Merge remote-tracking branch 'origin/dev'

.github/workflows/ci.yml

@@ -106,15 +106,6 @@ jobs:
       matrix:
         python-version: ['3.8', '3.9', '3.10']
       fail-fast: false
-    env:
-      # Enable Tika end to end testing
-      TIKA_LIVE: 1
-      # Enable paperless_mail testing against real server
-      PAPERLESS_MAIL_TEST_HOST: ${{ secrets.TEST_MAIL_HOST }}
-      PAPERLESS_MAIL_TEST_USER: ${{ secrets.TEST_MAIL_USER }}
-      PAPERLESS_MAIL_TEST_PASSWD: ${{ secrets.TEST_MAIL_PASSWD }}
-      # Enable Gotenberg end to end testing
-      GOTENBERG_LIVE: 1
     steps:
       -
         name: Checkout
@@ -156,12 +147,18 @@ jobs:
           pipenv --python ${{ steps.setup-python.outputs.python-version }} run pip list
       -
         name: Tests
+        env:
+          PAPERLESS_CI_TEST: 1
+          # Enable paperless_mail testing against real server
+          PAPERLESS_MAIL_TEST_HOST: ${{ secrets.TEST_MAIL_HOST }}
+          PAPERLESS_MAIL_TEST_USER: ${{ secrets.TEST_MAIL_USER }}
+          PAPERLESS_MAIL_TEST_PASSWD: ${{ secrets.TEST_MAIL_PASSWD }}
         run: |
           cd src/
           pipenv --python ${{ steps.setup-python.outputs.python-version }} run pytest -ra
       -
         name: Upload coverage to Codecov
-        if: ${{ matrix.python-version == env.DEFAULT_PYTHON_VERSION && github.event_name == 'push'}}
+        if: ${{ matrix.python-version == env.DEFAULT_PYTHON_VERSION }}
         uses: codecov/codecov-action@v3
         with:
           # not required for public repos, but intermittently fails otherwise
@@ -309,7 +306,7 @@ jobs:
           push: ${{ github.event_name != 'pull_request' }}
           tags: ${{ steps.docker-meta.outputs.tags }}
           labels: ${{ steps.docker-meta.outputs.labels }}
-          # Get cache layers from this branch, then dev, then main
+          # Get cache layers from this branch, then dev
           # This allows new branches to get at least some cache benefits, generally from dev
           cache-from: |
             type=registry,ref=ghcr.io/${{ steps.set-ghcr-repository.outputs.ghcr-repository }}/builder/cache/app:${{ github.ref_name }}

.github/workflows/cleanup-tags.yml

@@ -51,14 +51,6 @@ jobs:
         include:
           - primary-name: "paperless-ngx"
           - primary-name: "paperless-ngx/builder/cache/app"
-          - primary-name: "paperless-ngx/builder/qpdf"
-          - primary-name: "paperless-ngx/builder/cache/qpdf"
-          - primary-name: "paperless-ngx/builder/pikepdf"
-          - primary-name: "paperless-ngx/builder/cache/pikepdf"
-          - primary-name: "paperless-ngx/builder/jbig2enc"
-          - primary-name: "paperless-ngx/builder/cache/jbig2enc"
-          - primary-name: "paperless-ngx/builder/psycopg2"
-          - primary-name: "paperless-ngx/builder/cache/psycopg2"
           # TODO: Remove the above and replace with the below
           # - primary-name: "builder/qpdf"
           # - primary-name: "builder/cache/qpdf"

.pre-commit-config.yaml

@@ -37,7 +37,7 @@ repos:
         exclude: "(^Pipfile\\.lock$)"
   # Python hooks
   - repo: https://github.com/charliermarsh/ruff-pre-commit
-    rev: 'v0.0.265'
+    rev: 'v0.0.272'
     hooks:
       - id: ruff
   - repo: https://github.com/psf/black
@@ -57,6 +57,6 @@ repos:
         args:
           - "--tab"
   - repo: https://github.com/shellcheck-py/shellcheck-py
-    rev: "v0.9.0.2"
+    rev: "v0.9.0.5"
     hooks:
       - id: shellcheck

Dockerfile

@@ -5,7 +5,7 @@
 # Purpose: Compiles the frontend
 # Notes:
 #  - Does NPM stuff with Typescript and such
-FROM --platform=$BUILDPLATFORM node:16-bullseye-slim AS compile-frontend
+FROM --platform=$BUILDPLATFORM docker.io/node:16-bookworm-slim AS compile-frontend
 
 COPY ./src-ui /src/src-ui
 
@@ -21,7 +21,7 @@ RUN set -eux \
 # Comments:
 #  - pipenv dependencies are not left in the final image
 #  - pipenv can't touch the final image somehow
-FROM --platform=$BUILDPLATFORM python:3.9-alpine as pipenv-base
+FROM --platform=$BUILDPLATFORM docker.io/python:3.9-alpine as pipenv-base
 
 WORKDIR /usr/src/pipenv
 
@@ -37,7 +37,7 @@ RUN set -eux \
 # Purpose: The final image
 # Comments:
 #  - Don't leave anything extra in here
-FROM python:3.9-slim-bullseye as main-app
+FROM docker.io/python:3.9-slim-bookworm as main-app
 
 LABEL org.opencontainers.image.authors="paperless-ngx team <hello@paperless-ngx.com>"
 LABEL org.opencontainers.image.documentation="https://docs.paperless-ngx.com/"
@@ -70,9 +70,9 @@ ARG RUNTIME_PACKAGES="\
   # Image processing
   liblept5 \
   liblcms2-2 \
-  libtiff5 \
+  libtiff6 \
   libfreetype6 \
-  libwebp6 \
+  libwebp7 \
   libopenjp2-7 \
   libimagequant0 \
   libraqm0 \
@@ -98,6 +98,8 @@ ARG RUNTIME_PACKAGES="\
   libxml2 \
   libxslt1.1 \
   libgnutls30 \
+  libqpdf29 \
+  qpdf \
   # Mime type detection
   file \
   libmagic1 \
@@ -181,7 +183,7 @@ ARG PSYCOPG2_VERSION=2.9.6
 RUN set -eux \
   && echo "Getting binaries" \
     && mkdir paperless-ngx \
-    && curl --fail --silent --show-error --output paperless-ngx.tar.gz --location https://github.com/paperless-ngx/builder/archive/3d6574e2dbaa8b8cdced864a256b0de59015f605.tar.gz \
+    && curl --fail --silent --show-error --output paperless-ngx.tar.gz --location https://github.com/paperless-ngx/builder/archive/1f0e6665ba1b144f70fd6dfc8d0e8ba3b7a578ee.tar.gz \
     && tar -xf paperless-ngx.tar.gz --directory paperless-ngx --strip-components=1 \
     && cd paperless-ngx \
     # Setting a specific revision ensures we know what this installed
@@ -189,9 +191,7 @@ RUN set -eux \
   && echo "Installing jbig2enc" \
     && cp ./jbig2enc/${JBIG2ENC_VERSION}/${TARGETARCH}${TARGETVARIANT}/jbig2 /usr/local/bin/ \
     && cp ./jbig2enc/${JBIG2ENC_VERSION}/${TARGETARCH}${TARGETVARIANT}/libjbig2enc* /usr/local/lib/ \
-  && echo "Installing qpdf" \
-    && apt-get install --yes --no-install-recommends ./qpdf/${QPDF_VERSION}/${TARGETARCH}${TARGETVARIANT}/libqpdf29_*.deb \
-    && apt-get install --yes --no-install-recommends ./qpdf/${QPDF_VERSION}/${TARGETARCH}${TARGETVARIANT}/qpdf_*.deb \
+    && chmod a+x /usr/local/bin/jbig2 \
   && echo "Installing pikepdf and dependencies" \
     && python3 -m pip install --no-cache-dir ./pikepdf/${PIKEPDF_VERSION}/${TARGETARCH}${TARGETVARIANT}/*.whl \
     && python3 -m pip list \
@@ -214,8 +214,7 @@ COPY --from=pipenv-base /usr/src/pipenv/requirements.txt ./
 ARG BUILD_PACKAGES="\
   build-essential \
   git \
-  default-libmysqlclient-dev \
-  python3-dev"
+  default-libmysqlclient-dev"
 
 RUN set -eux \
   && echo "Installing build system packages" \

Pipfile

@@ -37,14 +37,13 @@ psycopg2 = "*"
 rapidfuzz = "*"
 redis = {extras = ["hiredis"], version = "*"}
 scikit-learn = "~=1.2"
-numpy = "*"
 whitenoise = "~=6.3"
 watchdog = "~=2.2"
 whoosh="~=2.7"
 inotifyrecursive = "~=0.3"
 ocrmypdf = "~=14.0"
 tqdm = "*"
-tika = "*"
+tika-client = "*"
 channels = "~=4.0"
 channels-redis = "*"
 uvicorn = {extras = ["standard"], version = "*"}
@@ -78,6 +77,7 @@ factory-boy = "*"
 pytest = "*"
 pytest-cov = "*"
 pytest-django = "*"
+pytest-httpx = "*"
 pytest-env = "*"
 pytest-sugar = "*"
 pytest-xdist = "*"

docker/docker-prepare.sh

@@ -80,7 +80,7 @@ django_checks() {
 
 search_index() {
 
-	local -r index_version=5
+	local -r index_version=6
 	local -r index_version_file=${DATA_DIR}/.index_version
 
 	if [[ (! -f "${index_version_file}") || $(<"${index_version_file}") != "$index_version" ]]; then

docker/wait-for-redis.py

@@ -28,7 +28,7 @@ if __name__ == "__main__":
             except Exception as e:
                 print(
                     f"Redis ping #{attempt} failed.\n"
-                    f"Error: {str(e)}.\n"
+                    f"Error: {e!s}.\n"
                     f"Waiting {RETRY_SLEEP_SECONDS}s",
                     flush=True,
                 )

docs/configuration.md

@@ -136,11 +136,11 @@ changed here.
 
     Defaults to unset, using the documented path in the home directory.
 
-`PAPERLESS_DB_TIMEOUT=<float>`
+`PAPERLESS_DB_TIMEOUT=<int>`
 
 : Amount of time for a database connection to wait for the database to
-unlock. Mostly applicable for an sqlite based installation, consider
-changing to postgresql if you need to increase this.
+unlock. Mostly applicable for sqlite based installation. Consider changing
+to postgresql if you are having concurrency problems with sqlite.
 
     Defaults to unset, keeping the Django defaults.
 

src-ui/messages.xlf

@@ -2106,6 +2106,10 @@
           <context context-type="sourcefile">src/app/components/common/tag/tag.component.html</context>
           <context context-type="linenumber">8</context>
         </context-group>
+        <context-group purpose="location">
+          <context context-type="sourcefile">src/app/components/document-list/document-card-small/document-card-small.component.ts</context>
+          <context context-type="linenumber">80</context>
+        </context-group>
       </trans-unit>
       <trans-unit id="6560126119609945418" datatype="html">
         <source>Add tag</source>

src-ui/src/app/components/document-list/document-card-small/document-card-small.component.html

@@ -29,7 +29,7 @@
     <div class="card-body bg-light p-2">
       <p class="card-text">
         <ng-container *ngIf="document.correspondent">
-          <a title="Toggle correspondent filter" i18n-title (click)="clickCorrespondent.emit(document.correspondent);$event.stopPropagation()" class="fw-bold btn-link">{{(document.correspondent$ | async)?.name}}</a>:
+          <a title="Toggle correspondent filter" i18n-title (click)="clickCorrespondent.emit(document.correspondent);$event.stopPropagation()" class="fw-bold btn-link">{{(document.correspondent$ | async)?.name ?? privateName}}</a>:
         </ng-container>
         {{document.title | documentTitle}}
       </p>
@@ -41,14 +41,14 @@
           <svg class="metadata-icon me-2 text-muted" fill="currentColor">
             <use xlink:href="assets/bootstrap-icons.svg#file-earmark"/>
           </svg>
-          <small>{{(document.document_type$ | async)?.name}}</small>
+          <small>{{(document.document_type$ | async)?.name ?? privateName}}</small>
         </button>
         <button *ngIf="document.storage_path" type="button" class="list-group-item list-group-item-action bg-transparent ps-0 p-1 border-0" title="Toggle storage path filter" i18n-title
          (click)="clickStoragePath.emit(document.storage_path);$event.stopPropagation()">
           <svg class="metadata-icon me-2 text-muted" fill="currentColor">
             <use xlink:href="assets/bootstrap-icons.svg#folder"/>
           </svg>
-          <small>{{(document.storage_path$ | async)?.name}}</small>
+          <small>{{(document.storage_path$ | async)?.name ?? privateName}}</small>
         </button>
         <div class="list-group-item bg-transparent p-0 border-0 d-flex flex-wrap-reverse justify-content-between">
           <ng-template #dateTooltip>

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

@@ -76,6 +76,10 @@ export class DocumentCardSmallComponent extends ComponentWithPermissions {
     return this.documentService.getPreviewUrl(this.document.id)
   }
 
+  get privateName() {
+    return $localize`Private`
+  }
+
   getTagsLimited$() {
     const limit = this.document.notes.length > 0 ? 6 : 7
     return this.document.tags$.pipe(

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

@@ -5,7 +5,7 @@ export const environment = {
   apiBaseUrl: document.baseURI + 'api/',
   apiVersion: '3',
   appTitle: 'Paperless-ngx',
-  version: '1.15.1',
+  version: '1.15.1-dev',
   webSocketHost: window.location.host,
   webSocketProtocol: window.location.protocol == 'https:' ? 'wss:' : 'ws:',
   webSocketBaseUrl: base_url.pathname + 'ws/',

src/documents/barcodes.py

@@ -121,7 +121,7 @@ class BarcodeReader:
             if barcode.text:
                 barcodes.append(barcode.text)
                 logger.debug(
-                    f"Barcode of type {str(barcode.format)} found: {barcode.text}",
+                    f"Barcode of type {barcode.format} found: {barcode.text}",
                 )
 
         return barcodes
@@ -141,7 +141,7 @@ class BarcodeReader:
                 decoded_barcode = barcode.data.decode("utf-8")
                 barcodes.append(decoded_barcode)
                 logger.debug(
-                    f"Barcode of type {str(barcode.type)} found: {decoded_barcode}",
+                    f"Barcode of type {barcode.type} found: {decoded_barcode}",
                 )
 
         return barcodes
@@ -180,6 +180,9 @@ class BarcodeReader:
         with scratch_image.open("rb") as img_file, self.pdf_file.open("wb") as pdf_file:
             pdf_file.write(img2pdf.convert(img_file))
 
+        # Copy what file stat is possible
+        shutil.copystat(self.file, self.pdf_file)
+
     def detect(self) -> None:
         """
         Scan all pages of the PDF as images, updating barcodes and the pages
@@ -292,6 +295,9 @@ class BarcodeReader:
                 savepath = Path(self.temp_dir.name) / output_filename
                 with open(savepath, "wb") as out:
                     dst.save(out)
+
+                shutil.copystat(self.file, savepath)
+
                 document_paths.append(savepath)
 
             return document_paths
@@ -342,7 +348,7 @@ class BarcodeReader:
 
         for idx, document_path in enumerate(doc_paths):
             if override_name is not None:
-                newname = f"{str(idx)}_{override_name}"
+                newname = f"{idx}_{override_name}"
                 dest = save_to_dir / newname
             else:
                 dest = save_to_dir

src/documents/consumer.py

@@ -69,7 +69,7 @@ class Consumer(LoggingMixin):
         status,
         message=None,
         document_id=None,
-    ):
+    ):  # pragma: no cover
         payload = {
             "filename": os.path.basename(self.filename) if self.filename else None,
             "task_id": self.task_id,
@@ -326,7 +326,7 @@ class Consumer(LoggingMixin):
             dir=settings.SCRATCH_DIR,
         )
         self.path = Path(tempdir.name) / Path(self.filename)
-        shutil.copy(self.original_path, self.path)
+        shutil.copy2(self.original_path, self.path)
 
