OR' | 'NOT' })
+      i += keyword[1].length
+      continue
+    }
+    const fieldMatch = FIELD_PREFIX.exec(rest)
+    const field = fieldMatch ? fieldMatch[1] : ''
+    i += fieldMatch ? fieldMatch[0].length : 0
+
+    const value = query.slice(i)
+    if (value.startsWith('"')) {
+      const end = query.indexOf('"', i + 1)
+      if (end < 0) throw new UnreadableQuery()
+      tokens.push({
+        type: 'term',
+        field,
+        value: query.slice(i + 1, end),
+        quoted: true,
+      })
+      i = end + 1
+    } else if (value.startsWith('[')) {
+      const end = query.indexOf(']', i + 1)
+      if (end < 0) throw new UnreadableQuery()
+      tokens.push({
+        type: 'term',
+        field,
+        value: query.slice(i + 1, end),
+        range: true,
+      })
+      i = end + 1
+    } else {
+      const bare = /^[^\s()]+/.exec(value)
+      if (!bare) throw new UnreadableQuery()
+      tokens.push({ type: 'term', field, value: bare[0] })
+      i += bare[0].length
+    }
+    // Nothing may run on directly after a value, e.g. title:"a"b
+    if (i < query.length && !/[\s)]/.test(query[i])) throw new UnreadableQuery()
+  }
+  return tokens
+}
+
+function resolveField(name: string): AdvancedSearchField {
+  const fields = Object.values(AdvancedSearchField) as string[]
+  // Aliases (type:, path:, notes:) are left to the text box on purpose:
+  // reading one would mean rewriting the user's query as it was read
+  if (!fields.includes(name)) throw new UnreadableQuery()
+  return name as AdvancedSearchField
+}
+
+function atomFrom(
+  field: AdvancedSearchField,
+  operator: AdvancedSearchOperator,
+  value: string,
+  extra: Partial<AdvancedSearchQueryAtom> = {}
+): AdvancedSearchQueryAtom {
+  return {
+    type: AdvancedSearchQueryElementType.Atom,
+    field,
+    operator,
+    value,
+    ...extra,
+  }
+}
+
+function parseRange(
+  field: AdvancedSearchField,
+  kind: AdvancedSearchFieldKind,
+  body: string
+): AdvancedSearchQueryAtom {
+  const bounds = RANGE_BOUNDS.exec(body)
+  if (!bounds) throw new UnreadableQuery()
+  const lo = bounds[1] ?? ''
+  const hi = bounds[2] ?? ''
+
+  if (kind === AdvancedSearchFieldKind.Date) {
+    const relative = RELATIVE_BOUND.exec(lo)
+    if (relative && hi === 'now') {
+      return atomFrom(field, AdvancedSearchOperator.WithinLast, relative[1], {
+        unit: relative[2] as AdvancedSearchDateUnit,
+      })
+    }
+  } else if (kind !== AdvancedSearchFieldKind.Number) {
+    throw new UnreadableQuery()
+  }
+
+  const isValid =
+    kind === AdvancedSearchFieldKind.Date
+      ? (v: string) => ISO_DATE.test(v)
+      : (v: string) => WHOLE_NUMBER.test(v)
+
+  if (lo && hi) {
+    if (!isValid(lo) || !isValid(hi)) throw new UnreadableQuery()
+    return atomFrom(field, AdvancedSearchOperator.Between, lo, { valueTo: hi })
+  }
+  if (lo && isValid(lo))
+    return atomFrom(field, AdvancedSearchOperator.AtLeast, lo)
+  if (hi && isValid(hi))
+    return atomFrom(field, AdvancedSearchOperator.AtMost, hi)
+  throw new UnreadableQuery()
+}
+
+function parseTerm(token: Token): Parsed {
+  const field = resolveField(token.field)
+  const kind = ADVANCED_SEARCH_FIELD_KINDS[field]
+  const value = token.value
+  const isKeyword = (
+    ADVANCED_SEARCH_DATE_KEYWORDS as readonly string[]
+  ).includes(value)
+
+  if (token.range) {
+    return { element: parseRange(field, kind, value) }
+  }
+
+  if (kind === AdvancedSearchFieldKind.Date) {
+    if (!isKeyword) throw new UnreadableQuery()
+    return {
+      element: atomFrom(field, AdvancedSearchOperator.DateKeyword, value),
+    }
+  }
+
+  if (token.quoted) {
+    if (kind !== AdvancedSearchFieldKind.Text) throw new UnreadableQuery()
+    return { element: atomFrom(field, AdvancedSearchOperator.Phrase, value) }
+  }
+
+  const wildcard = TRAILING_WILDCARD.exec(value)
+  if (wildcard) {
+    if (kind === AdvancedSearchFieldKind.Number) throw new UnreadableQuery()
+    return {
+      element: atomFrom(field, AdvancedSearchOperator.StartsWith, wildcard[1]),
+    }
+  }
+
+  if (kind === AdvancedSearchFieldKind.Number) {
+    if (!WHOLE_NUMBER.test(value)) throw new UnreadableQuery()
+    return { element: atomFrom(field, AdvancedSearchOperator.Equals, value) }
+  }
+  // A checksum is only ever searched by its first characters
+  if (kind === AdvancedSearchFieldKind.Checksum) throw new UnreadableQuery()
+
+  return {
+    element: atomFrom(field, AdvancedSearchOperator.AllWords, value),
+    word: { field, text: value },
+  }
+}
+
+// The serializer repeats the field for every word, because a field applies
+// only to the word after it. Put those back together into one condition.
+function mergeWords(
+  parts: Parsed[],
+  operator: AdvancedSearchLogicalOperator.And | AdvancedSearchLogicalOperator.Or
+): AdvancedSearchQueryElement[] {
+  const merged: AdvancedSearchQueryElement[] = []
+  for (let i = 0; i < parts.length; i++) {
+    const run = [parts[i]]
+    while (
+      parts[i].word &&
+      parts[i + 1]?.word &&
+      parts[i + 1].word.field === parts[i].word.field
+    ) {
+      run.push(parts[++i])
+    }
+    if (run.length === 1) {
+      merged.push(run[0].element)
+      continue
+    }
+    merged.push(
+      atomFrom(
+        run[0].word.field,
+        operator === AdvancedSearchLogicalOperator.And
+          ? AdvancedSearchOperator.AllWords
+          : AdvancedSearchOperator.AnyWord,
+        run.map((part) => part.word.text).join(' ')
+      )
+    )
+  }
+  return merged
+}
+
+interface Cursor {
+  tokens: Token[]
+  at: number
+}
+
+function parseExpression(cursor: Cursor): Parsed {
+  const parts: Parsed[] = [parseOperand(cursor)]
+  let operator:
+    AdvancedSearchLogicalOperator.And | AdvancedSearchLogicalOperator.Or
+  while (
+    cursor.tokens[cursor.at]?.type === 'AND' ||
+    cursor.tokens[cursor.at]?.type === 'OR'
+  ) {
+    const next = cursor.tokens[cursor.at++].type as
+      AdvancedSearchLogicalOperator.And | AdvancedSearchLogicalOperator.Or
+    // One level mixing AND and OR is never something the editor wrote
+    if (operator && next !== operator) throw new UnreadableQuery()
+    operator = next
+    parts.push(parseOperand(cursor))
+  }
+  if (parts.length === 1) return parts[0]
+
+  const children = mergeWords(parts, operator)
+  if (children.length === 1) return { element: children[0] }
+  return {
+    element: {
+      type: AdvancedSearchQueryElementType.Group,
+      operator,
+      children,
+    },
+  }
+}
+
+function parseOperand(cursor: Cursor): Parsed {
+  const token = cursor.tokens[cursor.at++]
+  if (!token) throw new UnreadableQuery()
+
+  if (token.type === 'NOT') {
+    const child = parseOperand(cursor)
+    return {
+      element: {
+        type: AdvancedSearchQueryElementType.Group,
+        operator: AdvancedSearchLogicalOperator.Not,
+        children: [child.element],
+      },
+    }
+  }
+  if (token.type === '(') {
+    const inner = parseExpression(cursor)
+    if (cursor.tokens[cursor.at++]?.type !== ')') throw new UnreadableQuery()
+    return { element: inner.element }
+  }
+  if (token.type !== 'term') throw new UnreadableQuery()
+  return parseTerm(token)
+}
+
+/**
+ * Reads a query the editor could have written back into an editor tree, or
+ * returns null when the editor cannot show it, in which case the query stays
+ * text. Never returns a tree that would be written back differently.
+ */
+export function parseAdvancedSearchQuery(
+  query: string
+): AdvancedSearchQueryGroup | null {
+  const trimmed = query?.trim() ?? ''
+  if (!trimmed) return null
+
+  let parsed: Parsed
+  try {
+    const cursor: Cursor = { tokens: tokenize(trimmed), at: 0 }
+    parsed = parseExpression(cursor)
+    if (cursor.at !== cursor.tokens.length) throw new UnreadableQuery()
+  } catch {
+    return null
+  }
+
+  const root =
+    parsed.element.type === AdvancedSearchQueryElementType.Group
+      ? parsed.element
+      : {
+          type: AdvancedSearchQueryElementType.Group as const,
+          operator: AdvancedSearchLogicalOperator.And,
+          children: [parsed.element],
+        }
+
+  return serializeAdvancedSearchQuery(root) === trimmed ? root : null
+}

src-ui/src/environments/environment.prod.ts

@@ -8,7 +8,7 @@ export const environment = {
   apiVersion: '10', // match src/paperless/settings.py
   appTitle: DEFAULT_APP_TITLE,
   tag: 'prod',
-  version: '3.2.1',
+  version: '3.2.0',
   webSocketHost: window.location.host,
   webSocketProtocol: window.location.protocol == 'https:' ? 'wss:' : 'ws:',
   webSocketBaseUrl: base_url.pathname + 'ws/',

src-ui/src/locale/messages.de_DE.xlf

@@ -2855,7 +2855,7 @@
           <context context-type="sourcefile">src/app/components/admin/tasks/tasks.component.ts</context>
           <context context-type="linenumber">104</context>
         </context-group>
-        <target state="translated">KI-Vorschläge anwenden</target>
+        <target state="needs-translation">Apply AI Suggestions</target>
       </trans-unit>
       <trans-unit id="9172233176401579786" datatype="html" approved="yes">
         <source>Scheduled</source>
@@ -3907,7 +3907,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">2,3</context>
         </context-group>
-        <target state="translated">Share Links</target>
+        <target state="needs-translation">Share links</target>
       </trans-unit>
       <trans-unit id="7844706011418789951" datatype="html" approved="yes">
         <source>Administration</source>
@@ -5835,7 +5835,7 @@
           <context context-type="sourcefile">src/app/components/common/edit-dialog/mail-rule-edit-dialog/mail-rule-edit-dialog.component.ts</context>
           <context context-type="linenumber">182</context>
         </context-group>
-        <target state="translated">Fehler beim Abrufen der Korrespondenten</target>
+        <target state="needs-translation">Error retrieving correspondents</target>
       </trans-unit>
       <trans-unit id="3407606224361082860" datatype="html">
         <source>Error retrieving document types</source>
@@ -9623,7 +9623,7 @@
           <context context-type="sourcefile">src/app/components/document-detail/document-detail.component.ts</context>
           <context context-type="linenumber">1051</context>
         </context-group>
-        <target state="translated">Sie haben ungespeicherte Änderungen am Inhalt dieser Version.</target>
+        <target state="needs-translation">You have unsaved changes to the content of this version.</target>
       </trans-unit>
       <trans-unit id="85184271222513014" datatype="html">
         <source>Switching versions will discard them.</source>
@@ -12574,7 +12574,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">4,5</context>
         </context-group>
-        <target state="translated">Verwalten Sie öffentliche Links zu einzelnen Dokumenten und Dokumentenbündeln.</target>
+        <target state="needs-translation">Manage public links to individual documents and document bundles.</target>
       </trans-unit>
       <trans-unit id="3740311348735113099" datatype="html">
         <source>Document links</source>

src-ui/src/locale/messages.it_IT.xlf

@@ -3907,7 +3907,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">2,3</context>
         </context-group>
-        <target state="translated">Condividi link</target>
+        <target state="needs-translation">Share links</target>
       </trans-unit>
       <trans-unit id="7844706011418789951" datatype="html">
         <source>Administration</source>
@@ -12411,7 +12411,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">40,41</context>
         </context-group>
-        <target state="translated">Scaduto</target>
+        <target state="needs-translation">Expired</target>
       </trans-unit>
       <trans-unit id="2057574872309338088" datatype="html">
         <source>Copy share link</source>
@@ -12455,7 +12455,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-bundle-list/share-link-bundle-list.component.html</context>
           <context context-type="linenumber">164,165</context>
         </context-group>
-        <target state="translated">Paginazione dei pacchetti di link di condivisione</target>
+        <target state="needs-translation">Share link bundles pagination</target>
       </trans-unit>
       <trans-unit id="1541995860059283227" datatype="html">
         <source>Failed to load share link bundles.</source>
@@ -12511,7 +12511,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">7,8</context>
         </context-group>
-        <target state="translated"> Attualmente non sono presenti link per la condivisione dei documenti. </target>
+        <target state="needs-translation"> No document share links currently exist. </target>
       </trans-unit>
       <trans-unit id="894965666917873756" datatype="html">
         <source>Document</source>
@@ -12519,7 +12519,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">15,16</context>
         </context-group>
-        <target state="translated">Documento</target>
+        <target state="needs-translation">Document</target>
       </trans-unit>
       <trans-unit id="7420713064779772535" datatype="html">
         <source>Delete share link</source>
@@ -12527,7 +12527,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">75,76</context>
         </context-group>
-        <target state="translated">Elimina link di condivisione</target>
+        <target state="needs-translation">Delete share link</target>
       </trans-unit>
       <trans-unit id="5198271726831144068" datatype="html">
         <source>Share links pagination</source>
@@ -12535,7 +12535,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">104,105</context>
         </context-group>
-        <target state="translated">Paginazione link condivisi</target>
+        <target state="needs-translation">Share links pagination</target>
       </trans-unit>
       <trans-unit id="2976350639944567244" datatype="html">
         <source>Failed to load share links.</source>
@@ -12543,7 +12543,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">105</context>
         </context-group>
-        <target state="translated">Impossibile caricare i link di condivisione.</target>
+        <target state="needs-translation">Failed to load share links.</target>
       </trans-unit>
       <trans-unit id="4100863040928608880" datatype="html">
         <source>Error retrieving share links.</source>
@@ -12551,7 +12551,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">107</context>
         </context-group>
-        <target state="translated">Errore nel recuperare i link di condivisione.</target>
+        <target state="needs-translation">Error retrieving share links.</target>
       </trans-unit>
       <trans-unit id="326120283142363299" datatype="html">
         <source>Share link deleted.</source>
@@ -12559,7 +12559,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">154</context>
         </context-group>
-        <target state="translated">Link di condivisione eliminato.</target>
+        <target state="needs-translation">Share link deleted.</target>
       </trans-unit>
       <trans-unit id="4964931417181298145" datatype="html">
         <source>Error deleting share link.</source>
@@ -12567,7 +12567,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">159</context>
         </context-group>
-        <target state="translated">Errore nell'eliminare il link condivisione.</target>
+        <target state="needs-translation">Error deleting share link.</target>
       </trans-unit>
       <trans-unit id="3439104787814858626" datatype="html">
         <source>Manage public links to individual documents and document bundles.</source>
@@ -12575,7 +12575,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">4,5</context>
         </context-group>
-        <target state="translated">Gestisci i link pubblici ai singoli documenti e pacchetti di documenti.</target>
+        <target state="needs-translation">Manage public links to individual documents and document bundles.</target>
       </trans-unit>
       <trans-unit id="3740311348735113099" datatype="html">
         <source>Document links</source>
@@ -12583,7 +12583,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">18,19</context>
         </context-group>
-        <target state="translated">Link al documento</target>
+        <target state="needs-translation">Document links</target>
       </trans-unit>
       <trans-unit id="8216780576869093031" datatype="html">
         <source>Bundles</source>
@@ -12591,7 +12591,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">26,27</context>
         </context-group>
-        <target state="translated">Pacchetti</target>
+        <target state="needs-translation">Bundles</target>
       </trans-unit>
       <trans-unit id="1229748338333965418" datatype="html">
         <source>Use workflows to customize the behavior of Paperless-ngx when events &apos;trigger&apos; a workflow.</source>

src-ui/src/locale/messages.ru_RU.xlf

@@ -3907,7 +3907,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">2,3</context>
         </context-group>
-        <target state="translated">Ссылки для обмена</target>
+        <target state="needs-translation">Share links</target>
       </trans-unit>
       <trans-unit id="7844706011418789951" datatype="html">
         <source>Administration</source>
@@ -12410,7 +12410,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">40,41</context>
         </context-group>
-        <target state="translated">Истёкшие</target>
+        <target state="needs-translation">Expired</target>
       </trans-unit>
       <trans-unit id="2057574872309338088" datatype="html">
         <source>Copy share link</source>
@@ -12454,7 +12454,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-bundle-list/share-link-bundle-list.component.html</context>
           <context context-type="linenumber">164,165</context>
         </context-group>
-        <target state="translated">Пагинация ссылок</target>
+        <target state="needs-translation">Share link bundles pagination</target>
       </trans-unit>
       <trans-unit id="1541995860059283227" datatype="html">
         <source>Failed to load share link bundles.</source>
@@ -12510,7 +12510,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">7,8</context>
         </context-group>
-        <target state="translated"> В настоящее время нет общих ссылок. </target>
+        <target state="needs-translation"> No document share links currently exist. </target>
       </trans-unit>
       <trans-unit id="894965666917873756" datatype="html">
         <source>Document</source>
@@ -12518,7 +12518,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">15,16</context>
         </context-group>
-        <target state="translated">Документ</target>
+        <target state="needs-translation">Document</target>
       </trans-unit>
       <trans-unit id="7420713064779772535" datatype="html">
         <source>Delete share link</source>
@@ -12526,7 +12526,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">75,76</context>
         </context-group>
-        <target state="translated">Удалить ссылку для обмена</target>
+        <target state="needs-translation">Delete share link</target>
       </trans-unit>
       <trans-unit id="5198271726831144068" datatype="html">
         <source>Share links pagination</source>
@@ -12534,7 +12534,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">104,105</context>
         </context-group>
-        <target state="translated">Пагинация ссылок</target>
+        <target state="needs-translation">Share links pagination</target>
       </trans-unit>
       <trans-unit id="2976350639944567244" datatype="html">
         <source>Failed to load share links.</source>
@@ -12542,7 +12542,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">105</context>
         </context-group>
-        <target state="translated">Не удалось загрузить ссылки.</target>
+        <target state="needs-translation">Failed to load share links.</target>
       </trans-unit>
       <trans-unit id="4100863040928608880" datatype="html">
         <source>Error retrieving share links.</source>
@@ -12550,7 +12550,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">107</context>
         </context-group>
-        <target state="translated">Ошибка при получении ссылок.</target>
+        <target state="needs-translation">Error retrieving share links.</target>
       </trans-unit>
       <trans-unit id="326120283142363299" datatype="html">
         <source>Share link deleted.</source>
@@ -12558,7 +12558,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">154</context>
         </context-group>
-        <target state="translated">Ссылка для обмена удалена.</target>
+        <target state="needs-translation">Share link deleted.</target>
       </trans-unit>
       <trans-unit id="4964931417181298145" datatype="html">
         <source>Error deleting share link.</source>
@@ -12566,7 +12566,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">159</context>
         </context-group>
-        <target state="translated">Ошибка при удалении ссылки.</target>
+        <target state="needs-translation">Error deleting share link.</target>
       </trans-unit>
       <trans-unit id="3439104787814858626" datatype="html">
         <source>Manage public links to individual documents and document bundles.</source>
@@ -12574,7 +12574,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">4,5</context>
         </context-group>
-        <target state="translated">Управление публичными ссылками на отдельные документы и комплекты документов.</target>
+        <target state="needs-translation">Manage public links to individual documents and document bundles.</target>
       </trans-unit>
       <trans-unit id="3740311348735113099" datatype="html">
         <source>Document links</source>
@@ -12582,7 +12582,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">18,19</context>
         </context-group>
-        <target state="translated">Ссылки на документы</target>
+        <target state="needs-translation">Document links</target>
       </trans-unit>
       <trans-unit id="8216780576869093031" datatype="html">
         <source>Bundles</source>
@@ -12590,7 +12590,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">26,27</context>
         </context-group>
-        <target state="translated">Комплекты</target>
+        <target state="needs-translation">Bundles</target>
       </trans-unit>
       <trans-unit id="1229748338333965418" datatype="html">
         <source>Use workflows to customize the behavior of Paperless-ngx when events &apos;trigger&apos; a workflow.</source>

src-ui/src/locale/messages.sl_SI.xlf

@@ -2855,7 +2855,7 @@
           <context context-type="sourcefile">src/app/components/admin/tasks/tasks.component.ts</context>
           <context context-type="linenumber">104</context>
         </context-group>
-        <target state="translated">Uporabi predloge umetne inteligence</target>
+        <target state="needs-translation">Apply AI Suggestions</target>
       </trans-unit>
       <trans-unit id="9172233176401579786" datatype="html">
         <source>Scheduled</source>
@@ -3907,7 +3907,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">2,3</context>
         </context-group>
-        <target state="translated">Delite povezave</target>
+        <target state="needs-translation">Share links</target>
       </trans-unit>
       <trans-unit id="7844706011418789951" datatype="html">
         <source>Administration</source>
@@ -12411,7 +12411,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">40,41</context>
         </context-group>
-        <target state="translated">Poteklo</target>
+        <target state="needs-translation">Expired</target>
       </trans-unit>
       <trans-unit id="2057574872309338088" datatype="html">
         <source>Copy share link</source>
@@ -12455,7 +12455,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-bundle-list/share-link-bundle-list.component.html</context>
           <context context-type="linenumber">164,165</context>
         </context-group>
-        <target state="translated">Deljenje povezav – strani v sklopih</target>
+        <target state="needs-translation">Share link bundles pagination</target>
       </trans-unit>
       <trans-unit id="1541995860059283227" datatype="html">
         <source>Failed to load share link bundles.</source>
@@ -12511,7 +12511,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">7,8</context>
         </context-group>
-        <target state="translated"> Trenutno ni povezav za skupno rabo dokumentov. </target>
+        <target state="needs-translation"> No document share links currently exist. </target>
       </trans-unit>
       <trans-unit id="894965666917873756" datatype="html">
         <source>Document</source>
@@ -12519,7 +12519,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">15,16</context>
         </context-group>
-        <target state="translated">Dokument</target>
+        <target state="needs-translation">Document</target>
       </trans-unit>
       <trans-unit id="7420713064779772535" datatype="html">
         <source>Delete share link</source>
@@ -12527,7 +12527,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">75,76</context>
         </context-group>
-        <target state="translated">Izbriši povezavo za deljenje</target>
+        <target state="needs-translation">Delete share link</target>
       </trans-unit>
       <trans-unit id="5198271726831144068" datatype="html">
         <source>Share links pagination</source>
@@ -12535,7 +12535,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.html</context>
           <context context-type="linenumber">104,105</context>
         </context-group>
-        <target state="translated">Deljenje povezav – strani v sklopih</target>
+        <target state="needs-translation">Share links pagination</target>
       </trans-unit>
       <trans-unit id="2976350639944567244" datatype="html">
         <source>Failed to load share links.</source>
@@ -12543,7 +12543,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">105</context>
         </context-group>
-        <target state="translated">Povezav za deljenje ni bilo mogoče naložiti.</target>
+        <target state="needs-translation">Failed to load share links.</target>
       </trans-unit>
       <trans-unit id="4100863040928608880" datatype="html">
         <source>Error retrieving share links.</source>
@@ -12551,7 +12551,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">107</context>
         </context-group>
-        <target state="translated">Napaka pri pridobivanju povezav za deljenje.</target>
+        <target state="needs-translation">Error retrieving share links.</target>
       </trans-unit>
       <trans-unit id="326120283142363299" datatype="html">
         <source>Share link deleted.</source>
@@ -12559,7 +12559,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">154</context>
         </context-group>
-        <target state="translated">Povezava za deljenje je bila izbrisana.</target>
+        <target state="needs-translation">Share link deleted.</target>
       </trans-unit>
       <trans-unit id="4964931417181298145" datatype="html">
         <source>Error deleting share link.</source>
@@ -12567,7 +12567,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-link-list/share-link-list.component.ts</context>
           <context context-type="linenumber">159</context>
         </context-group>
-        <target state="translated">Napaka pri brisanju povezave za deljenje.</target>
+        <target state="needs-translation">Error deleting share link.</target>
       </trans-unit>
       <trans-unit id="3439104787814858626" datatype="html">
         <source>Manage public links to individual documents and document bundles.</source>
@@ -12575,7 +12575,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">4,5</context>
         </context-group>
-        <target state="translated">Upravljajte javne povezave do posameznih dokumentov in paketov dokumentov.</target>
+        <target state="needs-translation">Manage public links to individual documents and document bundles.</target>
       </trans-unit>
       <trans-unit id="3740311348735113099" datatype="html">
         <source>Document links</source>
@@ -12583,7 +12583,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">18,19</context>
         </context-group>
-        <target state="translated">Povezave do dokumentov</target>
+        <target state="needs-translation">Document links</target>
       </trans-unit>
       <trans-unit id="8216780576869093031" datatype="html">
         <source>Bundles</source>
@@ -12591,7 +12591,7 @@
           <context context-type="sourcefile">src/app/components/manage/share-links/share-links.component.html</context>
           <context context-type="linenumber">26,27</context>
         </context-group>
-        <target state="translated">Paketi</target>
+        <target state="needs-translation">Bundles</target>
       </trans-unit>
       <trans-unit id="1229748338333965418" datatype="html">
         <source>Use workflows to customize the behavior of Paperless-ngx when events &apos;trigger&apos; a workflow.</source>

