fix: add RasterisedDocumentParser to new-style parser shim checks

The new RasterisedDocumentParser uses __enter__/__exit__ for resource
management instead of cleanup(). Update all existing new-style shims to
include it in the isinstance checks:

- documents/consumer.py: _parser_cleanup(), parser_is_new_style
- documents/tasks.py: parser_is_new_style, finally cleanup branch
  (also adds RemoteDocumentParser which was missing from the latter)
- documents/management/commands/document_thumbnails.py: adds new-style
  handling from scratch (enter/exit + 2-arg get_thumbnail signature)

Fix stale import paths in three test files that were still importing
from paperless_tesseract.parsers instead of paperless.parsers.tesseract.

Fix two registry tests that used application/pdf as a proxy for "no
handler" — now that RasterisedDocumentParser is registered, PDF always
has a handler, so switch to a truly unsupported MIME type.

Signal infrastructure and shims remain intact; this is plumbing only.

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

pyproject.toml

@@ -248,15 +248,13 @@ lint.per-file-ignores."docker/wait-for-redis.py" = [
 lint.per-file-ignores."src/documents/models.py" = [
   "SIM115",
 ]
-lint.per-file-ignores."src/paperless_tesseract/tests/test_parser.py" = [
-  "RUF001",
-]
+
 lint.isort.force-single-line = true
 
 [tool.codespell]
 write-changes = true
 ignore-words-list = "criterias,afterall,valeu,ureue,equest,ure,assertIn,Oktober,commitish"
-skip = "src-ui/src/locale/*,src-ui/pnpm-lock.yaml,src-ui/e2e/*,src/paperless_mail/tests/samples/*,src/documents/tests/samples/*,*.po,*.json"
+skip = "src-ui/src/locale/*,src-ui/pnpm-lock.yaml,src-ui/e2e/*,src/paperless_mail/tests/samples/*,src/paperless/tests/samples/mail/*,src/documents/tests/samples/*,*.po,*.json"
 
 [tool.pytest]
 minversion = "9.0"

src/documents/consumer.py

@@ -51,10 +51,12 @@ from documents.templating.workflows import parse_w_workflow_placeholders
 from documents.utils import copy_basic_file_stats
 from documents.utils import copy_file_with_basic_stats
 from documents.utils import run_subprocess
+from paperless.parsers import ParserContext
+from paperless.parsers.mail import MailDocumentParser
 from paperless.parsers.remote import RemoteDocumentParser
+from paperless.parsers.tesseract import RasterisedDocumentParser
 from paperless.parsers.text import TextDocumentParser
 from paperless.parsers.tika import TikaDocumentParser
-from paperless_mail.parsers import MailDocumentParser
 
 LOGGING_NAME: Final[str] = "paperless.consumer"
 
@@ -71,7 +73,13 @@ def _parser_cleanup(parser: DocumentParser) -> None:
     """
     if isinstance(
         parser,
-        (TextDocumentParser, RemoteDocumentParser, TikaDocumentParser),
+        (
+            MailDocumentParser,
+            RasterisedDocumentParser,
+            RemoteDocumentParser,
+            TextDocumentParser,
+            TikaDocumentParser,
+        ),
     ):
         parser.__exit__(None, None, None)
     else:
@@ -453,13 +461,21 @@ class ConsumerPlugin(
             progress_callback=progress_callback,
         )
 
+        parser_is_new_style = isinstance(
+            document_parser,
+            (
+                MailDocumentParser,
+                RasterisedDocumentParser,
+                RemoteDocumentParser,
+                TextDocumentParser,
+                TikaDocumentParser,
+            ),
+        )
+
         # New-style parsers use __enter__/__exit__ for resource management.
         # _parser_cleanup (below) handles __exit__; call __enter__ here.
         # TODO(stumpylog): Remove me in the future
-        if isinstance(
-            document_parser,
-            (TextDocumentParser, RemoteDocumentParser, TikaDocumentParser),
-        ):
+        if parser_is_new_style:
             document_parser.__enter__()
 
         self.log.debug(f"Parser: {type(document_parser).__name__}")
@@ -480,20 +496,12 @@ class ConsumerPlugin(
                 ConsumerStatusShortMessage.PARSING_DOCUMENT,
             )
             self.log.debug(f"Parsing {self.filename}...")
-            if (
-                isinstance(document_parser, MailDocumentParser)
-                and self.input_doc.mailrule_id
-            ):
-                document_parser.parse(
-                    self.working_copy,
-                    mime_type,
-                    self.filename,
-                    self.input_doc.mailrule_id,
+
+            # TODO(stumpylog): Remove me in the future when all parsers use new protocol
+            if parser_is_new_style:
+                document_parser.configure(
+                    ParserContext(mailrule_id=self.input_doc.mailrule_id),
                 )
-            elif isinstance(
-                document_parser,
-                (TextDocumentParser, RemoteDocumentParser, TikaDocumentParser),
-            ):
                 # TODO(stumpylog): Remove me in the future
                 document_parser.parse(self.working_copy, mime_type)
             else:
@@ -506,11 +514,8 @@ class ConsumerPlugin(
                 ProgressStatusOptions.WORKING,
                 ConsumerStatusShortMessage.GENERATING_THUMBNAIL,
             )
-            if isinstance(
-                document_parser,
-                (TextDocumentParser, RemoteDocumentParser, TikaDocumentParser),
-            ):
-                # TODO(stumpylog): Remove me in the future
+            # TODO(stumpylog): Remove me in the future when all parsers use new protocol
+            if parser_is_new_style:
                 thumbnail = document_parser.get_thumbnail(self.working_copy, mime_type)
             else:
                 thumbnail = document_parser.get_thumbnail(

src/documents/management/commands/document_thumbnails.py

@@ -4,6 +4,11 @@ import shutil
 from documents.management.commands.base import PaperlessCommand
 from documents.models import Document
 from documents.parsers import get_parser_class_for_mime_type
+from paperless.parsers.mail import MailDocumentParser
+from paperless.parsers.remote import RemoteDocumentParser
+from paperless.parsers.tesseract import RasterisedDocumentParser
+from paperless.parsers.text import TextDocumentParser
+from paperless.parsers.tika import TikaDocumentParser
 
 logger = logging.getLogger("paperless.management.thumbnails")
 
@@ -22,16 +27,38 @@ def _process_document(doc_id: int) -> None:
 
     parser = parser_class(logging_group=None)
 
+    parser_is_new_style = isinstance(
+        parser,
+        (
+            MailDocumentParser,
+            RasterisedDocumentParser,
+            RemoteDocumentParser,
+            TextDocumentParser,
+            TikaDocumentParser,
+        ),
+    )
+
+    # TODO(stumpylog): Remove branch in the future when all parsers use new protocol
+    if parser_is_new_style:
+        parser.__enter__()
+
     try:
-        thumb = parser.get_thumbnail(
-            document.source_path,
-            document.mime_type,
-            document.get_public_filename(),
-        )
+        # TODO(stumpylog): Remove branch in the future when all parsers use new protocol
+        if parser_is_new_style:
+            thumb = parser.get_thumbnail(document.source_path, document.mime_type)
+        else:
+            thumb = parser.get_thumbnail(
+                document.source_path,
+                document.mime_type,
+                document.get_public_filename(),
+            )
         shutil.move(thumb, document.thumbnail_path)
     finally:
         # TODO(stumpylog): Cleanup once all parsers are handled
-        parser.cleanup()
+        if parser_is_new_style:
+            parser.__exit__(None, None, None)
+        else:
+            parser.cleanup()
 
 
 class Command(PaperlessCommand):

src/documents/tasks.py

@@ -65,6 +65,12 @@ from documents.signals.handlers import run_workflows
 from documents.signals.handlers import send_websocket_document_updated
 from documents.workflows.utils import get_workflows_for_trigger
 from paperless.config import AIConfig
+from paperless.parsers import ParserContext
+from paperless.parsers.mail import MailDocumentParser
+from paperless.parsers.remote import RemoteDocumentParser
+from paperless.parsers.tesseract import RasterisedDocumentParser
+from paperless.parsers.text import TextDocumentParser
+from paperless.parsers.tika import TikaDocumentParser
 from paperless_ai.indexing import llm_index_add_or_update_document
 from paperless_ai.indexing import llm_index_remove_document
 from paperless_ai.indexing import update_llm_index
@@ -304,7 +310,9 @@ def update_document_content_maybe_archive_file(document_id) -> None:
 
     mime_type = document.mime_type
 
-    parser_class: type[DocumentParser] = get_parser_class_for_mime_type(mime_type)
+    parser_class: type[DocumentParser] | None = get_parser_class_for_mime_type(
+        mime_type,
+    )
 
     if not parser_class:
         logger.error(
@@ -315,14 +323,42 @@ def update_document_content_maybe_archive_file(document_id) -> None:
 
     parser: DocumentParser = parser_class(logging_group=uuid.uuid4())
 
-    try:
-        parser.parse(document.source_path, mime_type, document.get_public_filename())
+    parser_is_new_style = isinstance(
+        parser,
+        (
+            MailDocumentParser,
+            RasterisedDocumentParser,
+            RemoteDocumentParser,
+            TextDocumentParser,
+            TikaDocumentParser,
+        ),
+    )
 
-        thumbnail = parser.get_thumbnail(
-            document.source_path,
-            mime_type,
-            document.get_public_filename(),
-        )
+    # TODO(stumpylog): Remove branch in the future when all parsers use new protocol
+    if parser_is_new_style:
+        parser.__enter__()
+
+    try:
+        # TODO(stumpylog): Remove branch in the future when all parsers use new protocol
+        if parser_is_new_style:
+            parser.configure(ParserContext())
+            parser.parse(document.source_path, mime_type)
+        else:
+            parser.parse(
+                document.source_path,
+                mime_type,
+                document.get_public_filename(),
+            )
+
+        # TODO(stumpylog): Remove branch in the future when all parsers use new protocol
+        if parser_is_new_style:
+            thumbnail = parser.get_thumbnail(document.source_path, mime_type)
+        else:
+            thumbnail = parser.get_thumbnail(
+                document.source_path,
+                mime_type,
+                document.get_public_filename(),
+            )
 
         with transaction.atomic():
             oldDocument = Document.objects.get(pk=document.pk)
@@ -403,8 +439,20 @@ def update_document_content_maybe_archive_file(document_id) -> None:
             f"Error while parsing document {document} (ID: {document_id})",
         )
     finally:
-        # TODO(stumpylog): Cleanup once all parsers are handled
-        parser.cleanup()
+        # TODO(stumpylog): Remove branch in the future when all parsers use new protocol
+        if isinstance(
+            parser,
+            (
+                MailDocumentParser,
+                RasterisedDocumentParser,
+                RemoteDocumentParser,
+                TextDocumentParser,
+                TikaDocumentParser,
+            ),
+        ):
+            parser.__exit__(None, None, None)
+        else:
+            parser.cleanup()
 
 
 @shared_task

src/documents/tests/test_consumer.py

@@ -36,7 +36,6 @@ from documents.tests.utils import DummyProgressManager
 from documents.tests.utils import FileSystemAssertsMixin
 from documents.tests.utils import GetConsumerMixin
 from paperless_mail.models import MailRule
-from paperless_mail.parsers import MailDocumentParser
 
 
 class _BaseTestParser(DocumentParser):
@@ -1091,7 +1090,7 @@ class TestConsumer(
             self.assertEqual(command[1], "--replace-input")
 
     @mock.patch("paperless_mail.models.MailRule.objects.get")
-    @mock.patch("paperless_mail.parsers.MailDocumentParser.parse")
+    @mock.patch("paperless.parsers.mail.MailDocumentParser.parse")
     @mock.patch("documents.parsers.document_consumer_declaration.send")
     def test_mail_parser_receives_mailrule(
         self,
@@ -1107,11 +1106,13 @@ class TestConsumer(
         THEN:
             - The mail parser should receive the mail rule
         """
+        from paperless_mail.signals import get_parser as mail_get_parser
+
         mock_consumer_declaration_send.return_value = [
             (
                 None,
                 {
-                    "parser": MailDocumentParser,
+                    "parser": mail_get_parser,
                     "mime_types": {"message/rfc822": ".eml"},
                     "weight": 0,
                 },
@@ -1123,9 +1124,10 @@ class TestConsumer(
         with self.get_consumer(
             filepath=(
                 Path(__file__).parent.parent.parent
-                / Path("paperless_mail")
+                / Path("paperless")
                 / Path("tests")
                 / Path("samples")
+                / Path("mail")
             ).resolve()
             / "html.eml",
             source=DocumentSource.MailFetch,
@@ -1136,12 +1138,10 @@ class TestConsumer(
                 ConsumerError,
             ):
                 consumer.run()
-                mock_mail_parser_parse.assert_called_once_with(
-                    consumer.working_copy,
-                    "message/rfc822",
-                    file_name="sample.pdf",
-                    mailrule=mock_mailrule_get.return_value,
-                )
+            mock_mail_parser_parse.assert_called_once_with(
+                consumer.working_copy,
+                "message/rfc822",
+            )
 
 
 @mock.patch("documents.consumer.magic.from_file", fake_magic_from_file)

src/documents/tests/test_parsers.py

@@ -9,9 +9,9 @@ from documents.parsers import get_default_file_extension
 from documents.parsers import get_parser_class_for_mime_type
 from documents.parsers import get_supported_file_extensions
 from documents.parsers import is_file_ext_supported
+from paperless.parsers.tesseract import RasterisedDocumentParser
 from paperless.parsers.text import TextDocumentParser
 from paperless.parsers.tika import TikaDocumentParser
-from paperless_tesseract.parsers import RasterisedDocumentParser
 
 
 class TestParserDiscovery(TestCase):

src/paperless/parsers/__init__.py

@@ -35,6 +35,7 @@ Usage example (third-party parser)::
 
 from __future__ import annotations
 
+from dataclasses import dataclass
 from typing import TYPE_CHECKING
 from typing import Protocol
 from typing import Self
@@ -48,6 +49,7 @@ if TYPE_CHECKING:
 
 __all__ = [
     "MetadataEntry",
+    "ParserContext",
     "ParserProtocol",
 ]
 
@@ -73,6 +75,44 @@ class MetadataEntry(TypedDict):
     """String representation of the field value."""
 
