add support for suggestions

src/documents/tests/test_api.py

@@ -590,6 +590,10 @@ class TestDocumentApi(DirectoriesMixin, APITestCase):
         self.assertEqual(len(meta['original_metadata']), 0)
         self.assertGreater(len(meta['archive_metadata']), 0)
 
+    def test_get_metadata_invalid_doc(self):
+        response = self.client.get(f"/api/documents/34576/metadata/")
+        self.assertEqual(response.status_code, 404)
+
     def test_get_metadata_no_archive(self):
         doc = Document.objects.create(title="test", filename="file.pdf", mime_type="application/pdf")
 
@@ -605,6 +609,30 @@ class TestDocumentApi(DirectoriesMixin, APITestCase):
         self.assertGreater(len(meta['original_metadata']), 0)
         self.assertIsNone(meta['archive_metadata'])
 
+    def test_get_empty_suggestions(self):
+        doc = Document.objects.create(title="test", mime_type="application/pdf")
+
+        response = self.client.get(f"/api/documents/{doc.pk}/suggestions/")
+
+        self.assertEqual(response.status_code, 200)
+        self.assertEqual(response.data, {'correspondents': [], 'tags': [], 'document_types': []})
+
+    def test_get_suggestions_invalid_doc(self):
+        response = self.client.get(f"/api/documents/34676/suggestions/")
+        self.assertEqual(response.status_code, 404)
+
+    @mock.patch("documents.views.match_correspondents")
+    @mock.patch("documents.views.match_tags")
+    @mock.patch("documents.views.match_document_types")
+    def test_get_suggestions(self, match_document_types, match_tags, match_correspondents):
+        doc = Document.objects.create(title="test", mime_type="application/pdf", content="this is an invoice!")
+        match_tags.return_value = [Tag(id=56), Tag(id=123)]
+        match_document_types.return_value = [DocumentType(id=23)]
+        match_correspondents.return_value = [Correspondent(id=88), Correspondent(id=2)]
+
+        response = self.client.get(f"/api/documents/{doc.pk}/suggestions/")
+        self.assertEqual(response.data, {'correspondents': [88,2], 'tags': [56,123], 'document_types': [23]})
+
     def test_saved_views(self):
         u1 = User.objects.create_user("user1")
         u2 = User.objects.create_user("user2")

src/documents/views.py

@@ -34,6 +34,7 @@ from rest_framework.viewsets import (
 import documents.index as index
 from paperless.db import GnuPG
 from paperless.views import StandardPagination
+from .classifier import DocumentClassifier, IncompatibleClassifierVersionError
 from .filters import (
     CorrespondentFilterSet,
     DocumentFilterSet,
@@ -41,6 +42,7 @@ from .filters import (
     DocumentTypeFilterSet,
     LogFilterSet
 )
+from .matching import match_correspondents, match_tags, match_document_types
 from .models import Correspondent, Document, Log, Tag, DocumentType, SavedView
 from .parsers import get_parser_class_for_mime_type
 from .serialisers import (
@@ -225,31 +227,58 @@ class DocumentViewSet(RetrieveModelMixin,
     def metadata(self, request, pk=None):
         try:
             doc = Document.objects.get(pk=pk)
-
-            meta = {
-                "original_checksum": doc.checksum,
-                "original_size": os.stat(doc.source_path).st_size,
-                "original_mime_type": doc.mime_type,
-                "media_filename": doc.filename,
-                "has_archive_version": os.path.isfile(doc.archive_path),
-                "original_metadata": self.get_metadata(
-                    doc.source_path, doc.mime_type)
-            }
-
-            if doc.archive_checksum and os.path.isfile(doc.archive_path):
-                meta['archive_checksum'] = doc.archive_checksum
-                meta['archive_size'] = os.stat(doc.archive_path).st_size,
-                meta['archive_metadata'] = self.get_metadata(
-                    doc.archive_path, "application/pdf")
-            else:
-                meta['archive_checksum'] = None
-                meta['archive_size'] = None
-                meta['archive_metadata'] = None
-
-            return Response(meta)
         except Document.DoesNotExist:
             raise Http404()
 
+        meta = {
+            "original_checksum": doc.checksum,
+            "original_size": os.stat(doc.source_path).st_size,
+            "original_mime_type": doc.mime_type,
+            "media_filename": doc.filename,
+            "has_archive_version": os.path.isfile(doc.archive_path),
+            "original_metadata": self.get_metadata(
+                doc.source_path, doc.mime_type)
+        }
+
+        if doc.archive_checksum and os.path.isfile(doc.archive_path):
+            meta['archive_checksum'] = doc.archive_checksum
+            meta['archive_size'] = os.stat(doc.archive_path).st_size,
+            meta['archive_metadata'] = self.get_metadata(
+                doc.archive_path, "application/pdf")
+        else:
+            meta['archive_checksum'] = None
+            meta['archive_size'] = None
+            meta['archive_metadata'] = None
+
+        return Response(meta)
+
+    @action(methods=['get'], detail=True)
+    def suggestions(self, request, pk=None):
+        try:
+            doc = Document.objects.get(pk=pk)
+        except Document.DoesNotExist:
+            raise Http404()
+
+        try:
+            classifier = DocumentClassifier()
+            classifier.reload()
+        except (OSError, EOFError, IncompatibleClassifierVersionError) as e:
+            logging.getLogger(__name__).warning(
+                "Cannot load classifier: Not providing auto matching "
+                "suggestions"
+            )
+            classifier = None
+
+        return Response({
+            "correspondents": [
+                c.id for c in match_correspondents(doc, classifier)
+            ],
+            "tags": [t.id for t in match_tags(doc, classifier)],
+            "document_types": [
+                dt.id for dt in match_document_types(doc, classifier)
+            ]
+        })
+
     @action(methods=['get'], detail=True)
     def preview(self, request, pk=None):
         try: