initial translations support using deepl

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

@@ -199,6 +199,10 @@
             <div>
               <textarea class="form-control" id="content" rows="20" formControlName='content' [class.rtl]="isRTL"></textarea>
             </div>
+            <h3>Translation</h3>
+             <div>
+              <textarea class="form-control" id="translation" rows="20" formControlName='translation' [class.rtl]="isRTL"></textarea>
+            </div>
           </ng-template>
         </li>
 

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

@@ -86,6 +86,7 @@ const doc: Document = {
   storage_path: 31,
   tags: [41, 42, 43],
   content: 'text content',
+  translation: 'text content',
   added: new Date('May 4, 2014 03:24:00'),
   created: new Date('May 4, 2014 03:24:00'),
   modified: new Date('May 4, 2014 03:24:00'),

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

@@ -138,6 +138,7 @@ export class DocumentDetailComponent
   documentForm: FormGroup = new FormGroup({
     title: new FormControl(''),
     content: new FormControl(''),
+    translation: new FormControl(''),
     created_date: new FormControl(),
     correspondent: new FormControl(),
     document_type: new FormControl(),
@@ -404,6 +405,7 @@ export class DocumentDetailComponent
           this.store = new BehaviorSubject({
             title: doc.title,
             content: doc.content,
+            translation: doc.translation,
             created_date: doc.created_date,
             correspondent: doc.correspondent,
             document_type: doc.document_type,

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

@@ -32,6 +32,8 @@ export interface Document extends ObjectWithPermissions {
 
   content?: string
 
+  translation?: string
+
   tags$?: Observable<Tag[]>
 
   tags?: number[]

src/documents/filters.py

@@ -104,7 +104,11 @@ class InboxFilter(Filter):
 class TitleContentFilter(Filter):
     def filter(self, qs, value):
         if value:
-            return qs.filter(Q(title__icontains=value) | Q(content__icontains=value))
+            return qs.filter(
+                Q(title__icontains=value)
+                | Q(translation__icontains=value)
+                | Q(content__icontains=value),
+            )
         else:
             return qs
 

src/documents/migrations/1047_document_translation.py

@@ -0,0 +1,23 @@
+# Generated by Django 4.2.11 on 2024-04-11 18:24
+
+from django.db import migrations
+from django.db import models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("documents", "1046_workflowaction_remove_all_correspondents_and_more"),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name="document",
+            name="translation",
+            field=models.TextField(
+                blank=True,
+                help_text="The translated version of the content field. This field can also be used for searching.",
+                verbose_name="translation",
+            ),
+        ),
+    ]

src/documents/models.py

@@ -176,6 +176,15 @@ class Document(ModelWithOwner):
         ),
     )
 
+    translation = models.TextField(
+        _("translation"),
+        blank=True,
+        help_text=_(
+            "The translated version of the content field. This field can "
+            "also be used for searching.",
+        ),
+    )
+
     mime_type = models.CharField(_("mime type"), max_length=256, editable=False)
 
     tags = models.ManyToManyField(

src/documents/serialisers.py

@@ -772,6 +772,7 @@ class DocumentSerializer(
             "storage_path",
             "title",
             "content",
+            "translation",
             "tags",
             "created",
             "created_date",

src/documents/tasks.py

@@ -6,6 +6,7 @@ from pathlib import Path
 from tempfile import TemporaryDirectory
 from typing import Optional
 
+import requests
 import tqdm
 from celery import Task
 from celery import shared_task
@@ -49,6 +50,29 @@ if settings.AUDIT_LOG_ENABLED:
 logger = logging.getLogger("paperless.tasks")
 
 
+def translate_content(content):
+
+    headers = {
+        "Authorization": "DeepL-Auth-Key " + settings.DEEPL_TOKEN,
+        "Content-Type": "application/json",
+    }
+
+    json_data = {
+        "text": [
+            content,
+        ],
+        "target_lang": settings.TRANSLATION_TARGET,
+    }
+
+    response = requests.post(
+        "https://api-free.deepl.com/v2/translate",
+        headers=headers,
+        json=json_data,
+    )
+
+    return response.json()["translations"][0]["text"]
+
+
 @shared_task
 def index_optimize():
     ix = index.open_index()
@@ -243,9 +267,11 @@ def update_document_archive_file(document_id):
                     archive_filename=True,
                 )
                 oldDocument = Document.objects.get(pk=document.pk)
+                content = parser.get_text()
                 Document.objects.filter(pk=document.pk).update(
                     archive_checksum=checksum,
-                    content=parser.get_text(),
+                    content=content,
+                    translation=translate_content(content),
                     archive_filename=document.archive_filename,
                 )
                 newDocument = Document.objects.get(pk=document.pk)