 
+@dataclass(frozen=True, slots=True)
+class ParserContext:
+    """Immutable context passed to a parser before parse().
+
+    The consumer assembles this from the ingestion event and Django
+    settings, then calls ``parser.configure(context)`` before
+    ``parser.parse()``.  Parsers read only the fields relevant to them;
+    unneeded fields are ignored.
+
+    ``frozen=True`` prevents accidental mutation after the consumer
+    hands the context off.  ``slots=True`` keeps instances lightweight.
+
+    Fields
+    ------
+    mailrule_id : int | None
+        Primary key of the ``MailRule`` that triggered this ingestion,
+        or ``None`` when the document did not arrive via a mail rule.
+        Used by ``MailDocumentParser`` to select the PDF layout.
+
+    Notes
+    -----
+    Future fields (not yet implemented):
+
+    * ``output_type`` — PDF/A variant for archive generation
+      (replaces ``settings.OCR_OUTPUT_TYPE`` reads inside parsers).
+    * ``ocr_mode`` — skip-text, redo, force, etc.
+      (replaces ``settings.OCR_MODE`` reads inside parsers).
+    * ``ocr_language`` — Tesseract language string.
+      (replaces ``settings.OCR_LANGUAGE`` reads inside parsers).
+
+    When those fields are added the consumer will read from Django
+    settings once and populate them here, decoupling parsers from
+    ``settings.*`` entirely.
+    """
+
+    mailrule_id: int | None = None
+
+
 @runtime_checkable
 class ParserProtocol(Protocol):
     """Structural contract for all Paperless-ngx document parsers.
@@ -191,6 +231,21 @@ class ParserProtocol(Protocol):
     # Core parsing interface
     # ------------------------------------------------------------------
 
+    def configure(self, context: ParserContext) -> None:
+        """Apply source context before parse().
+
+        Called by the consumer after instantiation and before parse().
+        The default implementation is a no-op; parsers override only the
+        fields they need.
+
+        Parameters
+        ----------
+        context:
+            Immutable context assembled by the consumer for this
+            specific ingestion event.
+        """
+        ...
+
     def parse(
         self,
         document_path: Path,

src/paperless/parsers/mail.py

@@ -0,0 +1,834 @@
+"""
+Built-in mail document parser.
+
+Handles message/rfc822 (EML) MIME type by:
+- Parsing the email using imap_tools
+- Generating a PDF via Gotenberg (for display and archive)
+- Extracting text via Tika for HTML content
+- Extracting metadata from email headers
+
+The parser always produces a PDF because EML files cannot be rendered
+natively in a browser (requires_pdf_rendition=True).
+"""
+
+from __future__ import annotations
+
+import logging
+import re
+import shutil
+import tempfile
+from html import escape
+from pathlib import Path
+from typing import TYPE_CHECKING
+from typing import Self
+
+from bleach import clean
+from bleach import linkify
+from django.conf import settings
+from django.utils import timezone
+from django.utils.timezone import is_naive
+from django.utils.timezone import make_aware
+from gotenberg_client import GotenbergClient
+from gotenberg_client.constants import A4
+from gotenberg_client.options import Measurement
+from gotenberg_client.options import MeasurementUnitType
+from gotenberg_client.options import PageMarginsType
+from gotenberg_client.options import PdfAFormat
+from humanize import naturalsize
+from imap_tools import MailAttachment
+from imap_tools import MailMessage
+from tika_client import TikaClient
+
+from documents.parsers import ParseError
+from documents.parsers import make_thumbnail_from_pdf
+from paperless.models import OutputTypeChoices
+from paperless.version import __full_version_str__
+from paperless_mail.models import MailRule
+
+if TYPE_CHECKING:
+    import datetime
+    from types import TracebackType
+
+    from paperless.parsers import MetadataEntry
+    from paperless.parsers import ParserContext
+
+logger = logging.getLogger("paperless.parsing.mail")
+
+_SUPPORTED_MIME_TYPES: dict[str, str] = {
+    "message/rfc822": ".eml",
+}
+
+
+class MailDocumentParser:
+    """Parse .eml email files for Paperless-ngx.
+
+    Uses imap_tools to parse .eml files, generates a PDF using Gotenberg,
+    and sends the HTML part to a Tika server for text extraction.  Because
+    EML files cannot be rendered natively in a browser, the parser always
+    produces a PDF rendition (requires_pdf_rendition=True).
+
+    Pass a ``ParserContext`` to ``configure()`` before ``parse()`` to
+    apply mail-rule-specific PDF layout options:
+
+        parser.configure(ParserContext(mailrule_id=rule.pk))
+        parser.parse(path, mime_type)
+
+    Class attributes
+    ----------------
+    name : str
+        Human-readable parser name.
+    version : str
+        Semantic version string, kept in sync with Paperless-ngx releases.
+    author : str
+        Maintainer name.
+    url : str
+        Issue tracker / source URL.
+    """
+
+    name: str = "Paperless-ngx Mail Parser"
+    version: str = __full_version_str__
+    author: str = "Paperless-ngx Contributors"
+    url: str = "https://github.com/paperless-ngx/paperless-ngx"
+
+    # ------------------------------------------------------------------
+    # Class methods
+    # ------------------------------------------------------------------
+
+    @classmethod
+    def supported_mime_types(cls) -> dict[str, str]:
+        """Return the MIME types this parser handles.
+
+        Returns
+        -------
+        dict[str, str]
+            Mapping of MIME type to preferred file extension.
+        """
+        return _SUPPORTED_MIME_TYPES
+
+    @classmethod
+    def score(
+        cls,
+        mime_type: str,
+        filename: str,
+        path: Path | None = None,
+    ) -> int | None:
+        """Return the priority score for handling this file.
+
+        Parameters
+        ----------
+        mime_type:
+            Detected MIME type of the file.
+        filename:
+            Original filename including extension.
+        path:
+            Optional filesystem path. Not inspected by this parser.
+
+        Returns
+        -------
+        int | None
+            10 if the MIME type is supported, otherwise None.
+        """
+        if mime_type in _SUPPORTED_MIME_TYPES:
+            return 10
+        return None
+
+    # ------------------------------------------------------------------
+    # Properties
+    # ------------------------------------------------------------------
+
+    @property
+    def can_produce_archive(self) -> bool:
+        """Whether this parser can produce a searchable PDF archive copy.
+
+        Returns
+        -------
+        bool
+            Always False — the mail parser produces a display PDF
+            (requires_pdf_rendition=True), not an optional OCR archive.
+        """
+        return False
+
+    @property
+    def requires_pdf_rendition(self) -> bool:
+        """Whether the parser must produce a PDF for the frontend to display.
+
+        Returns
+        -------
+        bool
+            Always True — EML files cannot be rendered natively in a browser,
+            so a PDF conversion is always required for display.
+        """
+        return True
+
+    # ------------------------------------------------------------------
+    # Lifecycle
+    # ------------------------------------------------------------------
+
+    def __init__(self, logging_group: object = None) -> None:
+        settings.SCRATCH_DIR.mkdir(parents=True, exist_ok=True)
+        self._tempdir = Path(
+            tempfile.mkdtemp(prefix="paperless-", dir=settings.SCRATCH_DIR),
+        )
+        self._text: str | None = None
+        self._date: datetime.datetime | None = None
+        self._archive_path: Path | None = None
+        self._mailrule_id: int | None = None
+
+    def __enter__(self) -> Self:
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_val: BaseException | None,
+        exc_tb: TracebackType | None,
+    ) -> None:
+        logger.debug("Cleaning up temporary directory %s", self._tempdir)
+        shutil.rmtree(self._tempdir, ignore_errors=True)
+
+    # ------------------------------------------------------------------
+    # Core parsing interface
+    # ------------------------------------------------------------------
+
+    def configure(self, context: ParserContext) -> None:
+        self._mailrule_id = context.mailrule_id
+
+    def parse(
+        self,
+        document_path: Path,
+        mime_type: str,
+        *,
+        produce_archive: bool = True,
+    ) -> None:
+        """Parse the given .eml into formatted text and a PDF archive.
+
+        Call ``configure(ParserContext(mailrule_id=...))`` before this method
+        to apply mail-rule-specific PDF layout options.  The ``produce_archive``
+        flag is accepted for protocol compatibility but is always honoured —
+        the mail parser always produces a PDF since EML files cannot be
+        displayed natively.
+
+        Parameters
+        ----------
+        document_path:
+            Absolute path to the .eml file.
+        mime_type:
+            Detected MIME type of the document (should be "message/rfc822").
+        produce_archive:
+            Accepted for protocol compatibility. The PDF rendition is always
+            produced since EML files cannot be displayed natively in a browser.
+
+        Raises
+        ------
+        documents.parsers.ParseError
+            If the file cannot be parsed or PDF generation fails.
+        """
+
+        def strip_text(text: str) -> 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()
+
+        def build_formatted_text(mail_message: MailMessage) -> str:
+            """Constructs a formatted string based on the given email."""
+            fmt_text = f"Subject: {mail_message.subject}\n\n"
+            fmt_text += f"From: {mail_message.from_values.full if mail_message.from_values else ''}\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 mail_message.bcc_values:
+                fmt_text += (
+                    f"BCC: {', '.join(address.full for address in mail.bcc_values)}\n\n"
+                )
+            if mail_message.attachments:
+                att = []
+                for a in mail.attachments:
+                    attachment_size = naturalsize(a.size, binary=True, format="%.2f")
+                    att.append(
+                        f"{a.filename} ({attachment_size})",
+                    )
+                fmt_text += f"Attachments: {', '.join(att)}\n\n"
+
+            if mail.html:
+                fmt_text += "HTML content: " + strip_text(self.tika_parse(mail.html))
+
+            fmt_text += f"\n\n{strip_text(mail.text)}"
+
+            return fmt_text
+
+        logger.debug("Parsing file %s into an email", document_path.name)
+        mail = self.parse_file_to_message(document_path)
+
+        logger.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
+
+        logger.debug("Creating a PDF from the email")
+        if self._mailrule_id:
+            rule = MailRule.objects.get(pk=self._mailrule_id)
+            self._archive_path = self.generate_pdf(
+                mail,
+                MailRule.PdfLayout(rule.pdf_layout),
+            )
+        else:
+            self._archive_path = self.generate_pdf(mail)
+
+    # ------------------------------------------------------------------
+    # Result accessors
+    # ------------------------------------------------------------------
+
+    def get_text(self) -> str | None:
+        """Return the plain-text content extracted during parse.
+
+        Returns
+        -------
+        str | None
+            Extracted text, or None if parse has not been called yet.
+        """
+        return self._text
+
+    def get_date(self) -> datetime.datetime | None:
+        """Return the document date detected during parse.
+
+        Returns
+        -------
+        datetime.datetime | None
+            Date from the email headers, or None if not detected.
+        """
+        return self._date
+
+    def get_archive_path(self) -> Path | None:
+        """Return the path to the generated archive PDF, or None.
+
+        Returns
+        -------
+        Path | None
+            Path to the PDF produced by Gotenberg, or None if parse has not
+            been called yet.
+        """
+        return self._archive_path
+
+    # ------------------------------------------------------------------
+    # Thumbnail and metadata
+    # ------------------------------------------------------------------
+
+    def get_thumbnail(
+        self,
+        document_path: Path,
+        mime_type: str,
+        file_name: str | None = None,
+    ) -> Path:
+        """Generate a thumbnail from the PDF rendition of the email.
+
+        Converts the document to PDF first if not already done.
+
+        Parameters
+        ----------
+        document_path:
+            Absolute path to the source document.
+        mime_type:
+            Detected MIME type of the document.
+        file_name:
+            Kept for backward compatibility; not used.
+
+        Returns
+        -------
+        Path
+            Path to the generated WebP thumbnail inside the temporary directory.
+        """
+        if not self._archive_path:
+            self._archive_path = self.generate_pdf(
+                self.parse_file_to_message(document_path),
+            )
+
+        return make_thumbnail_from_pdf(
+            self._archive_path,
+            self._tempdir,
+        )
+
+    def get_page_count(
+        self,
+        document_path: Path,
+        mime_type: str,
+    ) -> int | None:
+        """Return the number of pages in the document.
+
+        Counts pages in the archive PDF produced by a preceding parse()
+        call.  Returns ``None`` if parse() has not been called yet or if
+        no archive was produced.
+
+        Returns
+        -------
+        int | None
+            Page count of the archive PDF, or ``None``.
+        """
+        if self._archive_path is not None:
+            from paperless.parsers.utils import get_page_count_for_pdf
+
+            return get_page_count_for_pdf(self._archive_path, log=logger)
+        return None
+
+    def extract_metadata(
+        self,
+        document_path: Path,
+        mime_type: str,
+    ) -> list[MetadataEntry]:
+        """Extract metadata from the email headers.
+
+        Returns email headers as metadata entries with prefix "header",
+        plus summary entries for attachments and date.
+
+        Returns
+        -------
+        list[MetadataEntry]
+            Sorted list of metadata entries, or ``[]`` on parse failure.
+        """
+        result: list[MetadataEntry] = []
+
+        try:
+            mail = self.parse_file_to_message(document_path)
+        except ParseError as e:
+            logger.warning(
+                "Error while fetching document metadata for %s: %s",
+                document_path,
+                e,
+            )
+            return result
+
+        for key, header_values in mail.headers.items():
+            value = ", ".join(header_values)
+            try:
+                value.encode("utf-8")
+            except UnicodeEncodeError as e:  # pragma: no cover
+                logger.debug("Skipping header %s: %s", key, e)
+                continue
+
+            result.append(
+                {
+                    "namespace": "",
+                    "prefix": "header",
+                    "key": key,
+                    "value": value,
+                },
+            )
+
+        result.append(
+            {
+                "namespace": "",
+                "prefix": "",
+                "key": "attachments",
+                "value": ", ".join(
+                    f"{attachment.filename}"
+                    f"({naturalsize(attachment.size, binary=True, format='%.2f')})"
+                    for attachment in mail.attachments
+                ),
+            },
+        )
+
+        result.append(
+            {
+                "namespace": "",
+                "prefix": "",
+                "key": "date",
+                "value": mail.date.strftime("%Y-%m-%d %H:%M:%S %Z"),
+            },
+        )
+
+        result.sort(key=lambda item: (item["prefix"], item["key"]))
+        return result
+
+    # ------------------------------------------------------------------
+    # Email-specific methods
+    # ------------------------------------------------------------------
+
+    def _settings_to_gotenberg_pdfa(self) -> PdfAFormat | None:
+        """Convert the OCR output type setting to a Gotenberg PdfAFormat."""
+        if settings.OCR_OUTPUT_TYPE in {
+            OutputTypeChoices.PDF_A,
+            OutputTypeChoices.PDF_A2,
+        }:
+            return PdfAFormat.A2b
+        elif settings.OCR_OUTPUT_TYPE == OutputTypeChoices.PDF_A1:  # pragma: no cover
+            logger.warning(
+                "Gotenberg does not support PDF/A-1a, choosing PDF/A-2b instead",
+            )
+            return PdfAFormat.A2b
+        elif settings.OCR_OUTPUT_TYPE == OutputTypeChoices.PDF_A3:  # pragma: no cover
+            return PdfAFormat.A3b
+        return None
+
+    @staticmethod
+    def parse_file_to_message(filepath: Path) -> MailMessage:
+        """Parse the given .eml file into a MailMessage object.
+
+        Parameters
+        ----------
+        filepath:
+            Path to the .eml file.
+
+        Returns
+        -------
+        MailMessage
+            Parsed mail message.
+
+        Raises
+        ------
+        documents.parsers.ParseError
+            If the file cannot be parsed or is missing required fields.
+        """
+        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) -> str:
+        """Send HTML content to the Tika server for text extraction.
+
+        Parameters
+        ----------
+        html:
+            HTML string to parse.
+
+        Returns
+        -------
+        str
+            Extracted plain text.
+
+        Raises
+        ------
+        documents.parsers.ParseError
+            If the Tika server cannot be reached or returns an error.
+        """
+        logger.info("Sending content to Tika server")
+
+        try:
+            with TikaClient(tika_url=settings.TIKA_ENDPOINT) as client:
+                parsed = client.tika.as_text.from_buffer(html, "text/html")
+
+                if parsed.content is not None:
+                    return parsed.content.strip()
+                return ""
+        except Exception as err:
+            raise ParseError(
+                f"Could not parse content with tika server at "
+                f"{settings.TIKA_ENDPOINT}: {err}",
+            ) from err
+
+    def generate_pdf(
+        self,
+        mail_message: MailMessage,
+        pdf_layout: MailRule.PdfLayout | None = None,
+    ) -> Path:
+        """Generate a PDF from the email message.
+
+        Creates separate PDFs for the email body and HTML content, then
+        merges them according to the requested layout.
+
+        Parameters
+        ----------
+        mail_message:
+            Parsed email message.
+        pdf_layout:
+            Layout option for the PDF. Falls back to the
+            EMAIL_PARSE_DEFAULT_LAYOUT setting if not provided.
+
+        Returns
+        -------
+        Path
+            Path to the generated PDF inside the temporary directory.
+        """
+        archive_path = Path(self._tempdir) / "merged.pdf"
+
+        mail_pdf_file = self.generate_pdf_from_mail(mail_message)
+
+        if pdf_layout is None:
+            pdf_layout = MailRule.PdfLayout(settings.EMAIL_PARSE_DEFAULT_LAYOUT)
+
+        # 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:
+            pdf_of_html_content = self.generate_pdf_from_html(
+                mail_message.html,
+                mail_message.attachments,
+            )
+
+            logger.debug("Merging email text and HTML content into single PDF")
+
+            with (
+                GotenbergClient(
+                    host=settings.TIKA_GOTENBERG_ENDPOINT,
+                    timeout=settings.CELERY_TASK_TIME_LIMIT,
+                ) as client,
+                client.merge.merge() as route,
+            ):
+                # Configure requested PDF/A formatting, if any
+                pdf_a_format = self._settings_to_gotenberg_pdfa()
+                if pdf_a_format is not None:
+                    route.pdf_format(pdf_a_format)
+
+                match pdf_layout:
+                    case MailRule.PdfLayout.HTML_TEXT:
+                        route.merge([pdf_of_html_content, mail_pdf_file])
+                    case MailRule.PdfLayout.HTML_ONLY:
+                        route.merge([pdf_of_html_content])
+                    case MailRule.PdfLayout.TEXT_ONLY:
+                        route.merge([mail_pdf_file])
+                    case MailRule.PdfLayout.TEXT_HTML | _:
+                        route.merge([mail_pdf_file, pdf_of_html_content])
+
+                try:
+                    response = route.run()
+                    archive_path.write_bytes(response.content)
+                except Exception as err:
+                    raise ParseError(
+                        f"Error while merging email HTML into PDF: {err}",
+                    ) from err
+
+        return archive_path
+
+    def mail_to_html(self, mail: MailMessage) -> Path:
+        """Convert the given email into an HTML file using a template.
+
+        Parameters
+        ----------
+        mail:
+            Parsed mail message.
+
+        Returns
+        -------
+        Path
+            Path to the rendered HTML file inside the temporary directory.
+        """
+
+        def clean_html(text: str) -> str:
+            """Attempt to clean, escape, and linkify the given HTML string."""
+            if isinstance(text, list):
+                text = "\n".join([str(e) for e in text])
+            if not isinstance(text, str):
+                text = str(text)
+            text = escape(text)
+            text = clean(text)
+            text = linkify(text, parse_email=True)
+            text = text.replace("\n", "<br>")
+            return text
+
+        data = {}
+
+        data["subject"] = clean_html(mail.subject)
+        if data["subject"]:
+            data["subject_label"] = "Subject"
+        data["from"] = clean_html(mail.from_values.full if mail.from_values else "")
+        if data["from"]:
+            data["from_label"] = "From"
+        data["to"] = clean_html(", ".join(address.full for address in mail.to_values))
+        if data["to"]:
+            data["to_label"] = "To"
+        data["cc"] = clean_html(", ".join(address.full for address in mail.cc_values))
+        if data["cc"]:
+            data["cc_label"] = "CC"
+        data["bcc"] = clean_html(", ".join(address.full for address in mail.bcc_values))
+        if data["bcc"]:
+            data["bcc_label"] = "BCC"
+
+        att = []
+        for a in mail.attachments:
+            att.append(
+                f"{a.filename} ({naturalsize(a.size, binary=True, format='%.2f')})",
+            )
+        data["attachments"] = clean_html(", ".join(att))
+        if data["attachments"]:
+            data["attachments_label"] = "Attachments"
+
+        data["date"] = clean_html(
+            timezone.localtime(mail.date).strftime("%Y-%m-%d %H:%M"),
+        )
+        data["content"] = clean_html(mail.text.strip())
+
+        from django.template.loader import render_to_string
+
+        html_file = Path(self._tempdir) / "email_as_html.html"
+        html_file.write_text(render_to_string("email_msg_template.html", context=data))
+
+        return html_file
+
+    def generate_pdf_from_mail(self, mail: MailMessage) -> Path:
+        """Create a PDF from the email body using an HTML template and Gotenberg.
+
+        Parameters
+        ----------
+        mail:
+            Parsed mail message.
+
+        Returns
+        -------
+        Path
+            Path to the generated PDF inside the temporary directory.
+
+        Raises
+        ------
+        documents.parsers.ParseError
+            If Gotenberg returns an error.
+        """
+        logger.info("Converting mail to PDF")
+
+        css_file = (
+            Path(__file__).parent.parent.parent
+            / "paperless_mail"
+            / "templates"
+            / "output.css"
+        )
+        email_html_file = self.mail_to_html(mail)
+
+        with (
+            GotenbergClient(
+                host=settings.TIKA_GOTENBERG_ENDPOINT,
+                timeout=settings.CELERY_TASK_TIME_LIMIT,
+            ) as client,
+            client.chromium.html_to_pdf() as route,
+        ):
+            # Configure requested PDF/A formatting, if any
+            pdf_a_format = self._settings_to_gotenberg_pdfa()
+            if pdf_a_format is not None:
+                route.pdf_format(pdf_a_format)
+
+            try:
+                response = (
+                    route.index(email_html_file)
+                    .resource(css_file)
+                    .margins(
+                        PageMarginsType(
+                            top=Measurement(0.1, MeasurementUnitType.Inches),
+                            bottom=Measurement(0.1, MeasurementUnitType.Inches),
+                            left=Measurement(0.1, MeasurementUnitType.Inches),
+                            right=Measurement(0.1, MeasurementUnitType.Inches),
+                        ),
+                    )
+                    .size(A4)
+                    .scale(1.0)
+                    .run()
+                )
+            except Exception as err:
+                raise ParseError(
+                    f"Error while converting email to PDF: {err}",
+                ) from err
+
+        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:
+        """Generate a PDF from the HTML content of the email.
+
+        Parameters
+        ----------
+        orig_html:
+            Raw HTML string from the email body.
+        attachments:
+            List of email attachments (used as inline resources).
+
+        Returns
+        -------
+        Path
+            Path to the generated PDF inside the temporary directory.
+
+        Raises
+        ------
+        documents.parsers.ParseError
+            If Gotenberg returns an error.
+        """
+
+        def clean_html_script(text: str) -> str:
+            compiled_open = re.compile(re.escape("<script"), re.IGNORECASE)
+            text = compiled_open.sub("<div hidden ", text)
+
+            compiled_close = re.compile(re.escape("</script"), re.IGNORECASE)
+            text = compiled_close.sub("</div", text)
+            return text
+
+        logger.info("Converting message html to PDF")
+
+        tempdir = Path(self._tempdir)
+
+        html_clean = clean_html_script(orig_html)
+        html_clean_file = tempdir / "index.html"
+        html_clean_file.write_text(html_clean)
+
+        with (
+            GotenbergClient(
+                host=settings.TIKA_GOTENBERG_ENDPOINT,
+                timeout=settings.CELERY_TASK_TIME_LIMIT,
+            ) as client,
+            client.chromium.html_to_pdf() as route,
+        ):
+            # Configure requested PDF/A formatting, if any
+            pdf_a_format = self._settings_to_gotenberg_pdfa()
+            if pdf_a_format is not None:
+                route.pdf_format(pdf_a_format)
+
+            # Add attachments as resources, cleaning the filename and replacing
+            # it in the index file for inclusion
+            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)
+
+                route.resource(temp_file)
+
+                # 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)
+            # This is our index file, the main page basically
+            route.index(html_clean_file)
+
+            # Set page size, margins
+            route.margins(
+                PageMarginsType(
+                    top=Measurement(0.1, MeasurementUnitType.Inches),
+                    bottom=Measurement(0.1, MeasurementUnitType.Inches),
+                    left=Measurement(0.1, MeasurementUnitType.Inches),
+                    right=Measurement(0.1, MeasurementUnitType.Inches),
+                ),
+            ).size(A4).scale(1.0)
+
+            try:
+                response = route.run()
+
+            except Exception as err:
+                raise ParseError(
+                    f"Error while converting document to PDF: {err}",
+                ) from err
+
+        html_pdf = tempdir / "html.pdf"
+        html_pdf.write_bytes(response.content)
+        return html_pdf

