Feature: Switches to a new client to handle communication with Gotenberg (#4391)

Switches to a new client to handle communication with Gotenberg for merging and generating PDFs

src/paperless_tika/parsers.py

@@ -1,9 +1,10 @@
-import os
 from pathlib import Path
 
 import httpx
 from django.conf import settings
 from django.utils import timezone
+from gotenberg_client import GotenbergClient
+from gotenberg_client.options import PdfAFormat
 from tika_client import TikaClient
 
 from documents.parsers import DocumentParser
@@ -80,47 +81,33 @@ class TikaDocumentParser(DocumentParser):
 
         self.archive_path = self.convert_to_pdf(document_path, file_name)
 
-    def convert_to_pdf(self, document_path, file_name):
-        pdf_path = os.path.join(self.tempdir, "convert.pdf")
-        gotenberg_server = settings.TIKA_GOTENBERG_ENDPOINT
-        url = gotenberg_server + "/forms/libreoffice/convert"
+    def convert_to_pdf(self, document_path: Path, file_name):
+        pdf_path = Path(self.tempdir) / "convert.pdf"
 
         self.log.info(f"Converting {document_path} to PDF as {pdf_path}")
-        with open(document_path, "rb") as document_handle:
-            files = {
-                "files": (
-                    "convert" + os.path.splitext(document_path)[-1],
-                    document_handle,
-                ),
-            }
-            headers = {}
-            data = {}
 
+        with GotenbergClient(
+            host=settings.TIKA_GOTENBERG_ENDPOINT,
+            timeout=settings.CELERY_TASK_TIME_LIMIT,
+        ) as client, client.libre_office.to_pdf() as route:
             # Set the output format of the resulting PDF
-            # Valid inputs: https://gotenberg.dev/docs/modules/pdf-engines#uno
             if settings.OCR_OUTPUT_TYPE in {"pdfa", "pdfa-2"}:
-                data["pdfFormat"] = "PDF/A-2b"
+                route.pdf_format(PdfAFormat.A2b)
             elif settings.OCR_OUTPUT_TYPE == "pdfa-1":
-                data["pdfFormat"] = "PDF/A-1a"
+                route.pdf_format(PdfAFormat.A1a)
             elif settings.OCR_OUTPUT_TYPE == "pdfa-3":
-                data["pdfFormat"] = "PDF/A-3b"
+                route.pdf_format(PdfAFormat.A3b)
+
+            route.convert(document_path)
 
             try:
-                response = httpx.post(
-                    url,
-                    files=files,
-                    headers=headers,
-                    data=data,
-                    timeout=settings.CELERY_TASK_TIME_LIMIT,
-                )
-                response.raise_for_status()  # ensure we notice bad responses
+                response = route.run()
+
+                pdf_path.write_bytes(response.content)
+
+                return pdf_path
+
             except Exception as err:
                 raise ParseError(
                     f"Error while converting document to PDF: {err}",
                 ) from err
-
-        with open(pdf_path, "wb") as file:
-            file.write(response.content)
-            file.close()
-
-        return pdf_path

src/paperless_tika/tests/test_tika_parser.py

@@ -2,12 +2,11 @@ import datetime
 import os
 import zoneinfo
 from pathlib import Path
-from unittest import mock
 
 from django.test import TestCase
 from django.test import override_settings
-from httpx import Request
-from httpx import Response
+from httpx import codes
+from httpx._multipart import DataField
 from rest_framework import status
 
 from documents.parsers import ParseError
@@ -95,8 +94,7 @@ class TestTikaParser(HttpxMockMixin, TestCase):
         with self.assertRaises(ParseError):
             self.parser.convert_to_pdf(file, None)
 
-    @mock.patch("paperless_tika.parsers.httpx.post")
-    def test_request_pdf_a_format(self, post: mock.Mock):
+    def test_request_pdf_a_format(self):
         """
         GIVEN:
             - Document needs to be converted to PDF
@@ -108,10 +106,6 @@ class TestTikaParser(HttpxMockMixin, TestCase):
         file = Path(os.path.join(self.parser.tempdir, "input.odt"))
         file.touch()
 
-        response = Response(status_code=status.HTTP_200_OK)
-        response.request = Request("POST", "/somewhere/")
-        post.return_value = response
-
         for setting, expected_key in [
             ("pdfa", "PDF/A-2b"),
             ("pdfa-2", "PDF/A-2b"),
@@ -119,11 +113,20 @@ class TestTikaParser(HttpxMockMixin, TestCase):
             ("pdfa-3", "PDF/A-3b"),
         ]:
             with override_settings(OCR_OUTPUT_TYPE=setting):
+                self.httpx_mock.add_response(
+                    status_code=codes.OK,
+                    content=b"PDF document",
+                    method="POST",
+                )
+
                 self.parser.convert_to_pdf(file, None)
 
-                post.assert_called_once()
-                _, kwargs = post.call_args
+                request = self.httpx_mock.get_request()
+                found = False
+                for field in request.stream.fields:
+                    if isinstance(field, DataField) and field.name == "pdfFormat":
+                        self.assertEqual(field.value, expected_key)
+                        found = True
+                self.assertTrue(found)
 
-                self.assertEqual(kwargs["data"]["pdfFormat"], expected_key)
-
-                post.reset_mock()
+                self.httpx_mock.reset(assert_all_responses_were_requested=False)