         # Determine the parser class.
 
@@ -352,7 +352,7 @@ class Consumer(LoggingMixin):
 
         self.run_pre_consume_script()
 
-        def progress_callback(current_progress, max_progress):
+        def progress_callback(current_progress, max_progress):  # pragma: no cover
             # recalculate progress to be within 20 and 80
             p = int((current_progress / max_progress) * 50 + 20)
             self._send_progress(p, 100, "WORKING")
@@ -582,6 +582,7 @@ class Consumer(LoggingMixin):
     def _write(self, storage_type, source, target):
         with open(source, "rb") as read_file, open(target, "wb") as write_file:
             write_file.write(read_file.read())
+        shutil.copystat(source, target)
 
     def _log_script_outputs(self, completed_process: CompletedProcess):
         """

src/documents/filters.py

@@ -116,6 +116,8 @@ class DocumentFilterSet(FilterSet):
             "created": DATE_KWARGS,
             "added": DATE_KWARGS,
             "modified": DATE_KWARGS,
+            "original_filename": CHAR_KWARGS,
+            "checksum": CHAR_KWARGS,
             "correspondent": ["isnull"],
             "correspondent__id": ID_KWARGS,
             "correspondent__name": CHAR_KWARGS,

src/documents/index.py

@@ -64,6 +64,8 @@ def get_schema():
         owner_id=NUMERIC(),
         has_owner=BOOLEAN(),
         viewer_id=KEYWORD(commas=True),
+        checksum=TEXT(),
+        original_filename=TEXT(sortable=True),
     )
 
 
@@ -149,6 +151,8 @@ def update_document(writer: AsyncWriter, doc: Document):
         owner_id=doc.owner.id if doc.owner else None,
         has_owner=doc.owner is not None,
         viewer_id=viewer_ids if viewer_ids else None,
+        checksum=doc.checksum,
+        original_filename=doc.original_filename,
     )
 
 
@@ -171,91 +175,85 @@ def remove_document_from_index(document):
 
 
 class DelayedQuery:
+    param_map = {
+        "correspondent": ("correspondent", ["id", "id__in", "id__none", "isnull"]),
+        "document_type": ("type", ["id", "id__in", "id__none", "isnull"]),
+        "storage_path": ("path", ["id", "id__in", "id__none", "isnull"]),
+        "owner": ("owner", ["id", "id__in", "id__none", "isnull"]),
+        "tags": ("tag", ["id__all", "id__in", "id__none"]),
+        "added": ("added", ["date__lt", "date__gt"]),
+        "created": ("created", ["date__lt", "date__gt"]),
+        "checksum": ("checksum", ["icontains", "istartswith"]),
+        "original_filename": ("original_filename", ["icontains", "istartswith"]),
+    }
+
     def _get_query(self):
         raise NotImplementedError
 
     def _get_query_filter(self):
         criterias = []
-        for k, v in self.query_params.items():
-            if k == "correspondent__id":
-                criterias.append(query.Term("correspondent_id", v))
-            elif k == "correspondent__id__in":
-                correspondents_in = []
-                for correspondent_id in v.split(","):
-                    correspondents_in.append(
-                        query.Term("correspondent_id", correspondent_id),
+        for key, value in self.query_params.items():
+            # is_tagged is a special case
+            if key == "is_tagged":
+                criterias.append(query.Term("has_tag", self.evalBoolean(value)))
+                continue
+
+            # Don't process query params without a filter
+            if "__" not in key:
+                continue
+
+            # All other query params consist of a parameter and a query filter
+            param, query_filter = key.split("__", 1)
+            try:
+                field, supported_query_filters = self.param_map[param]
+            except KeyError:
+                logger.error(f"Unable to build a query filter for parameter {key}")
+                continue
+
+            # We only support certain filters per parameter
+            if query_filter not in supported_query_filters:
+                logger.info(
+                    f"Query filter {query_filter} not supported for parameter {param}",
                 )
-                criterias.append(query.Or(correspondents_in))
-            elif k == "correspondent__id__none":
-                for correspondent_id in v.split(","):
+                continue
+
+            if query_filter == "id":
+                criterias.append(query.Term(f"{field}_id", value))
+            elif query_filter == "id__in":
+                in_filter = []
+                for object_id in value.split(","):
+                    in_filter.append(
+                        query.Term(f"{field}_id", object_id),
+                    )
+                criterias.append(query.Or(in_filter))
+            elif query_filter == "id__none":
+                for object_id in value.split(","):
                     criterias.append(
-                        query.Not(query.Term("correspondent_id", correspondent_id)),
+                        query.Not(query.Term(f"{field}_id", object_id)),
                     )
-            elif k == "tags__id__all":
-                for tag_id in v.split(","):
-                    criterias.append(query.Term("tag_id", tag_id))
-            elif k == "tags__id__none":
-                for tag_id in v.split(","):
-                    criterias.append(query.Not(query.Term("tag_id", tag_id)))
-            elif k == "tags__id__in":
-                tags_in = []
-                for tag_id in v.split(","):
-                    tags_in.append(query.Term("tag_id", tag_id))
-                criterias.append(query.Or(tags_in))
-            elif k == "document_type__id":
-                criterias.append(query.Term("type_id", v))
-            elif k == "document_type__id__in":
-                document_types_in = []
-                for document_type_id in v.split(","):
-                    document_types_in.append(query.Term("type_id", document_type_id))
-                criterias.append(query.Or(document_types_in))
-            elif k == "document_type__id__none":
-                for document_type_id in v.split(","):
-                    criterias.append(query.Not(query.Term("type_id", document_type_id)))
-            elif k == "correspondent__isnull":
+            elif query_filter == "isnull":
                 criterias.append(
-                    query.Term("has_correspondent", self.evalBoolean(v) is False),
+                    query.Term(f"has_{field}", self.evalBoolean(value) is False),
                 )
-            elif k == "is_tagged":
-                criterias.append(query.Term("has_tag", self.evalBoolean(v)))
-            elif k == "document_type__isnull":
-                criterias.append(query.Term("has_type", self.evalBoolean(v) is False))
-            elif k == "created__date__lt":
+            elif query_filter == "id__all":
+                for object_id in value.split(","):
+                    criterias.append(query.Term(f"{field}_id", object_id))
+            elif query_filter == "date__lt":
                 criterias.append(
-                    query.DateRange("created", start=None, end=isoparse(v)),
+                    query.DateRange(field, start=None, end=isoparse(value)),
                 )
-            elif k == "created__date__gt":
+            elif query_filter == "date__gt":
                 criterias.append(
-                    query.DateRange("created", start=isoparse(v), end=None),
+                    query.DateRange(field, start=isoparse(value), end=None),
+                )
+            elif query_filter == "icontains":
+                criterias.append(
+                    query.Term(field, value),
+                )
+            elif query_filter == "istartswith":
+                criterias.append(
+                    query.Prefix(field, value),
                 )
-            elif k == "added__date__gt":
-                criterias.append(query.DateRange("added", start=isoparse(v), end=None))
-            elif k == "added__date__lt":
-                criterias.append(query.DateRange("added", start=None, end=isoparse(v)))
-            elif k == "storage_path__id":
-                criterias.append(query.Term("path_id", v))
-            elif k == "storage_path__id__in":
-                storage_paths_in = []
-                for storage_path_id in v.split(","):
-                    storage_paths_in.append(query.Term("path_id", storage_path_id))
-                criterias.append(query.Or(storage_paths_in))
-            elif k == "storage_path__id__none":
-                for storage_path_id in v.split(","):
-                    criterias.append(query.Not(query.Term("path_id", storage_path_id)))
-            elif k == "storage_path__isnull":
-                criterias.append(query.Term("has_path", self.evalBoolean(v) is False))
-            elif k == "owner__isnull":
-                criterias.append(query.Term("has_owner", self.evalBoolean(v) is False))
-            elif k == "owner__id":
-                criterias.append(query.Term("owner_id", v))
-            elif k == "owner__id__in":
-                owners_in = []
-                for owner_id in v.split(","):
-                    owners_in.append(query.Term("owner_id", owner_id))
-                criterias.append(query.Or(owners_in))
-            elif k == "owner__id__none":
-                for owner_id in v.split(","):
-                    criterias.append(query.Not(query.Term("owner_id", owner_id)))
 
         user_criterias = get_permissions_criterias(
             user=self.user,

src/documents/signals/handlers.py

@@ -346,7 +346,7 @@ def cleanup_document_deletion(sender, instance, using, **kwargs):
                     logger.debug(f"Deleted file {filename}.")
                 except OSError as e:
                     logger.warning(
-                        f"While deleting document {str(instance)}, the file "
+                        f"While deleting document {instance!s}, the file "
                         f"{filename} could not be deleted: {e}",
                     )
 
@@ -369,13 +369,13 @@ class CannotMoveFilesException(Exception):
 def validate_move(instance, old_path, new_path):
     if not os.path.isfile(old_path):
         # Can't do anything if the old file does not exist anymore.
-        logger.fatal(f"Document {str(instance)}: File {old_path} has gone.")
+        logger.fatal(f"Document {instance!s}: File {old_path} has gone.")
         raise CannotMoveFilesException
 
     if os.path.isfile(new_path):
         # Can't do anything if the new file already exists. Skip updating file.
         logger.warning(
-            f"Document {str(instance)}: Cannot rename file "
+            f"Document {instance!s}: Cannot rename file "
             f"since target path {new_path} already exists.",
         )
         raise CannotMoveFilesException

src/documents/tasks.py

@@ -116,7 +116,7 @@ def consume_file(
                         {"type": "status_update", "data": payload},
                     )
                 except ConnectionError as e:
-                    logger.warning(f"ConnectionError on status send: {str(e)}")
+                    logger.warning(f"ConnectionError on status send: {e!s}")
                 # consuming stops here, since the original document with
                 # the barcodes has been split and will be consumed separately
 

src/documents/tests/test_api.py

@@ -420,6 +420,74 @@ class TestDocumentApi(DirectoriesMixin, DocumentConsumeDelayMixin, APITestCase):
         results = response.data["results"]
         self.assertEqual(len(results), 0)
 
+    def test_document_checksum_filter(self):
+        Document.objects.create(
+            title="none1",
+            checksum="A",
+            mime_type="application/pdf",
+        )
+        doc2 = Document.objects.create(
+            title="none2",
+            checksum="B",
+            mime_type="application/pdf",
+        )
+        Document.objects.create(
+            title="none3",
+            checksum="C",
+            mime_type="application/pdf",
+        )
+
+        response = self.client.get("/api/documents/?checksum__iexact=B")
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        results = response.data["results"]
+        self.assertEqual(len(results), 1)
+        self.assertEqual(results[0]["id"], doc2.id)
+
+        response = self.client.get("/api/documents/?checksum__iexact=X")
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        results = response.data["results"]
+        self.assertEqual(len(results), 0)
+
+    def test_document_original_filename_filter(self):
+        doc1 = Document.objects.create(
+            title="none1",
+            checksum="A",
+            mime_type="application/pdf",
+            original_filename="docA.pdf",
+        )
+        doc2 = Document.objects.create(
+            title="none2",
+            checksum="B",
+            mime_type="application/pdf",
+            original_filename="docB.pdf",
+        )
+        doc3 = Document.objects.create(
+            title="none3",
+            checksum="C",
+            mime_type="application/pdf",
+            original_filename="docC.pdf",
+        )
+
+        response = self.client.get("/api/documents/?original_filename__iexact=DOCa.pdf")
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        results = response.data["results"]
+        self.assertEqual(len(results), 1)
+        self.assertEqual(results[0]["id"], doc1.id)
+
+        response = self.client.get("/api/documents/?original_filename__iexact=docx.pdf")
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        results = response.data["results"]
+        self.assertEqual(len(results), 0)
+
+        response = self.client.get("/api/documents/?original_filename__istartswith=dOc")
+        self.assertEqual(response.status_code, status.HTTP_200_OK)
+        results = response.data["results"]
+        self.assertEqual(len(results), 3)
+        self.assertCountEqual(
+            [results[0]["id"], results[1]["id"], results[2]["id"]],
+            [doc1.id, doc2.id, doc3.id],
+        )
+
     def test_documents_title_content_filter(self):
         doc1 = Document.objects.create(
             title="title A",
@@ -1086,17 +1154,19 @@ class TestDocumentApi(DirectoriesMixin, DocumentConsumeDelayMixin, APITestCase):
             checksum="4",
             created=timezone.make_aware(datetime.datetime(2020, 7, 13)),
             content="test",
+            original_filename="doc4.pdf",
         )
         d4.tags.add(t2)
         d5 = Document.objects.create(
             checksum="5",
             added=timezone.make_aware(datetime.datetime(2020, 7, 13)),
             content="test",
+            original_filename="doc5.pdf",
         )
         Document.objects.create(checksum="6", content="test2")
         d7 = Document.objects.create(checksum="7", storage_path=sp, content="test")
         d8 = Document.objects.create(
-            checksum="8",
+            checksum="foo",
             correspondent=c2,
             document_type=dt2,
             storage_path=sp2,
@@ -1239,6 +1309,16 @@ class TestDocumentApi(DirectoriesMixin, DocumentConsumeDelayMixin, APITestCase):
             ),
         )
 
+        self.assertEqual(
+            search_query("&checksum__icontains=foo"),
+            [d8.id],
+        )
+
+        self.assertCountEqual(
+            search_query("&original_filename__istartswith=doc"),
+            [d4.id, d5.id],
+        )
+
     def test_search_filtering_respect_owner(self):
         """
         GIVEN:
@@ -2514,11 +2594,25 @@ class TestApiUiSettings(DirectoriesMixin, APITestCase):
     def setUp(self):
         super().setUp()
         self.test_user = User.objects.create_superuser(username="test")
+        self.test_user.first_name = "Test"
+        self.test_user.last_name = "User"
+        self.test_user.save()
         self.client.force_authenticate(user=self.test_user)
 
     def test_api_get_ui_settings(self):
         response = self.client.get(self.ENDPOINT, format="json")
         self.assertEqual(response.status_code, status.HTTP_200_OK)
+        self.assertDictEqual(
+            response.data["user"],
+            {
+                "id": self.test_user.id,
+                "username": self.test_user.username,
+                "is_superuser": True,
+                "groups": [],
+                "first_name": self.test_user.first_name,
+                "last_name": self.test_user.last_name,
+            },
+        )
         self.assertDictEqual(
             response.data["settings"],
             {

src/documents/tests/test_barcodes.py

@@ -12,6 +12,7 @@ from documents.barcodes import BarcodeReader
 from documents.consumer import ConsumerError
 from documents.data_models import ConsumableDocument
 from documents.data_models import DocumentSource
+from documents.models import Document
 from documents.tests.utils import DirectoriesMixin
 from documents.tests.utils import FileSystemAssertsMixin
 
@@ -764,7 +765,7 @@ class TestAsnBarcode(DirectoriesMixin, TestCase):
             self.assertEqual(reader.pdf_file, test_file)
             self.assertEqual(asn, 123)
 
-    def test_scan_file_for_asn_not_existing(self):
+    def test_scan_file_for_asn_not_found(self):
         """
         GIVEN:
             - PDF without an ASN barcode
@@ -781,6 +782,49 @@ class TestAsnBarcode(DirectoriesMixin, TestCase):
             self.assertEqual(reader.pdf_file, test_file)
             self.assertEqual(asn, None)
 
+    @override_settings(CONSUMER_ENABLE_ASN_BARCODE=True)
+    def test_scan_file_for_asn_already_exists(self):
+        """
+        GIVEN:
+            - PDF with an ASN barcode
+            - ASN value already exists
+        WHEN:
+            - File is scanned for barcodes
+        THEN:
+            - ASN is retrieved from the document
+            - Consumption fails
+        """
+
+        Document.objects.create(
+            title="WOW",
+            content="the content",
+            archive_serial_number=123,
+            checksum="456",
+            mime_type="application/pdf",
+        )
+
+        test_file = self.BARCODE_SAMPLE_DIR / "barcode-39-asn-123.pdf"
+
+        dst = settings.SCRATCH_DIR / "barcode-39-asn-123.pdf"
+        shutil.copy(test_file, dst)
+
+        with mock.patch("documents.consumer.Consumer._send_progress"):
+            with self.assertRaises(ConsumerError) as cm, self.assertLogs(
+                "paperless.consumer",
+                level="ERROR",
+            ) as logs_cm:
+                tasks.consume_file(
+                    ConsumableDocument(
+                        source=DocumentSource.ConsumeFolder,
+                        original_file=dst,
+                    ),
+                    None,
+                )
+            self.assertIn("Not consuming barcode-39-asn-123.pdf", str(cm.exception))
+            error_str = logs_cm.output[0]
+            expected_str = "ERROR:paperless.consumer:Not consuming barcode-39-asn-123.pdf: Given ASN already exists!"
+            self.assertEqual(expected_str, error_str)
+
     def test_scan_file_for_asn_barcode_invalid(self):
         """
         GIVEN:

src/documents/tests/test_delayedquery.py

@@ -0,0 +1,219 @@
+from dateutil.parser import isoparse
+from django.test import TestCase
+from whoosh import query
+
+from documents.index import DelayedQuery
+from documents.index import get_permissions_criterias
+from documents.models import User
+
+
+class TestDelayedQuery(TestCase):
+    def setUp(self):
+        super().setUp()
+        # all tests run without permission criteria, so has_no_owner query will always
+        # be appended.
+        self.has_no_owner = query.Or([query.Term("has_owner", False)])
+
+    def _get_testset__id__in(self, param, field):
+        return (
+            {f"{param}__id__in": "42,43"},
+            query.And(
+                [
+                    query.Or(
+                        [
+                            query.Term(f"{field}_id", "42"),
+                            query.Term(f"{field}_id", "43"),
+                        ],
+                    ),
+                    self.has_no_owner,
+                ],
+            ),
+        )
+
+    def _get_testset__id__none(self, param, field):
+        return (
+            {f"{param}__id__none": "42,43"},
+            query.And(
+                [
+                    query.Not(query.Term(f"{field}_id", "42")),
+                    query.Not(query.Term(f"{field}_id", "43")),
+                    self.has_no_owner,
+                ],
+            ),
+        )
+
+    def test_get_permission_criteria(self):
+        # tests contains touples of user instances and the expected filter
+        tests = (
+            (None, [query.Term("has_owner", False)]),
+            (User(42, username="foo", is_superuser=True), []),
+            (
+                User(42, username="foo", is_superuser=False),
+                [
+                    query.Term("has_owner", False),
+                    query.Term("owner_id", 42),
+                    query.Term("viewer_id", "42"),
+                ],
+            ),
+        )
+        for user, expected in tests:
+            self.assertEqual(get_permissions_criterias(user), expected)
+
+    def test_no_query_filters(self):
+        dq = DelayedQuery(None, {}, None, None)
+        self.assertEqual(dq._get_query_filter(), self.has_no_owner)
+
+    def test_date_query_filters(self):
+        def _get_testset(param: str):
+            date_str = "1970-01-01T02:44"
+            date_obj = isoparse(date_str)
+            return (
+                (
+                    {f"{param}__date__lt": date_str},
+                    query.And(
+                        [
+                            query.DateRange(param, start=None, end=date_obj),
+                            self.has_no_owner,
+                        ],
+                    ),
+                ),
+                (
+                    {f"{param}__date__gt": date_str},
+                    query.And(
+                        [
+                            query.DateRange(param, start=date_obj, end=None),
+                            self.has_no_owner,
+                        ],
+                    ),
+                ),
+            )
+
+        query_params = ["created", "added"]
+        for param in query_params:
+            for params, expected in _get_testset(param):
+                dq = DelayedQuery(None, params, None, None)
+                got = dq._get_query_filter()
+                self.assertCountEqual(got, expected)
+
+    def test_is_tagged_query_filter(self):
+        tests = (
+            ("True", True),
+            ("true", True),
+            ("1", True),
+            ("False", False),
+            ("false", False),
+            ("0", False),
+            ("foo", False),
+        )
+        for param, expected in tests:
+            dq = DelayedQuery(None, {"is_tagged": param}, None, None)
+            self.assertEqual(
+                dq._get_query_filter(),
+                query.And([query.Term("has_tag", expected), self.has_no_owner]),
+            )
+
+    def test_tags_query_filters(self):
+        # tests contains touples of query_parameter dics and the expected whoosh query
+        param = "tags"
+        field, _ = DelayedQuery.param_map[param]
+        tests = (
+            (
+                {f"{param}__id__all": "42,43"},
+                query.And(
+                    [
+                        query.Term(f"{field}_id", "42"),
+                        query.Term(f"{field}_id", "43"),
+                        self.has_no_owner,
+                    ],
+                ),
+            ),
+            # tags does not allow __id
+            (
+                {f"{param}__id": "42"},
+                self.has_no_owner,
+            ),
+            # tags does not allow __isnull
+            (
+                {f"{param}__isnull": "true"},
+                self.has_no_owner,
+            ),
+            self._get_testset__id__in(param, field),
+            self._get_testset__id__none(param, field),
+        )
+
+        for params, expected in tests:
+            dq = DelayedQuery(None, params, None, None)
+            got = dq._get_query_filter()
+            self.assertCountEqual(got, expected)
+
+    def test_generic_query_filters(self):
+        def _get_testset(param: str):
+            field, _ = DelayedQuery.param_map[param]
+            return (
+                (
+                    {f"{param}__id": "42"},
+                    query.And(
+                        [
+                            query.Term(f"{field}_id", "42"),
+                            self.has_no_owner,
+                        ],
+                    ),
+                ),
+                self._get_testset__id__in(param, field),
+                self._get_testset__id__none(param, field),
+                (
+                    {f"{param}__isnull": "true"},
+                    query.And(
+                        [
+                            query.Term(f"has_{field}", False),
+                            self.has_no_owner,
+                        ],
+                    ),
+                ),
+                (
+                    {f"{param}__isnull": "false"},
+                    query.And(
+                        [
+                            query.Term(f"has_{field}", True),
+                            self.has_no_owner,
+                        ],
+                    ),
+                ),
+            )
+
+        query_params = ["correspondent", "document_type", "storage_path", "owner"]
+        for param in query_params:
+            for params, expected in _get_testset(param):
+                dq = DelayedQuery(None, params, None, None)
+                got = dq._get_query_filter()
+                self.assertCountEqual(got, expected)
+
+    def test_char_query_filter(self):
+        def _get_testset(param: str):
+            return (
+                (
+                    {f"{param}__icontains": "foo"},
+                    query.And(
+                        [
+                            query.Term(f"{param}", "foo"),
+                            self.has_no_owner,
+                        ],
+                    ),
+                ),
+                (
+                    {f"{param}__istartswith": "foo"},
+                    query.And(
+                        [
+                            query.Prefix(f"{param}", "foo"),
+                            self.has_no_owner,
+                        ],
+                    ),
+                ),
+            )
+
+        query_params = ["checksum", "original_filename"]
+        for param in query_params:
+            for params, expected in _get_testset(param):
+                dq = DelayedQuery(None, params, None, None)
+                got = dq._get_query_filter()
+                self.assertCountEqual(got, expected)

src/documents/tests/utils.py

@@ -105,6 +105,20 @@ class FileSystemAssertsMixin:
     def assertIsNotDir(self, path: Union[PathLike, str]):
         self.assertFalse(Path(path).resolve().is_dir(), f"Dir does exist: {path}")
 
+    def assertFilesEqual(
+        self,
+        path1: Union[PathLike, str],
+        path2: Union[PathLike, str],
+    ):
+        path1 = Path(path1)
+        path2 = Path(path2)
+        import hashlib
+
+        hash1 = hashlib.sha256(path1.read_bytes()).hexdigest()
+        hash2 = hashlib.sha256(path2.read_bytes()).hexdigest()
+
+        self.assertEqual(hash1, hash2, "File SHA256 mismatch")
+
 
 class ConsumerProgressMixin:
     def setUp(self) -> None:

src/documents/views.py

@@ -519,7 +519,7 @@ class DocumentViewSet(
             try:
                 return Response(self.getNotes(doc))
             except Exception as e:
-                logger.warning(f"An error occurred retrieving notes: {str(e)}")
+                logger.warning(f"An error occurred retrieving notes: {e!s}")
                 return Response(
                     {"error": "Error retreiving notes, check logs for more detail."},
                 )
@@ -538,7 +538,7 @@ class DocumentViewSet(
 
                 return Response(self.getNotes(doc))
             except Exception as e:
-                logger.warning(f"An error occurred saving note: {str(e)}")
+                logger.warning(f"An error occurred saving note: {e!s}")
                 return Response(
                     {
                         "error": "Error saving note, check logs for more detail.",
@@ -628,7 +628,7 @@ class UnifiedSearchViewSet(DocumentViewSet):
             except NotFound:
                 raise
             except Exception as e:
-                logger.warning(f"An error occurred listing search results: {str(e)}")
+                logger.warning(f"An error occurred listing search results: {e!s}")
                 return HttpResponseBadRequest(
                     "Error listing search results, check logs for more detail.",
                 )
@@ -699,7 +699,7 @@ class BulkEditView(GenericAPIView):
             result = method(documents, **parameters)
             return Response({"result": result})
         except Exception as e:
-            logger.warning(f"An error occurred performing bulk edit: {str(e)}")
+            logger.warning(f"An error occurred performing bulk edit: {e!s}")
             return HttpResponseBadRequest(
                 "Error performing bulk edit, check logs for more detail.",
             )
@@ -1028,16 +1028,23 @@ class UiSettingsView(GenericAPIView):
             ui_settings["update_checking"] = {
                 "backend_setting": settings.ENABLE_UPDATE_CHECK,
             }
+        user_resp = {
+            "id": user.id,
+            "username": user.username,
+            "is_superuser": user.is_superuser,
+            "groups": list(user.groups.values_list("id", flat=True)),
+        }
+
+        if len(user.first_name) > 0:
+            user_resp["first_name"] = user.first_name
+        if len(user.last_name) > 0:
+            user_resp["last_name"] = user.last_name
+
         # strip <app_label>.
         roles = map(lambda perm: re.sub(r"^\w+.", "", perm), user.get_all_permissions())
         return Response(
             {
-                "user": {
-                    "id": user.id,
-                    "username": user.username,
-                    "is_superuser": user.is_superuser,
-                    "groups": user.groups.values_list("id", flat=True),
-                },
+                "user": user_resp,
                 "settings": ui_settings,
                 "permissions": roles,
             },

src/paperless/settings.py

@@ -476,24 +476,24 @@ CSRF_COOKIE_NAME = f"{COOKIE_PREFIX}csrftoken"
 SESSION_COOKIE_NAME = f"{COOKIE_PREFIX}sessionid"
 LANGUAGE_COOKIE_NAME = f"{COOKIE_PREFIX}django_language"
 
+
 ###############################################################################
 # Database                                                                    #
 ###############################################################################
-
-DATABASES = {
+def _parse_db_settings() -> Dict:
+    databases = {
         "default": {
             "ENGINE": "django.db.backends.sqlite3",
             "NAME": os.path.join(DATA_DIR, "db.sqlite3"),
             "OPTIONS": {},
         },
-}
-
-if os.getenv("PAPERLESS_DBHOST"):
+    }
+    if os.getenv("PAPERLESS_DBHOST"):
         # Have sqlite available as a second option for management commands
         # This is important when migrating to/from sqlite
-    DATABASES["sqlite"] = DATABASES["default"].copy()
+        databases["sqlite"] = databases["default"].copy()
 
-    DATABASES["default"] = {
+        databases["default"] = {
             "HOST": os.getenv("PAPERLESS_DBHOST"),
             "NAME": os.getenv("PAPERLESS_DBNAME", "paperless"),
             "USER": os.getenv("PAPERLESS_DBUSER", "paperless"),
@@ -501,7 +501,7 @@ if os.getenv("PAPERLESS_DBHOST"):
             "OPTIONS": {},
         }
         if os.getenv("PAPERLESS_DBPORT"):
-        DATABASES["default"]["PORT"] = os.getenv("PAPERLESS_DBPORT")
+            databases["default"]["PORT"] = os.getenv("PAPERLESS_DBPORT")
 
         # Leave room for future extensibility
         if os.getenv("PAPERLESS_DBENGINE") == "mariadb":
@@ -517,12 +517,6 @@ if os.getenv("PAPERLESS_DBHOST"):
                 },
             }
 
-        # Silence Django error on old MariaDB versions.
-        # VARCHAR can support > 255 in modern versions
-        # https://docs.djangoproject.com/en/4.1/ref/checks/#database
-        # https://mariadb.com/kb/en/innodb-system-variables/#innodb_large_prefix
-        SILENCED_SYSTEM_CHECKS = ["mysql.W003"]
-
         else:  # Default to PostgresDB
             engine = "django.db.backends.postgresql_psycopg2"
             options = {
@@ -532,13 +526,32 @@ if os.getenv("PAPERLESS_DBHOST"):
                 "sslkey": os.getenv("PAPERLESS_DBSSLKEY", None),
             }
 
-    DATABASES["default"]["ENGINE"] = engine
-    DATABASES["default"]["OPTIONS"].update(options)
+        databases["default"]["ENGINE"] = engine
+        databases["default"]["OPTIONS"].update(options)
 
-if os.getenv("PAPERLESS_DB_TIMEOUT") is not None:
-    DATABASES["default"]["OPTIONS"].update(
-        {"timeout": float(os.getenv("PAPERLESS_DB_TIMEOUT"))},
+    if os.getenv("PAPERLESS_DB_TIMEOUT") is not None:
+        if databases["default"]["ENGINE"] == "django.db.backends.sqlite3":
+            databases["default"]["OPTIONS"].update(
+                {"timeout": int(os.getenv("PAPERLESS_DB_TIMEOUT"))},
             )
+        else:
+            databases["default"]["OPTIONS"].update(
+                {"connect_timeout": int(os.getenv("PAPERLESS_DB_TIMEOUT"))},
+            )
+            databases["sqlite"]["OPTIONS"].update(
+                {"timeout": int(os.getenv("PAPERLESS_DB_TIMEOUT"))},
+            )
+    return databases
+
+
+DATABASES = _parse_db_settings()
+
+if os.getenv("PAPERLESS_DBENGINE") == "mariadb":
+    # Silence Django error on old MariaDB versions.
+    # VARCHAR can support > 255 in modern versions
+    # https://docs.djangoproject.com/en/4.1/ref/checks/#database
+    # https://mariadb.com/kb/en/innodb-system-variables/#innodb_large_prefix
+    SILENCED_SYSTEM_CHECKS = ["mysql.W003"]
 