src/paperless/parsers/registry.py

@@ -193,13 +193,17 @@ class ParserRegistry:
         that log output is predictable; scoring determines which parser wins
         at runtime regardless of registration order.
         """
+        from paperless.parsers.mail import MailDocumentParser
         from paperless.parsers.remote import RemoteDocumentParser
+        from paperless.parsers.tesseract import RasterisedDocumentParser
         from paperless.parsers.text import TextDocumentParser
         from paperless.parsers.tika import TikaDocumentParser
 
         self.register_builtin(TextDocumentParser)
         self.register_builtin(RemoteDocumentParser)
         self.register_builtin(TikaDocumentParser)
+        self.register_builtin(MailDocumentParser)
+        self.register_builtin(RasterisedDocumentParser)
 
     # ------------------------------------------------------------------
     # Discovery

src/paperless/parsers/remote.py

@@ -28,6 +28,7 @@ if TYPE_CHECKING:
     from types import TracebackType
 
     from paperless.parsers import MetadataEntry
+    from paperless.parsers import ParserContext
 
 logger = logging.getLogger("paperless.parsing.remote")
 
@@ -204,6 +205,9 @@ class RemoteDocumentParser:
     # Core parsing interface
     # ------------------------------------------------------------------
 
+    def configure(self, context: ParserContext) -> None:
+        pass
+
     def parse(
         self,
         document_path: Path,

src/paperless/parsers/tesseract.py

@@ -1,13 +1,18 @@
+from __future__ import annotations
+
+import logging
 import os
 import re
+import shutil
 import tempfile
 from pathlib import Path
 from typing import TYPE_CHECKING
+from typing import Any
+from typing import Self
 
 from django.conf import settings
 from PIL import Image
 
-from documents.parsers import DocumentParser
 from documents.parsers import ParseError
 from documents.parsers import make_thumbnail_from_pdf
 from documents.utils import maybe_override_pixel_limit
@@ -16,6 +21,28 @@ from paperless.config import OcrConfig
 from paperless.models import ArchiveFileChoices
 from paperless.models import CleanChoices
 from paperless.models import ModeChoices
+from paperless.parsers.utils import read_file_handle_unicode_errors
+from paperless.version import __full_version_str__
+
+if TYPE_CHECKING:
+    import datetime
+    from types import TracebackType
+
+    from paperless.parsers import MetadataEntry
+    from paperless.parsers import ParserContext
+
+logger = logging.getLogger("paperless.parsing.tesseract")
+
+_SUPPORTED_MIME_TYPES: dict[str, str] = {
+    "application/pdf": ".pdf",
+    "image/jpeg": ".jpg",
+    "image/png": ".png",
+    "image/tiff": ".tif",
+    "image/gif": ".gif",
+    "image/bmp": ".bmp",
+    "image/webp": ".webp",
+    "image/heic": ".heic",
+}
 
 
 class NoTextFoundException(Exception):
@@ -26,81 +53,125 @@ class RtlLanguageException(Exception):
     pass
 
 
-class RasterisedDocumentParser(DocumentParser):
+class RasterisedDocumentParser:
     """
     This parser uses Tesseract to try and get some text out of a rasterised
     image, whether it's a PDF, or other graphical format (JPEG, TIFF, etc.)
     """
 
-    logging_name = "paperless.parsing.tesseract"
+    name: str = "Paperless-ngx Tesseract OCR Parser"
+    version: str = __full_version_str__
+    author: str = "Paperless-ngx Contributors"
+    url: str = "https://github.com/paperless-ngx/paperless-ngx"
 
-    def get_settings(self) -> OcrConfig:
-        """
-        This parser uses the OCR configuration settings to parse documents
-        """
-        return OcrConfig()
+    # ------------------------------------------------------------------
+    # Class methods
+    # ------------------------------------------------------------------
 
-    def get_page_count(self, document_path, mime_type):
-        page_count = None
-        if mime_type == "application/pdf":
-            try:
-                import pikepdf
+    @classmethod
+    def supported_mime_types(cls) -> dict[str, str]:
+        return _SUPPORTED_MIME_TYPES
 
-                with pikepdf.Pdf.open(document_path) as pdf:
-                    page_count = len(pdf.pages)
-            except Exception as e:
-                self.log.warning(
-                    f"Unable to determine PDF page count {document_path}: {e}",
-                )
-        return page_count
+    @classmethod
+    def score(
+        cls,
+        mime_type: str,
+        filename: str,
+        path: Path | None = None,
+    ) -> int | None:
+        if mime_type in _SUPPORTED_MIME_TYPES:
+            return 10
+        return None
 
-    def extract_metadata(self, document_path, mime_type):
-        result = []
-        if mime_type == "application/pdf":
-            import pikepdf
+    # ------------------------------------------------------------------
+    # Properties
+    # ------------------------------------------------------------------
 
-            namespace_pattern = re.compile(r"\{(.*)\}(.*)")
+    @property
+    def can_produce_archive(self) -> bool:
+        return True
 
-            pdf = pikepdf.open(document_path)
-            meta = pdf.open_metadata()
-            for key, value in meta.items():
-                if isinstance(value, list):
-                    value = " ".join([str(e) for e in value])
-                value = str(value)
-                try:
-                    m = namespace_pattern.match(key)
-                    if m is None:  # pragma: no cover
-                        continue
-                    namespace = m.group(1)
-                    key_value = m.group(2)
-                    try:
-                        namespace.encode("utf-8")
-                        key_value.encode("utf-8")
-                    except UnicodeEncodeError as e:  # pragma: no cover
-                        self.log.debug(f"Skipping metadata key {key}: {e}")
-                        continue
-                    result.append(
-                        {
-                            "namespace": namespace,
-                            "prefix": meta.REVERSE_NS[namespace],
-                            "key": key_value,
-                            "value": value,
-                        },
-                    )
-                except Exception as e:
-                    self.log.warning(
-                        f"Error while reading metadata {key}: {value}. Error: {e}",
-                    )
-        return result
+    @property
+    def requires_pdf_rendition(self) -> bool:
+        return False
 
-    def get_thumbnail(self, document_path, mime_type, file_name=None):
+    # ------------------------------------------------------------------
+    # Lifecycle
+    # ------------------------------------------------------------------
+
+    def __init__(self, logging_group: object = None) -> None:
+        settings.SCRATCH_DIR.mkdir(parents=True, exist_ok=True)
+        self.tempdir = Path(
+            tempfile.mkdtemp(prefix="paperless-", dir=settings.SCRATCH_DIR),
+        )
+        self.settings = OcrConfig()
+        self.archive_path: Path | None = None
+        self.text: str | None = None
+        self.date: datetime.datetime | None = None
+        self.log = logger
+
+    def __enter__(self) -> Self:
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_val: BaseException | None,
+        exc_tb: TracebackType | None,
+    ) -> None:
+        logger.debug("Cleaning up temporary directory %s", self.tempdir)
+        shutil.rmtree(self.tempdir, ignore_errors=True)
+
+    # ------------------------------------------------------------------
+    # Core parsing interface
+    # ------------------------------------------------------------------
+
+    def configure(self, context: ParserContext) -> None:
+        pass
+
+    # ------------------------------------------------------------------
+    # Result accessors
+    # ------------------------------------------------------------------
+
+    def get_text(self) -> str | None:
+        return self.text
+
+    def get_date(self) -> datetime.datetime | None:
+        return self.date
+
+    def get_archive_path(self) -> Path | None:
+        return self.archive_path
+
+    # ------------------------------------------------------------------
+    # Thumbnail, page count, and metadata
+    # ------------------------------------------------------------------
+
+    def get_thumbnail(self, document_path: Path, mime_type: str) -> Path:
         return make_thumbnail_from_pdf(
-            self.archive_path or document_path,
+            self.archive_path or Path(document_path),
             self.tempdir,
-            self.logging_group,
         )
 
