Merge branch 'export-content-type' into 'master'

Add an SQLite content type to uploaded exports

See merge request !1379

arkindex/documents/export/__init__.py

@@ -4,9 +4,11 @@ import os
 import sqlite3
 import tempfile
 import uuid
+from datetime import datetime, timezone
 from pathlib import Path
 
 from django.db import connections
+from django.utils.text import slugify
 from django_rq import job
 from rq import get_current_job
 
@@ -119,7 +121,17 @@ def export_corpus(corpus_export: CorpusExport) -> None:
         if rq_job:
             rq_job.set_description(f'Uploading export for corpus {corpus_export.corpus.name}')
             rq_job.set_progress(0.0)
-        corpus_export.s3_object.upload_file(db_path, Callback=progress_callback)
+
+        # Set the filename in a HTTP header to make downloaded exports nicer
+        filename = f'{slugify(corpus_export.corpus.name)[:100]}-{datetime.now(timezone.utc).strftime("%Y%m%d-%H%M%S")}.sqlite'
+        corpus_export.s3_object.upload_file(
+            db_path,
+            Callback=progress_callback,
+            ExtraArgs={
+                'ContentType': 'application/vnd.sqlite3',
+                'ContentDisposition': f'attachment; filename="{filename}"'
+            },
+        )
 
         corpus_export.state = CorpusExportState.Done
         corpus_export.save()

arkindex/documents/tests/tasks/test_export.py

 import json
 import os
 import sqlite3
+from datetime import datetime, timezone
 from unittest.mock import call, patch
 
 from arkindex.dataimport.models import WorkerVersion
@@ -21,9 +22,12 @@ from arkindex.project.tests import FixtureTestCase
 
 class TestExport(FixtureTestCase):
 
+    @patch('arkindex.documents.export.datetime')
     @patch('arkindex.documents.export.os.unlink')
     @patch('arkindex.project.aws.s3.Object')
-    def test_export(self, s3_object_mock, unlink_mock):
+    def test_export(self, s3_object_mock, unlink_mock, datetime_mock):
+        datetime_mock.now.return_value = datetime(2042, 1, 1, 12, 34, 56, tzinfo=timezone.utc)
+
         element = self.corpus.elements.get(name='Volume 1')
         transcription = Transcription.objects.first()
         version = WorkerVersion.objects.get(worker__slug='reco')
@@ -68,8 +72,12 @@ class TestExport(FixtureTestCase):
 
         self.assertEqual(s3_object_mock().upload_file.call_count, 1)
         args, kwargs = s3_object_mock().upload_file.call_args
-        self.assertCountEqual(kwargs.keys(), {'Callback'})
+        self.assertCountEqual(kwargs.keys(), {'Callback', 'ExtraArgs'})
         self.assertTrue(callable(kwargs['Callback']))
+        self.assertDictEqual(kwargs['ExtraArgs'], {
+            'ContentType': 'application/vnd.sqlite3',
+            'ContentDisposition': 'attachment; filename="unit-tests-20420101-123456.sqlite"'
+        })
         self.assertEqual(len(args), 1)
         # Retrieve the database path from the S3 upload argument
         db_path = args[0]