src-ui/src/locale/messages.sv_SE.xlf

@@ -11518,7 +11518,7 @@
           <context context-type="sourcefile">src/app/components/manage/document-attributes/custom-fields/custom-fields.component.html</context>
           <context context-type="linenumber">70,72</context>
         </context-group>
-        <target state="translated">Inga fält har definierats.</target>
+        <target state="translated"/>
       </trans-unit>
       <trans-unit id="3032792139967609806" datatype="html">
         <source>Confirm delete field</source>

src-ui/src/main.ts

@@ -74,7 +74,6 @@ import {
   clipboardCheckFill,
   clipboardFill,
   clockHistory,
-  cpu,
   creditCard,
   dash,
   dashCircle,
@@ -119,7 +118,6 @@ import {
   infoCircle,
   journalBookmarkFill,
   journals,
-  lightbulb,
   link,
   list,
   listNested,
@@ -150,6 +148,7 @@ import {
   send,
   shop,
   slashCircle,
+  sliders,
   sliders2Vertical,
   sortAlphaDown,
   sortAlphaUpAlt,
@@ -324,7 +323,6 @@ const icons = {
   clipboardCheckFill,
   clipboardFill,
   clockHistory,
-  cpu,
   cash,
   creditCard,
   dash,
@@ -370,7 +368,6 @@ const icons = {
   infoCircle,
   journalBookmarkFill,
   journals,
-  lightbulb,
   link,
   list,
   listNested,
@@ -401,6 +398,7 @@ const icons = {
   send,
   slashCircle,
   shop,
+  sliders,
   sliders2Vertical,
   sortAlphaDown,
   sortAlphaUpAlt,

src-ui/src/styles.scss

@@ -292,7 +292,6 @@ a.btn-link:active,
 a.btn-link:focus-visible,
 .btn-link:hover,
 .btn-link:active,
-.btn-link.show,
 .btn-link:focus-visible {
   color: var(--pngx-primary-lighten-10) !important;
   .primary-light & {

src-ui/src/test/mocks/pdfjs-web-pdf_viewer.ts

@@ -25,20 +25,10 @@ export class PDFFindController {
   onIsPageVisible?: () => boolean
 }
 
-export const LinkTarget = {
-  NONE: 0,
-  SELF: 1,
-  BLANK: 2,
-  PARENT: 3,
-  TOP: 4,
-}
-
 export class PDFLinkService {
   private document?: unknown
   private viewer?: unknown
 
-  constructor(readonly options: Record<string, unknown> = {}) {}
-
   setDocument(document: unknown): void {
     this.document = document
   }

src/conftest.py

@@ -1,151 +0,0 @@
-"""Fixtures available to every Paperless-ngx app.
-
-Loaded automatically for every test path. Keep module-scope imports minimal:
-this file is imported for every session,  so anything heavy belongs inside
-the fixture body that needs it.
-"""
-
-from __future__ import annotations
-
-from typing import TYPE_CHECKING
-
-import pytest
-
-if TYPE_CHECKING:
-    from collections.abc import Generator
-    from pathlib import Path
-
-    from django.contrib.auth.models import User
-    from pytest_django.fixtures import Settings
-    from rest_framework.test import APIClient
-
-    from paperless_testing.dirs import PaperlessDirs
-    from paperless_testing.fakes.progress import FakeProgressManager
-
-
-@pytest.fixture(scope="session", autouse=True)
-def faker_session_locale() -> str:
-    """Pin Faker's locale so generated data does not follow the host locale.
-
-    The seed itself is left to pytest-randomly, which derives one per run.
-    """
-    return "en_US"
-
-
-@pytest.fixture(autouse=True)
-def _fast_password_hasher(settings: Settings) -> None:
-    """Hash test passwords with MD5 instead of Django's default PBKDF2.
-
-    PBKDF2 is deliberately slow, and every ``admin_user`` or
-    ``create_superuser`` call pays for it: about 600 ms each. No test depends
-    on the hash format, only on ``check_password`` and on the stored value
-    changing when the password does.
-    """
-    settings.PASSWORD_HASHERS = ["django.contrib.auth.hashers.MD5PasswordHasher"]
-
-
-@pytest.fixture(autouse=True)
-def _clear_content_type_caches() -> None:
-    """Clear Django's ContentType cache and guardian's lru_cache before each test.
-
-    Tests that delete and reinsert ContentType/Permission rows (e.g. the
-    importer) corrupt both caches. Without this fixture a subsequent test on
-    the same xdist worker sees stale ContentType objects and guardian raises
-    MixedContentTypeError.
-    """
-    from django.contrib.contenttypes.models import ContentType
-    from guardian.shortcuts import clear_ct_cache
-
-    ContentType.objects.clear_cache()
-    clear_ct_cache()
-
-
-@pytest.fixture(autouse=True)
-def _clear_django_caches() -> None:
-    """Clear every configured cache before each test.
-
-    Cached values outlive the test that wrote them: the classifier keys its
-    vectorized content on a hash of the content itself, so a second test
-    generating the same fixture data takes the cache-hit path and never calls
-    the code it is asserting against.
-    """
-    from django.core.cache import caches
-
-    for cache in caches.all(initialized_only=False):
-        cache.clear()
-
-
-@pytest.fixture
-def paperless_dirs(
-    tmp_path: Path,
-    settings: Settings,
-) -> Generator[PaperlessDirs, None, None]:
-    """The standard temp directory layout, applied to Django settings."""
-    from documents.search import reset_backend
-    from paperless_testing.dirs import build_paperless_dirs
-    from paperless_testing.dirs import dirs_settings
-
-    dirs = build_paperless_dirs(tmp_path)
-    for name, value in dirs_settings(dirs).items():
-        setattr(settings, name, value)
-
-    # Not directory settings, but they are needed alongside the layout by the
-    # sanity checker tests.
-    settings.IGNORABLE_FILES = {".DS_Store", "Thumbs.db", "desktop.ini"}
-    settings.APP_LOGO = ""
-
-    reset_backend()
-    yield dirs
-    reset_backend()
-
-
-@pytest.fixture
-def rest_api_client() -> APIClient:
-    """The basic DRF APIClient, unauthenticated."""
-    from rest_framework.test import APIClient
-
-    return APIClient()
-
-
-@pytest.fixture
-def regular_user(db: None) -> User:
-    """Unprivileged user for permission boundary tests."""
-    from paperless_testing.factories import UserFactory
-
-    return UserFactory(username="regular")
-
-
-@pytest.fixture
-def admin_client(rest_api_client: APIClient, admin_user: User) -> 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: User) -> 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: User) -> 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 fake_progress_manager(
-    monkeypatch: pytest.MonkeyPatch,
-) -> type[FakeProgressManager]:
-    """Replace documents.tasks.ProgressManager with the fake, so consuming a file
-    in a test never tries to reach a broker."""
-    from paperless_testing.fakes.progress import FakeProgressManager
-
-    monkeypatch.setattr("documents.tasks.ProgressManager", FakeProgressManager)
-    return FakeProgressManager

src/documents/apps.py

@@ -26,6 +26,7 @@ class DocumentsConfig(AppConfig):
         document_consumption_finished.connect(set_document_type)
         document_consumption_finished.connect(set_tags)
         document_consumption_finished.connect(set_storage_path)
+        document_consumption_finished.connect(add_to_index)
         document_consumption_finished.connect(run_workflows_added)
         document_consumption_finished.connect(add_to_index)
         document_consumption_finished.connect(add_or_update_document_in_llm_index)

src/documents/consumer.py

@@ -857,9 +857,8 @@ class ConsumerPlugin(
             self.log.debug(f"Creation date from parse_date: {create_date}")
         else:
             stats = Path(self.input_doc.original_file).stat()
-            create_date = datetime.datetime.fromtimestamp(
+            create_date = timezone.make_aware(
-                stats.st_mtime,
+                datetime.datetime.fromtimestamp(stats.st_mtime),
-                tz=timezone.get_current_timezone(),
             )
             self.log.debug(f"Creation date from st_mtime: {create_date}")
 

src/documents/search/_backend.py

@@ -196,49 +196,52 @@ class WriteBatch:
         return self._raw_writer
 
     def __enter__(self) -> Self:
-        lock_path = self._backend._path / ".tantivy.lock"
+        if self._backend._path is not None:
-        self._lock = filelock.FileLock(str(lock_path))
+            lock_path = self._backend._path / ".tantivy.lock"
-        for attempt in range(_LOCK_RETRY_ATTEMPTS):
+            self._lock = filelock.FileLock(str(lock_path))
-            try:
+            for attempt in range(_LOCK_RETRY_ATTEMPTS):
-                self._lock.acquire(timeout=self._lock_timeout)
+                try:
-                break
+                    self._lock.acquire(timeout=self._lock_timeout)
-            except filelock.Timeout:
+                    break
-                if attempt == _LOCK_RETRY_ATTEMPTS - 1:
+                except filelock.Timeout:
-                    raise SearchIndexLockError(
+                    if attempt == _LOCK_RETRY_ATTEMPTS - 1:
-                        f"Could not acquire index lock after {_LOCK_RETRY_ATTEMPTS} "
+                        raise SearchIndexLockError(
-                        f"attempts (timeout={self._lock_timeout}s each)",
+                            f"Could not acquire index lock after {_LOCK_RETRY_ATTEMPTS} "
+                            f"attempts (timeout={self._lock_timeout}s each)",
+                        )
+                    sleep_s = random.uniform(
+                        0,
+                        min(_LOCK_BACKOFF_CAP, _LOCK_BACKOFF_BASE * (2**attempt)),
                     )
-                sleep_s = random.uniform(
+                    logger.debug(
-                    0,
+                        "Index lock contention; retrying in %.2fs (attempt %d/%d)",
-                    min(_LOCK_BACKOFF_CAP, _LOCK_BACKOFF_BASE * (2**attempt)),
+                        sleep_s,
-                )
+                        attempt + 1,
-                logger.debug(
+                        _LOCK_RETRY_ATTEMPTS,
-                    "Index lock contention; retrying in %.2fs (attempt %d/%d)",
+                    )
-                    sleep_s,
+                    time.sleep(sleep_s)
-                    attempt + 1,
-                    _LOCK_RETRY_ATTEMPTS,
-                )
-                time.sleep(sleep_s)
 
-        # Open a fresh Index (and thus a fresh Tantivy ManagedDirectory)
+            # Open a fresh Index (and thus a fresh Tantivy ManagedDirectory)
-        # for the write, rather than reusing the process-local cached
+            # for the write, rather than reusing the process-local cached
-        # index. ManagedDirectory loads its GC bookkeeping (.managed.json)
+            # index. ManagedDirectory loads its GC bookkeeping (.managed.json)
-        # once, at construction, and never re-reads it; paperless runs
+            # once, at construction, and never re-reads it; paperless runs
-        # several long-lived processes (Granian workers, Celery workers)
+            # several long-lived processes (Granian workers, Celery workers)
-        # that take turns writing under the file lock above. A cached,
+            # that take turns writing under the file lock above. A cached,
-        # long-lived writer index would carry a stale managed-files view
+            # long-lived writer index would carry a stale managed-files view
-        # and, on commit, overwrite .managed.json with that stale view -
+            # and, on commit, overwrite .managed.json with that stale view -
-        # permanently losing track of segment files other processes
+            # permanently losing track of segment files other processes
-        # registered in the meantime, so they can never be garbage
+            # registered in the meantime, so they can never be garbage
-        # collected. Reopening fresh here always picks up the current
+            # collected. Reopening fresh here always picks up the current
-        # on-disk state. The long-lived self._backend._index is used for
+            # on-disk state. The long-lived self._backend._index is used for
-        # reads only and is reloaded (not reopened) after commit below.
+            # reads only and is reloaded (not reopened) after commit below.
-        write_index = tantivy.Index(
+            write_index = tantivy.Index(
-            build_schema(),
+                build_schema(),
-            path=str(self._backend._path),
+                path=str(self._backend._path),
-        )
+            )
-        register_tokenizers(write_index, settings.SEARCH_LANGUAGE)
+            register_tokenizers(write_index, settings.SEARCH_LANGUAGE)
-        self._raw_writer = write_index.writer()
+            self._raw_writer = write_index.writer()
+        else:
+            self._raw_writer = self._backend._index.writer()
         return self
 
     def __exit__(self, exc_type, exc_val, exc_tb):
@@ -369,8 +372,9 @@ class TantivyBackend:
     Tantivy search backend with explicit lifecycle management.
 
     Provides full-text search capabilities using the Tantivy search engine.
-    Keeps a persistent on-disk index. Handles document indexing, search queries,
+    Supports in-memory indexes (for testing) and persistent on-disk indexes
-    autocompletion, and "more like this" functionality.
+    (for production use). Handles document indexing, search queries, autocompletion,
+    and "more like this" functionality.
 
     The backend manages its own connection lifecycle and can be reset when
     the underlying index directory changes (e.g., during test isolation).
@@ -404,7 +408,9 @@ class TantivyBackend:
         },
     )
 
-    def __init__(self, path: Path):
+    def __init__(self, path: Path | None = None):
+        # path=None → in-memory index (for tests)
+        # path=some_dir → on-disk index (for production)
         self._path = path
         self._raw_index: tantivy.Index | None = None
         self._raw_schema: tantivy.Schema | None = None
@@ -423,13 +429,16 @@ class TantivyBackend:
         """
         Open or rebuild the index as needed.
 
-        Checks if rebuilding is needed due to schema version or language
+        For disk-based indexes, checks if rebuilding is needed due to schema
-        changes. Registers custom tokenizers after opening.
+        version or language changes. Registers custom tokenizers after opening.
         Safe to call multiple times - subsequent calls are no-ops.
         """
         if self._raw_index is not None:
             return  # pragma: no cover
-        self._raw_index = open_or_rebuild_index(self._path)
+        if self._path is not None:
+            self._raw_index = open_or_rebuild_index(self._path)
+        else:
+            self._raw_index = tantivy.Index(build_schema())
         register_tokenizers(self._raw_index, settings.SEARCH_LANGUAGE)
         self._raw_schema = self._raw_index.schema
 
@@ -1093,9 +1102,13 @@ class TantivyBackend:
                 writer's threads). Larger values buffer more docs in RAM before
                 flushing a segment, deferring merge work; they do not avoid it.
         """
-        wipe_index(self._path)
+        # Create new index (on-disk or in-memory)
-        new_index = tantivy.Index(build_schema(), path=str(self._path))
+        if self._path is not None:
-        _write_sentinels(self._path)
+            wipe_index(self._path)
+            new_index = tantivy.Index(build_schema(), path=str(self._path))
+            _write_sentinels(self._path)
+        else:
+            new_index = tantivy.Index(build_schema())
         register_tokenizers(new_index, settings.SEARCH_LANGUAGE)
 