-    def is_image(self, mime_type) -> bool:
+    def get_page_count(self, document_path: Path, mime_type: str) -> int | None:
+        if mime_type == "application/pdf":
+            from paperless.parsers.utils import get_page_count_for_pdf
+
+            return get_page_count_for_pdf(Path(document_path), log=self.log)
+        return None
+
+    def extract_metadata(
+        self,
+        document_path: Path,
+        mime_type: str,
+    ) -> list[MetadataEntry]:
+        if mime_type != "application/pdf":
+            return []
+
+        from paperless.parsers.utils import extract_pdf_metadata
+
+        return extract_pdf_metadata(Path(document_path), log=self.log)
+
+    def is_image(self, mime_type: str) -> bool:
         return mime_type in [
             "image/png",
             "image/jpeg",
@@ -111,25 +182,25 @@ class RasterisedDocumentParser(DocumentParser):
             "image/heic",
         ]
 
-    def has_alpha(self, image) -> bool:
+    def has_alpha(self, image: Path) -> bool:
         with Image.open(image) as im:
             return im.mode in ("RGBA", "LA")
 
-    def remove_alpha(self, image_path: str) -> Path:
+    def remove_alpha(self, image_path: Path) -> Path:
         no_alpha_image = Path(self.tempdir) / "image-no-alpha"
         run_subprocess(
             [
                 settings.CONVERT_BINARY,
                 "-alpha",
                 "off",
-                image_path,
-                no_alpha_image,
+                str(image_path),
+                str(no_alpha_image),
             ],
             logger=self.log,
         )
         return no_alpha_image
 
-    def get_dpi(self, image) -> int | None:
+    def get_dpi(self, image: Path) -> int | None:
         try:
             with Image.open(image) as im:
                 x, _ = im.info["dpi"]
@@ -138,7 +209,7 @@ class RasterisedDocumentParser(DocumentParser):
             self.log.warning(f"Error while getting DPI from image {image}: {e}")
             return None
 
-    def calculate_a4_dpi(self, image) -> int | None:
+    def calculate_a4_dpi(self, image: Path) -> int | None:
         try:
             with Image.open(image) as im:
                 width, _ = im.size
@@ -156,6 +227,7 @@ class RasterisedDocumentParser(DocumentParser):
         sidecar_file: Path | None,
         pdf_file: Path,
     ) -> str | None:
+        text: str | None = None
         # When re-doing OCR, the sidecar contains ONLY the new text, not
         # the whole text, so do not utilize it in that case
         if (
@@ -163,7 +235,7 @@ class RasterisedDocumentParser(DocumentParser):
             and sidecar_file.is_file()
             and self.settings.mode != "redo"
         ):
-            text = self.read_file_handle_unicode_errors(sidecar_file)
+            text = read_file_handle_unicode_errors(sidecar_file)
 
             if "[OCR skipped on page" not in text:
                 # This happens when there's already text in the input file.
@@ -191,12 +263,12 @@ class RasterisedDocumentParser(DocumentParser):
                         "-layout",
                         "-enc",
                         "UTF-8",
-                        pdf_file,
+                        str(pdf_file),
                         tmp.name,
                     ],
                     logger=self.log,
                 )
-                text = self.read_file_handle_unicode_errors(Path(tmp.name))
+                text = read_file_handle_unicode_errors(Path(tmp.name))
 
             return post_process_text(text)
 
@@ -211,16 +283,14 @@ class RasterisedDocumentParser(DocumentParser):
 
     def construct_ocrmypdf_parameters(
         self,
-        input_file,
-        mime_type,
-        output_file,
-        sidecar_file,
+        input_file: Path,
+        mime_type: str,
+        output_file: Path,
+        sidecar_file: Path,
         *,
-        safe_fallback=False,
-    ):
-        if TYPE_CHECKING:
-            assert isinstance(self.settings, OcrConfig)
-        ocrmypdf_args = {
+        safe_fallback: bool = False,
+    ) -> dict[str, Any]:
+        ocrmypdf_args: dict[str, Any] = {
             "input_file_or_options": input_file,
             "output_file": output_file,
             # need to use threads, since this will be run in daemonized
@@ -330,7 +400,13 @@ class RasterisedDocumentParser(DocumentParser):
 
         return ocrmypdf_args
 
-    def parse(self, document_path: Path, mime_type, file_name=None) -> None:
+    def parse(
+        self,
+        document_path: Path,
+        mime_type: str,
+        *,
+        produce_archive: bool = True,
+    ) -> None:
         # This forces tesseract to use one core per page.
         os.environ["OMP_THREAD_LIMIT"] = "1"
         VALID_TEXT_LENGTH = 50
@@ -458,7 +534,7 @@ class RasterisedDocumentParser(DocumentParser):
                 self.text = ""
 
 
-def post_process_text(text):
+def post_process_text(text: str | None) -> str | None:
     if not text:
         return None
 

src/paperless/parsers/text.py

@@ -27,6 +27,7 @@ if TYPE_CHECKING:
     from types import TracebackType
 
     from paperless.parsers import MetadataEntry
+    from paperless.parsers import ParserContext
 
 logger = logging.getLogger("paperless.parsing.text")
 
@@ -156,6 +157,9 @@ class TextDocumentParser:
     # Core parsing interface
     # ------------------------------------------------------------------
 
+    def configure(self, context: ParserContext) -> None:
+        pass
+
     def parse(
         self,
         document_path: Path,

src/paperless/parsers/tika.py

@@ -35,6 +35,7 @@ if TYPE_CHECKING:
     from types import TracebackType
 
     from paperless.parsers import MetadataEntry
+    from paperless.parsers import ParserContext
 
 logger = logging.getLogger("paperless.parsing.tika")
 
@@ -205,6 +206,9 @@ class TikaDocumentParser:
     # Core parsing interface
     # ------------------------------------------------------------------
 
+    def configure(self, context: ParserContext) -> None:
+        pass
+
     def parse(
         self,
         document_path: Path,
@@ -340,11 +344,19 @@ class TikaDocumentParser:
     ) -> int | None:
         """Return the number of pages in the document.
 
+        Counts pages in the archive PDF produced by a preceding parse()
+        call.  Returns ``None`` if parse() has not been called yet or if
+        no archive was produced.
+
         Returns
         -------
         int | None
-            Always None — page count is not available from Tika.
+            Page count of the archive PDF, or ``None``.
         """
+        if self._archive_path is not None:
+            from paperless.parsers.utils import get_page_count_for_pdf
+
+            return get_page_count_for_pdf(self._archive_path, log=logger)
         return None
 
     def extract_metadata(

src/paperless/parsers/utils.py

@@ -20,6 +20,34 @@ if TYPE_CHECKING:
 logger = logging.getLogger("paperless.parsers.utils")
 
 
+def read_file_handle_unicode_errors(
+    filepath: Path,
+    log: logging.Logger | None = None,
+) -> str:
+    """Read a file as UTF-8 text, replacing invalid bytes rather than raising.
+
+    Parameters
+    ----------
+    filepath:
+        Absolute path to the file to read.
+    log:
+        Logger to use for warnings.  Falls back to the module-level logger
+        when omitted.
+
+    Returns
+    -------
+    str
+        File content as a string, with any invalid UTF-8 sequences replaced
+        by the Unicode replacement character.
+    """
+    _log = log or logger
+    try:
+        return filepath.read_text(encoding="utf-8")
+    except UnicodeDecodeError as e:
+        _log.warning("Unicode error during text reading, continuing: %s", e)
+        return filepath.read_bytes().decode("utf-8", errors="replace")
+
+
 def get_page_count_for_pdf(
     document_path: Path,
     log: logging.Logger | None = None,

src/paperless/tests/parsers/conftest.py

@@ -6,19 +6,29 @@ so it is easy to see which files belong to which test module.
 
 from __future__ import annotations
 
+from contextlib import contextmanager
 from typing import TYPE_CHECKING
 
 import pytest
+from django.test import override_settings
 
+from paperless.parsers.mail import MailDocumentParser
 from paperless.parsers.remote import RemoteDocumentParser
+from paperless.parsers.tesseract import RasterisedDocumentParser
 from paperless.parsers.text import TextDocumentParser
 from paperless.parsers.tika import TikaDocumentParser
 
 if TYPE_CHECKING:
+    from collections.abc import Callable
     from collections.abc import Generator
     from pathlib import Path
+    from unittest.mock import MagicMock
 
     from pytest_django.fixtures import SettingsWrapper
+    from pytest_mock import MockerFixture
+
+    #: Type for the ``make_tesseract_parser`` fixture factory.
+    MakeTesseractParser = Callable[..., Generator[RasterisedDocumentParser, None, None]]
 
 
 # ------------------------------------------------------------------
@@ -247,3 +257,544 @@ def tika_parser() -> Generator[TikaDocumentParser, None, None]:
     """
     with TikaDocumentParser() as parser:
         yield parser
+
+
+# ------------------------------------------------------------------
+# Mail parser sample files
+# ------------------------------------------------------------------
+
+
+@pytest.fixture(scope="session")
+def mail_samples_dir(samples_dir: Path) -> Path:
+    """Absolute path to the mail parser sample files directory.
+
+    Returns
+    -------
+    Path
+        ``<samples_dir>/mail/``
+    """
+    return samples_dir / "mail"
+
+
+@pytest.fixture(scope="session")
+def broken_email_file(mail_samples_dir: Path) -> Path:
+    """Path to a broken/malformed EML sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``mail/broken.eml``.
+    """
+    return mail_samples_dir / "broken.eml"
+
+
+@pytest.fixture(scope="session")
+def simple_txt_email_file(mail_samples_dir: Path) -> Path:
+    """Path to a plain-text email sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``mail/simple_text.eml``.
+    """
+    return mail_samples_dir / "simple_text.eml"
+
+
+@pytest.fixture(scope="session")
+def simple_txt_email_pdf_file(mail_samples_dir: Path) -> Path:
+    """Path to the expected PDF rendition of the plain-text email.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``mail/simple_text.eml.pdf``.
+    """
+    return mail_samples_dir / "simple_text.eml.pdf"
+
+
+@pytest.fixture(scope="session")
+def simple_txt_email_thumbnail_file(mail_samples_dir: Path) -> Path:
+    """Path to the expected thumbnail for the plain-text email.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``mail/simple_text.eml.pdf.webp``.
+    """
+    return mail_samples_dir / "simple_text.eml.pdf.webp"
+
+
+@pytest.fixture(scope="session")
+def html_email_file(mail_samples_dir: Path) -> Path:
+    """Path to an HTML email sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``mail/html.eml``.
+    """
+    return mail_samples_dir / "html.eml"
+
+
+@pytest.fixture(scope="session")
+def html_email_pdf_file(mail_samples_dir: Path) -> Path:
+    """Path to the expected PDF rendition of the HTML email.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``mail/html.eml.pdf``.
+    """
+    return mail_samples_dir / "html.eml.pdf"
+
+
+@pytest.fixture(scope="session")
+def html_email_thumbnail_file(mail_samples_dir: Path) -> Path:
+    """Path to the expected thumbnail for the HTML email.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``mail/html.eml.pdf.webp``.
+    """
+    return mail_samples_dir / "html.eml.pdf.webp"
+
+
+@pytest.fixture(scope="session")
+def html_email_html_file(mail_samples_dir: Path) -> Path:
+    """Path to the HTML body of the HTML email sample.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``mail/html.eml.html``.
+    """
+    return mail_samples_dir / "html.eml.html"
+
+
+@pytest.fixture(scope="session")
+def merged_pdf_first(mail_samples_dir: Path) -> Path:
+    """Path to the first PDF used in PDF-merge tests.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``mail/first.pdf``.
+    """
+    return mail_samples_dir / "first.pdf"
+
+
+@pytest.fixture(scope="session")
+def merged_pdf_second(mail_samples_dir: Path) -> Path:
+    """Path to the second PDF used in PDF-merge tests.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``mail/second.pdf``.
+    """
+    return mail_samples_dir / "second.pdf"
+
+
+# ------------------------------------------------------------------
+# Mail parser instance
+# ------------------------------------------------------------------
+
+
+@pytest.fixture()
+def mail_parser() -> Generator[MailDocumentParser, None, None]:
+    """Yield a MailDocumentParser and clean up its temporary directory afterwards.
+
+    Yields
+    ------
+    MailDocumentParser
+        A ready-to-use parser instance.
+    """
+    with MailDocumentParser() as parser:
+        yield parser
+
+
+@pytest.fixture(scope="session")
+def nginx_base_url() -> Generator[str, None, None]:
+    """
+    The base URL for the nginx HTTP server we expect to be alive
+    """
+    yield "http://localhost:8080"
+
+
+# ------------------------------------------------------------------
+# Tesseract parser sample files
+# ------------------------------------------------------------------
+
+
+@pytest.fixture(scope="session")
+def tesseract_samples_dir(samples_dir: Path) -> Path:
+    """Absolute path to the tesseract parser sample files directory.
+
+    Returns
+    -------
+    Path
+        ``<samples_dir>/tesseract/``
+    """
+    return samples_dir / "tesseract"
+
+
+@pytest.fixture(scope="session")
+def document_webp_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a WebP document sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/document.webp``.
+    """
+    return tesseract_samples_dir / "document.webp"
+
+
+@pytest.fixture(scope="session")
+def encrypted_pdf_file(tesseract_samples_dir: Path) -> Path:
+    """Path to an encrypted PDF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/encrypted.pdf``.
+    """
+    return tesseract_samples_dir / "encrypted.pdf"
+
+
+@pytest.fixture(scope="session")
+def multi_page_digital_pdf_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a multi-page digital PDF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/multi-page-digital.pdf``.
+    """
+    return tesseract_samples_dir / "multi-page-digital.pdf"
+
+
+@pytest.fixture(scope="session")
+def multi_page_images_alpha_rgb_tiff_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a multi-page TIFF with alpha channel in RGB.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/multi-page-images-alpha-rgb.tiff``.
+    """
+    return tesseract_samples_dir / "multi-page-images-alpha-rgb.tiff"
+
+
+@pytest.fixture(scope="session")
+def multi_page_images_alpha_tiff_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a multi-page TIFF with alpha channel.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/multi-page-images-alpha.tiff``.
+    """
+    return tesseract_samples_dir / "multi-page-images-alpha.tiff"
+
+
+@pytest.fixture(scope="session")
+def multi_page_images_pdf_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a multi-page PDF with images.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/multi-page-images.pdf``.
+    """
+    return tesseract_samples_dir / "multi-page-images.pdf"
+
+
+@pytest.fixture(scope="session")
+def multi_page_images_tiff_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a multi-page TIFF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/multi-page-images.tiff``.
+    """
+    return tesseract_samples_dir / "multi-page-images.tiff"
+
+
+@pytest.fixture(scope="session")
+def multi_page_mixed_pdf_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a multi-page mixed PDF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/multi-page-mixed.pdf``.
+    """
+    return tesseract_samples_dir / "multi-page-mixed.pdf"
+
+
+@pytest.fixture(scope="session")
+def no_text_alpha_png_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a PNG with alpha channel and no text.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/no-text-alpha.png``.
+    """
+    return tesseract_samples_dir / "no-text-alpha.png"
+
+
+@pytest.fixture(scope="session")
+def rotated_pdf_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a rotated PDF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/rotated.pdf``.
+    """
+    return tesseract_samples_dir / "rotated.pdf"
+
+
+@pytest.fixture(scope="session")
+def rtl_test_pdf_file(tesseract_samples_dir: Path) -> Path:
+    """Path to an RTL test PDF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/rtl-test.pdf``.
+    """
+    return tesseract_samples_dir / "rtl-test.pdf"
+
+
+@pytest.fixture(scope="session")
+def signed_pdf_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a signed PDF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/signed.pdf``.
+    """
+    return tesseract_samples_dir / "signed.pdf"
+
+
+@pytest.fixture(scope="session")
+def simple_alpha_png_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a simple PNG with alpha channel.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/simple-alpha.png``.
+    """
+    return tesseract_samples_dir / "simple-alpha.png"
+
+
+@pytest.fixture(scope="session")
+def simple_digital_pdf_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a simple digital PDF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/simple-digital.pdf``.
+    """
+    return tesseract_samples_dir / "simple-digital.pdf"
+
+
+@pytest.fixture(scope="session")
+def simple_no_dpi_png_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a simple PNG without DPI information.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/simple-no-dpi.png``.
+    """
+    return tesseract_samples_dir / "simple-no-dpi.png"
+
+
+@pytest.fixture(scope="session")
+def simple_bmp_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a simple BMP sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/simple.bmp``.
+    """
+    return tesseract_samples_dir / "simple.bmp"
+
+
+@pytest.fixture(scope="session")
+def simple_gif_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a simple GIF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/simple.gif``.
+    """
+    return tesseract_samples_dir / "simple.gif"
+
+
+@pytest.fixture(scope="session")
+def simple_heic_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a simple HEIC sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/simple.heic``.
+    """
+    return tesseract_samples_dir / "simple.heic"
+
+
+@pytest.fixture(scope="session")
+def simple_jpg_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a simple JPG sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/simple.jpg``.
+    """
+    return tesseract_samples_dir / "simple.jpg"
+
+
+@pytest.fixture(scope="session")
+def simple_png_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a simple PNG sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/simple.png``.
+    """
+    return tesseract_samples_dir / "simple.png"
+
+
+@pytest.fixture(scope="session")
+def simple_tif_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a simple TIF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/simple.tif``.
+    """
+    return tesseract_samples_dir / "simple.tif"
+
+
+@pytest.fixture(scope="session")
+def single_page_mixed_pdf_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a single-page mixed PDF sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/single-page-mixed.pdf``.
+    """
+    return tesseract_samples_dir / "single-page-mixed.pdf"
+
+
+@pytest.fixture(scope="session")
+def with_form_pdf_file(tesseract_samples_dir: Path) -> Path:
+    """Path to a PDF with form sample file.
+
+    Returns
+    -------
+    Path
+        Absolute path to ``tesseract/with-form.pdf``.
+    """
+    return tesseract_samples_dir / "with-form.pdf"
+
+
+# ------------------------------------------------------------------
+# Tesseract parser instance and settings helpers
+# ------------------------------------------------------------------
+
+
+@pytest.fixture()
+def null_app_config(mocker: MockerFixture) -> MagicMock:
+    """Return a MagicMock with all OcrConfig fields set to None.
+
+    This allows the parser to fall back to Django settings instead of
+    hitting the database.
+
+    Returns
+    -------
+    MagicMock
+        Mock config with all fields as None
+    """
+    return mocker.MagicMock(
+        output_type=None,
+        pages=None,
+        language=None,
+        mode=None,
+        skip_archive_file=None,
+        image_dpi=None,
+        unpaper_clean=None,
+        deskew=None,
+        rotate_pages=None,
+        rotate_pages_threshold=None,
+        max_image_pixels=None,
+        color_conversion_strategy=None,
+        user_args=None,
+    )
+
+
+@pytest.fixture()
+def tesseract_parser(
+    mocker: MockerFixture,
+    null_app_config: MagicMock,
+) -> Generator[RasterisedDocumentParser, None, None]:
+    """Yield a RasterisedDocumentParser and clean up its temporary directory afterwards.
+
+    Patches the config system to avoid database access.
+
+    Yields
+    ------
+    RasterisedDocumentParser
+        A ready-to-use parser instance.
+    """
+    mocker.patch(
+        "paperless.config.BaseConfig._get_config_instance",
+        return_value=null_app_config,
+    )
+    with RasterisedDocumentParser() as parser:
+        yield parser
+
+
+@pytest.fixture()
+def make_tesseract_parser(
+    mocker: MockerFixture,
+    null_app_config: MagicMock,
+) -> MakeTesseractParser:
+    """Return a factory for creating RasterisedDocumentParser with Django settings overrides.
+
+    This fixture is useful for tests that need to create parsers with different
+    settings configurations.
+
+    Returns
+    -------
+    Callable[..., contextmanager[RasterisedDocumentParser]]
+        A context manager factory that accepts Django settings overrides
+    """
+    mocker.patch(
+        "paperless.config.BaseConfig._get_config_instance",
+        return_value=null_app_config,
+    )
+
+    @contextmanager
+    def _make_parser(**django_settings_overrides):
+        with override_settings(**django_settings_overrides):
+            with RasterisedDocumentParser() as parser:
+                yield parser
+
+    return _make_parser