 DEFAULT_AUTO_FIELD = "django.db.models.AutoField"
 
@@ -662,6 +675,8 @@ CELERY_WORKER_MAX_TASKS_PER_CHILD = 1
 CELERY_WORKER_SEND_TASK_EVENTS = True
 CELERY_TASK_SEND_SENT_EVENT = True
 CELERY_SEND_TASK_SENT_EVENT = True
+CELERY_BROKER_CONNECTION_RETRY = True
+CELERY_BROKER_CONNECTION_RETRY_ON_STARTUP = True
 
 CELERY_TASK_TRACK_STARTED = True
 CELERY_TASK_TIME_LIMIT: Final[int] = __get_int("PAPERLESS_WORKER_TIMEOUT", 1800)

src/paperless/tests/test_settings.py

@@ -6,6 +6,7 @@ from unittest import mock
 from celery.schedules import crontab
 
 from paperless.settings import _parse_beat_schedule
+from paperless.settings import _parse_db_settings
 from paperless.settings import _parse_ignore_dates
 from paperless.settings import _parse_redis_url
 from paperless.settings import default_threads_per_worker
@@ -291,3 +292,60 @@ class TestCeleryScheduleParsing(TestCase):
             {},
             schedule,
         )
+
+
+class TestDBSettings(TestCase):
+    def test_db_timeout_with_sqlite(self):
+        """
+        GIVEN:
+            - PAPERLESS_DB_TIMEOUT is set
+        WHEN:
+            - Settings are parsed
+        THEN:
+            - PAPERLESS_DB_TIMEOUT set for sqlite
+        """
+        with mock.patch.dict(
+            os.environ,
+            {
+                "PAPERLESS_DB_TIMEOUT": "10",
+            },
+        ):
+            databases = _parse_db_settings()
+
+            self.assertDictEqual(
+                {
+                    "timeout": 10.0,
+                },
+                databases["default"]["OPTIONS"],
+            )
+
+    def test_db_timeout_with_not_sqlite(self):
+        """
+        GIVEN:
+            - PAPERLESS_DB_TIMEOUT is set but db is not sqlite
+        WHEN:
+            - Settings are parsed
+        THEN:
+            - PAPERLESS_DB_TIMEOUT set correctly in non-sqlite db & for fallback sqlite db
+        """
+        with mock.patch.dict(
+            os.environ,
+            {
+                "PAPERLESS_DBHOST": "127.0.0.1",
+                "PAPERLESS_DB_TIMEOUT": "10",
+            },
+        ):
+            databases = _parse_db_settings()
+
+            self.assertDictContainsSubset(
+                {
+                    "connect_timeout": 10.0,
+                },
+                databases["default"]["OPTIONS"],
+            )
+            self.assertDictEqual(
+                {
+                    "timeout": 10.0,
+                },
+                databases["sqlite"]["OPTIONS"],
+            )

src/paperless_mail/mail.py

@@ -384,6 +384,8 @@ def make_criterias(rule: MailRule, supports_gmail_labels: bool):
     if isinstance(rule_query, dict):
         if len(rule_query) or len(criterias):
             return AND(**rule_query, **criterias)
+        else:
+            return "ALL"
     else:
         return AND(rule_query, **criterias)
 
@@ -542,7 +544,7 @@ class MailAccountHandler(LoggingMixin):
         criterias = make_criterias(rule, supports_gmail_labels)
 
         self.log.debug(
-            f"Rule {rule}: Searching folder with criteria {str(criterias)}",
+            f"Rule {rule}: Searching folder with criteria {criterias}",
         )
 
         try:

src/paperless_mail/parsers.py

@@ -1,18 +1,18 @@
-import os
 import re
 from html import escape
-from io import BytesIO
-from io import StringIO
+from pathlib import Path
+from typing import List
 
-import requests
+import httpx
 from bleach import clean
 from bleach import linkify
 from django.conf import settings
 from django.utils.timezone import is_naive
 from django.utils.timezone import make_aware
 from humanfriendly import format_size
+from imap_tools import MailAttachment
 from imap_tools import MailMessage
-from tika import parser
+from tika_client import TikaClient
 
 from documents.parsers import DocumentParser
 from documents.parsers import ParseError
@@ -22,33 +22,15 @@ from documents.parsers import make_thumbnail_from_pdf
 class MailDocumentParser(DocumentParser):
     """
     This parser uses imap_tools to parse .eml files, generates pdf using
-    gotenbergs and sends the html part to a local tika server for text extraction.
+    Gotenberg and sends the html part to a Tika server for text extraction.
     """
 
     gotenberg_server = settings.TIKA_GOTENBERG_ENDPOINT
     tika_server = settings.TIKA_ENDPOINT
 
     logging_name = "paperless.parsing.mail"
-    _parsed = None
 
-    def get_parsed(self, document_path) -> MailMessage:
-        if not self._parsed:
-            try:
-                with open(document_path, "rb") as eml:
-                    self._parsed = MailMessage.from_bytes(eml.read())
-            except Exception as err:
-                raise ParseError(
-                    f"Could not parse {document_path}: {err}",
-                ) from err
-            if not self._parsed.from_values:
-                self._parsed = None
-                raise ParseError(
-                    f"Could not parse {document_path}: Missing 'from'",
-                )
-
-        return self._parsed
-
-    def get_thumbnail(self, document_path, mime_type, file_name=None):
+    def get_thumbnail(self, document_path: Path, mime_type: str, file_name=None):
         if not self.archive_path:
             self.archive_path = self.generate_pdf(document_path)
 
@@ -58,11 +40,11 @@ class MailDocumentParser(DocumentParser):
             self.logging_group,
         )
 
-    def extract_metadata(self, document_path, mime_type):
+    def extract_metadata(self, document_path: Path, mime_type: str):
         result = []
 
         try:
-            mail = self.get_parsed(document_path)
+            mail = self.parse_file_to_message(document_path)
         except ParseError as e:
             self.log.warning(
                 f"Error while fetching document metadata for {document_path}: {e}",
@@ -106,101 +88,157 @@ class MailDocumentParser(DocumentParser):
         result.sort(key=lambda item: (item["prefix"], item["key"]))
         return result
 
-    def parse(self, document_path, mime_type, file_name=None):
+    def parse(self, document_path: Path, mime_type: str, file_name=None):
+        """
+        Parses the given .eml into formatted text, based on the decoded email.
+
+        """
+
         def strip_text(text: str):
+            """
+            Reduces the spacing of the given text string
+            """
             text = re.sub(r"\s+", " ", text)
             text = re.sub(r"(\n *)+", "\n", text)
             return text.strip()
 
-        mail = self.get_parsed(document_path)
-
-        self.text = f"Subject: {mail.subject}\n\n"
-        self.text += f"From: {mail.from_values.full}\n\n"
-        self.text += f"To: {', '.join(address.full for address in mail.to_values)}\n\n"
-        if len(mail.cc_values) >= 1:
-            self.text += (
+        def build_formatted_text(mail_message: MailMessage) -> str:
+            """
+            Constructs a formatted string, based on the given email.  Basically tries
+            to get most of the email content, included front matter, into a nice string
+            """
+            fmt_text = f"Subject: {mail_message.subject}\n\n"
+            fmt_text += f"From: {mail_message.from_values.full}\n\n"
+            to_list = [address.full for address in mail_message.to_values]
+            fmt_text += f"To: {', '.join(to_list)}\n\n"
+            if mail_message.cc_values:
+                fmt_text += (
                     f"CC: {', '.join(address.full for address in mail.cc_values)}\n\n"
                 )
-        if len(mail.bcc_values) >= 1:
-            self.text += (
+            if mail_message.bcc_values:
+                fmt_text += (
                     f"BCC: {', '.join(address.full for address in mail.bcc_values)}\n\n"
                 )
-        if len(mail.attachments) >= 1:
+            if mail_message.attachments:
                 att = []
                 for a in mail.attachments:
                     att.append(f"{a.filename} ({format_size(a.size, binary=True)})")
-
-            self.text += f"Attachments: {', '.join(att)}\n\n"
+                fmt_text += f"Attachments: {', '.join(att)}\n\n"
 
             if mail.html:
-            self.text += "HTML content: " + strip_text(self.tika_parse(mail.html))
+                fmt_text += "HTML content: " + strip_text(self.tika_parse(mail.html))
 
-        self.text += f"\n\n{strip_text(mail.text)}"
+            fmt_text += f"\n\n{strip_text(mail.text)}"
+
+            return fmt_text
+
+        self.log.debug(f"Parsing file {document_path.name} into an email")
+        mail = self.parse_file_to_message(document_path)
+
+        self.log.debug("Building formatted text from email")
+        self.text = build_formatted_text(mail)
 
         if is_naive(mail.date):
             self.date = make_aware(mail.date)
         else:
             self.date = mail.date
 
-        self.archive_path = self.generate_pdf(document_path)
+        self.log.debug("Creating a PDF from the email")
+        self.archive_path = self.generate_pdf(mail)
+
+    @staticmethod
+    def parse_file_to_message(filepath: Path) -> MailMessage:
+        """
+        Parses the given .eml file into a MailMessage object
+        """
+        try:
+            with filepath.open("rb") as eml:
+                parsed = MailMessage.from_bytes(eml.read())
+                if parsed.from_values is None:
+                    raise ParseError(
+                        f"Could not parse {filepath}: Missing 'from'",
+                    )
+        except Exception as err:
+            raise ParseError(
+                f"Could not parse {filepath}: {err}",
+            ) from err
+
+        return parsed
 
     def tika_parse(self, html: str):
         self.log.info("Sending content to Tika server")
 
         try:
-            parsed = parser.from_buffer(html, self.tika_server)
+            with TikaClient(tika_url=self.tika_server) as client:
+                parsed = client.tika.as_text.from_buffer(html, "text/html")
+
+                if "X-TIKA:content" in parsed.data:
+                    return parsed.data["X-TIKA:content"].strip()
+                return ""
         except Exception as err:
             raise ParseError(
                 f"Could not parse content with tika server at "
                 f"{self.tika_server}: {err}",
             ) from err
-        if parsed["content"]:
-            return parsed["content"]
-        else:
-            return ""
 
-    def generate_pdf(self, document_path):
-        pdf_collection = []
+    def generate_pdf(self, mail_message: MailMessage) -> Path:
+        archive_path = Path(self.tempdir) / "merged.pdf"
+
+        mail_pdf_file = self.generate_pdf_from_mail(mail_message)
+
+        # If no HTML content, create the PDF from the message
+        # Otherwise, create 2 PDFs and merge them with Gotenberg
+        if not mail_message.html:
+            archive_path.write_bytes(mail_pdf_file.read_bytes())
+        else:
             url_merge = self.gotenberg_server + "/forms/pdfengines/merge"
-        pdf_path = os.path.join(self.tempdir, "merged.pdf")
-        mail = self.get_parsed(document_path)
 
-        pdf_collection.append(("1_mail.pdf", self.generate_pdf_from_mail(mail)))
-
-        if not mail.html:
-            with open(pdf_path, "wb") as file:
-                file.write(pdf_collection[0][1])
-                file.close()
-            return pdf_path
-        else:
-            pdf_collection.append(
-                (
-                    "2_html.pdf",
-                    self.generate_pdf_from_html(mail.html, mail.attachments),
-                ),
+            pdf_of_html_content = self.generate_pdf_from_html(
+                mail_message.html,
+                mail_message.attachments,
             )
 
-        files = {}
-        for name, content in pdf_collection:
-            files[name] = (name, BytesIO(content))
-        headers = {}
+            pdf_collection = {
+                "1_mail.pdf": ("1_mail.pdf", mail_pdf_file, "application/pdf"),
+                "2_html.pdf": ("2_html.pdf", pdf_of_html_content, "application/pdf"),
+            }
+
             try:
-            response = requests.post(url_merge, files=files, headers=headers)
+                # Open a handle to each file, replacing the tuple
+                for filename in pdf_collection:
+                    file_multi_part = pdf_collection[filename]
+                    pdf_collection[filename] = (
+                        file_multi_part[0],
+                        file_multi_part[1].open("rb"),
+                        file_multi_part[2],
+                    )
+
+                response = httpx.post(url_merge, files=pdf_collection)
                 response.raise_for_status()  # ensure we notice bad responses
+
+                archive_path.write_bytes(response.content)
+
             except Exception as err:
-            raise ParseError(f"Error while converting document to PDF: {err}") from err
+                raise ParseError(
+                    f"Error while merging email HTML into PDF: {err}",
+                ) from err
+            finally:
+                for filename in pdf_collection:
+                    file_multi_part_handle = pdf_collection[filename][1]
+                    file_multi_part_handle.close()
 
-        with open(pdf_path, "wb") as file:
-            file.write(response.content)
-            file.close()
+        return archive_path
 
-        return pdf_path
+    def mail_to_html(self, mail: MailMessage) -> Path:
+        """
+        Converts the given email into an HTML file, formatted
+        based on the given template
+        """
 
-    @staticmethod
-    def mail_to_html(mail: MailMessage) -> StringIO:
-        data = {}
-
-        def clean_html(text: str):
+        def clean_html(text: str) -> str:
+            """
+            Attempts to clean, escape and linkify the given HTML string
+            """
             if isinstance(text, list):
                 text = "\n".join([str(e) for e in text])
             if type(text) != str:
@@ -211,6 +249,8 @@ class MailDocumentParser(DocumentParser):
             text = text.replace("\n", "<br>")
             return text
 
+        data = {}
+
         data["subject"] = clean_html(mail.subject)
         if data["subject"]:
             data["subject_label"] = "Subject"
@@ -237,27 +277,33 @@ class MailDocumentParser(DocumentParser):
         data["date"] = clean_html(mail.date.astimezone().strftime("%Y-%m-%d %H:%M"))
         data["content"] = clean_html(mail.text.strip())
 
-        html = StringIO()
-
         from django.template.loader import render_to_string
 
-        rendered = render_to_string("email_msg_template.html", context=data)
+        html_file = Path(self.tempdir) / "email_as_html.html"
+        html_file.write_text(render_to_string("email_msg_template.html", context=data))
 
-        html.write(rendered)
-        html.seek(0)
+        return html_file
 
-        return html
-
-    def generate_pdf_from_mail(self, mail):
+    def generate_pdf_from_mail(self, mail: MailMessage) -> Path:
+        """
+        Creates a PDF based on the given email, using the email's values in a
+        an HTML template
+        """
         url = self.gotenberg_server + "/forms/chromium/convert/html"
         self.log.info("Converting mail to PDF")
 
-        css_file = os.path.join(os.path.dirname(__file__), "templates/output.css")
+        css_file = Path(__file__).parent / "templates" / "output.css"
+        email_html_file = self.mail_to_html(mail)
 