         # Point instance at the new index so _build_tantivy_doc uses it

src/documents/serialisers/base.py

@@ -1,568 +0,0 @@
-from __future__ import annotations
-
-import logging
-from typing import TYPE_CHECKING
-from typing import Any
-from typing import Literal
-
-from django.contrib.auth.models import Group
-from django.contrib.auth.models import User
-from django.contrib.contenttypes.models import ContentType
-from django.utils.text import slugify
-from django.utils.translation import gettext as _
-from drf_spectacular.utils import extend_schema_field
-from guardian.core import ObjectPermissionChecker
-from guardian.shortcuts import get_users_with_perms
-from guardian.utils import get_group_obj_perms_model
-from guardian.utils import get_user_obj_perms_model
-from rest_framework import serializers
-from rest_framework.exceptions import PermissionDenied
-from rest_framework.fields import SerializerMethodField
-from rest_framework.utils import model_meta
-
-from documents import bulk_edit
-from documents.models import Document
-from documents.models import MatchingModel
-from documents.models import Note
-from documents.permissions import get_groups_with_only_permission
-from documents.permissions import set_permissions_for_object
-from documents.regex import validate_regex_pattern
-
-if TYPE_CHECKING:
-    from collections.abc import Iterable
-
-
-logger = logging.getLogger("paperless.serializers")
-
-
-# https://www.django-rest-framework.org/api-guide/serializers/#example
-class DynamicFieldsModelSerializer(serializers.ModelSerializer[Any]):
-    """
-    A ModelSerializer that takes an additional `fields` argument that
-    controls which fields should be displayed.
-    """
-
-    def __init__(self, *args, **kwargs) -> None:
-        # Don't pass the 'fields' arg up to the superclass
-        fields = kwargs.pop("fields", None)
-
-        # Instantiate the superclass normally
-        super().__init__(*args, **kwargs)
-
-        if fields is not None:
-            # Drop any fields that are not specified in the `fields` argument.
-            allowed = set(fields)
-            existing = set(self.fields)
-            for field_name in existing - allowed:
-                self.fields.pop(field_name)
-
-
-class DocumentUpdateFieldsModelSerializer(DynamicFieldsModelSerializer):
-    stale_update_excluded_fields = frozenset({"filename", "archive_filename"})
-
-    def _get_update_fields(self, validated_data) -> list[str]:
-        model_fields = {
-            field.name
-            for field in self.Meta.model._meta.concrete_fields
-            if field.name not in self.stale_update_excluded_fields
-        }
-        update_fields = [
-            field_name for field_name in validated_data if field_name in model_fields
-        ]
-        if "modified" in model_fields and "modified" not in update_fields:
-            update_fields.append("modified")
-        return update_fields
-
-    def update(self, instance, validated_data):
-        serializers.raise_errors_on_nested_writes("update", self, validated_data)
-        info = model_meta.get_field_info(instance)
-
-        m2m_fields = []
-        for attr, value in validated_data.items():
-            if attr in info.relations and info.relations[attr].to_many:
-                m2m_fields.append((attr, value))
-            else:
-                setattr(instance, attr, value)
-
-        # File names are managed by post-save file handling.  Saving only the
-        # serializer-updated fields prevents stale in-memory path values from
-        # overwriting a concurrent move.
-        instance.save(update_fields=self._get_update_fields(validated_data))
-
-        for attr, value in m2m_fields:
-            field = getattr(instance, attr)
-            field.set(value)
-
-        return instance
-
-
-class MatchingModelSerializer(serializers.ModelSerializer[Any]):
-    document_count = serializers.IntegerField(read_only=True)
-
-    def get_slug(self, obj) -> str:
-        return slugify(obj.name)
-
-    slug = SerializerMethodField()
-
-    def validate(self, data):
-        # TODO: remove pending https://github.com/encode/django-rest-framework/issues/7173
-        name = data.get(
-            "name",
-            self.instance.name if hasattr(self.instance, "name") else None,
-        )
-        owner = (
-            data["owner"]
-            if "owner" in data
-            else self.user
-            if hasattr(self, "user")
-            else None
-        )
-        pk = self.instance.pk if hasattr(self.instance, "pk") else None
-        if ("name" in data or "owner" in data) and self.Meta.model.objects.filter(
-            name=name,
-            owner=owner,
-        ).exclude(pk=pk).exists():
-            raise serializers.ValidationError(
-                {"error": "Object violates owner / name unique constraint"},
-            )
-        return data
-
-    def validate_match(self, match):
-        if (
-            "matching_algorithm" in self.initial_data
-            and self.initial_data["matching_algorithm"] == MatchingModel.MATCH_REGEX
-        ):
-            try:
-                validate_regex_pattern(match)
-            except ValueError as e:
-                logger.debug(f"Invalid regular expression: {e!s}")
-                raise serializers.ValidationError(
-                    "Invalid regular expression, see log for details.",
-                )
-        return match
-
-
-PERMISSION_ACTIONS = ("view", "change")
-
-
-class SetPermissionsMixin:
-    def _validate_user_ids(self, user_ids):
-        users = User.objects.none()
-        if user_ids is not None:
-            users = User.objects.filter(id__in=user_ids)
-            if not users.count() == len(user_ids):
-                raise serializers.ValidationError(
-                    "Some users in don't exist or were specified twice.",
-                )
-        return users
-
-    def _validate_group_ids(self, group_ids):
-        groups = Group.objects.none()
-        if group_ids is not None:
-            groups = Group.objects.filter(id__in=group_ids)
-            if not groups.count() == len(group_ids):
-                raise serializers.ValidationError(
-                    "Some groups in don't exist or were specified twice.",
-                )
-        return groups
-
-    def validate_set_permissions(self, set_permissions=None):
-        permissions_dict = {action: {} for action in PERMISSION_ACTIONS}
-        if set_permissions is not None:
-            for action in PERMISSION_ACTIONS:
-                if action in set_permissions:
-                    if "users" in set_permissions[action]:
-                        users = set_permissions[action]["users"]
-                        permissions_dict[action]["users"] = self._validate_user_ids(
-                            users,
-                        )
-                    if "groups" in set_permissions[action]:
-                        groups = set_permissions[action]["groups"]
-                        permissions_dict[action]["groups"] = self._validate_group_ids(
-                            groups,
-                        )
-                else:
-                    del permissions_dict[action]
-        return permissions_dict
-
-    def _set_permissions(self, permissions, object) -> None:
-        set_permissions_for_object(permissions, object)
-
-
-class SerializerWithPerms(serializers.Serializer[dict[str, Any]]):
-    def __init__(self, *args, **kwargs) -> None:
-        self.user = kwargs.pop("user", None)
-        self.full_perms = kwargs.pop("full_perms", False)
-        self.all_fields = kwargs.pop("all_fields", False)
-        super().__init__(*args, **kwargs)
-
-
-class PermissionSetSerializer(serializers.Serializer[dict[str, Any]]):
-    users = serializers.ListField(
-        child=serializers.IntegerField(),
-        required=False,
-        allow_null=True,
-    )
-    groups = serializers.ListField(
-        child=serializers.IntegerField(),
-        required=False,
-        allow_null=True,
-    )
-
-
-class SetPermissionsSerializer(serializers.Serializer[dict[str, Any]]):
-    view = PermissionSetSerializer(required=False)
-    change = PermissionSetSerializer(required=False)
-
-    def to_internal_value(self, data):
-        if isinstance(data, dict):
-            unknown_keys = set(data) - set(PERMISSION_ACTIONS)
-            if unknown_keys:
-                raise serializers.ValidationError(
-                    {key: "Unknown permission action." for key in sorted(unknown_keys)},
-                )
-        return super().to_internal_value(data)
-
-
-class OwnedObjectSerializer(
-    SerializerWithPerms,
-    serializers.ModelSerializer[Any],
-    SetPermissionsMixin,
-):
-    def __init__(self, *args, **kwargs) -> None:
-        super().__init__(*args, **kwargs)
-
-        if not self.all_fields:
-            try:
-                if self.full_perms:
-                    self.fields.pop("user_can_change")
-                    self.fields.pop("is_shared_by_requester")
-                else:
-                    self.fields.pop("permissions")
-            except KeyError:
-                pass
-
-    def _get_perms(self, obj, codename: str, target: Literal["users", "groups"]):
-        """
-        Get the given permissions from context or from django-guardian.
-
-        :param codename: The permission codename, e.g. 'view' or 'change'
-        :param target: 'users' or 'groups'
-        """
-        key = f"{target}_{codename}_perms"
-        cached = self.context.get(key, {}).get(obj.pk)
-        if cached is not None:
-            return list(cached)
-
-        # Permission not found in the context, get it from guardian
-        if target == "users":
-            return list(
-                get_users_with_perms(
-                    obj,
-                    only_with_perms_in=[f"{codename}_{obj.__class__.__name__.lower()}"],
-                    with_group_users=False,
-                ).values_list("id", flat=True),
-            )
-        else:  # groups
-            return list(
-                get_groups_with_only_permission(
-                    obj,
-                    codename=f"{codename}_{obj.__class__.__name__.lower()}",
-                ).values_list("id", flat=True),
-            )
-
-    @extend_schema_field(
-        field={
-            "type": "object",
-            "properties": {
-                "view": {
-                    "type": "object",
-                    "properties": {
-                        "users": {
-                            "type": "array",
-                            "items": {"type": "integer"},
-                        },
-                        "groups": {
-                            "type": "array",
-                            "items": {"type": "integer"},
-                        },
-                    },
-                },
-                "change": {
-                    "type": "object",
-                    "properties": {
-                        "users": {
-                            "type": "array",
-                            "items": {"type": "integer"},
-                        },
-                        "groups": {
-                            "type": "array",
-                            "items": {"type": "integer"},
-                        },
-                    },
-                },
-            },
-        },
-    )
-    def get_permissions(self, obj) -> dict:
-        return {
-            "view": {
-                "users": self._get_perms(obj, "view", "users"),
-                "groups": self._get_perms(obj, "view", "groups"),
-            },
-            "change": {
-                "users": self._get_perms(obj, "change", "users"),
-                "groups": self._get_perms(obj, "change", "groups"),
-            },
-        }
-
-    def get_user_can_change(self, obj) -> bool:
-        if obj.owner is None or obj.owner == self.user:
-            return True
-        if self.user is None:
-            return False
-        if self.user.is_active and self.user.is_superuser:
-            # Mirrors guardian's own ObjectPermissionChecker.has_perm() shortcut --
-            # superusers aren't necessarily granted explicit object permissions,
-            # so the batched context below would otherwise incorrectly say no.
-            return True
-
-        # Prefer the page-level batch computed by BulkPermissionMixin
-        # (get_serializer_context) over a fresh per-object guardian check,
-        # which would otherwise query the permission tables once per row.
-        users_change_perms = self.context.get("users_change_perms")
-        groups_change_perms = self.context.get("groups_change_perms")
-        if users_change_perms is not None and groups_change_perms is not None:
-            if self.user.pk in users_change_perms.get(obj.pk, []):
-                return True
-            user_group_ids = getattr(self, "_user_group_ids", None)
-            if user_group_ids is None:
-                user_group_ids = set(self.user.groups.values_list("id", flat=True))
-                self._user_group_ids = user_group_ids
-            return bool(
-                user_group_ids.intersection(groups_change_perms.get(obj.pk, [])),
-            )
-
-        checker = ObjectPermissionChecker(self.user)
-        return checker.has_perm(f"change_{obj.__class__.__name__.lower()}", obj)
-
-    @staticmethod
-    def get_shared_object_pks(objects: Iterable):
-        """
-        Return the primary keys of the subset of objects that are shared.
-        """
-        try:
-            first_obj = next(iter(objects))
-        except StopIteration:
-            return set()
-
-        ctype = ContentType.objects.get_for_model(first_obj)
-        object_pks = list(obj.pk for obj in objects)
-        pk_type = type(first_obj.pk)
-
-        def get_pks_for_permission_type(model):
-            return map(
-                pk_type,  # coerce the pk to be the same type of the provided objects
-                model.objects.filter(
-                    content_type=ctype,
-                    object_pk__in=object_pks,
-                )
-                .values_list("object_pk", flat=True)
-                .distinct(),
-            )
-
-        UserObjectPermission = get_user_obj_perms_model()
-        GroupObjectPermission = get_group_obj_perms_model()
-        user_permission_pks = get_pks_for_permission_type(UserObjectPermission)
-        group_permission_pks = get_pks_for_permission_type(GroupObjectPermission)
-
-        return set(user_permission_pks) | set(group_permission_pks)
-
-    def get_is_shared_by_requester(self, obj: Document) -> bool:
-        # First check the context to see if `shared_object_pks` is set by the parent.
-        shared_object_pks = self.context.get("shared_object_pks")
-        # If not just check if the current object is shared.
-        if shared_object_pks is None:
-            shared_object_pks = self.get_shared_object_pks([obj])
-        return obj.owner == self.user and obj.id in shared_object_pks
-
-    permissions = SerializerMethodField(read_only=True, required=False)
-    user_can_change = SerializerMethodField(read_only=True, required=False)
-    is_shared_by_requester = SerializerMethodField(read_only=True, required=False)
-
-    set_permissions = SetPermissionsSerializer(
-        label="Set permissions",
-        required=False,
-        write_only=True,
-    )
-    # other methods in mixin
-
-    def validate_unique_together(self, validated_data, instance=None) -> None:
-        # workaround for https://github.com/encode/django-rest-framework/issues/9358
-        if "owner" in validated_data and "name" in self.Meta.fields:
-            name = validated_data.get("name", instance.name if instance else None)
-            objects = (
-                self.Meta.model.objects.exclude(pk=instance.pk)
-                if instance
-                else self.Meta.model.objects.all()
-            )
-            not_unique = objects.filter(
-                owner=validated_data["owner"],
-                name=name,
-            ).exists()
-            if not_unique:
-                raise serializers.ValidationError(
-                    {"error": "Object violates owner / name unique constraint"},
-                )
-
-    def create(self, validated_data):
-        # default to current user if not set
-        request = self.context.get("request")
-        if (
-            "owner" not in validated_data
-            or (request is not None and "owner" not in request.data)
-        ) and self.user:
-            validated_data["owner"] = self.user
-        permissions = None
-        if "set_permissions" in validated_data:
-            permissions = validated_data.pop("set_permissions")
-        self.validate_unique_together(validated_data)
-        instance = super().create(validated_data)
-        if permissions is not None:
-            self._set_permissions(permissions, instance)
-        return instance
-
-    def update(self, instance, validated_data):
-        user = getattr(self, "user", None)
-        is_superuser = user.is_superuser if user is not None else False
-        is_owner = instance.owner == user if user is not None else False
-        is_unowned = instance.owner is None
-
-        if (
-            ("owner" in validated_data and validated_data["owner"] != instance.owner)
-            or "set_permissions" in validated_data
-        ) and not (is_superuser or is_owner or is_unowned):
-            raise PermissionDenied(
-                _("Insufficient permissions."),
-            )
-
-        if "set_permissions" in validated_data:
-            self._set_permissions(validated_data["set_permissions"], instance)
-        self.validate_unique_together(validated_data, instance)
-        return super().update(instance, validated_data)
-
-
-class OwnedObjectListSerializer(serializers.ListSerializer[Any]):
-    def to_representation(self, documents):
-        self.child.context["shared_object_pks"] = self.child.get_shared_object_pks(
-            documents,
-        )
-        return super().to_representation(documents)
-
-
-class ReadWriteSerializerMethodField(serializers.SerializerMethodField):
-    """
-    Based on https://stackoverflow.com/a/62579804
-    """
-
-    def __init__(self, method_name=None, *args, **kwargs) -> None:
-        self.method_name = method_name
-        kwargs["source"] = "*"
-        super(serializers.SerializerMethodField, self).__init__(*args, **kwargs)
-
-    def to_internal_value(self, data):
-        return {self.field_name: data}
-
-
-class DocumentListSerializer(serializers.Serializer[dict[str, list[int]]]):
-    documents = serializers.ListField(
-        required=True,
-        label="Documents",
-        write_only=True,
-        child=serializers.IntegerField(),
-    )
-
-    def _validate_document_id_list(self, documents, name="documents") -> None:
-        if not isinstance(documents, list):
-            raise serializers.ValidationError(f"{name} must be a list")
-        if not all(isinstance(i, int) for i in documents):
-            raise serializers.ValidationError(f"{name} must be a list of integers")
-        count = Document.objects.filter(id__in=documents).count()
-        if not count == len(documents):
-            raise serializers.ValidationError(
-                f"Some documents in {name} don't exist or were specified twice.",
-            )
-
-    def validate_documents(self, documents):
-        self._validate_document_id_list(documents)
-        return documents
-
-
-class DocumentSelectionSerializer(DocumentListSerializer):
-    documents = serializers.ListField(
-        required=False,
-        label="Documents",
-        write_only=True,
-        child=serializers.IntegerField(),
-    )
-
-    all = serializers.BooleanField(
-        default=False,
-        required=False,
-        write_only=True,
-    )
-
-    filters = serializers.DictField(
-        required=False,
-        allow_empty=True,
-        write_only=True,
-    )
-
-    excluded_documents = serializers.ListField(
-        required=False,
-        default=list,
-        write_only=True,
-        child=serializers.IntegerField(),
-    )
-
-    def validate(self, attrs):
-        if attrs.get("all", False):
-            attrs.setdefault("documents", [])
-            return attrs
-
-        if attrs["excluded_documents"]:
-            raise serializers.ValidationError(
-                "excluded_documents is only supported when all is true.",
-            )
-
-        if "documents" not in attrs:
-            raise serializers.ValidationError(
-                "documents is required unless all is true.",
-            )
-
-        documents = attrs["documents"]
-        self._validate_document_id_list(documents)
-        return attrs
-
-
-class SourceModeValidationMixin:
-    def validate_source_mode(self, source_mode: str) -> str:
-        if source_mode not in bulk_edit.SourceModeChoices.__dict__.values():
-            raise serializers.ValidationError("Invalid source_mode")
-        return source_mode
-
-
-class BasicUserSerializer(serializers.ModelSerializer[User]):
-    # Different than paperless.serializers.UserSerializer
-    class Meta:
-        model = User
-        fields = ["id", "username", "first_name", "last_name"]
-
-
-class NotesSerializer(serializers.ModelSerializer[Note]):
-    user = BasicUserSerializer(read_only=True)
-
-    class Meta:
-        model = Note
-        fields = ["id", "note", "created", "user"]
-        ordering = ["-created"]

src/documents/serialisers/bulk_edit.py

@@ -1,750 +0,0 @@
-from __future__ import annotations
-
-import logging
-
-from django.contrib.auth.models import User
-from rest_framework import serializers
-
-from documents import bulk_edit
-from documents.models import Correspondent
-from documents.models import CustomField
-from documents.models import Document
-from documents.models import DocumentType
-from documents.models import StoragePath
-from documents.models import Tag
-
-from .base import DocumentListSerializer
-from .base import DocumentSelectionSerializer
-from .base import SerializerWithPerms
-from .base import SetPermissionsMixin
-from .base import SetPermissionsSerializer
-from .base import SourceModeValidationMixin
-from .metadata import CustomFieldInstanceSerializer
-
-logger = logging.getLogger("paperless.serializers")
-
-
-def _validate_rotation_degrees(degrees: int, field: str = "degrees") -> int:
-    # QPDF refuses any other angle, which would otherwise fail inside the task
-    if degrees % 90 != 0:
-        raise serializers.ValidationError(f"{field} must be a multiple of 90")
-    return degrees
-
-
-class RotateDocumentsSerializer(DocumentSelectionSerializer, SourceModeValidationMixin):
-    degrees = serializers.IntegerField(required=True)
-    source_mode = serializers.CharField(
-        required=False,
-        default=bulk_edit.SourceModeChoices.LATEST_VERSION,
-    )
-    from_webui = serializers.BooleanField(required=False, default=False)
-
-    def validate_degrees(self, value: int) -> int:
-        return _validate_rotation_degrees(value)
-
-
-class MergeDocumentsSerializer(DocumentListSerializer, SourceModeValidationMixin):
-    metadata_document_id = serializers.IntegerField(
-        required=False,
-        allow_null=True,
-    )
-    delete_originals = serializers.BooleanField(required=False, default=False)
-    archive_fallback = serializers.BooleanField(required=False, default=False)
-    source_mode = serializers.CharField(
-        required=False,
-        default=bulk_edit.SourceModeChoices.LATEST_VERSION,
-    )
-    from_webui = serializers.BooleanField(required=False, default=False)
-
-
-class MergeDocumentsAsVersionsSerializer(DocumentListSerializer):
-    root_document_id = serializers.IntegerField(required=True)
-    version_label = serializers.CharField(
-        required=False,
-        allow_blank=True,
-        allow_null=True,
-        max_length=64,
-    )
-
-    def validate_version_label(self, value):
-        if value is None:
-            return None
-        normalized = value.strip()
-        return normalized or None
-
-    def validate(self, attrs):
-        documents = attrs["documents"]
-        if len(documents) < 2:
-            raise serializers.ValidationError(
-                "At least two documents are required.",
-            )
-        if attrs.get("version_label") is not None and len(documents) != 2:
-            raise serializers.ValidationError(
-                "version_label can only be used when merging one source document.",
-            )
-        if attrs["root_document_id"] not in documents:
-            raise serializers.ValidationError(
-                "root_document_id must be one of the selected documents.",
-            )
-
-        selected_documents = Document.objects.filter(id__in=documents)
-        if selected_documents.filter(root_document__isnull=False).exists():
-            raise serializers.ValidationError(
-                "Only top-level documents can be merged as versions.",
-            )
-
-        source_document_ids = set(documents) - {attrs["root_document_id"]}
-        if Document.global_objects.filter(
-            root_document_id__in=source_document_ids,
-        ).exists():
-            raise serializers.ValidationError(
-                "Documents with existing versions cannot be merged into another document.",
-            )
-        return attrs
-
-
-class PdfEditOperationSerializer(serializers.Serializer[dict[str, int]]):
-    page = serializers.IntegerField(min_value=1)
-    rotate = serializers.IntegerField(required=False)
-    doc = serializers.IntegerField(required=False, min_value=0)
-
-    def validate_rotate(self, value: int) -> int:
-        return _validate_rotation_degrees(value, field="rotate")
-
-
-class EditPdfDocumentsSerializer(DocumentListSerializer, SourceModeValidationMixin):
-    operations = serializers.ListField(
-        child=PdfEditOperationSerializer(),
-        required=True,
-        allow_empty=False,
-    )
-    delete_original = serializers.BooleanField(required=False, default=False)
-    update_document = serializers.BooleanField(required=False, default=False)
-    include_metadata = serializers.BooleanField(required=False, default=True)
-    source_mode = serializers.CharField(
-        required=False,
-        default=bulk_edit.SourceModeChoices.LATEST_VERSION,
-    )
-    from_webui = serializers.BooleanField(required=False, default=False)
-
-    def validate(self, attrs):
-        documents = attrs["documents"]
-        if len(documents) > 1:
-            raise serializers.ValidationError(
-                "Edit PDF method only supports one document",
-            )
-
-        operations = attrs["operations"]
-
-        if any(op.get("doc", 0) >= len(operations) for op in operations):
-            raise serializers.ValidationError("doc index is out of bounds")
-
-        if attrs["update_document"]:
-            max_idx = max(op.get("doc", 0) for op in operations)
-            if max_idx > 0:
-                raise serializers.ValidationError(
-                    "update_document only allowed with a single output document",
-                )
-
-        doc = Document.objects.get(id=documents[0])
-        if doc.page_count:
-            for op in operations:
-                if op["page"] > doc.page_count:
-                    raise serializers.ValidationError(
-                        f"Page {op['page']} is out of bounds for document with {doc.page_count} pages.",
-                    )
-        return attrs
-
-
-class RemovePasswordDocumentsSerializer(
-    DocumentListSerializer,
-    SourceModeValidationMixin,
-):
-    password = serializers.CharField(required=True)
-    update_document = serializers.BooleanField(required=False, default=False)
-    delete_original = serializers.BooleanField(required=False, default=False)
-    include_metadata = serializers.BooleanField(required=False, default=True)
-    source_mode = serializers.CharField(
-        required=False,
-        default=bulk_edit.SourceModeChoices.LATEST_VERSION,
-    )
-    from_webui = serializers.BooleanField(required=False, default=False)
-
-
-class DeleteDocumentsSerializer(DocumentSelectionSerializer):
-    pass
-
-
-class ReprocessDocumentsSerializer(DocumentSelectionSerializer):
-    remote_ocr = serializers.BooleanField(required=False, default=False)
-
-
-class BulkEditSerializer(
-    SerializerWithPerms,
-    DocumentSelectionSerializer,
-    SetPermissionsMixin,
-    SourceModeValidationMixin,
-):
-    # TODO: remove this and related backwards compatibility code when API v9 is dropped
-    # split, delete_pages can be removed entirely
-    MOVED_DOCUMENT_ACTION_ENDPOINTS = {
-        "delete": "/api/documents/delete/",
-        "reprocess": "/api/documents/reprocess/",
-        "rotate": "/api/documents/rotate/",
-        "merge": "/api/documents/merge/",
-        "edit_pdf": "/api/documents/edit_pdf/",
-        "remove_password": "/api/documents/remove_password/",
-        "split": "/api/documents/edit_pdf/",
-        "delete_pages": "/api/documents/edit_pdf/",
-    }
-    LEGACY_DOCUMENT_ACTION_METHODS = tuple(MOVED_DOCUMENT_ACTION_ENDPOINTS.keys())
-
-    method = serializers.ChoiceField(
-        choices=[
-            "set_correspondent",
-            "set_document_type",
-            "set_storage_path",
-            "add_tag",
-            "remove_tag",
-            "modify_tags",
-            "modify_custom_fields",
-            "set_permissions",
-            *LEGACY_DOCUMENT_ACTION_METHODS,
-        ],
-        label="Method",
-        write_only=True,
-    )
-
-    parameters = serializers.DictField(allow_empty=True, default={}, write_only=True)
-    from_webui = serializers.BooleanField(required=False, default=False)
-
-    def _validate_tag_id_list(self, tags, name="tags") -> None:
-        if not isinstance(tags, list):
-            raise serializers.ValidationError(f"{name} must be a list")
-        if not all(isinstance(i, int) for i in tags):
-            raise serializers.ValidationError(f"{name} must be a list of integers")
-        count = Tag.objects.filter(id__in=tags).count()
-        if not count == len(tags):
-            raise serializers.ValidationError(
-                f"Some tags in {name} don't exist or were specified twice.",
-            )
-
-    def _validate_custom_field_id_list_or_dict(
-        self,
-        custom_fields,
-        name="custom_fields",
-    ) -> None:
-        ids = custom_fields
-        if isinstance(custom_fields, dict):
-            try:
-                ids = [int(i[0]) for i in custom_fields.items()]
-            except Exception as e:
-                logger.exception(f"Error validating custom fields: {e}")
-                raise serializers.ValidationError(
-                    f"{name} must be a list of integers or a dict of id:value pairs, see the log for details",
-                )
-        elif not isinstance(custom_fields, list) or not all(
-            isinstance(i, int) for i in ids
-        ):
-            raise serializers.ValidationError(
-                f"{name} must be a list of integers or a dict of id:value pairs",
-            )
-        count = CustomField.objects.filter(id__in=ids).count()
-        if not count == len(ids):
-            raise serializers.ValidationError(
-                f"Some custom fields in {name} don't exist or were specified twice.",
-            )
-
-    def _validate_custom_field_values(self, custom_fields, name):
-        if not isinstance(custom_fields, dict):
-            return custom_fields
-
-        validated = {}
-        errors = {}
-        for raw_field_id, value in custom_fields.items():
-            field_id = int(raw_field_id)
-            validator = CustomFieldInstanceSerializer(
-                data={"field": field_id, "value": value},
-                context=self.context,
-            )
-            if validator.is_valid():
-                validated[field_id] = validator.validated_data["value"]
-            else:
-                errors[str(field_id)] = validator.errors
-
-        if errors:
-            raise serializers.ValidationError({name: errors})
-
-        return validated
-
-    def validate_method(self, method):
-        if method == "set_correspondent":
-            return bulk_edit.set_correspondent
-        elif method == "set_document_type":
-            return bulk_edit.set_document_type
-        elif method == "set_storage_path":
-            return bulk_edit.set_storage_path
-        elif method == "add_tag":
-            return bulk_edit.add_tag
-        elif method == "remove_tag":
-            return bulk_edit.remove_tag
-        elif method == "modify_tags":
-            return bulk_edit.modify_tags
-        elif method == "modify_custom_fields":
-            return bulk_edit.modify_custom_fields
-        elif method == "delete":
-            return bulk_edit.delete
-        elif method == "redo_ocr" or method == "reprocess":
-            return bulk_edit.reprocess
-        elif method == "set_permissions":
-            return bulk_edit.set_permissions
-        elif method == "rotate":
-            return bulk_edit.rotate
-        elif method == "merge":
-            return bulk_edit.merge
-        elif method == "split":
-            return bulk_edit.split
-        elif method == "delete_pages":
-            return bulk_edit.delete_pages
-        elif method == "edit_pdf":
-            return bulk_edit.edit_pdf
-        elif method == "remove_password":
-            return bulk_edit.remove_password
-        else:
-            raise serializers.ValidationError("Unsupported method.")
-
-    def _validate_parameters_tags(self, parameters) -> None:
-        if "tag" in parameters:
-            tag_id = parameters["tag"]
-            try:
-                Tag.objects.get(id=tag_id)
-            except Tag.DoesNotExist:
-                raise serializers.ValidationError("Tag does not exist")
-        else:
-            raise serializers.ValidationError("tag not specified")
-
-    def _validate_parameters_document_type(self, parameters) -> None:
-        if "document_type" in parameters:
-            document_type_id = parameters["document_type"]
-            if document_type_id is None:
-                # None is ok
-                return
-            try:
-                DocumentType.objects.get(id=document_type_id)
-            except DocumentType.DoesNotExist:
-                raise serializers.ValidationError("Document type does not exist")
-        else:
-            raise serializers.ValidationError("document_type not specified")
-
-    def _validate_parameters_correspondent(self, parameters) -> None:
-        if "correspondent" in parameters:
-            correspondent_id = parameters["correspondent"]
-            if correspondent_id is None:
-                return
-            try:
-                Correspondent.objects.get(id=correspondent_id)
-            except Correspondent.DoesNotExist:
-                raise serializers.ValidationError("Correspondent does not exist")
-        else:
-            raise serializers.ValidationError("correspondent not specified")
-
-    def _validate_storage_path(self, parameters) -> None:
-        if "storage_path" in parameters:
-            storage_path_id = parameters["storage_path"]
-            if storage_path_id is None:
-                return
-            try:
-                StoragePath.objects.get(id=storage_path_id)
-            except StoragePath.DoesNotExist:
-                raise serializers.ValidationError(
-                    "Storage path does not exist",
-                )
-        else:
-            raise serializers.ValidationError("storage path not specified")
-
-    def _validate_parameters_modify_tags(self, parameters) -> None:
-        if "add_tags" in parameters:
-            self._validate_tag_id_list(parameters["add_tags"], "add_tags")
-        else:
-            raise serializers.ValidationError("add_tags not specified")
-
-        if "remove_tags" in parameters:
-            self._validate_tag_id_list(parameters["remove_tags"], "remove_tags")
-        else:
-            raise serializers.ValidationError("remove_tags not specified")
-
-    def _validate_parameters_modify_custom_fields(self, parameters) -> None:
-        if "add_custom_fields" in parameters:
-            self._validate_custom_field_id_list_or_dict(
-                parameters["add_custom_fields"],
-                "add_custom_fields",
-            )
-            parameters["add_custom_fields"] = self._validate_custom_field_values(
-                parameters["add_custom_fields"],
-                "add_custom_fields",
-            )
-        else:
-            raise serializers.ValidationError("add_custom_fields not specified")
-
-        if "remove_custom_fields" in parameters:
-            self._validate_custom_field_id_list_or_dict(
-                parameters["remove_custom_fields"],
-                "remove_custom_fields",
-            )
-        else:
-            raise serializers.ValidationError("remove_custom_fields not specified")
-
-    def _validate_owner(self, owner) -> User:
-        owner_field = serializers.PrimaryKeyRelatedField(queryset=User.objects.all())
-        try:
-            return owner_field.run_validation(owner)
-        except serializers.ValidationError as e:
-            raise serializers.ValidationError(
-                "Specified owner cannot be found",
-            ) from e
-
-    def _validate_parameters_set_permissions(self, parameters) -> None:
-        if "set_permissions" not in parameters:
-            raise serializers.ValidationError("set_permissions not specified")
-        set_permissions = parameters["set_permissions"]
-        if set_permissions is not None:
-            set_permissions = SetPermissionsSerializer().run_validation(
-                set_permissions,
-            )
-        parameters["set_permissions"] = self.validate_set_permissions(
-            set_permissions,
-        )
-        if "owner" in parameters and parameters["owner"] is not None:
-            parameters["owner"] = self._validate_owner(parameters["owner"]).pk
-        if "merge" not in parameters:
-            parameters["merge"] = False
-
-    def _validate_parameters_rotate(self, parameters) -> None:
-        if "degrees" not in parameters:
-            raise serializers.ValidationError("invalid rotation degrees")
-        try:
-            degrees = serializers.IntegerField().run_validation(parameters["degrees"])
-        except serializers.ValidationError as e:
-            raise serializers.ValidationError("invalid rotation degrees") from e
-        parameters["degrees"] = _validate_rotation_degrees(degrees)
-
-    def _validate_source_mode(self, parameters) -> None:
-        source_mode = parameters.get(
-            "source_mode",
-            bulk_edit.SourceModeChoices.LATEST_VERSION,
-        )
-        parameters["source_mode"] = self.validate_source_mode(source_mode)
-
-    def _validate_parameters_split(self, parameters, document_id) -> None:
-        if "pages" not in parameters:
-            raise serializers.ValidationError("pages not specified")
-        if not isinstance(parameters["pages"], str):
-            raise serializers.ValidationError("invalid pages specified")
-        page_count = Document.objects.get(id=document_id).page_count
-        if not page_count:
-            raise serializers.ValidationError("document page count is unknown")
-        pages = []
-        for group in parameters["pages"].split(","):
-            start, is_range, end = group.partition("-")
-            try:
-                first = int(start)
-                last = int(end) if is_range else first
-            except ValueError as e:
-                raise serializers.ValidationError("invalid pages specified") from e
-            # Bound the range before building it, a huge one would exhaust memory
-            if not 1 <= first <= last <= page_count:
-                raise serializers.ValidationError("invalid pages specified")
-            pages.append(list(range(first, last + 1)))
-        parameters["pages"] = pages
-
-        if "delete_originals" in parameters:
-            if not isinstance(parameters["delete_originals"], bool):
-                raise serializers.ValidationError("delete_originals must be a boolean")
-        else:
-            parameters["delete_originals"] = False
-
-    def _validate_parameters_delete_pages(self, parameters) -> None:
-        if "pages" not in parameters:
-            raise serializers.ValidationError("pages not specified")
-        if not isinstance(parameters["pages"], list):
-            raise serializers.ValidationError("pages must be a list")
-        if not all(isinstance(i, int) for i in parameters["pages"]):
-            raise serializers.ValidationError("pages must be a list of integers")
-
-    def _validate_parameters_merge(self, parameters) -> None:
-        if "delete_originals" in parameters:
-            if not isinstance(parameters["delete_originals"], bool):
-                raise serializers.ValidationError("delete_originals must be a boolean")
-        else:
-            parameters["delete_originals"] = False
-        if "archive_fallback" in parameters:
-            if not isinstance(parameters["archive_fallback"], bool):
-                raise serializers.ValidationError("archive_fallback must be a boolean")
-        else:
-            parameters["archive_fallback"] = False
-
-    def _validate_parameters_edit_pdf(self, parameters, document_id) -> None:
-        if "operations" not in parameters:
-            raise serializers.ValidationError("operations not specified")
-        operations_field = serializers.ListField(
-            child=PdfEditOperationSerializer(),
-            allow_empty=False,
-        )
-        try:
-            operations = operations_field.run_validation(parameters["operations"])
-        except serializers.ValidationError as e:
-            # Key the errors under "operations" so they match what the
-            # dedicated edit_pdf endpoint returns
-            raise serializers.ValidationError({"operations": e.detail}) from e
-        parameters["operations"] = operations
-
-        if "update_document" in parameters:
-            if not isinstance(parameters["update_document"], bool):
-                raise serializers.ValidationError("update_document must be a boolean")
-        else:
-            parameters["update_document"] = False
-        if "include_metadata" in parameters:
-            if not isinstance(parameters["include_metadata"], bool):
-                raise serializers.ValidationError("include_metadata must be a boolean")
-        else:
-            parameters["include_metadata"] = True
-
-        if any(op.get("doc", 0) >= len(operations) for op in operations):
-            raise serializers.ValidationError("doc index is out of bounds")
-
-        if parameters["update_document"]:
-            max_idx = max(op.get("doc", 0) for op in operations)
-            if max_idx > 0:
-                raise serializers.ValidationError(
-                    "update_document only allowed with a single output document",
-                )
-
-        doc = Document.objects.get(id=document_id)
-        # doc existence is already validated
-        if doc.page_count:
-            for op in operations:
-                if op["page"] > doc.page_count:
-                    raise serializers.ValidationError(
-                        f"Page {op['page']} is out of bounds for document with {doc.page_count} pages.",
-                    )
-
-    def _validate_parameters_reprocess(self, parameters) -> None:
-        if "remote_ocr" in parameters:
-            if not isinstance(parameters["remote_ocr"], bool):
-                raise serializers.ValidationError("remote_ocr must be a boolean")
-        else:
-            parameters["remote_ocr"] = False
-
-    def validate_parameters_remove_password(self, parameters):
-        if "password" not in parameters:
-            raise serializers.ValidationError("password not specified")
-        if not isinstance(parameters["password"], str):
-            raise serializers.ValidationError("password must be a string")
-
-    def validate(self, attrs):
-        attrs = super().validate(attrs)
-
-        if attrs.get("all", False) and attrs["method"] in [
-            bulk_edit.merge,
-            bulk_edit.split,
-            bulk_edit.delete_pages,
-            bulk_edit.edit_pdf,
-            bulk_edit.remove_password,
-        ]:
-            raise serializers.ValidationError(
-                "This method does not support all=true.",
-            )
-
-        method = attrs["method"]
-        parameters = attrs["parameters"]
-
-        if "source_mode" in parameters:
-            self._validate_source_mode(parameters)
-
-        if method == bulk_edit.set_correspondent:
-            self._validate_parameters_correspondent(parameters)
-        elif method == bulk_edit.set_document_type:
-            self._validate_parameters_document_type(parameters)
-        elif method == bulk_edit.add_tag or method == bulk_edit.remove_tag:
-            self._validate_parameters_tags(parameters)
-        elif method == bulk_edit.modify_tags:
-            self._validate_parameters_modify_tags(parameters)
-        elif method == bulk_edit.set_storage_path:
-            self._validate_storage_path(parameters)
-        elif method == bulk_edit.modify_custom_fields:
-            self._validate_parameters_modify_custom_fields(parameters)
-        elif method == bulk_edit.set_permissions:
-            self._validate_parameters_set_permissions(parameters)
-        elif method == bulk_edit.rotate:
-            self._validate_parameters_rotate(parameters)
-        elif method == bulk_edit.split:
-            if len(attrs["documents"]) > 1:
-                raise serializers.ValidationError(
-                    "Split method only supports one document",
-                )
-            self._validate_parameters_split(parameters, attrs["documents"][0])
-        elif method == bulk_edit.delete_pages:
-            if len(attrs["documents"]) > 1:
-                raise serializers.ValidationError(
-                    "Delete pages method only supports one document",
-                )
-            self._validate_parameters_delete_pages(parameters)
-        elif method == bulk_edit.merge:
-            self._validate_parameters_merge(parameters)
-        elif method == bulk_edit.edit_pdf:
-            if len(attrs["documents"]) > 1:
-                raise serializers.ValidationError(
-                    "Edit PDF method only supports one document",
-                )
-            self._validate_parameters_edit_pdf(parameters, attrs["documents"][0])
-        elif method == bulk_edit.remove_password:
-            self.validate_parameters_remove_password(parameters)
-        elif method == bulk_edit.reprocess:
-            self._validate_parameters_reprocess(parameters)
-
-        return attrs
-
-
-class BulkDownloadSerializer(DocumentSelectionSerializer):
-    content = serializers.ChoiceField(
-        choices=["archive", "originals", "both"],
-        default="archive",
-    )
-
-    compression = serializers.ChoiceField(
-        choices=["none", "deflated", "bzip2", "lzma"],
-        default="none",
-    )
-
-    follow_formatting = serializers.BooleanField(
-        default=False,
-    )
-
-    def validate_compression(self, compression):
-        import zipfile
-
-        return {
-            "none": zipfile.ZIP_STORED,
-            "deflated": zipfile.ZIP_DEFLATED,
-            "bzip2": zipfile.ZIP_BZIP2,
-            "lzma": zipfile.ZIP_LZMA,
-        }[compression]
-
-
-class BulkEditObjectsSerializer(SerializerWithPerms, SetPermissionsMixin):
-    objects = serializers.ListField(
-        required=False,
-        allow_empty=True,
-        label="Objects",
-        write_only=True,
-        child=serializers.IntegerField(),
-    )
-
-    all = serializers.BooleanField(
-        default=False,
-        required=False,
-        write_only=True,
-    )
-
-    filters = serializers.DictField(
-        required=False,
-        allow_empty=True,
-        write_only=True,
-    )
-
-    object_type = serializers.ChoiceField(
-        choices=[
-            "tags",
-            "correspondents",
-            "document_types",
-            "storage_paths",
-        ],
-        label="Object Type",
-        write_only=True,
-    )
-
-    operation = serializers.ChoiceField(
-        choices=[
-            "set_permissions",
-            "delete",
-        ],
-        label="Operation",
-        required=True,
-        write_only=True,
-    )
-
-    owner = serializers.PrimaryKeyRelatedField(
-        queryset=User.objects.all(),
-        required=False,
-        allow_null=True,
-    )
-
-    permissions = SetPermissionsSerializer(
-        label="Set permissions",
-        required=False,
-        write_only=True,
-    )
-
-    merge = serializers.BooleanField(
-        default=False,
-        write_only=True,
-        required=False,
-    )
-
-    def get_object_class(self, object_type):
-        object_class = None
-        if object_type == "tags":
-            object_class = Tag
-        elif object_type == "correspondents":
-            object_class = Correspondent
-        elif object_type == "document_types":
-            object_class = DocumentType
-        elif object_type == "storage_paths":
-            object_class = StoragePath
-        return object_class
-
-    def _validate_objects(self, objects, object_type):
-        if not isinstance(objects, list):
-            raise serializers.ValidationError("objects must be a list")
-        if not all(isinstance(i, int) for i in objects):
-            raise serializers.ValidationError("objects must be a list of integers")
-        object_class = self.get_object_class(object_type)
-        count = object_class.objects.filter(id__in=objects).count()
-        if not count == len(objects):
-            raise serializers.ValidationError(
-                "Some ids in objects don't exist or were specified twice.",
-            )
-        return objects
-
-    def _validate_permissions(self, permissions) -> dict:
-        return self.validate_set_permissions(
-            permissions,
-        )
-
-    def validate(self, attrs):
-        object_type = attrs["object_type"]
-        objects = attrs.get("objects")
-        apply_to_all = attrs.get("all", False)
-        operation = attrs.get("operation")
-
-        if apply_to_all:
-            attrs.setdefault("objects", [])
-        else:
-            if objects is None:
-                raise serializers.ValidationError(
-                    "objects is required unless all is true.",
-                )
-            if len(objects) == 0:
-                raise serializers.ValidationError("objects must not be empty")
-            self._validate_objects(objects, object_type)
-
-        if operation == "set_permissions":
-            permissions = attrs.get("permissions")
-            if permissions is not None:
-                if not permissions:
-                    raise serializers.ValidationError(
-                        "permissions must not be empty",
-                    )
-                attrs["permissions"] = self._validate_permissions(permissions)
-
-        return attrs