src/paperless/tests/parsers/test_mail_parser.py

@@ -12,7 +12,64 @@ from pytest_httpx import HTTPXMock
 from pytest_mock import MockerFixture
 
 from documents.parsers import ParseError
-from paperless_mail.parsers import MailDocumentParser
+from paperless.parsers import ParserContext
+from paperless.parsers import ParserProtocol
+from paperless.parsers.mail import MailDocumentParser
+
+
+class TestMailParserProtocol:
+    """Verify that MailDocumentParser satisfies the ParserProtocol contract."""
+
+    def test_isinstance_satisfies_protocol(
+        self,
+        mail_parser: MailDocumentParser,
+    ) -> None:
+        assert isinstance(mail_parser, ParserProtocol)
+
+    def test_supported_mime_types(self) -> None:
+        mime_types = MailDocumentParser.supported_mime_types()
+        assert isinstance(mime_types, dict)
+        assert "message/rfc822" in mime_types
+
+    @pytest.mark.parametrize(
+        ("mime_type", "expected"),
+        [
+            ("message/rfc822", 10),
+            ("application/pdf", None),
+            ("text/plain", None),
+        ],
+    )
+    def test_score(self, mime_type: str, expected: int | None) -> None:
+        assert MailDocumentParser.score(mime_type, "email.eml") == expected
+
+    def test_can_produce_archive_is_false(
+        self,
+        mail_parser: MailDocumentParser,
+    ) -> None:
+        assert mail_parser.can_produce_archive is False
+
+    def test_requires_pdf_rendition_is_true(
+        self,
+        mail_parser: MailDocumentParser,
+    ) -> None:
+        assert mail_parser.requires_pdf_rendition is True
+
+    def test_get_page_count_returns_none_without_archive(
+        self,
+        mail_parser: MailDocumentParser,
+        html_email_file: Path,
+    ) -> None:
+        assert mail_parser.get_page_count(html_email_file, "message/rfc822") is None
+
+    def test_get_page_count_returns_int_with_pdf_archive(
+        self,
+        mail_parser: MailDocumentParser,
+        simple_txt_email_pdf_file: Path,
+    ) -> None:
+        mail_parser._archive_path = simple_txt_email_pdf_file
+        count = mail_parser.get_page_count(simple_txt_email_pdf_file, "message/rfc822")
+        assert isinstance(count, int)
+        assert count > 0
 
 
 class TestEmailFileParsing:
@@ -24,7 +81,7 @@ class TestEmailFileParsing:
     def test_parse_error_missing_file(
         self,
         mail_parser: MailDocumentParser,
-        sample_dir: Path,
+        mail_samples_dir: Path,
     ) -> None:
         """
         GIVEN:
@@ -35,7 +92,7 @@ class TestEmailFileParsing:
             - An Exception is thrown
         """
         # Check if exception is raised when parsing fails.
-        test_file = sample_dir / "doesntexist.eml"
+        test_file = mail_samples_dir / "doesntexist.eml"
 
         assert not test_file.exists()
 
@@ -246,12 +303,12 @@ class TestEmailThumbnailGenerate:
         """
         mocked_return = "Passing the return value through.."
         mock_make_thumbnail_from_pdf = mocker.patch(
-            "paperless_mail.parsers.make_thumbnail_from_pdf",
+            "paperless.parsers.mail.make_thumbnail_from_pdf",
         )
         mock_make_thumbnail_from_pdf.return_value = mocked_return
 
         mock_generate_pdf = mocker.patch(
-            "paperless_mail.parsers.MailDocumentParser.generate_pdf",
+            "paperless.parsers.mail.MailDocumentParser.generate_pdf",
         )
         mock_generate_pdf.return_value = "Mocked return value.."
 
@@ -260,8 +317,7 @@ class TestEmailThumbnailGenerate:
         mock_generate_pdf.assert_called_once()
         mock_make_thumbnail_from_pdf.assert_called_once_with(
             "Mocked return value..",
-            mail_parser.tempdir,
-            None,
+            mail_parser._tempdir,
         )
 
         assert mocked_return == thumb
@@ -373,7 +429,7 @@ class TestParser:
         """
         # Validate parsing returns the expected results
         mock_generate_pdf = mocker.patch(
-            "paperless_mail.parsers.MailDocumentParser.generate_pdf",
+            "paperless.parsers.mail.MailDocumentParser.generate_pdf",
         )
 
         mail_parser.parse(simple_txt_email_file, "message/rfc822")
@@ -385,7 +441,7 @@ class TestParser:
             "BCC: fdf@fvf.de\n\n"
             "\n\nThis is just a simple Text Mail."
         )
-        assert text_expected == mail_parser.text
+        assert text_expected == mail_parser.get_text()
         assert (
             datetime.datetime(
                 2022,
@@ -396,7 +452,7 @@ class TestParser:
                 43,
                 tzinfo=datetime.timezone(datetime.timedelta(seconds=7200)),
             )
-            == mail_parser.date
+            == mail_parser.get_date()
         )
 
         # Just check if tried to generate archive, the unittest for generate_pdf() goes deeper.
@@ -419,7 +475,7 @@ class TestParser:
         """
 
         mock_generate_pdf = mocker.patch(
-            "paperless_mail.parsers.MailDocumentParser.generate_pdf",
+            "paperless.parsers.mail.MailDocumentParser.generate_pdf",
         )
 
         # Validate parsing returns the expected results
@@ -443,7 +499,7 @@ class TestParser:
         mail_parser.parse(html_email_file, "message/rfc822")
 
         mock_generate_pdf.assert_called_once()
-        assert text_expected == mail_parser.text
+        assert text_expected == mail_parser.get_text()
         assert (
             datetime.datetime(
                 2022,
@@ -454,7 +510,7 @@ class TestParser:
                 19,
                 tzinfo=datetime.timezone(datetime.timedelta(seconds=7200)),
             )
-            == mail_parser.date
+            == mail_parser.get_date()
         )
 
     def test_generate_pdf_parse_error(
@@ -501,7 +557,7 @@ class TestParser:
 
         mail_parser.parse(simple_txt_email_file, "message/rfc822")
 
-        assert mail_parser.archive_path is not None
+        assert mail_parser.get_archive_path() is not None
 
     @pytest.mark.httpx_mock(can_send_already_matched_responses=True)
     def test_generate_pdf_html_email(
@@ -542,7 +598,7 @@ class TestParser:
         )
         mail_parser.parse(html_email_file, "message/rfc822")
 
-        assert mail_parser.archive_path is not None
+        assert mail_parser.get_archive_path() is not None
 
     def test_generate_pdf_html_email_html_to_pdf_failure(
         self,
@@ -712,10 +768,10 @@ class TestParser:
 
         def test_layout_option(layout_option, expected_calls, expected_pdf_names):
             mock_mailrule_get.return_value = mock.Mock(pdf_layout=layout_option)
+            mail_parser.configure(ParserContext(mailrule_id=1))
             mail_parser.parse(
                 document_path=html_email_file,
                 mime_type="message/rfc822",
-                mailrule_id=1,
             )
             args, _ = mock_merge_route.call_args
             assert len(args[0]) == expected_calls

src/paperless/tests/parsers/test_mail_parser_live.py

@@ -11,7 +11,7 @@ from PIL import Image
 from pytest_mock import MockerFixture
 
 from documents.tests.utils import util_call_with_backoff
-from paperless_mail.parsers import MailDocumentParser
+from paperless.parsers.mail import MailDocumentParser
 
 
 def extract_text(pdf_path: Path) -> str:
@@ -159,7 +159,7 @@ class TestParserLive:
             - The returned thumbnail image file shall match the expected hash
         """
         mock_generate_pdf = mocker.patch(
-            "paperless_mail.parsers.MailDocumentParser.generate_pdf",
+            "paperless.parsers.mail.MailDocumentParser.generate_pdf",
         )
         mock_generate_pdf.return_value = simple_txt_email_pdf_file
 
@@ -216,10 +216,10 @@ class TestParserLive:
             - The merged PDF shall contain text from both source PDFs
         """
         mock_generate_pdf_from_html = mocker.patch(
-            "paperless_mail.parsers.MailDocumentParser.generate_pdf_from_html",
+            "paperless.parsers.mail.MailDocumentParser.generate_pdf_from_html",
         )
         mock_generate_pdf_from_mail = mocker.patch(
-            "paperless_mail.parsers.MailDocumentParser.generate_pdf_from_mail",
+            "paperless.parsers.mail.MailDocumentParser.generate_pdf_from_mail",
         )
         mock_generate_pdf_from_mail.return_value = merged_pdf_first
         mock_generate_pdf_from_html.return_value = merged_pdf_second

src/paperless/tests/parsers/test_remote_parser.py

@@ -20,6 +20,7 @@ from unittest.mock import Mock
 
 import pytest
 
+from paperless.parsers import ParserContext
 from paperless.parsers import ParserProtocol
 from paperless.parsers.remote import RemoteDocumentParser
 
@@ -302,6 +303,7 @@ class TestRemoteParserParse:
         sample_pdf_file: Path,
         azure_client: Mock,
     ) -> None:
+        remote_parser.configure(ParserContext())
         remote_parser.parse(sample_pdf_file, "application/pdf")
 
         azure_client.close.assert_called_once()
@@ -479,12 +481,17 @@ class TestRemoteParserRegistry:
         assert parser_cls is RemoteDocumentParser
 
     @pytest.mark.usefixtures("no_engine_settings")
-    def test_get_parser_returns_none_for_pdf_when_not_configured(self) -> None:
-        """With no tesseract parser registered yet, PDF has no handler if remote is off."""
+    def test_get_parser_returns_none_for_unsupported_type_when_not_configured(
+        self,
+    ) -> None:
+        """With remote off and a truly unsupported MIME type, registry returns None."""
         from paperless.parsers.registry import ParserRegistry
 
         registry = ParserRegistry()
         registry.register_defaults()
-        parser_cls = registry.get_parser_for_file("application/pdf", "doc.pdf")
+        parser_cls = registry.get_parser_for_file(
+            "application/x-unknown-format",
+            "doc.xyz",
+        )
 
         assert parser_cls is None

src/paperless/tests/parsers/test_tesseract_custom_settings.py

@@ -10,7 +10,7 @@ from paperless.models import CleanChoices
 from paperless.models import ColorConvertChoices
 from paperless.models import ModeChoices
 from paperless.models import OutputTypeChoices
-from paperless_tesseract.parsers import RasterisedDocumentParser
+from paperless.parsers.tesseract import RasterisedDocumentParser
 
 
 class TestParserSettingsFromDb(DirectoriesMixin, FileSystemAssertsMixin, TestCase):

src/paperless/tests/parsers/test_text_parser.py

@@ -12,6 +12,7 @@ from pathlib import Path
 
 import pytest
 
+from paperless.parsers import ParserContext
 from paperless.parsers import ParserProtocol
 from paperless.parsers.text import TextDocumentParser
 