-        with open(css_file, "rb") as css_handle:
+        print(css_file)
+        print(email_html_file)
+
+        with css_file.open("rb") as css_handle, email_html_file.open(
+            "rb",
+        ) as email_html_handle:
             files = {
-                "html": ("index.html", self.mail_to_html(mail)),
-                "css": ("output.css", css_handle),
+                "html": ("index.html", email_html_handle, "text/html"),
+                "css": ("output.css", css_handle, "text/css"),
             }
             headers = {}
             data = {
@@ -280,7 +326,7 @@ class MailDocumentParser(DocumentParser):
                 data["pdfFormat"] = "PDF/A-3b"
 
             try:
-                response = requests.post(
+                response = httpx.post(
                     url,
                     files=files,
                     headers=headers,
@@ -289,13 +335,23 @@ class MailDocumentParser(DocumentParser):
                 response.raise_for_status()  # ensure we notice bad responses
             except Exception as err:
                 raise ParseError(
-                    f"Error while converting document to PDF: {err}",
+                    f"Error while converting email to PDF: {err}",
                 ) from err
 
-        return response.content
+        email_as_pdf_file = Path(self.tempdir) / "email_as_pdf.pdf"
+        email_as_pdf_file.write_bytes(response.content)
+
+        return email_as_pdf_file
+
+    def generate_pdf_from_html(
+        self,
+        orig_html: str,
+        attachments: List[MailAttachment],
+    ) -> Path:
+        """
+        Generates a PDF file based on the HTML and attachments of the email
+        """
 
-    @staticmethod
-    def transform_inline_html(html, attachments):
         def clean_html_script(text: str):
             compiled_open = re.compile(re.escape("<script"), re.IGNORECASE)
             text = compiled_open.sub("<div hidden ", text)
@@ -304,28 +360,36 @@ class MailDocumentParser(DocumentParser):
             text = compiled_close.sub("</div", text)
             return text
 
-        html_clean = clean_html_script(html)
-        files = []
-
-        for a in attachments:
-            name_cid = "cid:" + a.content_id
-            name_clean = "".join(e for e in name_cid if e.isalnum())
-            files.append((name_clean, BytesIO(a.payload)))
-            html_clean = html_clean.replace(name_cid, name_clean)
-
-        files.append(("index.html", StringIO(html_clean)))
-
-        return files
-
-    def generate_pdf_from_html(self, orig_html, attachments):
         url = self.gotenberg_server + "/forms/chromium/convert/html"
         self.log.info("Converting html to PDF")
 
-        files = {}
-        for name, file in self.transform_inline_html(orig_html, attachments):
-            files[name] = (name, file)
+        tempdir = Path(self.tempdir)
+
+        html_clean = clean_html_script(orig_html)
+
+        files = {}
+
+        for attachment in attachments:
+            # Clean the attachment name to be valid
+            name_cid = f"cid:{attachment.content_id}"
+            name_clean = "".join(e for e in name_cid if e.isalnum())
+
+            # Write attachment payload to a temp file
+            temp_file = tempdir / name_clean
+            temp_file.write_bytes(attachment.payload)
+
+            # Store the attachment for upload
+            files[name_clean] = (name_clean, temp_file, attachment.content_type)
+
+            # Replace as needed the name with the clean name
+            html_clean = html_clean.replace(name_cid, name_clean)
+
+        # Now store the cleaned up HTML version
+        html_clean_file = tempdir / "index.html"
+        html_clean_file.write_text(html_clean)
+
+        files["index.html"] = ("index.html", html_clean_file, "text/html")
 
-        headers = {}
         data = {
             "marginTop": "0.1",
             "marginBottom": "0.1",
@@ -336,14 +400,29 @@ class MailDocumentParser(DocumentParser):
             "scale": "1.0",
         }
         try:
-            response = requests.post(
+            # Open a handle to each file, replacing the tuple
+            for filename in files:
+                file_multi_part = files[filename]
+                files[filename] = (
+                    file_multi_part[0],
+                    file_multi_part[1].open("rb"),
+                    file_multi_part[2],
+                )
+
+            response = httpx.post(
                 url,
                 files=files,
-                headers=headers,
                 data=data,
             )
             response.raise_for_status()  # ensure we notice bad responses
         except Exception as err:
             raise ParseError(f"Error while converting document to PDF: {err}") from err
+        finally:
+            # Ensure all file handles as closed
+            for filename in files:
+                file_multi_part_handle = files[filename][1]
+                file_multi_part_handle.close()
 
-        return response.content
+        html_pdf = tempdir / "html.pdf"
+        html_pdf.write_bytes(response.content)
+        return html_pdf

src/paperless_mail/tests/test_mail.py

@@ -721,6 +721,31 @@ class TestMail(
         self.assertEqual(len(self.bogus_mailbox.messages), 2)
         self.assertEqual(len(self.bogus_mailbox.messages_spam), 1)
 
+    def test_handle_mail_account_move_no_filters(self):
+        account = MailAccount.objects.create(
+            name="test",
+            imap_server="",
+            username="admin",
+            password="secret",
+        )
+
+        _ = MailRule.objects.create(
+            name="testrule",
+            account=account,
+            action=MailRule.MailAction.MOVE,
+            action_parameter="spam",
+            maximum_age=0,
+        )
+
+        self.assertEqual(len(self.bogus_mailbox.messages), 3)
+        self.assertEqual(len(self.bogus_mailbox.messages_spam), 0)
+
+        self.mail_account_handler.handle_mail_account(account)
+        self.apply_mail_actions()
+
+        self.assertEqual(len(self.bogus_mailbox.messages), 0)
+        self.assertEqual(len(self.bogus_mailbox.messages_spam), 3)
+
     def test_handle_mail_account_tag(self):
         account = MailAccount.objects.create(
             name="test",

src/paperless_mail/tests/test_parsers.py

@@ -1,24 +1,39 @@
 import datetime
-import os
+from pathlib import Path
 from unittest import mock
 
+import httpx
 from django.test import TestCase
 
 from documents.parsers import ParseError
 from documents.tests.utils import FileSystemAssertsMixin
 from paperless_mail.parsers import MailDocumentParser
+from paperless_tika.tests.utils import HttpxMockMixin
 
 
-class TestParser(FileSystemAssertsMixin, TestCase):
-    SAMPLE_FILES = os.path.join(os.path.dirname(__file__), "samples")
+class BaseMailParserTestCase(TestCase):
+    """
+    Basic setup for the below test cases
+    """
+
+    SAMPLE_DIR = Path(__file__).parent / "samples"
 
     def setUp(self) -> None:
+        super().setUp()
         self.parser = MailDocumentParser(logging_group=None)
 
     def tearDown(self) -> None:
+        super().tearDown()
         self.parser.cleanup()
 
-    def test_get_parsed_missing_file(self):
+
+class TestEmailFileParsing(FileSystemAssertsMixin, BaseMailParserTestCase):
+    """
+    Tests around reading a file and parsing it into a
+    MailMessage
+    """
+
+    def test_parse_error_missing_file(self):
         """
         GIVEN:
             - Fresh parser
@@ -28,13 +43,17 @@ class TestParser(FileSystemAssertsMixin, TestCase):
             - An Exception is thrown
         """
         # Check if exception is raised when parsing fails.
+        test_file = self.SAMPLE_DIR / "doesntexist.eml"
+
+        self.assertIsNotFile(test_file)
         self.assertRaises(
             ParseError,
-            self.parser.get_parsed,
-            os.path.join(self.SAMPLE_FILES, "na"),
+            self.parser.parse,
+            test_file,
+            "messages/rfc822",
         )
 
-    def test_get_parsed_broken_file(self):
+    def test_parse_error_invalid_email(self):
         """
         GIVEN:
             - Fresh parser
@@ -46,11 +65,12 @@ class TestParser(FileSystemAssertsMixin, TestCase):
         # Check if exception is raised when the mail is faulty.
         self.assertRaises(
             ParseError,
-            self.parser.get_parsed,
-            os.path.join(self.SAMPLE_FILES, "broken.eml"),
+            self.parser.parse,
+            self.SAMPLE_DIR / "broken.eml",
+            "messages/rfc822",
         )
 
-    def test_get_parsed_simple_text_mail(self):
+    def test_parse_simple_text_email_file(self):
         """
         GIVEN:
             - Fresh parser
@@ -60,8 +80,8 @@ class TestParser(FileSystemAssertsMixin, TestCase):
             - The content of the mail should be available in the parse result.
         """
         # Parse Test file and check relevant content
-        parsed1 = self.parser.get_parsed(
-            os.path.join(self.SAMPLE_FILES, "simple_text.eml"),
+        parsed1 = self.parser.parse_file_to_message(
+            self.SAMPLE_DIR / "simple_text.eml",
         )
 
         self.assertEqual(parsed1.date.year, 2022)
@@ -76,58 +96,11 @@ class TestParser(FileSystemAssertsMixin, TestCase):
         self.assertEqual(parsed1.text, "This is just a simple Text Mail.\n")
         self.assertEqual(parsed1.to, ("some@one.de",))
 
-    def test_get_parsed_reparse(self):
-        """
-        GIVEN:
-            - An E-Mail was parsed
-        WHEN:
-            - Another .eml file should be parsed
-        THEN:
-            - The parser should not retry to parse and return the old results
-        """
-        # Parse Test file and check relevant content
-        parsed1 = self.parser.get_parsed(
-            os.path.join(self.SAMPLE_FILES, "simple_text.eml"),
-        )
-        # Check if same parsed object as before is returned, even if another file is given.
-        parsed2 = self.parser.get_parsed(
-            os.path.join(os.path.join(self.SAMPLE_FILES, "html.eml")),
-        )
-        self.assertEqual(parsed1, parsed2)
 
-    @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf")
-    @mock.patch("paperless_mail.parsers.make_thumbnail_from_pdf")
-    def test_get_thumbnail(
-        self,
-        mock_make_thumbnail_from_pdf: mock.MagicMock,
-        mock_generate_pdf: mock.MagicMock,
-    ):
+class TestEmailMetadataExtraction(BaseMailParserTestCase):
     """
-        GIVEN:
-            - An E-Mail was parsed
-        WHEN:
-            - The Thumbnail is requested
-        THEN:
-            - The parser should call the functions which generate the thumbnail
+    Tests extraction of metadata from an email
     """
-        mocked_return = "Passing the return value through.."
-        mock_make_thumbnail_from_pdf.return_value = mocked_return
-
-        mock_generate_pdf.return_value = "Mocked return value.."
-
-        thumb = self.parser.get_thumbnail(
-            os.path.join(self.SAMPLE_FILES, "simple_text.eml"),
-            "message/rfc822",
-        )
-        self.assertEqual(
-            self.parser.archive_path,
-            mock_make_thumbnail_from_pdf.call_args_list[0].args[0],
-        )
-        self.assertEqual(
-            self.parser.tempdir,
-            mock_make_thumbnail_from_pdf.call_args_list[0].args[1],
-        )
-        self.assertEqual(mocked_return, thumb)
 
     def test_extract_metadata_fail(self):
         """
@@ -157,7 +130,7 @@ class TestParser(FileSystemAssertsMixin, TestCase):
         """
         # Validate Metadata parsing returns the expected results
         metadata = self.parser.extract_metadata(
-            os.path.join(self.SAMPLE_FILES, "simple_text.eml"),
+            self.SAMPLE_DIR / "simple_text.eml",
             "message/rfc822",
         )
 
@@ -287,90 +260,53 @@ class TestParser(FileSystemAssertsMixin, TestCase):
             metadata,
         )
 
-    def test_parse_na(self):
+
+class TestEmailThumbnailGenerate(BaseMailParserTestCase):
+    """
+    Tests the correct generation of an thumbnail for an email
+    """
+
+    @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf")
+    @mock.patch("paperless_mail.parsers.make_thumbnail_from_pdf")
+    def test_get_thumbnail(
+        self,
+        mock_make_thumbnail_from_pdf: mock.MagicMock,
+        mock_generate_pdf: mock.MagicMock,
+    ):
         """
         GIVEN:
-            - Fresh start
+            - An E-Mail was parsed
         WHEN:
-            - parsing is attempted with nonexistent file
+            - The Thumbnail is requested
         THEN:
-            - Exception is thrown
+            - The parser should call the functions which generate the thumbnail
         """
-        # Check if exception is raised when parsing fails.
-        self.assertRaises(
-            ParseError,
-            self.parser.parse,
-            os.path.join(self.SAMPLE_FILES, "na"),
+        mocked_return = "Passing the return value through.."
+        mock_make_thumbnail_from_pdf.return_value = mocked_return
+
+        mock_generate_pdf.return_value = "Mocked return value.."
+
+        test_file = self.SAMPLE_DIR / "simple_text.eml"
+
+        thumb = self.parser.get_thumbnail(
+            test_file,
             "message/rfc822",
         )
 
-    @mock.patch("paperless_mail.parsers.MailDocumentParser.tika_parse")
-    @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf")
-    def test_parse_html_eml(self, n, mock_tika_parse: mock.MagicMock):
-        """
-        GIVEN:
-            - Fresh start
-        WHEN:
-            - parsing is done with html mail
-        THEN:
-            - Tika is called, parsed information from non html parts is available
-        """
-        # Validate parsing returns the expected results
-        text_expected = "Subject: HTML Message\n\nFrom: Name <someone@example.de>\n\nTo: someone@example.de\n\nAttachments: IntM6gnXFm00FEV5.png (6.89 KiB), 600+kbfile.txt (600.24 KiB)\n\nHTML content: tika return\n\nSome Text and an embedded image."
-        mock_tika_parse.return_value = "tika return"
-
-        self.parser.parse(os.path.join(self.SAMPLE_FILES, "html.eml"), "message/rfc822")
-
-        self.assertEqual(text_expected, self.parser.text)
-        self.assertEqual(
-            datetime.datetime(
-                2022,
-                10,
-                15,
-                11,
-                23,
-                19,
-                tzinfo=datetime.timezone(datetime.timedelta(seconds=7200)),
-            ),
-            self.parser.date,
+        mock_generate_pdf.assert_called_once_with(
+            test_file,
+        )
+        mock_make_thumbnail_from_pdf.assert_called_once_with(
+            "Mocked return value..",
+            self.parser.tempdir,
+            None,
         )
 
-    @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf")
-    def test_parse_simple_eml(self, m: mock.MagicMock):
-        """
-        GIVEN:
-            - Fresh start
-        WHEN:
-            - parsing is done with non html mail
-        THEN:
-            - parsed information is available
-        """
-        # Validate parsing returns the expected results
+        self.assertEqual(mocked_return, thumb)
 
-        self.parser.parse(
-            os.path.join(self.SAMPLE_FILES, "simple_text.eml"),
-            "message/rfc822",
-        )
-        text_expected = "Subject: Simple Text Mail\n\nFrom: Some One <mail@someserver.de>\n\nTo: some@one.de\n\nCC: asdasd@æsdasd.de, asdadasdasdasda.asdasd@æsdasd.de\n\nBCC: fdf@fvf.de\n\n\n\nThis is just a simple Text Mail."
-        self.assertEqual(text_expected, self.parser.text)
-        self.assertEqual(
-            datetime.datetime(
-                2022,
-                10,
-                12,
-                21,
-                40,
-                43,
-                tzinfo=datetime.timezone(datetime.timedelta(seconds=7200)),
-            ),
-            self.parser.date,
-        )
 