src/documents/serialisers/documents.py

@@ -1,484 +0,0 @@
-from __future__ import annotations
-
-import logging
-from typing import TYPE_CHECKING
-from typing import Any
-from typing import TypedDict
-
-from django.conf import settings
-from django.contrib.auth.models import User
-from django.db.models import Q
-from django.utils.dateparse import parse_datetime
-from django.utils.timezone import get_current_timezone
-from django.utils.timezone import is_naive
-from django.utils.timezone import make_aware
-from drf_spectacular.utils import extend_schema_field
-from drf_spectacular.utils import extend_schema_serializer
-from drf_writable_nested.serializers import NestedUpdateMixin
-from rest_framework import serializers
-from rest_framework.fields import SerializerMethodField
-
-if settings.AUDIT_LOG_ENABLED:
-    from auditlog.context import set_actor
-
-
-from documents import bulk_edit
-from documents.models import CustomField
-from documents.models import CustomFieldInstance
-from documents.models import Document
-from documents.models import Tag
-from documents.permissions import permitted_document_ids
-from documents.versioning import has_prefetched_effective_content
-from documents.versioning import sort_versions_newest_first
-
-from .base import DocumentUpdateFieldsModelSerializer
-from .base import NotesSerializer
-from .base import OwnedObjectListSerializer
-from .base import OwnedObjectSerializer
-from .metadata import CorrespondentField
-from .metadata import CustomFieldInstanceSerializer
-from .metadata import DocumentTypeField
-from .metadata import StoragePathField
-from .metadata import TagsField
-from .upload import PostDocumentSerializer
-
-if TYPE_CHECKING:
-    from datetime import datetime
-
-    from django.db.models.query import QuerySet
-    from rest_framework.relations import ManyRelatedField
-    from rest_framework.relations import RelatedField
-
-
-logger = logging.getLogger("paperless.serializers")
-
-
-def _get_viewable_duplicates(
-    document: Document,
-    user: User | None,
-) -> QuerySet[Document]:
-    checksums = {document.checksum}
-    if document.archive_checksum:
-        checksums.add(document.archive_checksum)
-    duplicates = Document.global_objects.filter(
-        Q(checksum__in=checksums) | Q(archive_checksum__in=checksums),
-    ).exclude(pk=document.pk)
-    duplicates = duplicates.filter(root_document__isnull=True)
-    duplicates = duplicates.order_by("-created")
-    allowed_ids = permitted_document_ids(user, include_deleted=True)
-    return duplicates.filter(id__in=allowed_ids)
-
-
-class DuplicateDocumentSummarySerializer(serializers.Serializer[dict[str, Any]]):
-    id = serializers.IntegerField()
-    title = serializers.CharField()
-    deleted_at = serializers.DateTimeField(allow_null=True)
-
-
-class _DocumentVersionInfo(TypedDict):
-    id: int
-    added: datetime
-    version_label: str | None
-    checksum: str | None
-    is_root: bool
-
-
-class DocumentVersionInfoSerializer(serializers.Serializer[_DocumentVersionInfo]):
-    id = serializers.IntegerField()
-    added = serializers.DateTimeField()
-    version_label = serializers.CharField(required=False, allow_null=True)
-    checksum = serializers.CharField(required=False, allow_null=True)
-    is_root = serializers.BooleanField()
-
-
-@extend_schema_serializer(
-    deprecate_fields=["created_date"],
-)
-class DocumentSerializer(
-    OwnedObjectSerializer,
-    NestedUpdateMixin,
-    DocumentUpdateFieldsModelSerializer,
-):
-    correspondent = CorrespondentField(allow_null=True)
-    tags = TagsField(many=True)
-    document_type = DocumentTypeField(allow_null=True)
-    storage_path = StoragePathField(allow_null=True)
-
-    original_file_name = SerializerMethodField()
-    archived_file_name = SerializerMethodField()
-    created_date = serializers.DateField(required=False)
-    page_count = SerializerMethodField()
-    duplicate_documents = SerializerMethodField()
-
-    notes = NotesSerializer(many=True, required=False, read_only=True)
-    root_document: RelatedField[Document, Document, Any] | ManyRelatedField = (
-        serializers.PrimaryKeyRelatedField(read_only=True)
-    )
-    versions = SerializerMethodField()
-
-    custom_fields = CustomFieldInstanceSerializer(
-        many=True,
-        allow_null=False,
-        required=False,
-    )
-
-    owner = serializers.PrimaryKeyRelatedField(
-        queryset=User.objects.all(),
-        required=False,
-        allow_null=True,
-    )
-
-    remove_inbox_tags = serializers.BooleanField(
-        default=False,
-        write_only=True,
-        allow_null=True,
-        required=False,
-    )
-
-    def get_page_count(self, obj) -> int | None:
-        # Like content versions get their own page count from the newest version,
-        # use the prefetched versions cache to avoid an extra query
-        prefetched_cache = getattr(obj, "_prefetched_objects_cache", None)
-        prefetched_versions = (
-            prefetched_cache.get("versions")
-            if isinstance(prefetched_cache, dict)
-            else None
-        )
-        if obj.root_document_id is None and prefetched_versions:
-            return sort_versions_newest_first(prefetched_versions)[0].page_count
-        return obj.page_count
-
-    @extend_schema_field(DuplicateDocumentSummarySerializer(many=True))
-    def get_duplicate_documents(self, obj):
-        view = self.context.get("view")
-        if view and getattr(view, "action", None) != "retrieve":
-            return []
-        request = self.context.get("request")
-        user = request.user if request else None
-        duplicates = _get_viewable_duplicates(obj, user)
-        return list(duplicates.values("id", "title", "deleted_at"))
-
-    @extend_schema_field(DocumentVersionInfoSerializer(many=True))
-    def get_versions(self, obj):
-        root_doc = obj if obj.root_document_id is None else obj.root_document
-        if root_doc is None:
-            return []
-
-        prefetched_cache = getattr(obj, "_prefetched_objects_cache", None)
-        prefetched_versions = (
-            prefetched_cache.get("versions")
-            if isinstance(prefetched_cache, dict)
-            else None
-        )
-
-        versions: list[Document]
-        if prefetched_versions is not None:
-            versions = [*prefetched_versions, root_doc]
-        else:
-            versions_qs = Document.objects.filter(root_document=root_doc).only(
-                "id",
-                "added",
-                "checksum",
-                "version_label",
-                "root_document_id",
-                "version_index",
-            )
-            versions = [*versions_qs, root_doc]
-
-        versions = sort_versions_newest_first(versions)
-
-        def build_info(doc: Document) -> _DocumentVersionInfo:
-            return {
-                "id": doc.id,
-                "added": doc.added,
-                "version_label": doc.version_label,
-                "checksum": doc.checksum,
-                "is_root": doc.id == root_doc.id,
-            }
-
-        return [build_info(doc) for doc in versions]
-
-    def get_original_file_name(self, obj) -> str | None:
-        return obj.original_filename
-
-    def get_archived_file_name(self, obj) -> str | None:
-        if obj.has_archive_version:
-            return obj.get_public_filename(archive=True)
-        else:
-            return None
-
-    def to_representation(self, instance):
-        doc = super().to_representation(instance)
-        if "content" in self.fields and has_prefetched_effective_content(instance):
-            # Only resolve version-aware content when it's cheap: an SQL
-            # annotation or a versions prefetch is already on the instance.
-            # A caller that set up neither (e.g. TrashView, GlobalSearchView,
-            # which build their own querysets) gets the document's own,
-            # unresolved content instead of paying for an extra per-instance
-            # query -- same as before effective_content resolution existed.
-            doc["content"] = instance.get_effective_content() or ""
-        if self.truncate_content and "content" in self.fields:
-            doc["content"] = doc.get("content")[0:550]
-        return doc
-
-    def to_internal_value(self, data):
-        if (
-            "created" in data
-            and isinstance(data["created"], str)
-            and ":" in data["created"]
-        ):
-            # Handle old format of isoformat datetime string
-            parsed = parse_datetime(data["created"])
-            if parsed:
-                if is_naive(parsed):
-                    parsed = make_aware(parsed, get_current_timezone())
-                data["created"] = parsed.astimezone().date()
-        return super().to_internal_value(data)
-
-    def validate(self, attrs):
-        if (
-            "archive_serial_number" in attrs
-            and attrs["archive_serial_number"] is not None
-            and len(str(attrs["archive_serial_number"])) > 0
-            and Document.deleted_objects.filter(
-                archive_serial_number=attrs["archive_serial_number"],
-            ).exists()
-        ):
-            raise serializers.ValidationError(
-                {
-                    "archive_serial_number": [
-                        "Document with this Archive Serial Number already exists in the trash.",
-                    ],
-                },
-            )
-        return super().validate(attrs)
-
-    def update(self, instance: Document, validated_data):
-        if "created_date" in validated_data:
-            if "created" not in validated_data:
-                validated_data["created"] = validated_data["created_date"]
-            logger.warning(
-                "created_date is deprecated, use created instead",
-            )
-            validated_data.pop("created_date")
-        if instance.custom_fields.count() > 0 and "custom_fields" in validated_data:
-            incoming_custom_fields = [
-                field["field"] for field in validated_data["custom_fields"]
-            ]
-            for custom_field_instance in instance.custom_fields.filter(
-                field__data_type=CustomField.FieldDataType.DOCUMENTLINK,
-            ):
-                if (
-                    custom_field_instance.field not in incoming_custom_fields
-                    and custom_field_instance.value is not None
-                ):
-                    # Doc link field is being removed entirely
-                    for doc_id in custom_field_instance.value:
-                        bulk_edit.remove_doclink(
-                            instance,
-                            custom_field_instance.field,
-                            doc_id,
-                        )
-        if "tags" in validated_data:
-            # Respect tag hierarchy on updates:
-            # - Adding a child adds its ancestors
-            # - Removing a parent removes all its descendants
-            prev_tags = set(instance.tags.all())
-            requested_tags = set(validated_data["tags"])
-
-            # Tags newly added in this update and the ancestors they require
-            added_tags = requested_tags - prev_tags
-            required_by_add_tags = set(added_tags)
-            for t in added_tags:
-                required_by_add_tags.update(t.get_ancestors())
-
-            # Tags being removed in this update and all descendants, except
-            # those required by a tag that is being added in this same update
-            removed_tags = prev_tags - requested_tags
-            blocked_tags = set(removed_tags)
-            for t in removed_tags:
-                blocked_tags.update(t.get_descendants())
-            blocked_tags.difference_update(required_by_add_tags)
-
-            # Add all parent tags
-            final_tags = set(requested_tags)
-            for t in requested_tags:
-                final_tags.update(t.get_ancestors())
-
-            # Drop removed parents and their descendants
-            final_tags.difference_update(blocked_tags)
-
-            validated_data["tags"] = list(final_tags)
-        if validated_data.get("remove_inbox_tags"):
-            current_tag_ids = {t.pk for t in instance.tags.all()}
-            tags = (
-                validated_data["tags"]
-                if "tags" in validated_data
-                else list(instance.tags.all())
-            )
-
-            # Tags newly added in this update, plus their ancestors, are kept
-            keep_ids: set[int] = set()
-            for tag in tags:
-                if tag.pk not in current_tag_ids:
-                    keep_ids.add(tag.pk)
-                    keep_ids.update(int(pk) for pk in tag.get_ancestors_pks())
-
-            # Remove inbox tags and their descendants, except those being kept
-            remove_ids: set[int] = set()
-            for inbox_tag in (
-                Tag.objects.filter(is_inbox_tag=True)
-                .exclude(pk__in=keep_ids)
-                .only("pk", "tn_descendants_pks")
-            ):
-                remove_ids.add(inbox_tag.pk)
-                remove_ids.update(int(pk) for pk in inbox_tag.get_descendants_pks())
-
-            validated_data["tags"] = [t for t in tags if t.pk not in remove_ids]
-
-        if settings.AUDIT_LOG_ENABLED:
-            with set_actor(self.user):
-                super().update(instance, validated_data)
-        else:
-            super().update(instance, validated_data)
-
-        # hard delete custom field instances that were soft deleted
-        CustomFieldInstance.deleted_objects.filter(document=instance).delete()
-        return instance
-
-    def __init__(self, *args, **kwargs) -> None:
-        self.truncate_content = kwargs.pop("truncate_content", False)
-
-        # return full permissions if we're doing a PATCH or PUT
-        context = kwargs.get("context")
-        if context is not None and (
-            context.get("request").method == "PATCH"
-            or context.get("request").method == "PUT"
-        ):
-            kwargs["full_perms"] = True
-
-        super().__init__(*args, **kwargs)
-
-    class Meta:
-        model = Document
-        fields = (
-            "id",
-            "correspondent",
-            "document_type",
-            "storage_path",
-            "title",
-            "content",
-            "tags",
-            "created",
-            "created_date",
-            "modified",
-            "added",
-            "deleted_at",
-            "archive_serial_number",
-            "original_file_name",
-            "archived_file_name",
-            "duplicate_documents",
-            "owner",
-            "permissions",
-            "user_can_change",
-            "is_shared_by_requester",
-            "set_permissions",
-            "notes",
-            "custom_fields",
-            "remove_inbox_tags",
-            "page_count",
-            "mime_type",
-            "root_document",
-            "versions",
-        )
-        read_only_fields = ("deleted_at",)
-        list_serializer_class = OwnedObjectListSerializer
-
-
-class SearchResultListSerializer(serializers.ListSerializer[Document]):
-    def to_representation(self, hits):
-        document_ids = [hit["id"] for hit in hits]
-        # Fetch all Document objects in the list in one SQL query.
-        documents = self.child.fetch_documents(document_ids)
-        self.child.context["documents"] = documents
-        # Also check if they are shared with other users / groups.
-        self.child.context["shared_object_pks"] = self.child.get_shared_object_pks(
-            documents.values(),
-        )
-
-        return super().to_representation(hits)
-
-
-class SearchResultSerializer(DocumentSerializer):
-    @staticmethod
-    def fetch_documents(ids):
-        """
-        Return a dict that maps given document IDs to Document objects.
-        """
-        return {
-            document.id: document
-            for document in Document.objects.select_related(
-                "correspondent",
-                "storage_path",
-                "document_type",
-                "owner",
-            )
-            .prefetch_related("tags", "custom_fields", "notes")
-            .filter(id__in=ids)
-        }
-
-    def to_representation(self, hit):
-        # Again we first check if the parent has already fetched the documents.
-        documents = self.context.get("documents")
-        # Otherwise we fetch this document.
-        if documents is None:  # pragma: no cover
-            # In practice we only serialize **lists** of SearchHit dicts.
-            # Keeping this check for completeness but marking it no cover for now.
-            documents = self.fetch_documents([hit["id"]])
-        document = documents[hit["id"]]
-
-        highlights = hit.get("highlights", {})
-        r = super().to_representation(document)
-        r["__search_hit__"] = {
-            "score": hit["score"],
-            "highlights": highlights.get("content", ""),
-            "note_highlights": highlights.get("notes") or None,
-            "rank": hit["rank"],
-        }
-
-        return r
-
-    class Meta(DocumentSerializer.Meta):
-        list_serializer_class = SearchResultListSerializer
-
-
-class DocumentVersionSerializer(serializers.Serializer[dict[str, Any]]):
-    document = serializers.FileField(
-        label="Document",
-        write_only=True,
-    )
-    version_label = serializers.CharField(
-        label="Version label",
-        required=False,
-        allow_blank=True,
-        allow_null=True,
-        max_length=64,
-    )
-
-    validate_document = PostDocumentSerializer().validate_document
-
-
-class DocumentVersionLabelSerializer(serializers.Serializer[dict[str, str | None]]):
-    version_label = serializers.CharField(
-        label="Version label",
-        required=True,
-        allow_blank=True,
-        allow_null=True,
-        max_length=64,
-    )
-
-    def validate_version_label(self, value):
-        if value is None:
-            return None
-        normalized = value.strip()
-        return normalized or None