@@ -93,6 +94,7 @@ class TestTextParserParse:
         text_parser: TextDocumentParser,
         sample_txt_file: Path,
     ) -> None:
+        text_parser.configure(ParserContext())
         text_parser.parse(sample_txt_file, "text/plain")
 
         assert text_parser.get_text() == "This is a test file.\n"
@@ -102,6 +104,7 @@ class TestTextParserParse:
         text_parser: TextDocumentParser,
         sample_txt_file: Path,
     ) -> None:
+        text_parser.configure(ParserContext())
         text_parser.parse(sample_txt_file, "text/plain")
 
         assert text_parser.get_archive_path() is None
@@ -111,6 +114,7 @@ class TestTextParserParse:
         text_parser: TextDocumentParser,
         sample_txt_file: Path,
     ) -> None:
+        text_parser.configure(ParserContext())
         text_parser.parse(sample_txt_file, "text/plain")
 
         assert text_parser.get_date() is None
@@ -129,6 +133,7 @@ class TestTextParserParse:
             - Parsing succeeds
             - Invalid bytes are replaced with the Unicode replacement character
         """
+        text_parser.configure(ParserContext())
         text_parser.parse(malformed_txt_file, "text/plain")
 
         assert text_parser.get_text() == "Pantothens\ufffdure\n"
@@ -251,6 +256,9 @@ class TestTextParserRegistry:
         from paperless.parsers.registry import get_parser_registry
 
         registry = get_parser_registry()
-        parser_cls = registry.get_parser_for_file("application/pdf", "doc.pdf")
+        parser_cls = registry.get_parser_for_file(
+            "application/x-unknown-format",
+            "doc.xyz",
+        )
 
         assert parser_cls is None

src/paperless/tests/parsers/test_tika_parser.py

@@ -9,6 +9,7 @@ from pytest_django.fixtures import SettingsWrapper
 from pytest_httpx import HTTPXMock
 
 from documents.parsers import ParseError
+from paperless.parsers import ParserContext
 from paperless.parsers import ParserProtocol
 from paperless.parsers.tika import TikaDocumentParser
 
@@ -60,6 +61,29 @@ class TestTikaParserRegistryInterface:
     def test_requires_pdf_rendition_is_true(self) -> None:
         assert TikaDocumentParser().requires_pdf_rendition is True
 
+    def test_get_page_count_returns_none_without_archive(
+        self,
+        tika_parser: TikaDocumentParser,
+        sample_odt_file: Path,
+    ) -> None:
+        assert (
+            tika_parser.get_page_count(
+                sample_odt_file,
+                "application/vnd.oasis.opendocument.text",
+            )
+            is None
+        )
+
+    def test_get_page_count_returns_int_with_pdf_archive(
+        self,
+        tika_parser: TikaDocumentParser,
+        sample_pdf_file: Path,
+    ) -> None:
+        tika_parser._archive_path = sample_pdf_file
+        count = tika_parser.get_page_count(sample_pdf_file, "application/pdf")
+        assert isinstance(count, int)
+        assert count > 0
+
 
 @pytest.mark.django_db()
 class TestTikaParser:
@@ -83,6 +107,7 @@ class TestTikaParser:
         # Pretend convert to PDF response
         httpx_mock.add_response(content=b"PDF document")
 
+        tika_parser.configure(ParserContext())
         tika_parser.parse(sample_odt_file, "application/vnd.oasis.opendocument.text")
 
         assert tika_parser.get_text() == "the content"

src/paperless/tests/test_registry.py

@@ -18,6 +18,7 @@ from unittest.mock import patch
 
 import pytest
 
+from paperless.parsers import ParserContext
 from paperless.parsers import ParserProtocol
 from paperless.parsers.registry import ParserRegistry
 from paperless.parsers.registry import get_parser_registry
@@ -103,6 +104,11 @@ def dummy_parser_cls() -> type:
         ) -> list:
             return []
 
+        def configure(self, context: ParserContext) -> None:
+            """
+            Required to exist, but doesn't need to do anything
+            """
+
         def __enter__(self) -> Self:
             return self
 
@@ -144,6 +150,7 @@ class TestParserProtocol:
     @pytest.mark.parametrize(
         "missing_method",
         [
+            pytest.param("configure", id="missing-configure"),
             pytest.param("parse", id="missing-parse"),
             pytest.param("get_text", id="missing-get_text"),
             pytest.param("get_thumbnail", id="missing-get_thumbnail"),

src/paperless_mail/parsers.py

@@ -1,481 +0,0 @@
-import re
-from html import escape
-from pathlib import Path
-
-from bleach import clean
-from bleach import linkify
-from django.conf import settings
-from django.utils import timezone
-from django.utils.timezone import is_naive
-from django.utils.timezone import make_aware
-from gotenberg_client import GotenbergClient
-from gotenberg_client.constants import A4
-from gotenberg_client.options import Measurement
-from gotenberg_client.options import MeasurementUnitType
-from gotenberg_client.options import PageMarginsType
-from gotenberg_client.options import PdfAFormat
-from humanize import naturalsize
-from imap_tools import MailAttachment
-from imap_tools import MailMessage
-from tika_client import TikaClient
-
-from documents.parsers import DocumentParser
-from documents.parsers import ParseError
-from documents.parsers import make_thumbnail_from_pdf
-from paperless.models import OutputTypeChoices
-from paperless_mail.models import MailRule
-
-
-class MailDocumentParser(DocumentParser):
-    """
-    This parser uses imap_tools to parse .eml files, generates pdf using
-    Gotenberg and sends the html part to a Tika server for text extraction.
-    """
-
-    logging_name = "paperless.parsing.mail"
-
-    def _settings_to_gotenberg_pdfa(self) -> PdfAFormat | None:
-        """
-        Converts our requested PDF/A output into the Gotenberg API
-        format
-        """
-        if settings.OCR_OUTPUT_TYPE in {
-            OutputTypeChoices.PDF_A,
-            OutputTypeChoices.PDF_A2,
-        }:
-            return PdfAFormat.A2b
-        elif settings.OCR_OUTPUT_TYPE == OutputTypeChoices.PDF_A1:  # pragma: no cover
-            self.log.warning(
-                "Gotenberg does not support PDF/A-1a, choosing PDF/A-2b instead",
-            )
-            return PdfAFormat.A2b
-        elif settings.OCR_OUTPUT_TYPE == OutputTypeChoices.PDF_A3:  # pragma: no cover
-            return PdfAFormat.A3b
-        return None
-
-    def get_thumbnail(
-        self,
-        document_path: Path,
-        mime_type: str,
-        file_name=None,
-    ) -> Path:
-        if not self.archive_path:
-            self.archive_path = self.generate_pdf(
-                self.parse_file_to_message(document_path),
-            )
-
-        return make_thumbnail_from_pdf(
-            self.archive_path,
-            self.tempdir,
-            self.logging_group,
-        )
-
-    def extract_metadata(self, document_path: Path, mime_type: str):
-        result = []
-
-        try:
-            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}",
-            )
-            return result
-
-        for key, value in mail.headers.items():
-            value = ", ".join(i for i in value)
-            try:
-                value.encode("utf-8")
-            except UnicodeEncodeError as e:  # pragma: no cover
-                self.log.debug(f"Skipping header {key}: {e}")
-                continue
-
-            result.append(
-                {
-                    "namespace": "",
-                    "prefix": "header",
-                    "key": key,
-                    "value": value,
-                },
-            )
-
-        result.append(
-            {
-                "namespace": "",
-                "prefix": "",
-                "key": "attachments",
-                "value": ", ".join(
-                    f"{attachment.filename}"
-                    f"({naturalsize(attachment.size, binary=True, format='%.2f')})"
-                    for attachment in mail.attachments
-                ),
-            },
-        )
-
-        result.append(
-            {
-                "namespace": "",
-                "prefix": "",
-                "key": "date",
-                "value": mail.date.strftime("%Y-%m-%d %H:%M:%S %Z"),
-            },
-        )
-
-        result.sort(key=lambda item: (item["prefix"], item["key"]))
-        return result
-
-    def parse(
-        self,
-        document_path: Path,
-        mime_type: str,
-        file_name=None,
-        mailrule_id: int | None = None,
-    ) -> 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()
-
-        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 mail_message.bcc_values:
-                fmt_text += (
-                    f"BCC: {', '.join(address.full for address in mail.bcc_values)}\n\n"
-                )
-            if mail_message.attachments:
-                att = []
-                for a in mail.attachments:
-                    attachment_size = naturalsize(a.size, binary=True, format="%.2f")
-                    att.append(
-                        f"{a.filename} ({attachment_size})",
-                    )
-                fmt_text += f"Attachments: {', '.join(att)}\n\n"
-
-            if mail.html:
-                fmt_text += "HTML content: " + strip_text(self.tika_parse(mail.html))
-
-            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.log.debug("Creating a PDF from the email")
-        if mailrule_id:
-            rule = MailRule.objects.get(pk=mailrule_id)
-            self.archive_path = self.generate_pdf(mail, rule.pdf_layout)
-        else:
-            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:
-            with TikaClient(tika_url=settings.TIKA_ENDPOINT) as client:
-                parsed = client.tika.as_text.from_buffer(html, "text/html")
-
-                if parsed.content is not None:
-                    return parsed.content.strip()
-                return ""
-        except Exception as err:
-            raise ParseError(
-                f"Could not parse content with tika server at "
-                f"{settings.TIKA_ENDPOINT}: {err}",
-            ) from err
-
-    def generate_pdf(
-        self,
-        mail_message: MailMessage,
-        pdf_layout: MailRule.PdfLayout | None = None,
-    ) -> Path:
-        archive_path = Path(self.tempdir) / "merged.pdf"
-
-        mail_pdf_file = self.generate_pdf_from_mail(mail_message)
-
-        pdf_layout = (
-            pdf_layout or settings.EMAIL_PARSE_DEFAULT_LAYOUT
-        )  # EMAIL_PARSE_DEFAULT_LAYOUT is a MailRule.PdfLayout
-
-        # 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:
-            pdf_of_html_content = self.generate_pdf_from_html(
-                mail_message.html,
-                mail_message.attachments,
-            )
-
-            self.log.debug("Merging email text and HTML content into single PDF")
-
-            with (
-                GotenbergClient(
-                    host=settings.TIKA_GOTENBERG_ENDPOINT,
-                    timeout=settings.CELERY_TASK_TIME_LIMIT,
-                ) as client,
-                client.merge.merge() as route,
-            ):
-                # Configure requested PDF/A formatting, if any
-                pdf_a_format = self._settings_to_gotenberg_pdfa()
-                if pdf_a_format is not None:
-                    route.pdf_format(pdf_a_format)
-
-                match pdf_layout:
-                    case MailRule.PdfLayout.HTML_TEXT:
-                        route.merge([pdf_of_html_content, mail_pdf_file])
-                    case MailRule.PdfLayout.HTML_ONLY:
-                        route.merge([pdf_of_html_content])
-                    case MailRule.PdfLayout.TEXT_ONLY:
-                        route.merge([mail_pdf_file])
-                    case MailRule.PdfLayout.TEXT_HTML | _:
-                        route.merge([mail_pdf_file, pdf_of_html_content])
-
-                try:
-                    response = route.run()
-                    archive_path.write_bytes(response.content)
-                except Exception as err:
-                    raise ParseError(
-                        f"Error while merging email HTML into PDF: {err}",
-                    ) from err
-
-        return archive_path
-
-    def mail_to_html(self, mail: MailMessage) -> Path:
-        """
-        Converts the given email into an HTML file, formatted
-        based on the given template
-        """
-
-        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 not isinstance(text, str):
-                text = str(text)
-            text = escape(text)
-            text = clean(text)
-            text = linkify(text, parse_email=True)
-            text = text.replace("\n", "<br>")
-            return text
-
-        data = {}
-
-        data["subject"] = clean_html(mail.subject)
-        if data["subject"]:
-            data["subject_label"] = "Subject"
-        data["from"] = clean_html(mail.from_values.full)
-        if data["from"]:
-            data["from_label"] = "From"
-        data["to"] = clean_html(", ".join(address.full for address in mail.to_values))
-        if data["to"]:
-            data["to_label"] = "To"
-        data["cc"] = clean_html(", ".join(address.full for address in mail.cc_values))
-        if data["cc"]:
-            data["cc_label"] = "CC"
-        data["bcc"] = clean_html(", ".join(address.full for address in mail.bcc_values))
-        if data["bcc"]:
-            data["bcc_label"] = "BCC"
-
-        att = []
-        for a in mail.attachments:
-            att.append(
-                f"{a.filename} ({naturalsize(a.size, binary=True, format='%.2f')})",
-            )
-        data["attachments"] = clean_html(", ".join(att))
-        if data["attachments"]:
-            data["attachments_label"] = "Attachments"
-
-        data["date"] = clean_html(
-            timezone.localtime(mail.date).strftime("%Y-%m-%d %H:%M"),
-        )
-        data["content"] = clean_html(mail.text.strip())
-
-        from django.template.loader import render_to_string
-
-        html_file = Path(self.tempdir) / "email_as_html.html"
-        html_file.write_text(render_to_string("email_msg_template.html", context=data))
-
-        return html_file
-
-    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
-        """
-        self.log.info("Converting mail to PDF")
-
-        css_file = Path(__file__).parent / "templates" / "output.css"
-        email_html_file = self.mail_to_html(mail)
-
-        with (
-            GotenbergClient(
-                host=settings.TIKA_GOTENBERG_ENDPOINT,
-                timeout=settings.CELERY_TASK_TIME_LIMIT,
-            ) as client,
-            client.chromium.html_to_pdf() as route,
-        ):
-            # Configure requested PDF/A formatting, if any
-            pdf_a_format = self._settings_to_gotenberg_pdfa()
-            if pdf_a_format is not None:
-                route.pdf_format(pdf_a_format)
-
-            try:
-                response = (
-                    route.index(email_html_file)
-                    .resource(css_file)
-                    .margins(
-                        PageMarginsType(
-                            top=Measurement(0.1, MeasurementUnitType.Inches),
-                            bottom=Measurement(0.1, MeasurementUnitType.Inches),
-                            left=Measurement(0.1, MeasurementUnitType.Inches),
-                            right=Measurement(0.1, MeasurementUnitType.Inches),
-                        ),
-                    )
-                    .size(A4)
-                    .scale(1.0)
-                    .run()
-                )
-            except Exception as err:
-                raise ParseError(
-                    f"Error while converting email to PDF: {err}",
-                ) from err
-
-        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
-        """
-
-        def clean_html_script(text: str):
-            compiled_open = re.compile(re.escape("<script"), re.IGNORECASE)
-            text = compiled_open.sub("<div hidden ", text)
-
-            compiled_close = re.compile(re.escape("</script"), re.IGNORECASE)
-            text = compiled_close.sub("</div", text)
-            return text
-
-        self.log.info("Converting message html to PDF")
-
-        tempdir = Path(self.tempdir)
-
-        html_clean = clean_html_script(orig_html)
-        html_clean_file = tempdir / "index.html"
-        html_clean_file.write_text(html_clean)
-
-        with (
-            GotenbergClient(
-                host=settings.TIKA_GOTENBERG_ENDPOINT,
-                timeout=settings.CELERY_TASK_TIME_LIMIT,
-            ) as client,
-            client.chromium.html_to_pdf() as route,
-        ):
-            # Configure requested PDF/A formatting, if any
-            pdf_a_format = self._settings_to_gotenberg_pdfa()
-            if pdf_a_format is not None:
-                route.pdf_format(pdf_a_format)
-
-            # Add attachments as resources, cleaning the filename and replacing
-            # it in the index file for inclusion
-            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)
-
-                route.resource(temp_file)
-
-                # 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)
-            # This is our index file, the main page basically
-            route.index(html_clean_file)
-
-            # Set page size, margins
-            route.margins(
-                PageMarginsType(
-                    top=Measurement(0.1, MeasurementUnitType.Inches),
-                    bottom=Measurement(0.1, MeasurementUnitType.Inches),
-                    left=Measurement(0.1, MeasurementUnitType.Inches),
-                    right=Measurement(0.1, MeasurementUnitType.Inches),
-                ),
-            ).size(A4).scale(1.0)
-
-            try:
-                response = route.run()
-
-            except Exception as err:
-                raise ParseError(
-                    f"Error while converting document to PDF: {err}",
-                ) from err
-
-        html_pdf = tempdir / "html.pdf"
-        html_pdf.write_bytes(response.content)
-        return html_pdf
-
-    def get_settings(self) -> None:
-        """
-        This parser does not implement additional settings yet
-        """
-        return None

src/paperless_mail/signals.py

@@ -1,7 +1,12 @@
 def get_parser(*args, **kwargs):
-    from paperless_mail.parsers import MailDocumentParser
+    from paperless.parsers.mail import MailDocumentParser
 