-        # Just check if tried to generate archive, the unittest for generate_pdf() goes deeper.
-        m.assert_called()
-
-    @mock.patch("paperless_mail.parsers.parser.from_buffer")
-    def test_tika_parse_unsuccessful(self, mock_from_buffer: mock.MagicMock):
+class TestTikaHtmlParse(HttpxMockMixin, BaseMailParserTestCase):
+    def test_tika_parse_unsuccessful(self):
         """
         GIVEN:
             - Fresh start
@@ -380,12 +316,13 @@ class TestParser(FileSystemAssertsMixin, TestCase):
             - the parser should return an empty string
         """
         # Check unsuccessful parsing
-        mock_from_buffer.return_value = {"content": None}
-        parsed = self.parser.tika_parse(None)
+        self.httpx_mock.add_response(
+            json={"Content-Type": "text/html", "X-TIKA:Parsed-By": []},
+        )
+        parsed = self.parser.tika_parse("None")
         self.assertEqual("", parsed)
 
-    @mock.patch("paperless_mail.parsers.parser.from_buffer")
-    def test_tika_parse(self, mock_from_buffer: mock.MagicMock):
+    def test_tika_parse(self):
         """
         GIVEN:
             - Fresh start
@@ -397,14 +334,18 @@ class TestParser(FileSystemAssertsMixin, TestCase):
         html = '<html><head><meta http-equiv="content-type" content="text/html; charset=UTF-8"></head><body><p>Some Text</p></body></html>'
         expected_text = "Some Text"
 
-        # Check successful parsing
-        mock_from_buffer.return_value = {"content": expected_text}
+        self.httpx_mock.add_response(
+            json={
+                "Content-Type": "text/html",
+                "X-TIKA:Parsed-By": [],
+                "X-TIKA:content": expected_text,
+            },
+        )
         parsed = self.parser.tika_parse(html)
         self.assertEqual(expected_text, parsed.strip())
-        mock_from_buffer.assert_called_with(html, self.parser.tika_server)
+        self.assertIn(self.parser.tika_server, str(self.httpx_mock.get_request().url))
 
-    @mock.patch("paperless_mail.parsers.parser.from_buffer")
-    def test_tika_parse_exception(self, mock_from_buffer: mock.MagicMock):
+    def test_tika_parse_exception(self):
         """
         GIVEN:
             - Fresh start
@@ -415,11 +356,8 @@ class TestParser(FileSystemAssertsMixin, TestCase):
         """
         html = '<html><head><meta http-equiv="content-type" content="text/html; charset=UTF-8"></head><body><p>Some Text</p></body></html>'
 
-        # Check ParseError
-        def my_side_effect():
-            raise Exception("Test")
+        self.httpx_mock.add_response(status_code=httpx.codes.INTERNAL_SERVER_ERROR)
 
-        mock_from_buffer.side_effect = my_side_effect
         self.assertRaises(ParseError, self.parser.tika_parse, html)
 
     def test_tika_parse_unreachable(self):
@@ -437,258 +375,285 @@ class TestParser(FileSystemAssertsMixin, TestCase):
         self.parser.tika_server = ""
         self.assertRaises(ParseError, self.parser.tika_parse, html)
 
-    @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf_from_mail")
-    @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf_from_html")
-    def test_generate_pdf_parse_error(self, m: mock.MagicMock, n: mock.MagicMock):
+
+class TestParser(FileSystemAssertsMixin, HttpxMockMixin, BaseMailParserTestCase):
+    def test_parse_no_file(self):
         """
         GIVEN:
             - Fresh start
         WHEN:
-            - pdf generation is requested but gotenberg can not be reached
+            - parsing is attempted with nonexistent file
+        THEN:
+            - Exception is thrown
+        """
+        # Check if exception is raised when parsing fails.
+        self.assertRaises(
+            ParseError,
+            self.parser.parse,
+            self.SAMPLE_DIR / "na.eml",
+            "message/rfc822",
+        )
+
+    @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf")
+    def test_parse_eml_simple(self, mock_generate_pdf: mock.MagicMock):
+        """
+        GIVEN:
+            - Fresh start
+        WHEN:
+            - parsing is done with non html mail
+        THEN:
+            - parsed information is available
+        """
+        # Validate parsing returns the expected results
+
+        self.parser.parse(
+            self.SAMPLE_DIR / "simple_text.eml",
+            "message/rfc822",
+        )
+        text_expected = (
+            "Subject: Simple Text Mail\n\n"
+            "From: Some One <mail@someserver.de>\n\n"
+            "To: some@one.de\n\n"
+            "CC: asdasd@æsdasd.de, asdadasdasdasda.asdasd@æsdasd.de\n\n"
+            "BCC: fdf@fvf.de\n\n"
+            "\n\nThis is just a simple Text Mail."
+        )
+        self.assertEqual(text_expected, self.parser.text)
+        self.assertEqual(
+            datetime.datetime(
+                2022,
+                10,
+                12,
+                21,
+                40,
+                43,
+                tzinfo=datetime.timezone(datetime.timedelta(seconds=7200)),
+            ),
+            self.parser.date,
+        )
+
+        # Just check if tried to generate archive, the unittest for generate_pdf() goes deeper.
+        mock_generate_pdf.assert_called()
+
+    @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf")
+    def test_parse_eml_html(self, mock_generate_pdf: mock.MagicMock):
+        """
+        GIVEN:
+            - Fresh start
+        WHEN:
+            - parsing is done with html mail
+        THEN:
+            - Tika is called, parsed information from non html parts is available
+        """
+        # Validate parsing returns the expected results
+        text_expected = (
+            "Subject: HTML Message\n\n"
+            "From: Name <someone@example.de>\n\n"
+            "To: someone@example.de\n\n"
+            "Attachments: IntM6gnXFm00FEV5.png (6.89 KiB), 600+kbfile.txt (600.24 KiB)\n\n"
+            "HTML content: tika return\n\n"
+            "Some Text and an embedded image."
+        )
+
+        self.httpx_mock.add_response(
+            json={
+                "Content-Type": "text/html",
+                "X-TIKA:Parsed-By": [],
+                "X-TIKA:content": "tika return",
+            },
+        )
+
+        self.parser.parse(self.SAMPLE_DIR / "html.eml", "message/rfc822")
+
+        mock_generate_pdf.assert_called_once()
+        self.assertEqual(text_expected, self.parser.text)
+        self.assertEqual(
+            datetime.datetime(
+                2022,
+                10,
+                15,
+                11,
+                23,
+                19,
+                tzinfo=datetime.timezone(datetime.timedelta(seconds=7200)),
+            ),
+            self.parser.date,
+        )
+
+    def test_generate_pdf_parse_error(self):
+        """
+        GIVEN:
+            - Fresh start
+        WHEN:
+            - pdf generation is requested but gotenberg fails
         THEN:
             - a ParseError Exception is thrown
         """
-        m.return_value = b""
-        n.return_value = b""
+        self.httpx_mock.add_response(status_code=httpx.codes.INTERNAL_SERVER_ERROR)
 
-        # Check if exception is raised when the pdf can not be created.
-        self.parser.gotenberg_server = ""
         self.assertRaises(
             ParseError,
-            self.parser.generate_pdf,
-            os.path.join(self.SAMPLE_FILES, "html.eml"),
+            self.parser.parse,
+            self.SAMPLE_DIR / "simple_text.eml",
+            "message/rfc822",
         )
 
-    def test_generate_pdf_exception(self):
+    def test_generate_pdf_simple_email(self):
         """
         GIVEN:
-            - Fresh start
+            - Simple text email with no HTML content
         WHEN:
-            - pdf generation is requested but parsing throws an exception
+            - Email is parsed
         THEN:
-            - a ParseError Exception is thrown
+            - Gotenberg is called to generate a PDF from HTML
+            - Archive file is generated
         """
-        # Check if exception is raised when the mail can not be parsed.
-        self.assertRaises(
-            ParseError,
-            self.parser.generate_pdf,
-            os.path.join(self.SAMPLE_FILES, "broken.eml"),
+
+        self.httpx_mock.add_response(
+            url="http://localhost:3000/forms/chromium/convert/html",
+            method="POST",
+            content=(self.SAMPLE_DIR / "simple_text.eml.pdf").read_bytes(),
         )
 
-    @mock.patch("paperless_mail.parsers.requests.post")
-    @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf_from_mail")
-    @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf_from_html")
-    def test_generate_pdf(
-        self,
-        mock_generate_pdf_from_html: mock.MagicMock,
-        mock_generate_pdf_from_mail: mock.MagicMock,
-        mock_post: mock.MagicMock,
-    ):
+        self.parser.parse(self.SAMPLE_DIR / "simple_text.eml", "message/rfc822")
+
+        self.assertIsNotNone(self.parser.archive_path)
+
+    def test_generate_pdf_html_email(self):
         """
         GIVEN:
-            - Fresh start
+            - email with HTML content
         WHEN:
-            - pdf generation is requested
+            - Email is parsed
         THEN:
-            - gotenberg is called and the resulting file is returned
+            - Gotenberg is called to generate a PDF from HTML
+            - Gotenberg is used to merge the two PDFs
+            - Archive file is generated
         """
-        mock_generate_pdf_from_mail.return_value = b"Mail Return"
-        mock_generate_pdf_from_html.return_value = b"HTML Return"
+        self.httpx_mock.add_response(
+            url="http://localhost:9998/tika/text",
+            method="PUT",
+            json={
+                "Content-Type": "text/html",
+                "X-TIKA:Parsed-By": [],
+                "X-TIKA:content": "This is some Tika HTML text",
+            },
+        )
+        self.httpx_mock.add_response(
+            url="http://localhost:3000/forms/chromium/convert/html",
+            method="POST",
+            content=(self.SAMPLE_DIR / "html.eml.pdf").read_bytes(),
+        )
+        self.httpx_mock.add_response(
+            url="http://localhost:3000/forms/pdfengines/merge",
+            method="POST",
+            content=b"Pretend merged PDF content",
+        )
+        self.parser.parse(self.SAMPLE_DIR / "html.eml", "message/rfc822")
 
-        mock_response = mock.MagicMock()
-        mock_response.content = b"Content"
-        mock_post.return_value = mock_response
-        pdf_path = self.parser.generate_pdf(os.path.join(self.SAMPLE_FILES, "html.eml"))
-        self.assertIsFile(pdf_path)
+        self.assertIsNotNone(self.parser.archive_path)
 
-        mock_generate_pdf_from_mail.assert_called_once_with(
-            self.parser.get_parsed(None),
+    def test_generate_pdf_html_email_html_to_pdf_failure(self):
+        """
+        GIVEN:
+            - email with HTML content
+        WHEN:
+            - Email is parsed
+            - Conversion of email HTML content to PDF fails
+        THEN:
+            - ParseError is raised
+        """
+        self.httpx_mock.add_response(
+            url="http://localhost:9998/tika/text",
+            method="PUT",
+            json={
+                "Content-Type": "text/html",
+                "X-TIKA:Parsed-By": [],
+                "X-TIKA:content": "This is some Tika HTML text",
+            },
         )
-        mock_generate_pdf_from_html.assert_called_once_with(
-            self.parser.get_parsed(None).html,
-            self.parser.get_parsed(None).attachments,
+        self.httpx_mock.add_response(
+            url="http://localhost:3000/forms/chromium/convert/html",
+            method="POST",
+            content=(self.SAMPLE_DIR / "html.eml.pdf").read_bytes(),
         )
-        self.assertEqual(
-            self.parser.gotenberg_server + "/forms/pdfengines/merge",
-            mock_post.call_args.args[0],
-        )
-        self.assertEqual({}, mock_post.call_args.kwargs["headers"])
-        self.assertEqual(
-            b"Mail Return",
-            mock_post.call_args.kwargs["files"]["1_mail.pdf"][1].read(),
-        )
-        self.assertEqual(
-            b"HTML Return",
-            mock_post.call_args.kwargs["files"]["2_html.pdf"][1].read(),
+        self.httpx_mock.add_response(
+            url="http://localhost:3000/forms/chromium/convert/html",
+            method="POST",
+            status_code=httpx.codes.INTERNAL_SERVER_ERROR,
         )
+        with self.assertRaises(ParseError):
+            self.parser.parse(self.SAMPLE_DIR / "html.eml", "message/rfc822")
 
-        mock_response.raise_for_status.assert_called_once()
-
-        with open(pdf_path, "rb") as file:
-            self.assertEqual(b"Content", file.read())
+    def test_generate_pdf_html_email_merge_failure(self):
+        """
+        GIVEN:
+            - email with HTML content
+        WHEN:
+            - Email is parsed
+            - Merging of PDFs fails
+        THEN:
+            - ParseError is raised
+        """
+        self.httpx_mock.add_response(
+            url="http://localhost:9998/tika/text",
+            method="PUT",
+            json={
+                "Content-Type": "text/html",
+                "X-TIKA:Parsed-By": [],
+                "X-TIKA:content": "This is some Tika HTML text",
+            },
+        )
+        self.httpx_mock.add_response(
+            url="http://localhost:3000/forms/chromium/convert/html",
+            method="POST",
+            content=(self.SAMPLE_DIR / "html.eml.pdf").read_bytes(),
+        )
+        self.httpx_mock.add_response(
+            url="http://localhost:3000/forms/pdfengines/merge",
+            method="POST",
+            status_code=httpx.codes.INTERNAL_SERVER_ERROR,
+        )
+        with self.assertRaises(ParseError):
+            self.parser.parse(self.SAMPLE_DIR / "html.eml", "message/rfc822")
 
     def test_mail_to_html(self):
         """
         GIVEN:
-            - Fresh start
+            - Email message with HTML content
         WHEN:
-            - conversion from eml to html is requested
+            - Email is parsed
         THEN:
-            - html should be returned
+            - Resulting HTML is as expected
         """
-        mail = self.parser.get_parsed(os.path.join(self.SAMPLE_FILES, "html.eml"))
-        html_handle = self.parser.mail_to_html(mail)
-        html_received = html_handle.read()
+        mail = self.parser.parse_file_to_message(self.SAMPLE_DIR / "html.eml")
+        html_file = self.parser.mail_to_html(mail)
+        expected_html_file = self.SAMPLE_DIR / "html.eml.html"
 
-        with open(
-            os.path.join(self.SAMPLE_FILES, "html.eml.html"),
-        ) as html_expected_handle:
-            html_expected = html_expected_handle.read()
+        self.assertHTMLEqual(expected_html_file.read_text(), html_file.read_text())
 
-            self.assertHTMLEqual(html_expected, html_received)
-
-    @mock.patch("paperless_mail.parsers.requests.post")
-    @mock.patch("paperless_mail.parsers.MailDocumentParser.mail_to_html")
     def test_generate_pdf_from_mail(
         self,
-        mock_mail_to_html: mock.MagicMock,
-        mock_post: mock.MagicMock,
     ):
         """
         GIVEN:
-            - Fresh start
+            - Email message with HTML content
         WHEN:
-            - conversion of PDF from .eml is requested
+            - Email is parsed
         THEN:
-            - gotenberg should be called with valid intermediary html files, the resulting pdf is returned
+            - Gotenberg is used to convert HTML to PDF
         """
-        mock_response = mock.MagicMock()
-        mock_response.content = b"Content"
-        mock_post.return_value = mock_response
 
-        mock_mail_to_html.return_value = "Testresponse"
+        self.httpx_mock.add_response(content=b"Content")
 