src/documents/serialisers/metadata.py

@@ -1,557 +0,0 @@
-from __future__ import annotations
-
-import logging
-import math
-import re
-from decimal import Decimal
-from typing import Any
-
-from django.core.exceptions import ValidationError
-from django.core.validators import DecimalValidator
-from django.core.validators import MaxLengthValidator
-from django.core.validators import MaxValueValidator
-from django.core.validators import MinValueValidator
-from django.core.validators import RegexValidator
-from django.core.validators import integer_validator
-from django.db.models import Count
-from django.db.models.functions import Lower
-from django.utils.crypto import get_random_string
-from django.utils.translation import gettext as _
-from drf_spectacular.utils import extend_schema_field
-from rest_framework import serializers
-from rest_framework.exceptions import PermissionDenied
-from rest_framework.filters import OrderingFilter
-
-from documents import bulk_edit
-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 PaperlessTask
-from documents.models import StoragePath
-from documents.models import Tag
-from documents.permissions import get_document_count_filter_for_user
-from documents.permissions import permitted_document_ids
-from documents.permissions import restrict_queryset_to_visible
-from documents.templating.filepath import validate_filepath_template_and_render
-from documents.templating.utils import convert_format_str_to_template_format
-from documents.validators import uri_validator
-
-from .base import MatchingModelSerializer
-from .base import OwnedObjectSerializer
-from .base import ReadWriteSerializerMethodField
-from .base import SerializerWithPerms
-
-logger = logging.getLogger("paperless.serializers")
-
-
-class CorrespondentSerializer(MatchingModelSerializer, OwnedObjectSerializer):
-    last_correspondence = serializers.DateField(read_only=True, required=False)
-
-    class Meta:
-        model = Correspondent
-        fields = (
-            "id",
-            "slug",
-            "name",
-            "match",
-            "matching_algorithm",
-            "is_insensitive",
-            "document_count",
-            "last_correspondence",
-            "owner",
-            "permissions",
-            "user_can_change",
-            "set_permissions",
-        )
-
-
-class DocumentTypeSerializer(MatchingModelSerializer, OwnedObjectSerializer):
-    class Meta:
-        model = DocumentType
-        fields = (
-            "id",
-            "slug",
-            "name",
-            "match",
-            "matching_algorithm",
-            "is_insensitive",
-            "document_count",
-            "owner",
-            "permissions",
-            "user_can_change",
-            "set_permissions",
-        )
-
-
-class DeprecatedColors:
-    COLOURS = (
-        (1, "#a6cee3"),
-        (2, "#1f78b4"),
-        (3, "#b2df8a"),
-        (4, "#33a02c"),
-        (5, "#fb9a99"),
-        (6, "#e31a1c"),
-        (7, "#fdbf6f"),
-        (8, "#ff7f00"),
-        (9, "#cab2d6"),
-        (10, "#6a3d9a"),
-        (11, "#b15928"),
-        (12, "#000000"),
-        (13, "#cccccc"),
-    )
-
-
-@extend_schema_field(
-    serializers.ChoiceField(
-        choices=DeprecatedColors.COLOURS,
-    ),
-)
-class ColorField(serializers.Field):
-    def to_internal_value(self, data):
-        for id, color in DeprecatedColors.COLOURS:
-            if id == data:
-                return color
-        raise serializers.ValidationError
-
-    def to_representation(self, value):
-        for id, color in DeprecatedColors.COLOURS:
-            if color == value:
-                return id
-        return 1
-
-
-class TagSerializer(MatchingModelSerializer, OwnedObjectSerializer):
-    def get_text_color(self, obj) -> str:
-        try:
-            h = obj.color.lstrip("#")
-            rgb = tuple(int(h[i : i + 2], 16) / 256 for i in (0, 2, 4))
-            luminance = math.sqrt(
-                0.299 * math.pow(rgb[0], 2)
-                + 0.587 * math.pow(rgb[1], 2)
-                + 0.114 * math.pow(rgb[2], 2),
-            )
-            return "#ffffff" if luminance < 0.53 else "#000000"
-        except ValueError:
-            return "#000000"
-
-    text_color = serializers.SerializerMethodField()
-
-    # map to treenode's tn_parent
-    parent = serializers.PrimaryKeyRelatedField(
-        queryset=Tag.objects.all(),
-        allow_null=True,
-        required=False,
-        source="tn_parent",
-    )
-
-    @extend_schema_field(
-        field=serializers.ListSerializer(
-            child=serializers.PrimaryKeyRelatedField(
-                queryset=Tag.objects.all(),
-            ),
-        ),
-    )
-    def get_children(self, obj):
-        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 = (
-                obj.get_children_queryset()
-                .select_related("owner")
-                .annotate(document_count=Count("documents", filter=filter_q))
-            )
-            user = getattr(request, "user", None) if request else self.user
-            children = restrict_queryset_to_visible(children, user, "view_tag")
-
-            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)
-
-        if not children:
-            return []
-
-        serializer = TagSerializer(
-            children,
-            many=True,
-            user=self.user,
-            full_perms=self.full_perms,
-            all_fields=self.all_fields,
-            context=self.context,
-        )
-        return serializer.data
-
-    # children as nested Tag objects
-    children = serializers.SerializerMethodField()
-
-    class Meta:
-        model = Tag
-        fields = (
-            "id",
-            "slug",
-            "name",
-            "color",
-            "text_color",
-            "match",
-            "matching_algorithm",
-            "is_insensitive",
-            "is_inbox_tag",
-            "document_count",
-            "owner",
-            "permissions",
-            "user_can_change",
-            "set_permissions",
-            "parent",
-            "children",
-        )
-
-    def validate_color(self, color):
-        regex = r"#[0-9a-fA-F]{6}"
-        if not re.match(regex, color):
-            raise serializers.ValidationError(_("Invalid color."))
-        return color
-
-    def validate(self, attrs):
-        # Validate when changing parent
-        parent = attrs.get(
-            "tn_parent",
-            self.instance.get_parent() if self.instance else None,
-        )
-
-        if self.instance:
-            # Temporarily set parent on the instance if updating and use model clean()
-            original_parent = self.instance.get_parent()
-            try:
-                # Temporarily set tn_parent in-memory to validate clean()
-                self.instance.tn_parent = parent
-                self.instance.clean()
-            except ValidationError as e:
-                logger.debug("Tag parent validation failed: %s", e)
-                raise e
-            finally:
-                self.instance.tn_parent = original_parent
-        else:
-            # For new instances, create a transient Tag and validate
-            temp = Tag(tn_parent=parent)
-            try:
-                temp.clean()
-            except ValidationError as e:
-                logger.debug("Tag parent validation failed: %s", e)
-                raise e
-
-        return super().validate(attrs)
-
-
-class CorrespondentField(serializers.PrimaryKeyRelatedField[Correspondent]):
-    def get_queryset(self):
-        return Correspondent.objects.all()
-
-
-class TagsField(serializers.PrimaryKeyRelatedField[Tag]):
-    def get_queryset(self):
-        return Tag.objects.all()
-
-
-class DocumentTypeField(serializers.PrimaryKeyRelatedField[DocumentType]):
-    def get_queryset(self):
-        return DocumentType.objects.all()
-
-
-class StoragePathField(serializers.PrimaryKeyRelatedField[StoragePath]):
-    def get_queryset(self):
-        return StoragePath.objects.all()
-
-
-class CustomFieldSerializer(serializers.ModelSerializer[CustomField]):
-    data_type = serializers.ChoiceField(
-        choices=CustomField.FieldDataType,
-        read_only=False,
-    )
-
-    document_count = serializers.IntegerField(read_only=True)
-
-    class Meta:
-        model = CustomField
-        fields = [
-            "id",
-            "name",
-            "data_type",
-            "extra_data",
-            "document_count",
-        ]
-
-    def validate(self, attrs):
-        # TODO: remove pending https://github.com/encode/django-rest-framework/issues/7173
-        name = attrs.get(
-            "name",
-            self.instance.name if hasattr(self.instance, "name") else None,
-        )
-        objects = (
-            self.Meta.model.objects.exclude(
-                pk=self.instance.pk,
-            )
-            if self.instance is not None
-            else self.Meta.model.objects.all()
-        )
-        if ("name" in attrs) and objects.filter(
-            name=name,
-        ).exists():
-            raise serializers.ValidationError(
-                {"error": "Object violates name unique constraint"},
-            )
-        if (
-            "data_type" in attrs
-            and attrs["data_type"] == CustomField.FieldDataType.SELECT
-        ) or (
-            self.instance
-            and self.instance.data_type == CustomField.FieldDataType.SELECT
-        ):
-            if (
-                "extra_data" not in attrs
-                or "select_options" not in attrs["extra_data"]
-                or not isinstance(attrs["extra_data"]["select_options"], list)
-                or len(attrs["extra_data"]["select_options"]) == 0
-                or not all(
-                    len(option.get("label", "")) > 0
-                    for option in attrs["extra_data"]["select_options"]
-                )
-            ):
-                raise serializers.ValidationError(
-                    {"error": "extra_data.select_options must be a valid list"},
-                )
-            # labels are valid, generate ids if not present
-            for option in attrs["extra_data"]["select_options"]:
-                if option.get("id") is None:
-                    option["id"] = get_random_string(length=16)
-        elif (
-            "data_type" in attrs
-            and attrs["data_type"] == CustomField.FieldDataType.MONETARY
-            and "extra_data" in attrs
-            and "default_currency" in attrs["extra_data"]
-            and attrs["extra_data"]["default_currency"] is not None
-            and (
-                not isinstance(attrs["extra_data"]["default_currency"], str)
-                or (
-                    len(attrs["extra_data"]["default_currency"]) > 0
-                    and len(attrs["extra_data"]["default_currency"]) != 3
-                )
-            )
-        ):
-            raise serializers.ValidationError(
-                {"error": "extra_data.default_currency must be a 3-character string"},
-            )
-        return super().validate(attrs)
-
-
-def validate_documentlink_targets(user, doc_ids):
-    if Document.objects.filter(id__in=doc_ids).count() != len(doc_ids):
-        raise serializers.ValidationError(
-            "Some documents in value don't exist or were specified twice.",
-        )
-
-    if user is None:
-        return
-
-    if (
-        Document.objects.filter(id__in=doc_ids)
-        .exclude(id__in=permitted_document_ids(user, perm="change_document"))
-        .exists()
-    ):
-        raise PermissionDenied(
-            _("Insufficient permissions."),
-        )
-
-
-class CustomFieldInstanceSerializer(serializers.ModelSerializer[CustomFieldInstance]):
-    field = serializers.PrimaryKeyRelatedField(queryset=CustomField.objects.all())
-    value = ReadWriteSerializerMethodField(allow_null=True)
-
-    def create(self, validated_data):
-        # An instance is attached to a document
-        document: Document = validated_data["document"]
-        # And to a CustomField
-        custom_field: CustomField = validated_data["field"]
-        # This key must exist, as it is validated
-        data_store_name = CustomFieldInstance.get_value_field_name(
-            custom_field.data_type,
-        )
-
-        if custom_field.data_type == CustomField.FieldDataType.DOCUMENTLINK:
-            # prior to update so we can look for any docs that are going to be removed
-            bulk_edit.reflect_doclinks(document, custom_field, validated_data["value"])
-
-        # Actually update or create the instance, providing the value
-        # to fill in the correct attribute based on the type
-        instance, _ = CustomFieldInstance.objects.update_or_create(
-            document=document,
-            field=custom_field,
-            defaults={data_store_name: validated_data["value"]},
-        )
-        return instance
-
-    def get_value(self, obj: CustomFieldInstance) -> str | int | float | dict | None:
-        return obj.value
-
-    def validate(self, data):
-        """
-        Probably because we're kind of doing it odd, validation from the model
-        doesn't run against the field "value", so we have to re-create it here.
-
-        Don't like it, but it is better than returning an HTTP 500 when the database
-        hates the value
-        """
-        data = super().validate(data)
-        field: CustomField = data["field"]
-        if "value" in data and data["value"] is not None:
-            if (
-                field.data_type == CustomField.FieldDataType.URL
-                and len(data["value"]) > 0
-            ):
-                uri_validator(data["value"])
-            elif field.data_type == CustomField.FieldDataType.INT:
-                integer_validator(data["value"])
-                try:
-                    value_int = int(data["value"])
-                except (TypeError, ValueError):
-                    raise serializers.ValidationError("Enter a valid integer.")
-                # Keep values within the PostgreSQL integer range
-                MinValueValidator(-2147483648)(value_int)
-                MaxValueValidator(2147483647)(value_int)
-            elif (
-                field.data_type == CustomField.FieldDataType.MONETARY
-                and data["value"] != ""
-            ):
-                try:
-                    # First try to validate as a number from legacy format
-                    DecimalValidator(max_digits=12, decimal_places=2)(
-                        Decimal(str(data["value"])),
-                    )
-                except Exception:
-                    # If that fails, try to validate as a monetary string
-                    RegexValidator(
-                        regex=r"^[A-Z]{3}-?\d+(\.\d{1,2})$",
-                        message="Must be a two-decimal number with optional currency code e.g. GBP123.45",
-                    )(data["value"])
-            elif field.data_type == CustomField.FieldDataType.STRING:
-                MaxLengthValidator(limit_value=128)(data["value"])
-            elif field.data_type == CustomField.FieldDataType.SELECT:
-                select_options = field.extra_data["select_options"]
-                try:
-                    next(
-                        option
-                        for option in select_options
-                        if option["id"] == data["value"]
-                    )
-                except Exception:
-                    raise serializers.ValidationError(
-                        f"Value must be an id of an element in {select_options}",
-                    )
-            elif field.data_type == CustomField.FieldDataType.DOCUMENTLINK:
-                if not (isinstance(data["value"], list) or data["value"] is None):
-                    raise serializers.ValidationError(
-                        "Value must be a list",
-                    )
-                doc_ids = data["value"]
-                request = self.context.get("request")
-                validate_documentlink_targets(
-                    getattr(request, "user", None) if request is not None else None,
-                    doc_ids,
-                )
-            elif field.data_type == CustomField.FieldDataType.DATE:
-                data["value"] = serializers.DateField().to_internal_value(data["value"])
-
-        return data
-
-    class Meta:
-        model = CustomFieldInstance
-        fields = [
-            "value",
-            "field",
-        ]
-
-
-class StoragePathSerializer(MatchingModelSerializer, OwnedObjectSerializer):
-    class Meta:
-        model = StoragePath
-        fields = (
-            "id",
-            "slug",
-            "name",
-            "path",
-            "match",
-            "matching_algorithm",
-            "is_insensitive",
-            "document_count",
-            "owner",
-            "permissions",
-            "user_can_change",
-            "set_permissions",
-        )
-
-    def validate_path(self, path: str):
-        converted_path = convert_format_str_to_template_format(path)
-        if converted_path != path:
-            logger.warning(
-                f"Storage path {path} is not using the new style format, consider updating",
-            )
-        result = validate_filepath_template_and_render(converted_path)
-
-        if result is None:
-            raise serializers.ValidationError(_("Invalid variable detected."))
-
-        return converted_path
-
-    def update(self, instance, validated_data):
-        """
-        When a storage path is updated, see if documents
-        using it require a rename/move
-        """
-        doc_ids = [doc.id for doc in instance.documents.all()]
-        if doc_ids:
-            bulk_edit.bulk_update_documents.apply_async(
-                kwargs={"document_ids": doc_ids},
-                headers={"trigger_source": PaperlessTask.TriggerSource.SYSTEM},
-            )
-
-        return super().update(instance, validated_data)
-
-
-class StoragePathTestSerializer(SerializerWithPerms):
-    path = serializers.CharField(
-        required=True,
-        label="Path",
-        write_only=True,
-    )
-
-    document = serializers.PrimaryKeyRelatedField(
-        queryset=Document.objects.none(),
-        required=True,
-        label="Document",
-        write_only=True,
-    )
-
-    def __init__(self, *args: Any, **kwargs: Any) -> None:
-        super().__init__(*args, **kwargs)
-        request = self.context.get("request")
-        user = getattr(request, "user", None) if request else None
-        if user is not None and user.is_authenticated:
-            document_field = self.fields.get("document")
-            if not isinstance(document_field, serializers.PrimaryKeyRelatedField):
-                return
-            document_field.queryset = Document.objects.filter(
-                id__in=permitted_document_ids(user),
-            )

src/documents/serialisers/saved_views.py

@@ -1,234 +0,0 @@
-from __future__ import annotations
-
-import logging
-import re
-
-from django.conf import settings
-from rest_framework import serializers
-
-from documents.models import CustomField
-from documents.models import SavedView
-from documents.models import SavedViewFilterRule
-from documents.models import UiSettings
-
-from .base import OwnedObjectSerializer
-
-logger = logging.getLogger("paperless.serializers")
-
-
-class SavedViewFilterRuleSerializer(serializers.ModelSerializer[SavedViewFilterRule]):
-    class Meta:
-        model = SavedViewFilterRule
-        fields = ["rule_type", "value"]
-
-
-class SavedViewSerializer(OwnedObjectSerializer):
-    filter_rules = SavedViewFilterRuleSerializer(many=True)
-
-    class Meta:
-        model = SavedView
-        fields = [
-            "id",
-            "name",
-            "icon",
-            "sort_field",
-            "sort_reverse",
-            "filter_rules",
-            "page_size",
-            "display_mode",
-            "display_fields",
-            "owner",
-            "permissions",
-            "user_can_change",
-            "set_permissions",
-        ]
-
-    def _get_api_version(self) -> int:
-        request = self.context.get("request")
-        return int(
-            request.version if request else settings.REST_FRAMEWORK["DEFAULT_VERSION"],
-        )
-
-    def _update_legacy_visibility_preferences(
-        self,
-        saved_view_id: int,
-        *,
-        show_on_dashboard: bool | None,
-        show_in_sidebar: bool | None,
-    ) -> UiSettings | None:
-        if show_on_dashboard is None and show_in_sidebar is None:
-            return None
-
-        request = self.context.get("request")
-        user = request.user if request else self.user
-        if user is None:
-            return None
-
-        ui_settings, _ = UiSettings.objects.get_or_create(
-            user=user,
-            defaults={"settings": {}},
-        )
-        current_settings = (
-            ui_settings.settings if isinstance(ui_settings.settings, dict) else {}
-        )
-        current_settings = dict(current_settings)
-
-        saved_views_settings = current_settings.get("saved_views")
-        if isinstance(saved_views_settings, dict):
-            saved_views_settings = dict(saved_views_settings)
-        else:
-            saved_views_settings = {}
-
-        dashboard_ids = {
-            int(raw_id)
-            for raw_id in saved_views_settings.get("dashboard_views_visible_ids", [])
-            if str(raw_id).isdigit()
-        }
-        sidebar_ids = {
-            int(raw_id)
-            for raw_id in saved_views_settings.get("sidebar_views_visible_ids", [])
-            if str(raw_id).isdigit()
-        }
-
-        if show_on_dashboard is not None:
-            if show_on_dashboard:
-                dashboard_ids.add(saved_view_id)
-            else:
-                dashboard_ids.discard(saved_view_id)
-        if show_in_sidebar is not None:
-            if show_in_sidebar:
-                sidebar_ids.add(saved_view_id)
-            else:
-                sidebar_ids.discard(saved_view_id)
-
-        saved_views_settings["dashboard_views_visible_ids"] = sorted(dashboard_ids)
-        saved_views_settings["sidebar_views_visible_ids"] = sorted(sidebar_ids)
-        current_settings["saved_views"] = saved_views_settings
-        ui_settings.settings = current_settings
-        ui_settings.save(update_fields=["settings"])
-        return ui_settings
-
-    def to_representation(self, instance):
-        # TODO: remove this and related backwards compatibility code when API v9 is dropped
-        ret = super().to_representation(instance)
-        request = self.context.get("request")
-        api_version = self._get_api_version()
-
-        if api_version < 10:
-            dashboard_ids = set()
-            sidebar_ids = set()
-            user = request.user if request else None
-            if user is not None and hasattr(user, "ui_settings"):
-                ui_settings = user.ui_settings.settings or None
-                saved_views = None
-                if isinstance(ui_settings, dict):
-                    saved_views = ui_settings.get("saved_views", {})
-                if isinstance(saved_views, dict):
-                    dashboard_ids = set(
-                        saved_views.get("dashboard_views_visible_ids", []),
-                    )
-                    sidebar_ids = set(
-                        saved_views.get("sidebar_views_visible_ids", []),
-                    )
-            ret["show_on_dashboard"] = instance.id in dashboard_ids
-            ret["show_in_sidebar"] = instance.id in sidebar_ids
-
-        return ret
-
-    def to_internal_value(self, data):
-        # TODO: remove this and related backwards compatibility code when API v9 is dropped
-        api_version = self._get_api_version()
-        if api_version >= 10:
-            return super().to_internal_value(data)
-
-        normalized_data = data.copy()
-        legacy_visibility_fields = {}
-        boolean_field = serializers.BooleanField()
-
-        for field_name in ("show_on_dashboard", "show_in_sidebar"):
-            if field_name in normalized_data:
-                try:
-                    legacy_visibility_fields[field_name] = (
-                        boolean_field.to_internal_value(
-                            normalized_data.get(field_name),
-                        )
-                    )
-                except serializers.ValidationError as exc:
-                    raise serializers.ValidationError({field_name: exc.detail})
-                del normalized_data[field_name]
-
-        ret = super().to_internal_value(normalized_data)
-        ret.update(legacy_visibility_fields)
-        return ret
-
-    def validate(self, attrs):
-        attrs = super().validate(attrs)
-        if "display_fields" in attrs and attrs["display_fields"] is not None:
-            for field in attrs["display_fields"]:
-                if (
-                    SavedView.DisplayFields.CUSTOM_FIELD[:-2] in field
-                ):  # i.e. check for 'custom_field_' prefix
-                    field_id = int(re.search(r"\d+", field)[0])
-                    if not CustomField.objects.filter(id=field_id).exists():
-                        raise serializers.ValidationError(
-                            f"Invalid field: {field}",
-                        )
-                elif field not in SavedView.DisplayFields.values:
-                    raise serializers.ValidationError(
-                        f"Invalid field: {field}",
-                    )
-        return attrs
-
-    def update(self, instance, validated_data):
-        request = self.context.get("request")
-        show_on_dashboard = validated_data.pop("show_on_dashboard", None)
-        show_in_sidebar = validated_data.pop("show_in_sidebar", None)
-        if "filter_rules" in validated_data:
-            rules_data = validated_data.pop("filter_rules")
-        else:
-            rules_data = None
-        if "user" in validated_data:
-            # backwards compatibility
-            validated_data["owner"] = validated_data.pop("user")
-        if (
-            "display_fields" in validated_data
-            and isinstance(
-                validated_data["display_fields"],
-                list,
-            )
-            and len(validated_data["display_fields"]) == 0
-        ):
-            validated_data["display_fields"] = None
-        instance = super().update(instance, validated_data)
-        if rules_data is not None:
-            SavedViewFilterRule.objects.filter(saved_view=instance).delete()
-            for rule_data in rules_data:
-                SavedViewFilterRule.objects.create(saved_view=instance, **rule_data)
-        ui_settings = self._update_legacy_visibility_preferences(
-            instance.id,
-            show_on_dashboard=show_on_dashboard,
-            show_in_sidebar=show_in_sidebar,
-        )
-        if request is not None and ui_settings is not None:
-            request.user.ui_settings = ui_settings
-        return instance
-
-    def create(self, validated_data):
-        request = self.context.get("request")
-        show_on_dashboard = validated_data.pop("show_on_dashboard", None)
-        show_in_sidebar = validated_data.pop("show_in_sidebar", None)
-        rules_data = validated_data.pop("filter_rules")
-        if "user" in validated_data:
-            # backwards compatibility
-            validated_data["owner"] = validated_data.pop("user")
-        saved_view = super().create(validated_data)
-        for rule_data in rules_data:
-            SavedViewFilterRule.objects.create(saved_view=saved_view, **rule_data)
-        ui_settings = self._update_legacy_visibility_preferences(
-            saved_view.id,
-            show_on_dashboard=show_on_dashboard,
-            show_in_sidebar=show_in_sidebar,
-        )
-        if request is not None and ui_settings is not None:
-            request.user.ui_settings = ui_settings
-        return saved_view