-    return MailDocumentParser(*args, **kwargs)
+    # MailDocumentParser accepts no constructor args in the new-style protocol.
+    # Pop legacy args that arrive from the signal-based consumer path.
+    # Phase 4 will replace this signal path with the ParserRegistry.
+    kwargs.pop("logging_group", None)
+    kwargs.pop("progress_callback", None)
+    return MailDocumentParser()
 
 
 def mail_consumer_declaration(sender, **kwargs):

src/paperless_mail/tests/conftest.py

@@ -1,71 +1,9 @@
 from collections.abc import Generator
-from pathlib import Path
 
 import pytest
 
 from paperless_mail.mail import MailAccountHandler
 from paperless_mail.models import MailAccount
-from paperless_mail.parsers import MailDocumentParser
-
-
-@pytest.fixture(scope="session")
-def sample_dir() -> Path:
-    return (Path(__file__).parent / Path("samples")).resolve()
-
-
-@pytest.fixture(scope="session")
-def broken_email_file(sample_dir: Path) -> Path:
-    return sample_dir / "broken.eml"
-
-
-@pytest.fixture(scope="session")
-def simple_txt_email_file(sample_dir: Path) -> Path:
-    return sample_dir / "simple_text.eml"
-
-
-@pytest.fixture(scope="session")
-def simple_txt_email_pdf_file(sample_dir: Path) -> Path:
-    return sample_dir / "simple_text.eml.pdf"
-
-
-@pytest.fixture(scope="session")
-def simple_txt_email_thumbnail_file(sample_dir: Path) -> Path:
-    return sample_dir / "simple_text.eml.pdf.webp"
-
-
-@pytest.fixture(scope="session")
-def html_email_file(sample_dir: Path) -> Path:
-    return sample_dir / "html.eml"
-
-
-@pytest.fixture(scope="session")
-def html_email_pdf_file(sample_dir: Path) -> Path:
-    return sample_dir / "html.eml.pdf"
-
-
-@pytest.fixture(scope="session")
-def html_email_thumbnail_file(sample_dir: Path) -> Path:
-    return sample_dir / "html.eml.pdf.webp"
-
-
-@pytest.fixture(scope="session")
-def html_email_html_file(sample_dir: Path) -> Path:
-    return sample_dir / "html.eml.html"
-
-
-@pytest.fixture(scope="session")
-def merged_pdf_first(sample_dir: Path) -> Path:
-    return sample_dir / "first.pdf"
-
-
-@pytest.fixture(scope="session")
-def merged_pdf_second(sample_dir: Path) -> Path:
-    return sample_dir / "second.pdf"
-
-
-@pytest.fixture()
-def mail_parser() -> MailDocumentParser:
-    return MailDocumentParser(logging_group=None)
 
 
 @pytest.fixture()
@@ -89,11 +27,3 @@ def greenmail_mail_account(db: None) -> Generator[MailAccount, None, None]:
 @pytest.fixture()
 def mail_account_handler() -> MailAccountHandler:
     return MailAccountHandler()
-
-
-@pytest.fixture(scope="session")
-def nginx_base_url() -> Generator[str, None, None]:
-    """
-    The base URL for the nginx HTTP server we expect to be alive
-    """
-    yield "http://localhost:8080"

src/paperless_tesseract/signals.py

@@ -1,10 +1,23 @@
-def get_parser(*args, **kwargs):
-    from paperless_tesseract.parsers import RasterisedDocumentParser
+from __future__ import annotations
 
+from typing import Any
+
+
+def get_parser(*args: Any, **kwargs: Any) -> Any:
+    from paperless.parsers.tesseract import RasterisedDocumentParser
+
+    # RasterisedDocumentParser accepts logging_group for constructor compatibility but
+    # does not store or use it (no legacy DocumentParser base class).
+    # progress_callback is also not used.  Both may arrive as a positional arg
+    # (consumer) or a keyword arg (views); *args absorbs the positional form,
+    # kwargs.pop handles the keyword form.  Phase 4 will replace this signal
+    # path with the new ParserRegistry so the shim can be removed at that point.
+    kwargs.pop("logging_group", None)
+    kwargs.pop("progress_callback", None)
     return RasterisedDocumentParser(*args, **kwargs)
 
 
-def tesseract_consumer_declaration(sender, **kwargs):
+def tesseract_consumer_declaration(sender: Any, **kwargs: Any) -> dict[str, Any]:
     return {
         "parser": get_parser,
         "weight": 0,