-        mail = self.parser.get_parsed(os.path.join(self.SAMPLE_FILES, "html.eml"))
+        mail = self.parser.parse_file_to_message(self.SAMPLE_DIR / "html.eml")
 
         retval = self.parser.generate_pdf_from_mail(mail)
-        self.assertEqual(b"Content", retval)
+        self.assertEqual(b"Content", retval.read_bytes())
+
+        request = self.httpx_mock.get_request()
 
-        mock_mail_to_html.assert_called_once_with(mail)
         self.assertEqual(
+            str(request.url),
             self.parser.gotenberg_server + "/forms/chromium/convert/html",
-            mock_post.call_args.args[0],
         )
-        self.assertDictEqual({}, mock_post.call_args.kwargs["headers"])
-        self.assertDictEqual(
-            {
-                "marginTop": "0.1",
-                "marginBottom": "0.1",
-                "marginLeft": "0.1",
-                "marginRight": "0.1",
-                "paperWidth": "8.27",
-                "paperHeight": "11.7",
-                "scale": "1.0",
-                "pdfFormat": "PDF/A-2b",
-            },
-            mock_post.call_args.kwargs["data"],
-        )
-        self.assertEqual(
-            "Testresponse",
-            mock_post.call_args.kwargs["files"]["html"][1],
-        )
-        self.assertEqual(
-            "output.css",
-            mock_post.call_args.kwargs["files"]["css"][0],
-        )
-
-        mock_response.raise_for_status.assert_called_once()
-
-    def test_transform_inline_html(self):
-        """
-        GIVEN:
-            - Fresh start
-        WHEN:
-            - transforming of html content from an email with an inline image attachment is requested
-        THEN:
-            - html is returned and sanitized
-        """
-
-        class MailAttachmentMock:
-            def __init__(self, payload, content_id):
-                self.payload = payload
-                self.content_id = content_id
-
-        result = None
-
-        with open(os.path.join(self.SAMPLE_FILES, "sample.html")) as html_file:
-            with open(os.path.join(self.SAMPLE_FILES, "sample.png"), "rb") as png_file:
-                html = html_file.read()
-                png = png_file.read()
-                attachments = [
-                    MailAttachmentMock(png, "part1.pNdUSz0s.D3NqVtPg@example.de"),
-                ]
-                result = self.parser.transform_inline_html(html, attachments)
-
-        resulting_html = result[-1][1].read()
-        self.assertTrue(result[-1][0] == "index.html")
-        self.assertIn(result[0][0], resulting_html)
-        self.assertNotIn("<script", resulting_html.lower())
-
-    @mock.patch("paperless_mail.parsers.requests.post")
-    def test_generate_pdf_from_html(self, mock_post: mock.MagicMock):
-        """
-        GIVEN:
-            - Fresh start
-        WHEN:
-            - generating pdf from html with inline attachments is attempted
-        THEN:
-            - gotenberg is called with the correct parameters and the resulting pdf is returned
-        """
-
-        class MailAttachmentMock:
-            def __init__(self, payload, content_id):
-                self.payload = payload
-                self.content_id = content_id
-
-        mock_response = mock.MagicMock()
-        mock_response.content = b"Content"
-        mock_post.return_value = mock_response
-
-        result = None
-
-        with open(os.path.join(self.SAMPLE_FILES, "sample.html")) as html_file:
-            with open(os.path.join(self.SAMPLE_FILES, "sample.png"), "rb") as png_file:
-                html = html_file.read()
-                png = png_file.read()
-                attachments = [
-                    MailAttachmentMock(png, "part1.pNdUSz0s.D3NqVtPg@example.de"),
-                ]
-                result = self.parser.generate_pdf_from_html(html, attachments)
-
-        self.assertEqual(
-            self.parser.gotenberg_server + "/forms/chromium/convert/html",
-            mock_post.call_args.args[0],
-        )
-        self.assertDictEqual({}, mock_post.call_args.kwargs["headers"])
-        self.assertDictEqual(
-            {
-                "marginTop": "0.1",
-                "marginBottom": "0.1",
-                "marginLeft": "0.1",
-                "marginRight": "0.1",
-                "paperWidth": "8.27",
-                "paperHeight": "11.7",
-                "scale": "1.0",
-            },
-            mock_post.call_args.kwargs["data"],
-        )
-
-        # read to assert it is a file like object.
-        mock_post.call_args.kwargs["files"]["cidpart1pNdUSz0sD3NqVtPgexamplede"][
-            1
-        ].read()
-        mock_post.call_args.kwargs["files"]["index.html"][1].read()
-
-        mock_response.raise_for_status.assert_called_once()
-
-        self.assertEqual(b"Content", result)

src/paperless_mail/tests/test_parsers_live.py

@@ -1,29 +1,80 @@
 import os
 import time
 from unittest import mock
-from urllib.error import HTTPError
-from urllib.request import urlopen
 
+import httpx
 import pytest
 from django.test import TestCase
 from imagehash import average_hash
 from pdfminer.high_level import extract_text
 from PIL import Image
 
-from documents.parsers import run_convert
 from documents.tests.utils import FileSystemAssertsMixin
-from paperless_mail.parsers import MailDocumentParser
+from paperless_mail.tests.test_parsers import BaseMailParserTestCase
 
 
-class TestParserLive(FileSystemAssertsMixin, TestCase):
-    SAMPLE_FILES = os.path.join(os.path.dirname(__file__), "samples")
+class MailAttachmentMock:
+    def __init__(self, payload, content_id):
+        self.payload = payload
+        self.content_id = content_id
+        self.content_type = "image/png"
 
-    def setUp(self) -> None:
-        self.parser = MailDocumentParser(logging_group=None)
 
-    def tearDown(self) -> None:
-        self.parser.cleanup()
+@pytest.mark.skipif(
+    "PAPERLESS_CI_TEST" not in os.environ,
+    reason="No Gotenberg/Tika servers to test with",
+)
+class TestUrlCanary(TestCase):
+    """
+    Verify certain URLs are still available so testing is valid still
+    """
 
+    def test_online_image_exception_on_not_available(self):
+        """
+        GIVEN:
+            - Fresh start
+        WHEN:
+            - nonexistent image is requested
+        THEN:
+            - An exception shall be thrown
+        """
+        """
+        A public image is used in the html sample file. We have no control
+        whether this image stays online forever, so here we check if we can detect if is not
+        available anymore.
+        """
+        with self.assertRaises(httpx.HTTPStatusError) as cm:
+            resp = httpx.get(
+                "https://upload.wikimedia.org/wikipedia/en/f/f7/nonexistent.png",
+            )
+            resp.raise_for_status()
+
+        self.assertEqual(cm.exception.response.status_code, httpx.codes.NOT_FOUND)
+
+    def test_is_online_image_still_available(self):
+        """
+        GIVEN:
+            - Fresh start
+        WHEN:
+            - A public image used in the html sample file is requested
+        THEN:
+            - No exception shall be thrown
+        """
+        """
+        A public image is used in the html sample file. We have no control
+        whether this image stays online forever, so here we check if it is still there
+        """
+
+        # Now check the URL used in samples/sample.html
+        resp = httpx.get("https://upload.wikimedia.org/wikipedia/en/f/f7/RickRoll.png")
+        resp.raise_for_status()
+
+
+@pytest.mark.skipif(
+    "PAPERLESS_CI_TEST" not in os.environ,
+    reason="No Gotenberg/Tika servers to test with",
+)
+class TestParserLive(FileSystemAssertsMixin, BaseMailParserTestCase):
     @staticmethod
     def imagehash(file, hash_size=18):
         return f"{average_hash(Image.open(file), hash_size)}"
@@ -54,13 +105,18 @@ class TestParserLive(FileSystemAssertsMixin, TestCase):
                 result = method_or_callable(*args)
 
                 succeeded = True
-            except Exception as e:
+            except httpx.HTTPError as e:
+                raise
+                # Retry on HTTP errors
                 print(f"{e} during try #{retry_count}", flush=True)
 
                 retry_count = retry_count + 1
 
                 time.sleep(retry_time)
                 retry_time = retry_time * 2.0
+            except Exception:
+                # Not on other error
+                raise
 
         self.assertTrue(
             succeeded,
@@ -79,17 +135,14 @@ class TestParserLive(FileSystemAssertsMixin, TestCase):
         THEN:
             - The returned thumbnail image file is as expected
         """
-        mock_generate_pdf.return_value = os.path.join(
-            self.SAMPLE_FILES,
-            "simple_text.eml.pdf",
-        )
+        mock_generate_pdf.return_value = self.SAMPLE_DIR / "simple_text.eml.pdf"
         thumb = self.parser.get_thumbnail(
-            os.path.join(self.SAMPLE_FILES, "simple_text.eml"),
+            self.SAMPLE_DIR / "simple_text.eml",
             "message/rfc822",
         )
         self.assertIsFile(thumb)
 
-        expected = os.path.join(self.SAMPLE_FILES, "simple_text.eml.pdf.webp")
+        expected = self.SAMPLE_DIR / "simple_text.eml.pdf.webp"
 
         self.assertEqual(
             self.imagehash(thumb),
@@ -97,10 +150,6 @@ class TestParserLive(FileSystemAssertsMixin, TestCase):
             f"Created Thumbnail {thumb} differs from expected file {expected}",
         )
 
-    @pytest.mark.skipif(
-        "TIKA_LIVE" not in os.environ,
-        reason="No tika server",
-    )
     def test_tika_parse_successful(self):
         """
         GIVEN:
@@ -117,27 +166,6 @@ class TestParserLive(FileSystemAssertsMixin, TestCase):
         parsed = self.parser.tika_parse(html)
         self.assertEqual(expected_text, parsed.strip())
 
-    @pytest.mark.skipif(
-        "TIKA_LIVE" not in os.environ,
-        reason="No tika server",
-    )
-    def test_tika_parse_unsuccessful(self):
-        """
-        GIVEN:
-            - Fresh start
-        WHEN:
-            - tika parsing fails
-        THEN:
-            - the parser should return an empty string
-        """
-        # Check unsuccessful parsing
-        parsed = self.parser.tika_parse(None)
-        self.assertEqual("", parsed)
-
-    @pytest.mark.skipif(
-        "GOTENBERG_LIVE" not in os.environ,
-        reason="No gotenberg server",
-    )
     @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf_from_mail")
     @mock.patch("paperless_mail.parsers.MailDocumentParser.generate_pdf_from_html")
     def test_generate_pdf_gotenberg_merging(
@@ -153,15 +181,16 @@ class TestParserLive(FileSystemAssertsMixin, TestCase):
         THEN:
             - gotenberg is called to merge files and the resulting file is returned
         """
-        with open(os.path.join(self.SAMPLE_FILES, "first.pdf"), "rb") as first:
-            mock_generate_pdf_from_mail.return_value = first.read()
+        mock_generate_pdf_from_mail.return_value = self.SAMPLE_DIR / "first.pdf"
+        mock_generate_pdf_from_html.return_value = self.SAMPLE_DIR / "second.pdf"
 
-        with open(os.path.join(self.SAMPLE_FILES, "second.pdf"), "rb") as second:
-            mock_generate_pdf_from_html.return_value = second.read()
+        msg = self.parser.parse_file_to_message(
+            self.SAMPLE_DIR / "html.eml",
+        )
 
         pdf_path = self.util_call_with_backoff(
             self.parser.generate_pdf,
-            [os.path.join(self.SAMPLE_FILES, "html.eml")],
+            [msg],
         )
         self.assertIsFile(pdf_path)
 
@@ -169,38 +198,9 @@ class TestParserLive(FileSystemAssertsMixin, TestCase):
         expected = (
             "first\tPDF\tto\tbe\tmerged.\n\n\x0csecond\tPDF\tto\tbe\tmerged.\n\n\x0c"
         )
+
         self.assertEqual(expected, extracted)
 
-    @pytest.mark.skipif(
-        "GOTENBERG_LIVE" not in os.environ,
-        reason="No gotenberg server",
-    )
-    def test_generate_pdf_from_mail_no_convert(self):
-        """
-        GIVEN:
-            - Fresh start
-        WHEN:
-            - pdf generation from simple eml file is requested
-        THEN:
-            - gotenberg is called and the resulting file is returned and contains the expected text.
-        """
-        mail = self.parser.get_parsed(os.path.join(self.SAMPLE_FILES, "html.eml"))
-
-        pdf_path = os.path.join(self.parser.tempdir, "html.eml.pdf")
-
-        with open(pdf_path, "wb") as file:
-            file.write(
-                self.util_call_with_backoff(self.parser.generate_pdf_from_mail, [mail]),
-            )
-
-        extracted = extract_text(pdf_path)
-        expected = extract_text(os.path.join(self.SAMPLE_FILES, "html.eml.pdf"))
-        self.assertEqual(expected, extracted)
-
-    @pytest.mark.skipif(
-        "GOTENBERG_LIVE" not in os.environ,
-        reason="No gotenberg server",
-    )
     def test_generate_pdf_from_mail(self):
         """
         GIVEN:
@@ -210,193 +210,32 @@ class TestParserLive(FileSystemAssertsMixin, TestCase):
         THEN:
             - gotenberg is called and the resulting file is returned and look as expected.
         """
-        mail = self.parser.get_parsed(os.path.join(self.SAMPLE_FILES, "html.eml"))
 
-        pdf_path = os.path.join(self.parser.tempdir, "html.eml.pdf")
-
-        with open(pdf_path, "wb") as file:
-            file.write(
-                self.util_call_with_backoff(self.parser.generate_pdf_from_mail, [mail]),
+        self.util_call_with_backoff(
+            self.parser.parse,
+            [self.SAMPLE_DIR / "html.eml", "message/rfc822"],
         )
 
-        converted = os.path.join(
-            self.parser.tempdir,
-            "html.eml.pdf.webp",
+        # Check the archive PDF
+        archive_path = self.parser.get_archive_path()
+        archive_text = extract_text(archive_path)
+        expected_archive_text = extract_text(self.SAMPLE_DIR / "html.eml.pdf")
+
+        # Archive includes the HTML content, so use in
+        self.assertIn(expected_archive_text, archive_text)
+
+        # Check the thumbnail
+        generated_thumbnail = self.parser.get_thumbnail(
+            self.SAMPLE_DIR / "html.eml",
+            "message/rfc822",
         )
-        run_convert(
-            density=300,
-            scale="500x5000>",
-            alpha="remove",
-            strip=True,
-            trim=False,
-            auto_orient=True,
-            input_file=f"{pdf_path}",  # Do net define an index to convert all pages.
-            output_file=converted,
-            logging_group=None,
-        )
-        self.assertIsFile(converted)
-        thumb_hash = self.imagehash(converted)
+        generated_thumbnail_hash = self.imagehash(generated_thumbnail)
 