src/documents/serialisers/sharing.py

@@ -1,204 +0,0 @@
-from __future__ import annotations
-
-import logging
-from datetime import timedelta
-
-from django.core.exceptions import ValidationError
-from django.core.validators import EmailValidator
-from django.utils import timezone
-from django.utils.crypto import get_random_string
-from django.utils.translation import gettext as _
-from rest_framework import serializers
-from rest_framework.exceptions import PermissionDenied
-from rest_framework.fields import SerializerMethodField
-
-from documents.models import Document
-from documents.models import ShareLink
-from documents.models import ShareLinkBundle
-from documents.permissions import has_perms_owner_aware
-
-from .base import DocumentListSerializer
-from .base import OwnedObjectSerializer
-
-logger = logging.getLogger("paperless.serializers")
-
-
-class EmailSerializer(DocumentListSerializer):
-    addresses = serializers.CharField(
-        required=True,
-        label="Email addresses",
-        help_text="Comma-separated email addresses",
-    )
-
-    subject = serializers.CharField(
-        required=True,
-        label="Email subject",
-    )
-
-    message = serializers.CharField(
-        required=True,
-        label="Email message",
-    )
-
-    use_archive_version = serializers.BooleanField(
-        default=True,
-        label="Use archive version",
-        help_text="Use archive version of documents if available",
-    )
-
-    def validate_addresses(self, addresses):
-        address_list = [addr.strip() for addr in addresses.split(",")]
-        if not address_list:
-            raise serializers.ValidationError("At least one email address is required")
-
-        email_validator = EmailValidator()
-        try:
-            for address in address_list:
-                email_validator(address)
-        except ValidationError:
-            raise serializers.ValidationError(f"Invalid email address: {address}")
-
-        return ",".join(address_list)
-
-    def validate_documents(self, documents):
-        super().validate_documents(documents)
-        if not documents:
-            raise serializers.ValidationError("At least one document is required")
-
-        return documents
-
-
-class ShareLinkSerializer(OwnedObjectSerializer):
-    document_title = serializers.CharField(
-        source="document.title",
-        read_only=True,
-    )
-
-    class Meta:
-        model = ShareLink
-        fields = (
-            "id",
-            "created",
-            "expiration",
-            "slug",
-            "document",
-            "document_title",
-            "file_version",
-        )
-
-    def create(self, validated_data):
-        validated_data["slug"] = get_random_string(50)
-        return super().create(validated_data)
-
-    def validate_document(self, document):
-        if (
-            self.user is not None
-            and self.user.has_perm("documents.view_document")
-            and has_perms_owner_aware(
-                self.user,
-                "view_document",
-                document,
-            )
-        ):
-            return document
-        raise PermissionDenied(
-            _("Insufficient permissions."),
-        )
-
-
-class ShareLinkBundleSerializer(OwnedObjectSerializer):
-    document_ids = serializers.ListField(
-        child=serializers.IntegerField(min_value=1),
-        allow_empty=False,
-        write_only=True,
-    )
-    expiration_days = serializers.IntegerField(
-        required=False,
-        allow_null=True,
-        min_value=1,
-        write_only=True,
-    )
-    documents = serializers.PrimaryKeyRelatedField(
-        many=True,
-        read_only=True,
-    )
-    document_count = SerializerMethodField()
-
-    class Meta:
-        model = ShareLinkBundle
-        fields = (
-            "id",
-            "created",
-            "expiration",
-            "expiration_days",
-            "slug",
-            "file_version",
-            "status",
-            "size_bytes",
-            "last_error",
-            "built_at",
-            "documents",
-            "document_ids",
-            "document_count",
-        )
-        read_only_fields = (
-            "id",
-            "created",
-            "expiration",
-            "slug",
-            "status",
-            "size_bytes",
-            "last_error",
-            "built_at",
-            "documents",
-            "document_count",
-        )
-
-    def validate_document_ids(self, value):
-        unique_ids = set(value)
-        if len(unique_ids) != len(value):
-            raise serializers.ValidationError(
-                _("Duplicate document identifiers are not allowed."),
-            )
-        return value
-
-    def create(self, validated_data):
-        document_ids = validated_data.pop("document_ids")
-        expiration_days = validated_data.pop("expiration_days", None)
-        validated_data["slug"] = get_random_string(50)
-        if expiration_days:
-            validated_data["expiration"] = timezone.now() + timedelta(
-                days=expiration_days,
-            )
-        else:
-            validated_data["expiration"] = None
-
-        share_link_bundle = super().create(validated_data)
-
-        documents = list(
-            Document.objects.filter(pk__in=document_ids).only(
-                "pk",
-            ),
-        )
-        documents_by_id = {doc.pk: doc for doc in documents}
-        missing = [
-            str(doc_id) for doc_id in document_ids if doc_id not in documents_by_id
-        ]
-        if missing:
-            raise serializers.ValidationError(
-                {
-                    "document_ids": _(
-                        "Documents not found: %(ids)s",
-                    )
-                    % {"ids": ", ".join(missing)},
-                },
-            )
-
-        ordered_documents = [documents_by_id[doc_id] for doc_id in document_ids]
-        share_link_bundle.documents.set(ordered_documents)
-        share_link_bundle.document_total = len(ordered_documents)
-
-        return share_link_bundle
-
-    def get_document_count(self, obj: ShareLinkBundle) -> int:
-        return getattr(obj, "document_total") or obj.documents.count()

src/documents/serialisers/system.py

@@ -1,63 +0,0 @@
-from __future__ import annotations
-
-import logging
-
-from rest_framework import serializers
-
-from documents.models import Document
-from documents.models import UiSettings
-
-from .base import SerializerWithPerms
-
-logger = logging.getLogger("paperless.serializers")
-
-
-class UiSettingsViewSerializer(serializers.ModelSerializer[UiSettings]):
-    settings = serializers.DictField(required=False, allow_null=True)
-
-    class Meta:
-        model = UiSettings
-        depth = 1
-        fields = [
-            "id",
-            "settings",
-        ]
-
-    def validate_settings(self, settings):
-        # we never save update checking backend setting
-        if "update_checking" in settings:
-            try:
-                settings["update_checking"].pop("backend_setting")
-            except KeyError:
-                pass
-        return settings
-
-    def create(self, validated_data):
-        ui_settings = UiSettings.objects.update_or_create(
-            user=validated_data.get("user"),
-            defaults={"settings": validated_data.get("settings", None)},
-        )
-        return ui_settings
-
-
-class TrashSerializer(SerializerWithPerms):
-    documents = serializers.ListField(
-        required=False,
-        label="Documents",
-        write_only=True,
-        child=serializers.IntegerField(),
-    )
-
-    action = serializers.ChoiceField(
-        choices=["restore", "empty"],
-        label="Action",
-        write_only=True,
-    )
-
-    def validate_documents(self, documents: list[int]) -> list[int]:
-        count = Document.deleted_objects.filter(id__in=documents).count()
-        if not count == len(documents):
-            raise serializers.ValidationError(
-                "Some documents in the list have not yet been deleted.",
-            )
-        return documents

src/documents/serialisers/tasks.py

@@ -1,246 +0,0 @@
-from __future__ import annotations
-
-import logging
-from typing import Any
-
-from rest_framework import serializers
-
-from documents.models import Document
-from documents.models import PaperlessTask
-from documents.permissions import permitted_document_ids
-
-from .base import OwnedObjectSerializer
-
-logger = logging.getLogger("paperless.serializers")
-
-
-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",
-            "related_document_ids",
-            "acknowledged",
-            "owner",
-        )
-        read_only_fields = fields
-
-
-class TaskSerializerV9(serializers.ModelSerializer[PaperlessTask]):
-    """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 (with value remapping for renamed tasks)
-    task_name = serializers.SerializerMethodField()
-
-    # 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: status -> uppercase Celery state strings
-    status = serializers.SerializerMethodField()
-
-    # v9 field: result -> derived from result_data
-    result = serializers.SerializerMethodField()
-
-    # 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 Meta:
-        model = PaperlessTask
-        fields = (
-            "id",
-            "task_id",
-            "task_name",
-            "task_file_name",
-            "type",
-            "status",
-            "date_created",
-            "date_done",
-            "result",
-            "acknowledged",
-            "related_document",
-            "duplicate_documents",
-            "owner",
-        )
-        read_only_fields = fields
-
-    _TASK_TYPE_TO_V9_NAME = {
-        PaperlessTask.TaskType.SANITY_CHECK: "check_sanity",
-        PaperlessTask.TaskType.LLM_INDEX: "llmindex_update",
-    }
-
-    def get_result(self, obj: PaperlessTask) -> str | None:
-        """Reconstruct a human-readable result string from result_data for v9 clients."""
-        if not obj.result_data:
-            return None
-        if doc_id := obj.result_data.get("document_id"):
-            return f"Success. New document id {doc_id} created"
-        if reason := obj.result_data.get("reason"):
-            return reason
-        if dup_id := obj.result_data.get("duplicate_of"):
-            return f"Not consuming: It is a duplicate of document #{dup_id}"
-        if error := obj.result_data.get("error_message"):
-            return error
-        return None
-
-    def get_task_name(self, obj: PaperlessTask) -> str:
-        return self._TASK_TYPE_TO_V9_NAME.get(obj.task_type, obj.task_type)
-
-    def get_task_file_name(self, obj: PaperlessTask) -> str | None:
-        if not obj.input_data:
-            return None
-        return obj.input_data.get("filename")
-
-    _STATUS_TO_V9 = {
-        PaperlessTask.Status.PENDING: "PENDING",
-        PaperlessTask.Status.STARTED: "STARTED",
-        PaperlessTask.Status.SUCCESS: "SUCCESS",
-        PaperlessTask.Status.FAILURE: "FAILURE",
-        PaperlessTask.Status.REVOKED: "REVOKED",
-    }
-
-    def get_status(self, obj: PaperlessTask) -> str:
-        return self._STATUS_TO_V9.get(obj.status, obj.status.upper())
-
-    _TRIGGER_SOURCE_TO_V9_TYPE = {
-        PaperlessTask.TriggerSource.SCHEDULED: "scheduled_task",
-        PaperlessTask.TriggerSource.SYSTEM: "auto_task",
-        # Email and folder-consumer documents are system-initiated, not manually triggered
-        PaperlessTask.TriggerSource.EMAIL_CONSUME: "auto_task",
-        PaperlessTask.TriggerSource.FOLDER_CONSUME: "auto_task",
-    }
-
-    def get_type(self, obj: PaperlessTask) -> str:
-        return self._TRIGGER_SOURCE_TO_V9_TYPE.get(obj.trigger_source, "manual_task")
-
-    def get_related_document(self, obj: PaperlessTask) -> int | None:
-        ids = obj.related_document_ids
-        return ids[0] if ids else None
-
-    def get_duplicate_documents(
-        self,
-        obj: PaperlessTask,
-    ) -> list[dict[str, Any]]:
-        if not obj.result_data:
-            return []
-        dup_of = obj.result_data.get("duplicate_of")
-        if dup_of is None:
-            return []
-        request = self.context.get("request")
-        if request is None:
-            return []
-        user = request.user
-        qs = Document.global_objects.filter(pk=dup_of)
-        if not user.is_staff:
-            allowed_ids = permitted_document_ids(user, include_deleted=True)
-            qs = qs.filter(pk__in=allowed_ids)
-        return list(qs.values("id", "title", "deleted_at"))
-
-
-class TaskSummarySerializer(serializers.Serializer[dict[str, Any]]):
-    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[dict[str, str]]):
-    task_type = serializers.ChoiceField(
-        choices=PaperlessTask.TaskType.choices,
-        label="Task Type",
-        write_only=True,
-    )
-
-
-class AcknowledgeTasksViewSerializer(serializers.Serializer[dict[str, Any]]):
-    tasks = serializers.ListField(
-        required=False,
-        label="Tasks",
-        write_only=True,
-        child=serializers.IntegerField(),
-    )
-    all = serializers.BooleanField(
-        required=False,
-        default=False,
-        label="All",
-        write_only=True,
-    )
-
-    def _validate_task_id_list(self, tasks, name="tasks") -> None:
-        if not isinstance(tasks, list):
-            raise serializers.ValidationError(f"{name} must be a list")
-        if not all(isinstance(i, int) for i in tasks):
-            raise serializers.ValidationError(f"{name} must be a list of integers")
-        queryset = self.context.get("queryset", PaperlessTask.objects.all())
-        count = queryset.filter(id__in=tasks).count()
-        if not count == len(tasks):
-            raise serializers.ValidationError(
-                f"Some tasks in {name} don't exist or were specified twice.",
-            )
-
-    def validate_tasks(self, tasks):
-        self._validate_task_id_list(tasks)
-        return tasks
-
-    def validate(self, attrs):
-        acknowledge_all = attrs.get("all", False)
-        task_ids = attrs.get("tasks")
-
-        if acknowledge_all and task_ids is not None:
-            raise serializers.ValidationError(
-                "Set either all or tasks, not both.",
-            )
-        if not acknowledge_all and task_ids is None:
-            raise serializers.ValidationError(
-                "Either all must be true or tasks must be provided.",
-            )
-
-        return attrs

src/documents/serialisers/upload.py

@@ -1,199 +0,0 @@
-from __future__ import annotations
-
-import logging
-from datetime import datetime
-from typing import Any
-
-import magic
-from django.conf import settings
-from django.utils.translation import gettext as _
-from rest_framework import serializers
-
-from documents.models import Correspondent
-from documents.models import CustomField
-from documents.models import Document
-from documents.models import DocumentType
-from documents.models import StoragePath
-from documents.models import Tag
-from documents.parsers import is_mime_type_supported
-
-from .metadata import CustomFieldInstanceSerializer
-
-logger = logging.getLogger("paperless.serializers")
-
-
-class PostDocumentSerializer(serializers.Serializer[dict[str, Any]]):
-    created = serializers.DateTimeField(
-        label="Created",
-        allow_null=True,
-        write_only=True,
-        required=False,
-    )
-
-    document = serializers.FileField(
-        label="Document",
-        write_only=True,
-    )
-
-    title = serializers.CharField(
-        label="Title",
-        write_only=True,
-        required=False,
-    )
-
-    correspondent = serializers.PrimaryKeyRelatedField(
-        queryset=Correspondent.objects.all(),
-        label="Correspondent",
-        allow_null=True,
-        write_only=True,
-        required=False,
-    )
-
-    document_type = serializers.PrimaryKeyRelatedField(
-        queryset=DocumentType.objects.all(),
-        label="Document type",
-        allow_null=True,
-        write_only=True,
-        required=False,
-    )
-
-    storage_path = serializers.PrimaryKeyRelatedField(
-        queryset=StoragePath.objects.all(),
-        label="Storage path",
-        allow_null=True,
-        write_only=True,
-        required=False,
-    )
-
-    tags = serializers.PrimaryKeyRelatedField(
-        many=True,
-        queryset=Tag.objects.all(),
-        label="Tags",
-        write_only=True,
-        required=False,
-    )
-
-    archive_serial_number = serializers.IntegerField(
-        label="ASN",
-        write_only=True,
-        required=False,
-        min_value=Document.ARCHIVE_SERIAL_NUMBER_MIN,
-        max_value=Document.ARCHIVE_SERIAL_NUMBER_MAX,
-    )
-
-    # Accept either a list of custom field ids or a dict mapping id -> value
-    custom_fields = serializers.JSONField(
-        label="Custom fields",
-        write_only=True,
-        required=False,
-    )
-
-    from_webui = serializers.BooleanField(
-        label="Documents are from Paperless-ngx WebUI",
-        write_only=True,
-        required=False,
-    )
-
-    def validate_document(self, document):
-        document_data = document.file.read()
-        mime_type = magic.from_buffer(document_data, mime=True)
-
-        if not is_mime_type_supported(mime_type):
-            if (
-                mime_type in settings.CONSUMER_PDF_RECOVERABLE_MIME_TYPES
-                and document.name.endswith(
-                    ".pdf",
-                )
-            ):
-                # If the file is an invalid PDF, we can try to recover it later in the consumer
-                mime_type = "application/pdf"
-            else:
-                raise serializers.ValidationError(
-                    _("File type %(type)s not supported") % {"type": mime_type},
-                )
-
-        return document.name, document_data
-
-    def validate_correspondent(self, correspondent):
-        if correspondent:
-            return correspondent.id
-        else:
-            return None
-
-    def validate_document_type(self, document_type):
-        if document_type:
-            return document_type.id
-        else:
-            return None
-
-    def validate_storage_path(self, storage_path):
-        if storage_path:
-            return storage_path.id
-        else:
-            return None
-
-    def validate_tags(self, tags):
-        if tags:
-            return [tag.id for tag in tags]
-        else:
-            return None
-
-    def validate_custom_fields(self, custom_fields):
-        if not custom_fields:
-            return None
-
-        # Normalize single values to a list
-        if isinstance(custom_fields, int):
-            custom_fields = [custom_fields]
-        if isinstance(custom_fields, dict):
-            custom_field_serializer = CustomFieldInstanceSerializer()
-            normalized = {}
-            for field_id, value in custom_fields.items():
-                try:
-                    field_id_int = int(field_id)
-                except (TypeError, ValueError):
-                    raise serializers.ValidationError(
-                        _("Custom field id must be an integer: %(id)s")
-                        % {"id": field_id},
-                    )
-                try:
-                    field = CustomField.objects.get(id=field_id_int)
-                except CustomField.DoesNotExist:
-                    raise serializers.ValidationError(
-                        _("Custom field with id %(id)s does not exist")
-                        % {"id": field_id_int},
-                    )
-                custom_field_serializer.validate(
-                    {
-                        "field": field,
-                        "value": value,
-                    },
-                )
-                normalized[field_id_int] = value
-            return normalized
-        elif isinstance(custom_fields, list):
-            try:
-                ids = [int(i) for i in custom_fields]
-            except (TypeError, ValueError):
-                raise serializers.ValidationError(
-                    _(
-                        "Custom fields must be a list of integers or an object mapping ids to values.",
-                    ),
-                )
-            if CustomField.objects.filter(id__in=ids).count() != len(set(ids)):
-                raise serializers.ValidationError(
-                    _("Some custom fields don't exist or were specified twice."),
-                )
-            return ids
-        raise serializers.ValidationError(
-            _(
-                "Custom fields must be a list of integers or an object mapping ids to values.",
-            ),
-        )
-
-    # custom_fields_w_values handled via validate_custom_fields
-
-    def validate_created(self, created):
-        # support datetime format for created for backwards compatibility
-        if isinstance(created, datetime):
-            return created.date()