src/paperless_tesseract/tests/test_parser.py

@@ -1,924 +0,0 @@
-import shutil
-import tempfile
-import unicodedata
-import uuid
-from pathlib import Path
-from unittest import mock
-
-from django.test import TestCase
-from django.test import override_settings
-from ocrmypdf import SubprocessOutputError
-
-from documents.parsers import ParseError
-from documents.parsers import run_convert
-from documents.tests.utils import DirectoriesMixin
-from documents.tests.utils import FileSystemAssertsMixin
-from paperless_tesseract.parsers import RasterisedDocumentParser
-from paperless_tesseract.parsers import post_process_text
-
-
-class TestParser(DirectoriesMixin, FileSystemAssertsMixin, TestCase):
-    SAMPLE_FILES = Path(__file__).resolve().parent / "samples"
-
-    def assertContainsStrings(self, content, strings) -> None:
-        # Asserts that all strings appear in content, in the given order.
-        indices = []
-        for s in strings:
-            if s in content:
-                indices.append(content.index(s))
-            else:
-                self.fail(f"'{s}' is not in '{content}'")
-        self.assertListEqual(indices, sorted(indices))
-
-    def test_post_process_text(self) -> None:
-        text_cases = [
-            ("simple     string", "simple string"),
-            ("simple    newline\n   testing string", "simple newline\ntesting string"),
-            (
-                "utf-8   строка с пробелами в конце  ",
-                "utf-8 строка с пробелами в конце",
-            ),
-        ]
-
-        for source, result in text_cases:
-            actual_result = post_process_text(source)
-            self.assertEqual(
-                result,
-                actual_result,
-                f"strip_exceess_whitespace({source}) != '{result}', but '{actual_result}'",
-            )
-
-    def test_get_text_from_pdf(self) -> None:
-        parser = RasterisedDocumentParser(uuid.uuid4())
-        text = parser.extract_text(
-            None,
-            self.SAMPLE_FILES / "simple-digital.pdf",
-        )
-
-        self.assertContainsStrings(text.strip(), ["This is a test document."])
-
-    def test_get_page_count(self) -> None:
-        """
-        GIVEN:
-            - PDF file with a single page
-            - PDF file with multiple pages
-        WHEN:
-            - The number of pages is requested
-        THEN:
-            - The method returns 1 as the expected number of pages
-            - The method returns the correct number of pages (6)
-        """
-        parser = RasterisedDocumentParser(uuid.uuid4())
-        page_count = parser.get_page_count(
-            str(self.SAMPLE_FILES / "simple-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertEqual(page_count, 1)
-
-        page_count = parser.get_page_count(
-            str(self.SAMPLE_FILES / "multi-page-mixed.pdf"),
-            "application/pdf",
-        )
-        self.assertEqual(page_count, 6)
-
-    def test_get_page_count_password_protected(self) -> None:
-        """
-        GIVEN:
-            - Password protected PDF file
-        WHEN:
-            - The number of pages is requested
-        THEN:
-            - The method returns None
-        """
-        parser = RasterisedDocumentParser(uuid.uuid4())
-        with self.assertLogs("paperless.parsing.tesseract", level="WARNING") as cm:
-            page_count = parser.get_page_count(
-                str(self.SAMPLE_FILES / "password-protected.pdf"),
-                "application/pdf",
-            )
-            self.assertEqual(page_count, None)
-            self.assertIn("Unable to determine PDF page count", cm.output[0])
-
-    def test_thumbnail(self) -> None:
-        parser = RasterisedDocumentParser(uuid.uuid4())
-        thumb = parser.get_thumbnail(
-            str(self.SAMPLE_FILES / "simple-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertIsFile(thumb)
-
-    @mock.patch("documents.parsers.run_convert")
-    def test_thumbnail_fallback(self, m) -> None:
-        def call_convert(input_file, output_file, **kwargs) -> None:
-            if ".pdf" in str(input_file):
-                raise ParseError("Does not compute.")
-            else:
-                run_convert(input_file=input_file, output_file=output_file, **kwargs)
-
-        m.side_effect = call_convert
-
-        parser = RasterisedDocumentParser(uuid.uuid4())
-        thumb = parser.get_thumbnail(
-            str(self.SAMPLE_FILES / "simple-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertIsFile(thumb)
-
-    def test_thumbnail_encrypted(self) -> None:
-        parser = RasterisedDocumentParser(uuid.uuid4())
-        thumb = parser.get_thumbnail(
-            str(self.SAMPLE_FILES / "encrypted.pdf"),
-            "application/pdf",
-        )
-        self.assertIsFile(thumb)
-
-    def test_get_dpi(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        dpi = parser.get_dpi(str(self.SAMPLE_FILES / "simple-no-dpi.png"))
-        self.assertEqual(dpi, None)
-
-        dpi = parser.get_dpi(str(self.SAMPLE_FILES / "simple.png"))
-        self.assertEqual(dpi, 72)
-
-    def test_simple_digital(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        parser.parse(
-            str(self.SAMPLE_FILES / "simple-digital.pdf"),
-            "application/pdf",
-        )
-
-        self.assertIsFile(parser.archive_path)
-
-        self.assertContainsStrings(parser.get_text(), ["This is a test document."])
-
-    def test_with_form(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        parser.parse(
-            str(self.SAMPLE_FILES / "with-form.pdf"),
-            "application/pdf",
-        )
-
-        self.assertIsFile(parser.archive_path)
-
-        self.assertContainsStrings(
-            parser.get_text(),
-            ["Please enter your name in here:", "This is a PDF document with a form."],
-        )
-
-    @override_settings(OCR_MODE="redo")
-    def test_with_form_error(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        parser.parse(
-            str(self.SAMPLE_FILES / "with-form.pdf"),
-            "application/pdf",
-        )
-
-        self.assertIsNone(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text(),
-            ["Please enter your name in here:", "This is a PDF document with a form."],
-        )
-
-    @override_settings(OCR_MODE="skip")
-    def test_signed(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        parser.parse(str(self.SAMPLE_FILES / "signed.pdf"), "application/pdf")
-
-        self.assertIsNone(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text(),
-            [
-                "This is a digitally signed PDF, created with Acrobat Pro for the Paperless project to enable",
-                "automated testing of signed/encrypted PDFs",
-            ],
-        )
-
-    @override_settings(OCR_MODE="skip")
-    def test_encrypted(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        parser.parse(
-            str(self.SAMPLE_FILES / "encrypted.pdf"),
-            "application/pdf",
-        )
-
-        self.assertIsNone(parser.archive_path)
-        self.assertEqual(parser.get_text(), "")
-
-    @override_settings(OCR_MODE="redo")
-    def test_with_form_error_notext(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "with-form.pdf"),
-            "application/pdf",
-        )
-
-        self.assertContainsStrings(
-            parser.get_text(),
-            ["Please enter your name in here:", "This is a PDF document with a form."],
-        )
-
-    @override_settings(OCR_MODE="force")
-    def test_with_form_force(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        parser.parse(
-            str(self.SAMPLE_FILES / "with-form.pdf"),
-            "application/pdf",
-        )
-
-        self.assertContainsStrings(
-            parser.get_text(),
-            ["Please enter your name in here:", "This is a PDF document with a form."],
-        )
-
-    def test_image_simple(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        parser.parse(str(self.SAMPLE_FILES / "simple.png"), "image/png")
-
-        self.assertIsFile(parser.archive_path)
-
-        self.assertContainsStrings(parser.get_text(), ["This is a test document."])
-
-    def test_image_simple_alpha(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        with tempfile.TemporaryDirectory() as tempdir:
-            # Copy sample file to temp directory, as the parsing changes the file
-            # and this makes it modified to Git
-            sample_file = self.SAMPLE_FILES / "simple-alpha.png"
-            dest_file = Path(tempdir) / "simple-alpha.png"
-            shutil.copy(sample_file, dest_file)
-
-            parser.parse(str(dest_file), "image/png")
-
-            self.assertIsFile(parser.archive_path)
-
-            self.assertContainsStrings(parser.get_text(), ["This is a test document."])
-
-    def test_image_calc_a4_dpi(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        dpi = parser.calculate_a4_dpi(
-            str(self.SAMPLE_FILES / "simple-no-dpi.png"),
-        )
-
-        self.assertEqual(dpi, 62)
-
-    @mock.patch("paperless_tesseract.parsers.RasterisedDocumentParser.calculate_a4_dpi")
-    def test_image_dpi_fail(self, m) -> None:
-        m.return_value = None
-        parser = RasterisedDocumentParser(None)
-
-        def f() -> None:
-            parser.parse(
-                str(self.SAMPLE_FILES / "simple-no-dpi.png"),
-                "image/png",
-            )
-
-        self.assertRaises(ParseError, f)
-
-    @override_settings(OCR_IMAGE_DPI=72, MAX_IMAGE_PIXELS=0)
-    def test_image_no_dpi_default(self) -> None:
-        parser = RasterisedDocumentParser(None)
-
-        parser.parse(str(self.SAMPLE_FILES / "simple-no-dpi.png"), "image/png")
-
-        self.assertIsFile(parser.archive_path)
-
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["this is a test document."],
-        )
-
-    def test_multi_page(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertIsFile(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_PAGES=2, OCR_MODE="skip")
-    def test_multi_page_pages_skip(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertIsFile(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_PAGES=2, OCR_MODE="redo")
-    def test_multi_page_pages_redo(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertIsFile(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_PAGES=2, OCR_MODE="force")
-    def test_multi_page_pages_force(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertIsFile(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_MODE="skip")
-    def test_multi_page_analog_pages_skip(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-images.pdf"),
-            "application/pdf",
-        )
-        self.assertIsFile(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_PAGES=2, OCR_MODE="redo")
-    def test_multi_page_analog_pages_redo(self) -> None:
-        """
-        GIVEN:
-            - File with text contained in images but no text layer
-            - OCR of only pages 1 and 2 requested
-            - OCR mode set to redo
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text of page 1 and 2 extracted
-            - An archive file is created
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-images.pdf"),
-            "application/pdf",
-        )
-        self.assertIsFile(parser.archive_path)
-        self.assertContainsStrings(parser.get_text().lower(), ["page 1", "page 2"])
-        self.assertNotIn("page 3", parser.get_text().lower())
-
-    @override_settings(OCR_PAGES=1, OCR_MODE="force")
-    def test_multi_page_analog_pages_force(self) -> None:
-        """
-        GIVEN:
-            - File with text contained in images but no text layer
-            - OCR of only page 1 requested
-            - OCR mode set to force
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Only text of page 1 is extracted
-            - An archive file is created
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-images.pdf"),
-            "application/pdf",
-        )
-        self.assertIsFile(parser.archive_path)
-        self.assertContainsStrings(parser.get_text().lower(), ["page 1"])
-        self.assertNotIn("page 2", parser.get_text().lower())
-        self.assertNotIn("page 3", parser.get_text().lower())
-
-    @override_settings(OCR_MODE="skip_noarchive")
-    def test_skip_noarchive_withtext(self) -> None:
-        """
-        GIVEN:
-            - File with existing text layer
-            - OCR mode set to skip_noarchive
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from images is extracted
-            - No archive file is created
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertIsNone(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_MODE="skip_noarchive")
-    def test_skip_noarchive_notext(self) -> None:
-        """
-        GIVEN:
-            - File with text contained in images but no text layer
-            - OCR mode set to skip_noarchive
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from images is extracted
-            - An archive file is created with the OCRd text
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-images.pdf"),
-            "application/pdf",
-        )
-
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-        self.assertIsNotNone(parser.archive_path)
-
-    @override_settings(OCR_SKIP_ARCHIVE_FILE="never")
-    def test_skip_archive_never_withtext(self) -> None:
-        """
-        GIVEN:
-            - File with existing text layer
-            - OCR_SKIP_ARCHIVE_FILE set to never
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from text layer is extracted
-            - Archive file is created
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertIsNotNone(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_SKIP_ARCHIVE_FILE="never")
-    def test_skip_archive_never_withimages(self) -> None:
-        """
-        GIVEN:
-            - File with text contained in images but no text layer
-            - OCR_SKIP_ARCHIVE_FILE set to never
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from images is extracted
-            - Archive file is created
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-images.pdf"),
-            "application/pdf",
-        )
-        self.assertIsNotNone(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_SKIP_ARCHIVE_FILE="with_text")
-    def test_skip_archive_withtext_withtext(self) -> None:
-        """
-        GIVEN:
-            - File with existing text layer
-            - OCR_SKIP_ARCHIVE_FILE set to with_text
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from text layer is extracted
-            - No archive file is created
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertIsNone(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_SKIP_ARCHIVE_FILE="with_text")
-    def test_skip_archive_withtext_withimages(self) -> None:
-        """
-        GIVEN:
-            - File with text contained in images but no text layer
-            - OCR_SKIP_ARCHIVE_FILE set to with_text
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from images is extracted
-            - Archive file is created
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-images.pdf"),
-            "application/pdf",
-        )
-        self.assertIsNotNone(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_SKIP_ARCHIVE_FILE="always")
-    def test_skip_archive_always_withtext(self) -> None:
-        """
-        GIVEN:
-            - File with existing text layer
-            - OCR_SKIP_ARCHIVE_FILE set to always
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from text layer is extracted
-            - No archive file is created
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-digital.pdf"),
-            "application/pdf",
-        )
-        self.assertIsNone(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_SKIP_ARCHIVE_FILE="always")
-    def test_skip_archive_always_withimages(self) -> None:
-        """
-        GIVEN:
-            - File with text contained in images but no text layer
-            - OCR_SKIP_ARCHIVE_FILE set to always
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from images is extracted
-            - No archive file is created
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-images.pdf"),
-            "application/pdf",
-        )
-        self.assertIsNone(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    @override_settings(OCR_MODE="skip")
-    def test_multi_page_mixed(self) -> None:
-        """
-        GIVEN:
-            - File with some text contained in images and some in text layer
-            - OCR mode set to skip
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from images is extracted
-            - An archive file is created with the OCRd text and the original text
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-mixed.pdf"),
-            "application/pdf",
-        )
-        self.assertIsNotNone(parser.archive_path)
-        self.assertIsFile(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3", "page 4", "page 5", "page 6"],
-        )
-
-        with (parser.tempdir / "sidecar.txt").open() as f:
-            sidecar = f.read()
-
-        self.assertIn("[OCR skipped on page(s) 4-6]", sidecar)
-
-    @override_settings(OCR_MODE="redo")
-    def test_single_page_mixed(self) -> None:
-        """
-        GIVEN:
-            - File with some text contained in images and some in text layer
-            - Text and images are mixed on the same page
-            - OCR mode set to redo
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from images is extracted
-            - Full content of the file is parsed (not just the image text)
-            - An archive file is created with the OCRd text and the original text
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "single-page-mixed.pdf"),
-            "application/pdf",
-        )
-        self.assertIsNotNone(parser.archive_path)
-        self.assertIsFile(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            [
-                "this is some normal text, present on page 1 of the document.",
-                "this is some text, but in an image, also on page 1.",
-                "this is further text on page 1.",
-            ],
-        )
-
-        with (parser.tempdir / "sidecar.txt").open() as f:
-            sidecar = f.read().lower()
-
-        self.assertIn("this is some text, but in an image, also on page 1.", sidecar)
-        self.assertNotIn(
-            "this is some normal text, present on page 1 of the document.",
-            sidecar,
-        )
-
-    @override_settings(OCR_MODE="skip_noarchive")
-    def test_multi_page_mixed_no_archive(self) -> None:
-        """
-        GIVEN:
-            - File with some text contained in images and some in text layer
-            - OCR mode set to skip_noarchive
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from images is extracted
-            - No archive file is created as original file contains text
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-mixed.pdf"),
-            "application/pdf",
-        )
-        self.assertIsNone(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 4", "page 5", "page 6"],
-        )
-
-    @override_settings(OCR_MODE="skip", OCR_ROTATE_PAGES=True)
-    def test_rotate(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(str(self.SAMPLE_FILES / "rotated.pdf"), "application/pdf")
-        self.assertContainsStrings(
-            parser.get_text(),
-            [
-                "This is the text that appears on the first page. It’s a lot of text.",
-                "Even if the pages are rotated, OCRmyPDF still gets the job done.",
-                "This is a really weird file with lots of nonsense text.",
-                "If you read this, it’s your own fault. Also check your screen orientation.",
-            ],
-        )
-
-    def test_multi_page_tiff(self) -> None:
-        """
-        GIVEN:
-            - Multi-page TIFF image
-        WHEN:
-            - Image is parsed
-        THEN:
-            - Text from all pages extracted
-        """
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "multi-page-images.tiff"),
-            "image/tiff",
-        )
-        self.assertIsFile(parser.archive_path)
-        self.assertContainsStrings(
-            parser.get_text().lower(),
-            ["page 1", "page 2", "page 3"],
-        )
-
-    def test_multi_page_tiff_alpha(self) -> None:
-        """
-        GIVEN:
-            - Multi-page TIFF image
-            - Image include an alpha channel
-        WHEN:
-            - Image is parsed
-        THEN:
-            - Text from all pages extracted
-        """
-        parser = RasterisedDocumentParser(None)
-        sample_file = self.SAMPLE_FILES / "multi-page-images-alpha.tiff"
-        with tempfile.NamedTemporaryFile() as tmp_file:
-            shutil.copy(sample_file, tmp_file.name)
-            parser.parse(
-                tmp_file.name,
-                "image/tiff",
-            )
-            self.assertIsFile(parser.archive_path)
-            self.assertContainsStrings(
-                parser.get_text().lower(),
-                ["page 1", "page 2", "page 3"],
-            )
-
-    def test_multi_page_tiff_alpha_srgb(self) -> None:
-        """
-        GIVEN:
-            - Multi-page TIFF image
-            - Image include an alpha channel
-            - Image is srgb colorspace
-        WHEN:
-            - Image is parsed
-        THEN:
-            - Text from all pages extracted
-        """
-        parser = RasterisedDocumentParser(None)
-        sample_file = str(
-            self.SAMPLE_FILES / "multi-page-images-alpha-rgb.tiff",
-        )
-        with tempfile.NamedTemporaryFile() as tmp_file:
-            shutil.copy(sample_file, tmp_file.name)
-            parser.parse(
-                tmp_file.name,
-                "image/tiff",
-            )
-            self.assertIsFile(parser.archive_path)
-            self.assertContainsStrings(
-                parser.get_text().lower(),
-                ["page 1", "page 2", "page 3"],
-            )
-
-    def test_ocrmypdf_parameters(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        params = parser.construct_ocrmypdf_parameters(
-            input_file="input.pdf",
-            output_file="output.pdf",
-            sidecar_file="sidecar.txt",
-            mime_type="application/pdf",
-            safe_fallback=False,
-        )
-
-        self.assertEqual(params["input_file_or_options"], "input.pdf")
-        self.assertEqual(params["output_file"], "output.pdf")
-        self.assertEqual(params["sidecar"], "sidecar.txt")
-
-        with override_settings(OCR_CLEAN="none"):
-            parser = RasterisedDocumentParser(None)
-            params = parser.construct_ocrmypdf_parameters("", "", "", "")
-            self.assertNotIn("clean", params)
-            self.assertNotIn("clean_final", params)
-
-        with override_settings(OCR_CLEAN="clean"):
-            parser = RasterisedDocumentParser(None)
-            params = parser.construct_ocrmypdf_parameters("", "", "", "")
-            self.assertTrue(params["clean"])
-            self.assertNotIn("clean_final", params)
-
-        with override_settings(OCR_CLEAN="clean-final", OCR_MODE="skip"):
-            parser = RasterisedDocumentParser(None)
-            params = parser.construct_ocrmypdf_parameters("", "", "", "")
-            self.assertTrue(params["clean_final"])
-            self.assertNotIn("clean", params)
-
-        with override_settings(OCR_CLEAN="clean-final", OCR_MODE="redo"):
-            parser = RasterisedDocumentParser(None)
-            params = parser.construct_ocrmypdf_parameters("", "", "", "")
-            self.assertTrue(params["clean"])
-            self.assertNotIn("clean_final", params)
-
-        with override_settings(OCR_DESKEW=True, OCR_MODE="skip"):
-            parser = RasterisedDocumentParser(None)
-            params = parser.construct_ocrmypdf_parameters("", "", "", "")
-            self.assertTrue(params["deskew"])
-
-        with override_settings(OCR_DESKEW=True, OCR_MODE="redo"):
-            parser = RasterisedDocumentParser(None)
-            params = parser.construct_ocrmypdf_parameters("", "", "", "")
-            self.assertNotIn("deskew", params)
-
-        with override_settings(OCR_DESKEW=False, OCR_MODE="skip"):
-            parser = RasterisedDocumentParser(None)
-            params = parser.construct_ocrmypdf_parameters("", "", "", "")
-            self.assertNotIn("deskew", params)
-
-        with override_settings(OCR_MAX_IMAGE_PIXELS=1_000_001.0):
-            parser = RasterisedDocumentParser(None)
-            params = parser.construct_ocrmypdf_parameters("", "", "", "")
-            self.assertIn("max_image_mpixels", params)
-            self.assertAlmostEqual(params["max_image_mpixels"], 1, places=4)
-
-        with override_settings(OCR_MAX_IMAGE_PIXELS=-1_000_001.0):
-            parser = RasterisedDocumentParser(None)
-            params = parser.construct_ocrmypdf_parameters("", "", "", "")
-            self.assertNotIn("max_image_mpixels", params)
-
-    def test_rtl_language_detection(self) -> None:
-        """
-        GIVEN:
-            - File with text in an RTL language
-        WHEN:
-            - Document is parsed
-        THEN:
-            - Text from the document is extracted
-        """
-        parser = RasterisedDocumentParser(None)
-
-        parser.parse(
-            str(self.SAMPLE_FILES / "rtl-test.pdf"),
-            "application/pdf",
-        )
-
-        # OCR output for RTL text varies across platforms/versions due to
-        # bidi controls and presentation forms; normalize before assertion.
-        normalized_text = "".join(
-            char
-            for char in unicodedata.normalize("NFKC", parser.get_text())
-            if unicodedata.category(char) != "Cf" and not char.isspace()
-        )
-
-        self.assertIn("ةرازو", normalized_text)
-        self.assertTrue(
-            any(token in normalized_text for token in ("ةیلخادلا", "الاخليد")),
-        )
-
-    @mock.patch("ocrmypdf.ocr")
-    def test_gs_rendering_error(self, m) -> None:
-        m.side_effect = SubprocessOutputError("Ghostscript PDF/A rendering failed")
-        parser = RasterisedDocumentParser(None)
-
-        self.assertRaises(
-            ParseError,
-            parser.parse,
-            str(self.SAMPLE_FILES / "simple-digital.pdf"),
-            "application/pdf",
-        )
-
-
-class TestParserFileTypes(DirectoriesMixin, FileSystemAssertsMixin, TestCase):
-    SAMPLE_FILES = Path(__file__).parent / "samples"
-
-    def test_bmp(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(str(self.SAMPLE_FILES / "simple.bmp"), "image/bmp")
-        self.assertIsFile(parser.archive_path)
-        self.assertIn("this is a test document", parser.get_text().lower())
-
-    def test_jpg(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(str(self.SAMPLE_FILES / "simple.jpg"), "image/jpeg")
-        self.assertIsFile(parser.archive_path)
-        self.assertIn("this is a test document", parser.get_text().lower())
-
-    def test_heic(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(str(self.SAMPLE_FILES / "simple.heic"), "image/heic")
-        self.assertIsFile(parser.archive_path)
-        self.assertIn("pizza", parser.get_text().lower())
-
-    @override_settings(OCR_IMAGE_DPI=200)
-    def test_gif(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(str(self.SAMPLE_FILES / "simple.gif"), "image/gif")
-        self.assertIsFile(parser.archive_path)
-        self.assertIn("this is a test document", parser.get_text().lower())
-
-    def test_tiff(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(str(self.SAMPLE_FILES / "simple.tif"), "image/tiff")
-        self.assertIsFile(parser.archive_path)
-        self.assertIn("this is a test document", parser.get_text().lower())
-
-    @override_settings(OCR_IMAGE_DPI=72)
-    def test_webp(self) -> None:
-        parser = RasterisedDocumentParser(None)
-        parser.parse(
-            str(self.SAMPLE_FILES / "document.webp"),
-            "image/webp",
-        )
-        self.assertIsFile(parser.archive_path)
-        # Older tesseracts consistently mangle the space between "a webp",
-        # tesseract 5.3.0 seems to do a better job, so we're accepting both
-        self.assertRegex(
-            parser.get_text().lower(),
-            r"this is a ?webp document, created 11/14/2022.",
-        )