         # The created pdf is not reproducible. But the converted image should always look the same.
-        expected_hash = self.imagehash(
-            os.path.join(self.SAMPLE_FILES, "html.eml.pdf.webp"),
-        )
-        self.assertEqual(
-            thumb_hash,
-            expected_hash,
-            f"PDF looks different. Check if {converted} looks weird.",
-        )
-
-    @pytest.mark.skipif(
-        "GOTENBERG_LIVE" not in os.environ,
-        reason="No gotenberg server",
-    )
-    def test_generate_pdf_from_html_no_convert(self):
-        """
-        GIVEN:
-            - Fresh start
-        WHEN:
-            - pdf generation from html eml file is requested
-        THEN:
-            - gotenberg is called and the resulting file is returned and contains the expected text.
-        """
-
-        class MailAttachmentMock:
-            def __init__(self, payload, content_id):
-                self.payload = payload
-                self.content_id = content_id
-
-        result = None
-
-        with open(os.path.join(self.SAMPLE_FILES, "sample.html")) as html_file:
-            with open(os.path.join(self.SAMPLE_FILES, "sample.png"), "rb") as png_file:
-                html = html_file.read()
-                png = png_file.read()
-                attachments = [
-                    MailAttachmentMock(png, "part1.pNdUSz0s.D3NqVtPg@example.de"),
-                ]
-                result = self.util_call_with_backoff(
-                    self.parser.generate_pdf_from_html,
-                    [html, attachments],
-                )
-
-        pdf_path = os.path.join(self.parser.tempdir, "sample.html.pdf")
-
-        with open(pdf_path, "wb") as file:
-            file.write(result)
-
-        extracted = extract_text(pdf_path)
-        expected = extract_text(os.path.join(self.SAMPLE_FILES, "sample.html.pdf"))
-        self.assertEqual(expected, extracted)
-
-    @pytest.mark.skipif(
-        "GOTENBERG_LIVE" not in os.environ,
-        reason="No gotenberg server",
-    )
-    def test_generate_pdf_from_html(self):
-        """
-        GIVEN:
-            - Fresh start
-        WHEN:
-            - pdf generation from html eml file is requested
-        THEN:
-            - gotenberg is called and the resulting file is returned and look as expected.
-        """
-
-        class MailAttachmentMock:
-            def __init__(self, payload, content_id):
-                self.payload = payload
-                self.content_id = content_id
-
-        result = None
-
-        with open(os.path.join(self.SAMPLE_FILES, "sample.html")) as html_file:
-            with open(os.path.join(self.SAMPLE_FILES, "sample.png"), "rb") as png_file:
-                html = html_file.read()
-                png = png_file.read()
-                attachments = [
-                    MailAttachmentMock(png, "part1.pNdUSz0s.D3NqVtPg@example.de"),
-                ]
-                result = self.util_call_with_backoff(
-                    self.parser.generate_pdf_from_html,
-                    [html, attachments],
-                )
-
-        pdf_path = os.path.join(self.parser.tempdir, "sample.html.pdf")
-
-        with open(pdf_path, "wb") as file:
-            file.write(result)
-
-        converted = os.path.join(self.parser.tempdir, "sample.html.pdf.webp")
-        run_convert(
-            density=300,
-            scale="500x5000>",
-            alpha="remove",
-            strip=True,
-            trim=False,
-            auto_orient=True,
-            input_file=f"{pdf_path}",  # Do net define an index to convert all pages.
-            output_file=converted,
-            logging_group=None,
-        )
-        self.assertIsFile(converted)
-        thumb_hash = self.imagehash(converted)
-
-        # The created pdf is not reproducible. But the converted image should always look the same.
-        expected_hash = self.imagehash(
-            os.path.join(self.SAMPLE_FILES, "sample.html.pdf.webp"),
-        )
+        expected_hash = self.imagehash(self.SAMPLE_DIR / "html.eml.pdf.webp")
 
         self.assertEqual(
-            thumb_hash,
+            generated_thumbnail_hash,
             expected_hash,
-            f"PDF looks different. Check if {converted} looks weird. "
-            f"If Rick Astley is shown, Gotenberg loads from web which is bad for Mail content.",
+            f"PDF looks different. Check if {generated_thumbnail} looks weird.",
         )
-
-    @pytest.mark.skipif(
-        "GOTENBERG_LIVE" not in os.environ,
-        reason="No gotenberg server",
-    )
-    def test_online_image_exception_on_not_available(self):
-        """
-        GIVEN:
-            - Fresh start
-        WHEN:
-            - nonexistent image is requested
-        THEN:
-            - An exception shall be thrown
-        """
-        """
-        A public image is used in the html sample file. We have no control
-        whether this image stays online forever, so here we check if we can detect if is not
-        available anymore.
-        """
-
-        # Start by Testing if nonexistent URL really throws an Exception
-        self.assertRaises(
-            HTTPError,
-            urlopen,
-            "https://upload.wikimedia.org/wikipedia/en/f/f7/nonexistent.png",
-        )
-
-    @pytest.mark.skipif(
-        "GOTENBERG_LIVE" not in os.environ,
-        reason="No gotenberg server",
-    )
-    def test_is_online_image_still_available(self):
-        """
-        GIVEN:
-            - Fresh start
-        WHEN:
-            - A public image used in the html sample file is requested
-        THEN:
-            - No exception shall be thrown
-        """
-        """
-        A public image is used in the html sample file. We have no control
-        whether this image stays online forever, so here we check if it is still there
-        """
-
-        # Now check the URL used in samples/sample.html
-        urlopen("https://upload.wikimedia.org/wikipedia/en/f/f7/RickRoll.png")

src/paperless_tesseract/parsers.py

@@ -335,7 +335,7 @@ class RasterisedDocumentParser(DocumentParser):
                 self.text = text_original
         except (NoTextFoundException, InputFileError) as e:
             self.log.warning(
-                f"Encountered an error while running OCR: {str(e)}. "
+                f"Encountered an error while running OCR: {e!s}. "
                 f"Attempting force OCR to get the text.",
             )
 
@@ -370,11 +370,11 @@ class RasterisedDocumentParser(DocumentParser):
 
             except Exception as e:
                 # If this fails, we have a serious issue at hand.
-                raise ParseError(f"{e.__class__.__name__}: {str(e)}") from e
+                raise ParseError(f"{e.__class__.__name__}: {e!s}") from e
 
         except Exception as e:
             # Anything else is probably serious.
-            raise ParseError(f"{e.__class__.__name__}: {str(e)}") from e
+            raise ParseError(f"{e.__class__.__name__}: {e!s}") from e
 
         # As a last resort, if we still don't have any text for any reason,
         # try to extract the text from the original document.

src/paperless_tika/parsers.py

@@ -1,10 +1,9 @@
 import os
 from pathlib import Path
 
-import dateutil.parser
-import requests
+import httpx
 from django.conf import settings
-from tika import parser
+from tika_client import TikaClient
 
 from documents.parsers import DocumentParser
 from documents.parsers import ParseError
@@ -29,55 +28,38 @@ class TikaDocumentParser(DocumentParser):
         )
 
     def extract_metadata(self, document_path, mime_type):
-        tika_server = settings.TIKA_ENDPOINT
-
-        # tika does not support a PathLike, only strings
-        # ensure this is a string
-        document_path = str(document_path)
-
         try:
-            parsed = parser.from_file(document_path, tika_server)
+            with TikaClient(tika_url=settings.TIKA_ENDPOINT) as client:
+                parsed = client.metadata.from_file(document_path, mime_type)
+                return [
+                    {
+                        "namespace": "",
+                        "prefix": "",
+                        "key": key,
+                        "value": parsed.data[key],
+                    }
+                    for key in parsed.data
+                ]
         except Exception as e:
             self.log.warning(
                 f"Error while fetching document metadata for {document_path}: {e}",
             )
             return []
 
-        return [
-            {
-                "namespace": "",
-                "prefix": "",
-                "key": key,
-                "value": parsed["metadata"][key],
-            }
-            for key in parsed["metadata"]
-        ]
-
-    def parse(self, document_path: Path, mime_type, file_name=None):
+    def parse(self, document_path: Path, mime_type: str, file_name=None):
         self.log.info(f"Sending {document_path} to Tika server")
-        tika_server = settings.TIKA_ENDPOINT
-
-        # tika does not support a PathLike, only strings
-        # ensure this is a string
-        document_path = str(document_path)
 
         try:
-            parsed = parser.from_file(document_path, tika_server)
+            with TikaClient(tika_url=settings.TIKA_ENDPOINT) as client:
+                parsed = client.tika.as_text.from_file(document_path, mime_type)
         except Exception as err:
             raise ParseError(
                 f"Could not parse {document_path} with tika server at "
-                f"{tika_server}: {err}",
+                f"{settings.TIKA_ENDPOINT}: {err}",
             ) from err
 
-        self.text = parsed["content"].strip()
-
-        try:
-            self.date = dateutil.parser.isoparse(parsed["metadata"]["Creation-Date"])
-        except Exception as e:
-            self.log.warning(
-                f"Unable to extract date for document {document_path}: {e}",
-            )
-
+        self.text = parsed.content.strip()
+        self.date = parsed.metadata.created
         self.archive_path = self.convert_to_pdf(document_path, file_name)
 
     def convert_to_pdf(self, document_path, file_name):
@@ -106,7 +88,7 @@ class TikaDocumentParser(DocumentParser):
                 data["pdfFormat"] = "PDF/A-3b"
 
             try:
-                response = requests.post(url, files=files, headers=headers, data=data)
+                response = httpx.post(url, files=files, headers=headers, data=data)
                 response.raise_for_status()  # ensure we notice bad responses
             except Exception as err:
                 raise ParseError(

src/paperless_tika/tests/test_live_tika.py

@@ -9,7 +9,10 @@ from django.test import TestCase
 from paperless_tika.parsers import TikaDocumentParser
 
 
-@pytest.mark.skipif("TIKA_LIVE" not in os.environ, reason="No tika server")
+@pytest.mark.skipif(
+    "PAPERLESS_CI_TEST" not in os.environ,
+    reason="No Gotenberg/Tika servers to test with",
+)
 class TestTikaParserAgainstServer(TestCase):
     """
     This test case tests the Tika parsing against a live tika server,
@@ -25,7 +28,7 @@ class TestTikaParserAgainstServer(TestCase):
     def tearDown(self) -> None:
         self.parser.cleanup()
 
-    def try_parse_with_wait(self, test_file, mime_type):
+    def try_parse_with_wait(self, test_file: Path, mime_type: str):
         """
         For whatever reason, the image started during the test pipeline likes to
         segfault sometimes, when run with the exact files that usually pass.

src/paperless_tika/tests/test_tika_parser.py

@@ -5,34 +5,38 @@ from unittest import mock
 
 from django.test import TestCase
 from django.test import override_settings
-from requests import Response
+from httpx import Request
+from httpx import Response
 from rest_framework import status
 
 from documents.parsers import ParseError
 from paperless_tika.parsers import TikaDocumentParser
+from paperless_tika.tests.utils import HttpxMockMixin
 
 
-class TestTikaParser(TestCase):
+class TestTikaParser(HttpxMockMixin, TestCase):
     def setUp(self) -> None:
         self.parser = TikaDocumentParser(logging_group=None)
 
     def tearDown(self) -> None:
         self.parser.cleanup()
 
-    @mock.patch("paperless_tika.parsers.parser.from_file")
-    @mock.patch("paperless_tika.parsers.requests.post")
-    def test_parse(self, post, from_file):
-        from_file.return_value = {
-            "content": "the content",
-            "metadata": {"Creation-Date": "2020-11-21"},
-        }
-        response = Response()
-        response._content = b"PDF document"
-        response.status_code = status.HTTP_200_OK
-        post.return_value = response
+    def test_parse(self):
+        # Pretend parse response
+        self.httpx_mock.add_response(
+            json={
+                "Content-Type": "application/vnd.oasis.opendocument.text",
+                "X-TIKA:Parsed-By": [],
+                "X-TIKA:content": "the content",
+                "dcterms:created": "2020-11-21T00:00:00",
+            },
+        )
+        # Pretend convert to PDF response
+        self.httpx_mock.add_response(content=b"PDF document")
+
+        file = Path(os.path.join(self.parser.tempdir, "input.odt"))
+        file.touch()
 
-        file = os.path.join(self.parser.tempdir, "input.odt")
-        Path(file).touch()
         self.parser.parse(file, "application/vnd.oasis.opendocument.text")
 
         self.assertEqual(self.parser.text, "the content")
@@ -42,26 +46,28 @@ class TestTikaParser(TestCase):
 
         self.assertEqual(self.parser.date, datetime.datetime(2020, 11, 21))
 
-    @mock.patch("paperless_tika.parsers.parser.from_file")
-    def test_metadata(self, from_file):
-        from_file.return_value = {
-            "metadata": {"Creation-Date": "2020-11-21", "Some-key": "value"},
-        }
+    def test_metadata(self):
+        self.httpx_mock.add_response(
+            json={
+                "Content-Type": "application/vnd.oasis.opendocument.text",
+                "X-TIKA:Parsed-By": [],
+                "Some-key": "value",
+                "dcterms:created": "2020-11-21T00:00:00",
+            },
+        )
 
-        file = os.path.join(self.parser.tempdir, "input.odt")
-        Path(file).touch()
+        file = Path(os.path.join(self.parser.tempdir, "input.odt"))
+        file.touch()
 
         metadata = self.parser.extract_metadata(
             file,
             "application/vnd.oasis.opendocument.text",
         )
 
-        self.assertTrue("Creation-Date" in [m["key"] for m in metadata])
+        self.assertTrue("dcterms:created" in [m["key"] for m in metadata])
         self.assertTrue("Some-key" in [m["key"] for m in metadata])
 
-    @mock.patch("paperless_tika.parsers.parser.from_file")
-    @mock.patch("paperless_tika.parsers.requests.post")
-    def test_convert_failure(self, post, from_file):
+    def test_convert_failure(self):
         """
         GIVEN:
             - Document needs to be converted to PDF
@@ -70,22 +76,16 @@ class TestTikaParser(TestCase):
         THEN:
             - Parse error is raised
         """
-        from_file.return_value = {
-            "content": "the content",
-            "metadata": {"Creation-Date": "2020-11-21"},
-        }
-        response = Response()
-        response._content = b"PDF document"
-        response.status_code = status.HTTP_500_INTERNAL_SERVER_ERROR
-        post.return_value = response
+        # Pretend convert to PDF response
+        self.httpx_mock.add_response(status_code=status.HTTP_500_INTERNAL_SERVER_ERROR)
 
-        file = os.path.join(self.parser.tempdir, "input.odt")
-        Path(file).touch()
+        file = Path(os.path.join(self.parser.tempdir, "input.odt"))
+        file.touch()
 
         with self.assertRaises(ParseError):
             self.parser.convert_to_pdf(file, None)
 
-    @mock.patch("paperless_tika.parsers.requests.post")
+    @mock.patch("paperless_tika.parsers.httpx.post")
     def test_request_pdf_a_format(self, post: mock.Mock):
         """
         GIVEN:
@@ -95,12 +95,11 @@ class TestTikaParser(TestCase):
         THEN:
             - Request to Gotenberg contains the expected PDF/A format string
         """
-        file = os.path.join(self.parser.tempdir, "input.odt")
-        Path(file).touch()
+        file = Path(os.path.join(self.parser.tempdir, "input.odt"))
+        file.touch()
 
-        response = Response()
-        response._content = b"PDF document"
-        response.status_code = status.HTTP_200_OK
+        response = Response(status_code=status.HTTP_200_OK)
+        response.request = Request("POST", "/somewhere/")
         post.return_value = response
 
         for setting, expected_key in [

src/paperless_tika/tests/utils.py

@@ -0,0 +1,11 @@
+import pytest
+from pytest_httpx import HTTPXMock
+
+
+class HttpxMockMixin:
+    @pytest.fixture(autouse=True)
+    def httpx_mock_auto(self, httpx_mock: HTTPXMock):
+        """
+        Workaround for allowing use of a fixture with unittest style testing
+        """
+        self.httpx_mock = httpx_mock