src/documents/serialisers/workflows.py

@@ -1,584 +0,0 @@
-from __future__ import annotations
-
-import logging
-
-from django.db.models import Count
-from rest_framework import fields
-from rest_framework import serializers
-
-from documents.data_models import DocumentSource
-from documents.filters import CustomFieldQueryParser
-from documents.models import Workflow
-from documents.models import WorkflowAction
-from documents.models import WorkflowActionEmail
-from documents.models import WorkflowActionWebhook
-from documents.models import WorkflowTrigger
-from documents.templating.workflows import validate_workflow_template
-from documents.validators import url_validator
-
-from .metadata import CorrespondentField
-from .metadata import DocumentTypeField
-from .metadata import StoragePathField
-from .metadata import TagsField
-
-logger = logging.getLogger("paperless.serializers")
-
-
-class WorkflowTriggerSerializer(serializers.ModelSerializer[WorkflowTrigger]):
-    id = serializers.IntegerField(required=False, allow_null=True)
-    sources = fields.MultipleChoiceField(
-        choices=WorkflowTrigger.DocumentSourceChoices.choices,
-        allow_empty=True,
-        default={
-            DocumentSource.ConsumeFolder,
-            DocumentSource.ApiUpload,
-            DocumentSource.MailFetch,
-        },
-    )
-
-    type = serializers.ChoiceField(
-        choices=WorkflowTrigger.WorkflowTriggerType.choices,
-        label="Trigger Type",
-    )
-
-    class Meta:
-        model = WorkflowTrigger
-        fields = [
-            "id",
-            "sources",
-            "type",
-            "filter_path",
-            "filter_filename",
-            "filter_mailrule",
-            "matching_algorithm",
-            "match",
-            "is_insensitive",
-            "filter_has_tags",
-            "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",
-            "filter_has_storage_path",
-            "schedule_offset_days",
-            "schedule_is_recurring",
-            "schedule_recurring_interval_days",
-            "schedule_date_field",
-            "schedule_date_custom_field",
-        ]
-
-    def validate(self, attrs):
-        # Empty strings treated as None to avoid unexpected behavior
-        if (
-            "filter_filename" in attrs
-            and attrs["filter_filename"] is not None
-            and len(attrs["filter_filename"]) == 0
-        ):
-            attrs["filter_filename"] = None
-        if (
-            "filter_path" in attrs
-            and attrs["filter_path"] is not None
-            and len(attrs["filter_path"]) == 0
-        ):
-            attrs["filter_path"] = None
-
-        if (
-            "filter_custom_field_query" in attrs
-            and attrs["filter_custom_field_query"] is not None
-            and len(attrs["filter_custom_field_query"]) == 0
-        ):
-            attrs["filter_custom_field_query"] = None
-
-        if (
-            "filter_custom_field_query" in attrs
-            and attrs["filter_custom_field_query"] is not None
-        ):
-            parser = CustomFieldQueryParser("filter_custom_field_query")
-            parser.parse(attrs["filter_custom_field_query"])
-
-        trigger_type = attrs.get("type", getattr(self.instance, "type", None))
-        if (
-            trigger_type == WorkflowTrigger.WorkflowTriggerType.CONSUMPTION
-            and "filter_mailrule" not in attrs
-            and ("filter_filename" not in attrs or attrs["filter_filename"] is None)
-            and ("filter_path" not in attrs or attrs["filter_path"] is None)
-        ):
-            raise serializers.ValidationError(
-                "File name, path or mail rule filter are required",
-            )
-
-        return attrs
-
-    @staticmethod
-    def normalize_workflow_trigger_sources(trigger) -> None:
-        """
-        Convert sources to strings to handle django-multiselectfield v1.0 changes
-        """
-        if trigger and "sources" in trigger:
-            trigger["sources"] = [
-                str(s.value if hasattr(s, "value") else s) for s in trigger["sources"]
-            ]
-
-    def create(self, validated_data):
-        WorkflowTriggerSerializer.normalize_workflow_trigger_sources(validated_data)
-        return super().create(validated_data)
-
-    def update(self, instance, validated_data):
-        WorkflowTriggerSerializer.normalize_workflow_trigger_sources(validated_data)
-        return super().update(instance, validated_data)
-
-
-class WorkflowActionEmailSerializer(serializers.ModelSerializer[WorkflowActionEmail]):
-    id = serializers.IntegerField(allow_null=True, required=False)
-
-    class Meta:
-        model = WorkflowActionEmail
-        fields = [
-            "id",
-            "subject",
-            "body",
-            "to",
-            "include_document",
-        ]
-
-
-class WorkflowActionWebhookSerializer(
-    serializers.ModelSerializer[WorkflowActionWebhook],
-):
-    id = serializers.IntegerField(allow_null=True, required=False)
-
-    def validate_url(self, url):
-        url_validator(url)
-        return url
-
-    class Meta:
-        model = WorkflowActionWebhook
-        fields = [
-            "id",
-            "url",
-            "use_params",
-            "as_json",
-            "params",
-            "body",
-            "headers",
-            "include_document",
-        ]
-
-
-class WorkflowActionSerializer(serializers.ModelSerializer[WorkflowAction]):
-    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)
-    assign_document_type = DocumentTypeField(allow_null=True, required=False)
-    assign_storage_path = StoragePathField(allow_null=True, required=False)
-    email = WorkflowActionEmailSerializer(allow_null=True, required=False)
-    webhook = WorkflowActionWebhookSerializer(allow_null=True, required=False)
-
-    class Meta:
-        model = WorkflowAction
-        fields = [
-            "id",
-            "type",
-            "assign_title",
-            "assign_tags",
-            "assign_correspondent",
-            "assign_document_type",
-            "assign_storage_path",
-            "assign_owner",
-            "assign_view_users",
-            "assign_view_groups",
-            "assign_change_users",
-            "assign_change_groups",
-            "assign_custom_fields",
-            "assign_custom_fields_values",
-            "remove_all_tags",
-            "remove_tags",
-            "remove_all_correspondents",
-            "remove_correspondents",
-            "remove_all_document_types",
-            "remove_document_types",
-            "remove_all_storage_paths",
-            "remove_storage_paths",
-            "remove_custom_fields",
-            "remove_all_custom_fields",
-            "remove_all_owners",
-            "remove_owners",
-            "remove_all_permissions",
-            "remove_view_users",
-            "remove_view_groups",
-            "remove_change_users",
-            "remove_change_groups",
-            "email",
-            "webhook",
-            "passwords",
-            "ai_suggestion_fields",
-            "ai_create_missing",
-            "ai_overwrite_existing",
-        ]
-
-    def validate(self, attrs):
-        if "assign_title" in attrs and attrs["assign_title"] is not None:
-            if len(attrs["assign_title"]) == 0:
-                # Empty strings treated as None to avoid unexpected behavior
-                attrs["assign_title"] = None
-            else:
-                try:
-                    validate_workflow_template(attrs["assign_title"])
-                except (ValueError, KeyError) as e:
-                    raise serializers.ValidationError(
-                        {"assign_title": f"{e.args[0]}"},
-                    )
-
-        if attrs.get("assign_custom_fields_values"):
-            # Empty strings treated as None to avoid unexpected behavior
-            attrs["assign_custom_fields_values"] = {
-                field_id: (None if value == "" else value)
-                for field_id, value in attrs["assign_custom_fields_values"].items()
-            }
-
-        if (
-            "type" in attrs
-            and attrs["type"] == WorkflowAction.WorkflowActionType.EMAIL
-            and "email" not in attrs
-        ):
-            raise serializers.ValidationError(
-                "Email data is required for email actions",
-            )
-
-        if (
-            "type" in attrs
-            and attrs["type"] == WorkflowAction.WorkflowActionType.WEBHOOK
-            and "webhook" not in attrs
-        ):
-            raise serializers.ValidationError(
-                "Webhook data is required for webhook actions",
-            )
-
-        if (
-            "type" in attrs
-            and attrs["type"] == WorkflowAction.WorkflowActionType.PASSWORD_REMOVAL
-        ):
-            passwords = attrs.get("passwords")
-            # ensure passwords is a non-empty list of non-empty strings
-            if (
-                passwords is None
-                or not isinstance(passwords, list)
-                or len(passwords) == 0
-                or any(not isinstance(pw, str) for pw in passwords)
-                or any(len(pw.strip()) == 0 for pw in passwords)
-            ):
-                raise serializers.ValidationError(
-                    "Passwords are required for password removal actions",
-                )
-
-        if (
-            "type" in attrs
-            and attrs["type"] == WorkflowAction.WorkflowActionType.APPLY_AI_SUGGESTIONS
-        ):
-            fields = attrs.get("ai_suggestion_fields")
-            valid_fields = set(WorkflowAction.AISuggestionField.values)
-            if (
-                fields is None
-                or not isinstance(fields, list)
-                or len(fields) == 0
-                or any(field not in valid_fields for field in fields)
-            ):
-                raise serializers.ValidationError(
-                    "At least one valid field is required for apply AI "
-                    f"suggestions actions, options are: {sorted(valid_fields)}",
-                )
-
-        return attrs
-
-
-class WorkflowSerializer(serializers.ModelSerializer[Workflow]):
-    order = serializers.IntegerField(required=False)
-
-    triggers = WorkflowTriggerSerializer(many=True)
-    actions = WorkflowActionSerializer(many=True)
-
-    class Meta:
-        model = Workflow
-        fields = [
-            "id",
-            "name",
-            "order",
-            "enabled",
-            "triggers",
-            "actions",
-        ]
-
-    def validate(self, attrs):
-        attrs = super().validate(attrs)
-
-        if "actions" in attrs:
-            has_remote_ocr_action = any(
-                action.get("type") == WorkflowAction.WorkflowActionType.REMOTE_OCR
-                for action in attrs["actions"]
-            )
-            has_ai_suggestions_action = any(
-                action.get("type")
-                == WorkflowAction.WorkflowActionType.APPLY_AI_SUGGESTIONS
-                for action in attrs["actions"]
-            )
-        else:
-            has_remote_ocr_action = self.instance is not None and (
-                self.instance.actions.filter(
-                    type=WorkflowAction.WorkflowActionType.REMOTE_OCR,
-                ).exists()
-            )
-            has_ai_suggestions_action = self.instance is not None and (
-                self.instance.actions.filter(
-                    type=WorkflowAction.WorkflowActionType.APPLY_AI_SUGGESTIONS,
-                ).exists()
-            )
-
-        if "triggers" in attrs:
-            has_consumption_trigger = any(
-                trigger.get("type") == WorkflowTrigger.WorkflowTriggerType.CONSUMPTION
-                for trigger in attrs["triggers"]
-            )
-            has_non_consumption_trigger = any(
-                trigger.get("type") != WorkflowTrigger.WorkflowTriggerType.CONSUMPTION
-                for trigger in attrs["triggers"]
-            )
-        else:
-            has_consumption_trigger = self.instance is not None and (
-                self.instance.triggers.filter(
-                    type=WorkflowTrigger.WorkflowTriggerType.CONSUMPTION,
-                ).exists()
-            )
-            has_non_consumption_trigger = self.instance is not None and (
-                self.instance.triggers.exclude(
-                    type=WorkflowTrigger.WorkflowTriggerType.CONSUMPTION,
-                ).exists()
-            )
-
-        # Remote OCR can only work with consumption triggers
-        if has_remote_ocr_action and not has_consumption_trigger:
-            raise serializers.ValidationError(
-                "Remote OCR actions require a consumption started trigger",
-            )
-
-        # Suggestions are made from the document content, which does not exist
-        # until after consumption has finished
-        if has_ai_suggestions_action and not has_non_consumption_trigger:
-            raise serializers.ValidationError(
-                "Apply AI suggestions actions require a trigger other than "
-                "consumption started",
-            )
-
-        return attrs
-
-    def update_triggers_and_actions(
-        self,
-        instance: Workflow,
-        triggers,
-        actions,
-    ) -> None:
-        set_triggers = []
-        set_actions = []
-
-        if triggers is not None and triggers is not serializers.empty:
-            for trigger in triggers:
-                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,
-                )
-                # Convert sources to strings to handle django-multiselectfield v1.0 changes
-                WorkflowTriggerSerializer.normalize_workflow_trigger_sources(trigger)
-                trigger_instance, _ = WorkflowTrigger.objects.update_or_create(
-                    id=trigger.get("id"),
-                    defaults=trigger,
-                )
-                if filter_has_tags is not None:
-                    trigger_instance.filter_has_tags.set(filter_has_tags)
-                if filter_has_all_tags is not None:
-                    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,
-                    )
-                set_triggers.append(trigger_instance)
-
-        if actions is not None and actions is not serializers.empty:
-            for index, action in enumerate(actions):
-                action["order"] = index
-                assign_tags = action.pop("assign_tags", None)
-                assign_view_users = action.pop("assign_view_users", None)
-                assign_view_groups = action.pop("assign_view_groups", None)
-                assign_change_users = action.pop("assign_change_users", None)
-                assign_change_groups = action.pop("assign_change_groups", None)
-                assign_custom_fields = action.pop("assign_custom_fields", None)
-                remove_tags = action.pop("remove_tags", None)
-                remove_correspondents = action.pop("remove_correspondents", None)
-                remove_document_types = action.pop("remove_document_types", None)
-                remove_storage_paths = action.pop("remove_storage_paths", None)
-                remove_custom_fields = action.pop("remove_custom_fields", None)
-                remove_owners = action.pop("remove_owners", None)
-                remove_view_users = action.pop("remove_view_users", None)
-                remove_view_groups = action.pop("remove_view_groups", None)
-                remove_change_users = action.pop("remove_change_users", None)
-                remove_change_groups = action.pop("remove_change_groups", None)
-
-                email_data = action.pop("email", None)
-                webhook_data = action.pop("webhook", None)
-
-                action_instance, _ = WorkflowAction.objects.update_or_create(
-                    id=action.get("id"),
-                    defaults=action,
-                )
-
-                if email_data is not None:
-                    serializer = WorkflowActionEmailSerializer(data=email_data)
-                    serializer.is_valid(raise_exception=True)
-                    email, _ = WorkflowActionEmail.objects.update_or_create(
-                        id=email_data.get("id"),
-                        defaults=serializer.validated_data,
-                    )
-                    action_instance.email = email
-                    action_instance.save()
-
-                if webhook_data is not None:
-                    serializer = WorkflowActionWebhookSerializer(data=webhook_data)
-                    serializer.is_valid(raise_exception=True)
-                    webhook, _ = WorkflowActionWebhook.objects.update_or_create(
-                        id=webhook_data.get("id"),
-                        defaults=serializer.validated_data,
-                    )
-                    action_instance.webhook = webhook
-                    action_instance.save()
-
-                if assign_tags is not None:
-                    action_instance.assign_tags.set(assign_tags)
-                if assign_view_users is not None:
-                    action_instance.assign_view_users.set(assign_view_users)
-                if assign_view_groups is not None:
-                    action_instance.assign_view_groups.set(assign_view_groups)
-                if assign_change_users is not None:
-                    action_instance.assign_change_users.set(assign_change_users)
-                if assign_change_groups is not None:
-                    action_instance.assign_change_groups.set(assign_change_groups)
-                if assign_custom_fields is not None:
-                    action_instance.assign_custom_fields.set(assign_custom_fields)
-                if remove_tags is not None:
-                    action_instance.remove_tags.set(remove_tags)
-                if remove_correspondents is not None:
-                    action_instance.remove_correspondents.set(remove_correspondents)
-                if remove_document_types is not None:
-                    action_instance.remove_document_types.set(remove_document_types)
-                if remove_storage_paths is not None:
-                    action_instance.remove_storage_paths.set(remove_storage_paths)
-                if remove_custom_fields is not None:
-                    action_instance.remove_custom_fields.set(remove_custom_fields)
-                if remove_owners is not None:
-                    action_instance.remove_owners.set(remove_owners)
-                if remove_view_users is not None:
-                    action_instance.remove_view_users.set(remove_view_users)
-                if remove_view_groups is not None:
-                    action_instance.remove_view_groups.set(remove_view_groups)
-                if remove_change_users is not None:
-                    action_instance.remove_change_users.set(remove_change_users)
-                if remove_change_groups is not None:
-                    action_instance.remove_change_groups.set(remove_change_groups)
-
-                set_actions.append(action_instance)
-
-        if triggers is not serializers.empty:
-            instance.triggers.set(set_triggers)
-        if actions is not serializers.empty:
-            instance.actions.set(set_actions)
-        instance.save()
-
-    def prune_triggers_and_actions(self) -> None:
-        """
-        ManyToMany fields dont support e.g. on_delete so we need to discard unattached
-        triggers and actions manually
-        """
-        WorkflowTrigger.objects.annotate(
-            workflow_count=Count("workflows"),
-        ).filter(workflow_count=0).delete()
-
-        WorkflowAction.objects.annotate(
-            workflow_count=Count("workflows"),
-        ).filter(workflow_count=0).delete()
-
-        WorkflowActionEmail.objects.filter(action=None).delete()
-        WorkflowActionWebhook.objects.filter(action=None).delete()
-
-    def create(self, validated_data) -> Workflow:
-        if "triggers" in validated_data:
-            triggers = validated_data.pop("triggers")
-
-        if "actions" in validated_data:
-            actions = validated_data.pop("actions")
-            for action in actions:
-                action.pop("id", None)
-
-        instance = super().create(validated_data)
-
-        self.update_triggers_and_actions(instance, triggers, actions)
-
-        return instance
-
-    def update(self, instance: Workflow, validated_data) -> Workflow:
-        triggers = validated_data.pop("triggers", serializers.empty)
-        actions = validated_data.pop("actions", serializers.empty)
-
-        instance = super().update(instance, validated_data)
-
-        self.update_triggers_and_actions(instance, triggers, actions)
-        self.prune_triggers_and_actions()
-
-        return instance

src/documents/signals/handlers.py

@@ -56,7 +56,6 @@ from documents.permissions import get_objects_for_user_owner_aware
 from documents.plugins.helpers import DocumentsStatusManager
 from documents.templating.utils import convert_format_str_to_template_format
 from documents.utils import compute_checksum
-from documents.utils import copy_file_with_basic_stats
 from documents.workflows.actions import build_workflow_action_context
 from documents.workflows.actions import execute_email_action
 from documents.workflows.actions import execute_move_to_trash_action
@@ -364,11 +363,7 @@ def cleanup_document_deletion(sender, instance, **kwargs) -> None:
 
             logger.debug(f"Moving {instance.source_path} to trash at {new_file_path}")
             try:
-                shutil.move(
+                shutil.move(instance.source_path, new_file_path)
-                    instance.source_path,
-                    new_file_path,
-                    copy_function=copy_file_with_basic_stats,
-                )
             except OSError as e:
                 logger.error(
                     f"Failed to move {instance.source_path} to trash at "

src/documents/templates/index.html

@@ -12,7 +12,7 @@
 	<meta name="robots" content="noindex,nofollow">
 	<meta name="author" content="The Paperless-ngx Team">
 	<link rel="icon" type="image/x-icon" href="favicon.ico">
-	<link rel="manifest" href="{% static webmanifest %}" crossorigin="use-credentials">
+	<link rel="manifest" href="{% static webmanifest %}">
 	<link rel="stylesheet" href="{% static styles_css %}">
 	<link rel="apple-touch-icon" href="{% static apple_touch_icon %}">
 </head>

src/documents/tests/conftest.py

@@ -1,41 +1,88 @@
 import shutil
+import zoneinfo
+from collections.abc import Generator
+from dataclasses import dataclass
 from pathlib import Path
 from typing import TYPE_CHECKING
 
 import filelock
 import pytest
+from django.contrib.auth import get_user_model
+from django.contrib.contenttypes.models import ContentType
+from guardian.shortcuts import clear_ct_cache
+from pytest_django.fixtures import Settings
+from rest_framework.test import APIClient
 
-from paperless_testing.factories import DocumentFactory
+from documents.tests.factories import DocumentFactory
+
+UserModelT = get_user_model()
 
 if TYPE_CHECKING:
     from documents.models import Document
-    from paperless_testing.dirs import PaperlessDirs
+
+
+@dataclass(frozen=True, slots=True)
+class PaperlessDirs:
+    """Standard Paperless-ngx directory layout for tests."""
+
+    media: Path
+    originals: Path
+    archive: Path
+    thumbnails: Path
 
 
 @pytest.fixture(scope="session")
-def document_samples_dir() -> Path:
+def samples_dir() -> Path:
     """Path to the shared test sample documents."""
     return Path(__file__).parent / "samples" / "documents"
 
 
+@pytest.fixture()
+def paperless_dirs(tmp_path: Path) -> PaperlessDirs:
+    """Create and return the directory structure for testing."""
+    media = tmp_path / "media"
+    dirs = PaperlessDirs(
+        media=media,
+        originals=media / "documents" / "originals",
+        archive=media / "documents" / "archive",
+        thumbnails=media / "documents" / "thumbnails",
+    )
+    for d in (dirs.originals, dirs.archive, dirs.thumbnails):
+        d.mkdir(parents=True)
+    return dirs
+
+
+@pytest.fixture()
+def _media_settings(paperless_dirs: PaperlessDirs, settings) -> None:
+    """Configure Django settings to point at temp directories."""
+    settings.MEDIA_ROOT = paperless_dirs.media
+    settings.ORIGINALS_DIR = paperless_dirs.originals
+    settings.ARCHIVE_DIR = paperless_dirs.archive
+    settings.THUMBNAIL_DIR = paperless_dirs.thumbnails
+    settings.MEDIA_LOCK = paperless_dirs.media / "media.lock"
+    settings.IGNORABLE_FILES = {".DS_Store", "Thumbs.db", "desktop.ini"}
+    settings.APP_LOGO = ""
+
+
 @pytest.fixture()
 def sample_doc(
-    paperless_dirs: "PaperlessDirs",
+    paperless_dirs: PaperlessDirs,
-    document_samples_dir: Path,
+    _media_settings: None,
+    samples_dir: Path,
 ) -> "Document":
     """Create a document with valid files and matching checksums."""
-    with filelock.FileLock(paperless_dirs.media_lock):
+    with filelock.FileLock(paperless_dirs.media / "media.lock"):
         shutil.copy(
-            document_samples_dir / "originals" / "0000001.pdf",
+            samples_dir / "originals" / "0000001.pdf",
-            paperless_dirs.originals_dir / "0000001.pdf",
+            paperless_dirs.originals / "0000001.pdf",
         )
         shutil.copy(
-            document_samples_dir / "archive" / "0000001.pdf",
+            samples_dir / "archive" / "0000001.pdf",
-            paperless_dirs.archive_dir / "0000001.pdf",
+            paperless_dirs.archive / "0000001.pdf",
         )
         shutil.copy(
-            document_samples_dir / "thumbnails" / "0000001.webp",
+            samples_dir / "thumbnails" / "0000001.webp",
-            paperless_dirs.thumbnail_dir / "0000001.webp",
+            paperless_dirs.thumbnails / "0000001.webp",
         )
 
     return DocumentFactory(
@@ -50,17 +97,95 @@ def sample_doc(
     )
 
 
-@pytest.fixture
+@pytest.fixture()
-def _search_index(paperless_dirs: "PaperlessDirs") -> None:
+def _search_index(
-    """Point the search backend at a fresh, empty index directory.
+    tmp_path: Path,
+    settings: Settings,
+) -> Generator[None, None, None]:
+    """Create a temp index directory and point INDEX_DIR at it.
 
-    paperless_dirs owns INDEX_DIR and resets the backend singleton on both
+    Resets the backend singleton before and after so each test gets a clean
-    sides of the test, so requesting it is all that is needed.
+    index rather than reusing a stale singleton from another test.
     """
+    from documents.search import reset_backend
+
+    index_dir = tmp_path / "index"
+    index_dir.mkdir()
+    settings.INDEX_DIR = index_dir
+    reset_backend()
+    yield
+    reset_backend()
+
+
+@pytest.fixture()
+def settings_timezone(settings: Settings) -> zoneinfo.ZoneInfo:
+    return zoneinfo.ZoneInfo(settings.TIME_ZONE)
 
 
 @pytest.fixture
-def searchable_document(_search_index: None) -> "Document":
+def rest_api_client():
+    """
+    The basic DRF ApiClient
+    """
+    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(autouse=True)
+def _clear_content_type_caches() -> None:
+    """Clear Django's ContentType cache and guardian's lru_cache before each test.
+
+    Tests that delete and reinsert ContentType/Permission rows (e.g. the
+    importer) corrupt both caches. Without this fixture a subsequent test on
+    the same xdist worker sees stale ContentType objects and guardian raises
+    MixedContentTypeError.
+    """
+    ContentType.objects.clear_cache()
+    clear_ct_cache()
+
+
+@pytest.fixture(scope="session", autouse=True)
+def faker_session_locale():
+    """Set Faker locale for reproducibility."""
+    return "en_US"
+
+
+@pytest.fixture(scope="session", autouse=True)
+def faker_seed():
+    return 12345
+
+
+@pytest.fixture
+def indexed_document(_search_index: None) -> "Document":
     """One searchable document, for tests about what the search endpoint
     returns rather than about what it finds.
     """

src/documents/tests/factories.py

@@ -5,7 +5,8 @@ Factory-boy factories for documents app models.
 from __future__ import annotations
 
 import factory
-from django.contrib.auth.models import User
+from django.contrib.auth import get_user_model
+from factory.django import DjangoModelFactory
 
 from documents.models import Correspondent
 from documents.models import Document
@@ -14,10 +15,11 @@ from documents.models import MatchingModel
 from documents.models import PaperlessTask
 from documents.models import StoragePath
 from documents.models import Tag
-from paperless_testing.typed_factory import TypedModelFactory
+
+UserModelT = get_user_model()
 
 
-class CorrespondentFactory(TypedModelFactory[Correspondent]):
+class CorrespondentFactory(DjangoModelFactory[Correspondent]):
     class Meta:
         model = Correspondent
 
@@ -26,7 +28,7 @@ class CorrespondentFactory(TypedModelFactory[Correspondent]):
     matching_algorithm = MatchingModel.MATCH_NONE
 
 
-class DocumentTypeFactory(TypedModelFactory[DocumentType]):
+class DocumentTypeFactory(DjangoModelFactory[DocumentType]):
     class Meta:
         model = DocumentType
 
@@ -35,7 +37,7 @@ class DocumentTypeFactory(TypedModelFactory[DocumentType]):
     matching_algorithm = MatchingModel.MATCH_NONE
 
 
-class TagFactory(TypedModelFactory[Tag]):
+class TagFactory(DjangoModelFactory[Tag]):
     class Meta:
         model = Tag
 
@@ -45,7 +47,7 @@ class TagFactory(TypedModelFactory[Tag]):
     is_inbox_tag = False
 
 
-class StoragePathFactory(TypedModelFactory[StoragePath]):
+class StoragePathFactory(DjangoModelFactory[StoragePath]):
     class Meta:
         model = StoragePath
 
@@ -57,7 +59,7 @@ class StoragePathFactory(TypedModelFactory[StoragePath]):
     matching_algorithm = MatchingModel.MATCH_NONE
 
 
-class DocumentFactory(TypedModelFactory[Document]):
+class DocumentFactory(DjangoModelFactory[Document]):
     class Meta:
         model = Document
 
@@ -69,24 +71,21 @@ class DocumentFactory(TypedModelFactory[Document]):
     storage_path = None
 
 
-class UserFactory(TypedModelFactory[User]):
+class UserFactory(DjangoModelFactory[UserModelT]):
     class Meta:
-        model = User
+        model = UserModelT
 
     username = factory.Sequence(lambda n: f"user{n}")
     is_staff = False
     is_superuser = False
-    # Hashing a real password costs about half a second per user, and no test
+    password = factory.django.Password("test")
-    # authenticates with its password. None gives the unusable password
-    # create_user() gives.
-    password = factory.django.Password(None)
 
     class Params:
         superuser = factory.Trait(is_staff=True, is_superuser=True)
         staff = factory.Trait(is_staff=True)
 
 
-class PaperlessTaskFactory(TypedModelFactory[PaperlessTask]):
+class PaperlessTaskFactory(DjangoModelFactory[PaperlessTask]):
     class Meta:
         model = PaperlessTask
 

src/documents/tests/helpers.py

@@ -1,10 +0,0 @@
-import re
-
-
-def dummy_preprocess(content: str) -> str:
-    """
-    Simpler, faster pre-processing for testing purposes
-    """
-    content = content.lower().strip()
-    content = re.sub(r"\s+", " ", content)
-    return content

src/documents/tests/management/test_management_sanity_checker.py

@@ -15,11 +15,11 @@ from rich.console import Console
 
 from documents.management.commands.document_sanity_checker import Command
 from documents.sanity_checker import SanityCheckMessages
-from paperless_testing.factories import DocumentFactory
+from documents.tests.factories import DocumentFactory
 
 if TYPE_CHECKING:
     from documents.models import Document
-    from paperless_testing.dirs import PaperlessDirs
+    from documents.tests.conftest import PaperlessDirs
 
 
 def _render_to_string(messages: SanityCheckMessages) -> str:
@@ -71,7 +71,7 @@ class TestRenderResultsWithIssues:
         assert "INFO" in output
         assert "No OCR data" in output
 
-    @pytest.mark.usefixtures("paperless_dirs")
+    @pytest.mark.usefixtures("_media_settings")
     def test_global_message(self) -> None:
         msgs = SanityCheckMessages()
         msgs.warning(None, "Orphaned file: /tmp/stray.pdf")
@@ -87,7 +87,7 @@ class TestRenderResultsWithIssues:
         assert "Thumbnail missing" in output
         assert "Checksum mismatch" in output
 
-    @pytest.mark.usefixtures("paperless_dirs")
+    @pytest.mark.usefixtures("_media_settings")
     def test_unknown_doc_pk(self) -> None:
         msgs = SanityCheckMessages()
         msgs.error(99999, "Ghost document")
@@ -184,6 +184,7 @@ class TestDocumentSanityCheckerCommand:
         assert "ERROR" in output
         assert "Original of document does not exist" in output
 
+    @pytest.mark.usefixtures("_media_settings")
     def test_checksum_mismatch(self, paperless_dirs: PaperlessDirs) -> None:
         """Lightweight document with zero-byte files triggers checksum mismatch."""
         doc = DocumentFactory(

src/documents/tests/search/conftest.py

@@ -9,21 +9,29 @@ from documents.search._backend import TantivyBackend
 from documents.search._backend import reset_backend
 from documents.search._schema import build_schema
 from documents.search._tokenizer import register_tokenizers
-from paperless_testing.factories import DocumentFactory
+from documents.tests.factories import DocumentFactory
 
 if TYPE_CHECKING:
     from collections.abc import Callable
     from collections.abc import Generator
+    from pathlib import Path
 
     from pytest_django.fixtures import Settings
 
     from documents.models import Document
-    from paperless_testing.dirs import PaperlessDirs
 
 
 @pytest.fixture
-def backend(paperless_dirs: PaperlessDirs) -> Generator[TantivyBackend, None, None]:
+def index_dir(tmp_path: Path, settings: Settings) -> Path:
-    b = TantivyBackend(path=paperless_dirs.index_dir)
+    path = tmp_path / "index"
+    path.mkdir()
+    settings.INDEX_DIR = path
+    return path
+
+
+@pytest.fixture
+def backend() -> Generator[TantivyBackend, None, None]:
+    b = TantivyBackend()  # path=None → in-memory index
     b.open()
     try:
         yield b

src/documents/tests/search/test_acceptance.py

@@ -14,6 +14,7 @@ from typing import TYPE_CHECKING
 
 import pytest
 import time_machine
+from django.contrib.auth.models import User
 
 from documents.models import CustomField
 from documents.models import CustomFieldInstance
@@ -21,8 +22,7 @@ from documents.models import DocumentType
 from documents.models import Note
 from documents.models import StoragePath
 from documents.search._query import parse_user_query
-from paperless_testing.factories import DocumentFactory
+from documents.tests.factories import DocumentFactory
-from paperless_testing.factories import UserFactory
 
 if TYPE_CHECKING:
     from collections.abc import Callable
@@ -148,7 +148,7 @@ class TestJsonSubpaths:
         THEN:
             - Only the document with alice's note matches
         """
-        alice = UserFactory(username="alice")
+        alice = User.objects.create_user(username="alice")
         doc_with_note = DocumentFactory(
             title="Has note",
             content="x",

src/documents/tests/search/test_backend.py

@@ -3,8 +3,10 @@ from pathlib import Path
 
 import pytest
 from django.contrib.auth.models import Group
+from django.contrib.auth.models import User
 from django.db import connection
 from django.test.utils import CaptureQueriesContext
+from guardian.shortcuts import assign_perm
 from pytest_mock import MockerFixture
 
 from documents.models import CustomField
@@ -17,12 +19,11 @@ from documents.search._backend import WriteBatch
 from documents.search._backend import get_backend
 from documents.search._backend import reset_backend
 from documents.signals.handlers import add_to_index
-from paperless_testing.factories import CorrespondentFactory
+from documents.tests.factories import CorrespondentFactory
-from paperless_testing.factories import DocumentFactory
+from documents.tests.factories import DocumentFactory
-from paperless_testing.factories import DocumentTypeFactory
+from documents.tests.factories import DocumentTypeFactory
-from paperless_testing.factories import TagFactory
+from documents.tests.factories import TagFactory
-from paperless_testing.factories import UserFactory
+from documents.tests.factories import UserFactory
-from paperless_testing.permissions import grant_object
 
 pytestmark = [pytest.mark.search, pytest.mark.django_db]
 
@@ -188,7 +189,7 @@ class TestAddOrUpdateIds:
             pk=1,
             owner=owner,
         )
-        grant_object(user, doc, "view_document")
+        assign_perm("view_document", user, doc)
 
         with backend.batch_update() as batch:
             batch.add_or_update_ids([doc.pk])
@@ -208,7 +209,7 @@ class TestAddOrUpdateIds:
             pk=1,
             owner=owner,
         )
-        grant_object(group, doc, "view_document")
+        assign_perm("view_document", group, doc)
 
         with backend.batch_update() as batch:
             batch.add_or_update_ids([doc.pk])
@@ -762,8 +763,8 @@ class TestSearchIds:
 
     def test_respects_permission_filter(self, backend: TantivyBackend) -> None:
         """search_ids must respect user permission filtering."""
-        owner = UserFactory(username="ids_owner")
+        owner = User.objects.create_user("ids_owner")
-        other = UserFactory(username="ids_other")
+        other = User.objects.create_user("ids_other")
         doc = Document.objects.create(
             title="private doc",
             content="secret keyword",
@@ -842,7 +843,7 @@ class TestRebuild:
             content="group secret keyword",
             owner=owner,
         )
-        grant_object(group, doc, "view_document")
+        assign_perm("view_document", group, doc)
 
         backend.rebuild(Document.objects.all())
 
@@ -947,8 +948,7 @@ class TestSingleton:
         yield
         reset_backend()
 
-    @pytest.mark.usefixtures("paperless_dirs")
+    def test_returns_same_instance_on_repeated_calls(self, index_dir) -> None:
-    def test_returns_same_instance_on_repeated_calls(self) -> None:
         """Singleton pattern: repeated calls to get_backend() must return the same instance."""
         assert get_backend() is get_backend()
 
@@ -965,8 +965,7 @@ class TestSingleton:
         assert b1 is not b2
         assert b2._path == tmp_path / "b"
 
-    @pytest.mark.usefixtures("paperless_dirs")
+    def test_reset_forces_new_instance(self, index_dir) -> None:
-    def test_reset_forces_new_instance(self) -> None:
         """reset_backend() must force creation of a new backend instance on next get_backend() call."""
         b1 = get_backend()
         reset_backend()
@@ -1072,7 +1071,7 @@ class TestFieldHandling:
 
     def test_notes_include_user_information(self, backend: TantivyBackend) -> None:
         """Notes must be indexed with user information when available for structured queries."""
-        user = UserFactory(username="notewriter")
+        user = User.objects.create_user("notewriter")
         doc = Document.objects.create(
             title="Doc with notes",
             content="test",
@@ -1174,7 +1173,7 @@ class TestHighlightHits:
         notes.note: prefix so the query targets notes content directly, but
         the snippet is generated from notes_text which stores the same text.
         """
-        user = UserFactory(username="hl_noteuser")
+        user = User.objects.create_user("hl_noteuser")
         doc = Document.objects.create(
             title="Doc with matching note",
             content="unrelated content",

src/documents/tests/search/test_documented_syntax.py

@@ -27,7 +27,7 @@ import time_machine
 from documents.models import Note
 from documents.models import Tag
 from documents.search._errors import InvalidDateQuery
-from paperless_testing.factories import DocumentFactory
+from documents.tests.factories import DocumentFactory
 
 if TYPE_CHECKING:
     from collections.abc import Callable
@@ -269,7 +269,7 @@ class TestDocumentedDateForms:
             yield
 
     @pytest.fixture
-    def dated(self, backend: TantivyBackend) -> dict[str, int]:
+    def dated(self, index_document: Callable[..., Document]) -> dict[str, int]:
         stamps = {
             "today": datetime(2026, 6, 15, 9, 0, tzinfo=UTC),
             "yesterday": datetime(2026, 6, 14, 9, 0, tzinfo=UTC),
@@ -279,14 +279,14 @@ class TestDocumentedDateForms:
             "january": datetime(2026, 1, 10, 10, 0, tzinfo=UTC),
             "old": datetime(2005, 3, 4, 15, 30, tzinfo=UTC),
         }
-        docs = {
+        return {
-            label: DocumentFactory(title=label, content="dated body", added=stamp)
+            label: index_document(
+                title=label,
+                content="dated body",
+                added=stamp,
+            ).pk
             for label, stamp in stamps.items()
         }
-        with backend.batch_update() as batch:
-            for doc in docs.values():
-                batch.add_or_update(doc)
-        return {label: doc.pk for label, doc in docs.items()}
 
     @pytest.mark.parametrize(
         ("query", "label"),

src/documents/tests/search/test_highlight_query_guard.py

@@ -29,7 +29,7 @@ from rest_framework import status
 
 from documents.search._backend import SearchMode
 from documents.search._query import parse_simple_text_highlight_query
-from paperless_testing.factories import DocumentFactory
+from documents.tests.factories import DocumentFactory
 
 if TYPE_CHECKING:
     from rest_framework.test import APIClient

src/documents/tests/search/test_json_field_prefixes.py

@@ -17,12 +17,12 @@ from __future__ import annotations
 from typing import TYPE_CHECKING
 
 import pytest
+from django.contrib.auth.models import User
 
 from documents.models import CustomField
 from documents.models import CustomFieldInstance
 from documents.models import Note
-from paperless_testing.factories import DocumentFactory
+from documents.tests.factories import DocumentFactory
-from paperless_testing.factories import UserFactory
 
 if TYPE_CHECKING:
     from collections.abc import Callable
@@ -53,7 +53,7 @@ class TestBareJsonFieldPrefixes:
               decoy's content match does not resurface through a demoted
               text search
         """
-        alice = UserFactory(username="alice")
+        alice = User.objects.create_user(username="alice")
         with_note = DocumentFactory(title="Has note", content="x")
         Note.objects.create(document=with_note, user=alice, note="crocodile")
         backend.add_or_update(with_note)
@@ -116,7 +116,7 @@ class TestBareJsonFieldPrefixes:
               document; the default-subpath resolution for the bare
               prefix does not interfere with explicit subpath addressing
         """
-        bob = UserFactory(username="bob")
+        bob = User.objects.create_user(username="bob")
         doc = DocumentFactory(title="Bob note", content="x")
         Note.objects.create(document=doc, user=bob, note="remark")
         backend.add_or_update(doc)

src/documents/tests/search/test_json_subpath_completeness.py

@@ -20,6 +20,7 @@ from typing import TYPE_CHECKING
 
 import pytest
 import tantivy
+from django.contrib.auth.models import User
 from whoosh_compat import FieldKind
 
 from documents.models import CustomField
@@ -27,7 +28,6 @@ from documents.models import CustomFieldInstance
 from documents.models import Document
 from documents.models import Note
 from documents.search._fields import PUBLIC_FIELDS
-from paperless_testing.factories import UserFactory
 
 if TYPE_CHECKING:
     from documents.search._backend import TantivyBackend
@@ -49,7 +49,7 @@ class TestJsonSubpathsAreWrittenAtIndexTime:
             - Every subpath PUBLIC_FIELDS declares for notes/custom_fields
               is present as a key in the document's stored JSON payload
         """
-        user = UserFactory(username="completeness-user")
+        user = User.objects.create_user(username="completeness-user")
         field = CustomField.objects.create(
             name="Completeness Field",
             data_type=CustomField.FieldDataType.STRING,

src/documents/tests/search/test_lock_backoff.py

@@ -15,7 +15,7 @@ from documents.search._backend import SearchIndexLockError
 from documents.search._backend import TantivyBackend
 from documents.tasks import index_document
 from documents.tasks import remove_document_from_index
-from paperless_testing.factories import DocumentFactory
+from documents.tests.factories import DocumentFactory
 
 if TYPE_CHECKING:
     from collections.abc import Generator

src/documents/tests/search/test_migration_fulltext_query_field_prefixes.py

@@ -1,6 +1,6 @@
 import pytest
 
-from paperless_testing.migrations import TestMigrations
+from documents.tests.utils import TestMigrations
 
 pytestmark = pytest.mark.search
 

src/documents/tests/search/test_permission_field_isolation.py

@@ -18,14 +18,13 @@ from typing import TYPE_CHECKING
 import pytest
 from django.contrib.auth.models import Group
 from django.contrib.auth.models import User
+from guardian.shortcuts import assign_perm
 
 from documents.models import Correspondent
 from documents.models import Document
 from documents.models import DocumentType
 from documents.models import StoragePath
 from documents.models import Tag
-from paperless_testing.factories import UserFactory
-from paperless_testing.permissions import grant_object
 
 if TYPE_CHECKING:
     from documents.search._backend import TantivyBackend
@@ -35,22 +34,22 @@ pytestmark = [pytest.mark.search, pytest.mark.django_db]
 
 @pytest.fixture
 def owner() -> User:
-    return UserFactory(username="owner")
+    return User.objects.create_user(username="owner")
 
 
 @pytest.fixture
 def stranger() -> User:
-    return UserFactory(username="stranger")
+    return User.objects.create_user(username="stranger")
 
 
 @pytest.fixture
 def viewer() -> User:
-    return UserFactory(username="viewer")
+    return User.objects.create_user(username="viewer")
 
 
 @pytest.fixture
 def group_member() -> User:
-    user = UserFactory(username="group_member")
+    user = User.objects.create_user(username="group_member")
     user.groups.add(Group.objects.create(name="accounting"))
     return user
 
@@ -128,7 +127,7 @@ class TestPermissionFilteringOnIndexedDocuments:
             checksum="perm-shared-user",
             owner=owner,
         )
-        grant_object(viewer, doc, "view_document")
+        assign_perm("view_document", viewer, doc)
         backend.add_or_update(doc)
 
         assert backend.search_ids("invoice", user=viewer) == [doc.pk]
@@ -158,7 +157,7 @@ class TestPermissionFilteringOnIndexedDocuments:
             checksum="perm-shared-group",
             owner=owner,
         )
-        grant_object(group_member.groups.first(), doc, "view_document")
+        assign_perm("view_document", group_member.groups.first(), doc)
         backend.add_or_update(doc)
 
         assert backend.search_ids("invoice", user=group_member) == [doc.pk]

src/documents/tests/search/test_schema.py

@@ -13,11 +13,11 @@ from documents.search._schema import needs_rebuild
 from documents.search._schema import schema_fingerprint
 
 if TYPE_CHECKING:
+    from pathlib import Path
+
     import tantivy
     from pytest_django.fixtures import Settings
 
-    from paperless_testing.dirs import PaperlessDirs
-
 
 pytestmark = pytest.mark.search
 
@@ -25,19 +25,16 @@ pytestmark = pytest.mark.search
 class TestNeedsRebuild:
     """needs_rebuild covers all sentinel-file states that require a full reindex."""
 
-    def test_returns_true_when_settings_file_missing(
+    def test_returns_true_when_settings_file_missing(self, index_dir: Path) -> None:
-        self,
+        assert needs_rebuild(index_dir) is True
-        paperless_dirs: PaperlessDirs,
-    ) -> None:
-        assert needs_rebuild(paperless_dirs.index_dir) is True
 
     def test_returns_false_when_version_and_language_match(
         self,
-        paperless_dirs: PaperlessDirs,
+        index_dir: Path,
         settings: Settings,
     ) -> None:
         settings.SEARCH_LANGUAGE = "en"
-        (paperless_dirs.index_dir / ".index_settings.json").write_text(
+        (index_dir / ".index_settings.json").write_text(
             json.dumps(
                 {
                     "schema_version": SCHEMA_VERSION,
@@ -46,51 +43,51 @@ class TestNeedsRebuild:
                 },
             ),
         )
-        assert needs_rebuild(paperless_dirs.index_dir) is False
+        assert needs_rebuild(index_dir) is False
 
     def test_returns_true_on_schema_version_mismatch(
         self,
-        paperless_dirs: PaperlessDirs,
+        index_dir: Path,
         settings: Settings,
     ) -> None:
         settings.SEARCH_LANGUAGE = None
-        (paperless_dirs.index_dir / ".index_settings.json").write_text(
+        (index_dir / ".index_settings.json").write_text(
             json.dumps({"schema_version": SCHEMA_VERSION - 1, "language": None}),
         )
-        assert needs_rebuild(paperless_dirs.index_dir) is True
+        assert needs_rebuild(index_dir) is True
 
     def test_returns_true_when_version_is_not_an_integer(
         self,
-        paperless_dirs: PaperlessDirs,
+        index_dir: Path,
         settings: Settings,
     ) -> None:
         settings.SEARCH_LANGUAGE = None
-        (paperless_dirs.index_dir / ".index_settings.json").write_text(
+        (index_dir / ".index_settings.json").write_text(
             json.dumps({"schema_version": "not-a-number", "language": None}),
         )
-        assert needs_rebuild(paperless_dirs.index_dir) is True
+        assert needs_rebuild(index_dir) is True
 
     def test_returns_true_when_language_key_missing(
         self,
-        paperless_dirs: PaperlessDirs,
+        index_dir: Path,
         settings: Settings,
     ) -> None:
         settings.SEARCH_LANGUAGE = "en"
-        (paperless_dirs.index_dir / ".index_settings.json").write_text(
+        (index_dir / ".index_settings.json").write_text(
             json.dumps({"schema_version": SCHEMA_VERSION}),
         )
-        assert needs_rebuild(paperless_dirs.index_dir) is True
+        assert needs_rebuild(index_dir) is True
 
     def test_returns_true_when_language_differs(
         self,
-        paperless_dirs: PaperlessDirs,
+        index_dir: Path,
         settings: Settings,
     ) -> None:
         settings.SEARCH_LANGUAGE = "de"
-        (paperless_dirs.index_dir / ".index_settings.json").write_text(
+        (index_dir / ".index_settings.json").write_text(
             json.dumps({"schema_version": SCHEMA_VERSION, "language": "en"}),
         )
-        assert needs_rebuild(paperless_dirs.index_dir) is True
+        assert needs_rebuild(index_dir) is True
 
 
 def _schema_fields(schema: tantivy.Schema) -> dict[str, dict]:

src/documents/tests/search/test_schema_fingerprint.py

@@ -35,8 +35,6 @@ if TYPE_CHECKING:
 
     from pytest_django.fixtures import SettingsWrapper
 
-    from paperless_testing.dirs import PaperlessDirs
-
 pytestmark = pytest.mark.search
 
 # The on-disk field layout of a v2 index, pinned as data. Any edit here is an
@@ -471,7 +469,7 @@ def _fingerprint_of(descriptors: list[FieldDescriptor]) -> str:
 class TestNeedsRebuildOnFingerprint:
     def test_matching_fingerprint_does_not_rebuild(
         self,
-        paperless_dirs: PaperlessDirs,
+        index_dir: Path,
         settings: SettingsWrapper,
     ) -> None:
         """
@@ -484,13 +482,13 @@ class TestNeedsRebuildOnFingerprint:
             - It returns False
         """
         settings.SEARCH_LANGUAGE = None
-        _sentinels(paperless_dirs.index_dir)
+        _sentinels(index_dir)
 
-        assert needs_rebuild(paperless_dirs.index_dir) is False
+        assert needs_rebuild(index_dir) is False
 
     def test_stale_fingerprint_rebuilds_despite_a_matching_version(
         self,
-        paperless_dirs: PaperlessDirs,
+        index_dir: Path,
         settings: SettingsWrapper,
         monkeypatch: pytest.MonkeyPatch,
     ) -> None:
@@ -507,7 +505,7 @@ class TestNeedsRebuildOnFingerprint:
               every subsequent write would raise
         """
         settings.SEARCH_LANGUAGE = None
-        _sentinels(paperless_dirs.index_dir)
+        _sentinels(index_dir)
         extended = [
             *field_descriptors(),
             FieldDescriptor(
@@ -521,11 +519,11 @@ class TestNeedsRebuildOnFingerprint:
         ]
         monkeypatch.setattr(_schema, "field_descriptors", lambda: extended)
 
-        assert needs_rebuild(paperless_dirs.index_dir) is True
+        assert needs_rebuild(index_dir) is True
 
     def test_reordered_schema_rebuilds(
         self,
-        paperless_dirs: PaperlessDirs,
+        index_dir: Path,
         settings: SettingsWrapper,
         monkeypatch: pytest.MonkeyPatch,
     ) -> None:
@@ -540,16 +538,16 @@ class TestNeedsRebuildOnFingerprint:
             - It returns True
         """
         settings.SEARCH_LANGUAGE = None
-        _sentinels(paperless_dirs.index_dir)
+        _sentinels(index_dir)
         reordered = field_descriptors()
         reordered[1], reordered[2] = reordered[2], reordered[1]
         monkeypatch.setattr(_schema, "field_descriptors", lambda: reordered)
 
-        assert needs_rebuild(paperless_dirs.index_dir) is True
+        assert needs_rebuild(index_dir) is True
 
     def test_missing_fingerprint_rebuilds(
         self,
-        paperless_dirs: PaperlessDirs,
+        index_dir: Path,
         settings: SettingsWrapper,
     ) -> None:
         """
@@ -563,15 +561,15 @@ class TestNeedsRebuildOnFingerprint:
               is rebuilt rather than trusted
         """
         settings.SEARCH_LANGUAGE = None
-        (paperless_dirs.index_dir / ".index_settings.json").write_text(
+        (index_dir / ".index_settings.json").write_text(
             json.dumps({"schema_version": SCHEMA_VERSION, "language": None}),
         )
 
-        assert needs_rebuild(paperless_dirs.index_dir) is True
+        assert needs_rebuild(index_dir) is True
 
     def test_written_sentinels_satisfy_the_check(
         self,
-        paperless_dirs: PaperlessDirs,
+        index_dir: Path,
         settings: SettingsWrapper,
     ) -> None:
         """
@@ -584,6 +582,6 @@ class TestNeedsRebuildOnFingerprint:
             - It returns False
         """
         settings.SEARCH_LANGUAGE = "en"
-        _write_sentinels(paperless_dirs.index_dir)
+        _write_sentinels(index_dir)
 
-        assert needs_rebuild(paperless_dirs.index_dir) is False
+        assert needs_rebuild(index_dir) is False