examiner: Replace md5 with sha1 in PDF hasher

examiner/migrations/0007_auto_20181215_2305.py

@@ -0,0 +1,25 @@
+# -*- coding: utf-8 -*-
+# Generated by Django 1.11.2 on 2018-12-15 22:05
+from __future__ import unicode_literals
+
+import django.core.validators
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('examiner', '0006_remove_scrapedpdf_filetype'),
+    ]
+
+    operations = [
+        migrations.RemoveField(
+            model_name='scrapedpdf',
+            name='md5_hash',
+        ),
+        migrations.AddField(
+            model_name='scrapedpdf',
+            name='sha1_hash',
+            field=models.CharField(help_text='Unik sha1 hash relativt til filinnhold.', max_length=40, null=True, unique=True, validators=[django.core.validators.RegexValidator(message='Not a valid SHA1 hash string.', regex='^[0-9a-f]{40}$')]),
+        ),
+    ]

examiner/migrations/0008_auto_20181215_2306.py

@@ -0,0 +1,21 @@
+# -*- coding: utf-8 -*-
+# Generated by Django 1.11.2 on 2018-12-15 22:06
+from __future__ import unicode_literals
+
+import django.core.validators
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('examiner', '0007_auto_20181215_2305'),
+    ]
+
+    operations = [
+        migrations.AlterField(
+            model_name='scrapedpdf',
+            name='sha1_hash',
+            field=models.CharField(help_text='Unik sha1 hash relativt til filinnhold.', max_length=40, unique=True, validators=[django.core.validators.RegexValidator(message='Not a valid SHA1 hash string.', regex='^[0-9a-f]{40}$')]),
+        ),
+    ]

examiner/models.py

@@ -3,7 +3,7 @@
 
 from django.contrib.auth.models import User
 from django.core.files.base import ContentFile
-from django.core.validators import URLValidator
+from django.core.validators import RegexValidator, URLValidator
 from django.db import models
 from django.utils import timezone
 
@@ -29,11 +29,15 @@ class ScrapedPdf(models.Model):
         default=None,
         help_text=_('Filinnhold i rent tekstformat.'),
     )
-    md5_hash = models.CharField(
-        max_length=32,
+    sha1_hash = models.CharField(
+        max_length=40,
         unique=True,
         null=False,
-        help_text=_('Unik md5 hash relativt til filinnhold.'),
+        help_text=_('Unik sha1 hash relativt til filinnhold.'),
+        validators=[RegexValidator(
+            regex='^[0-9a-f]{40}$',
+            message='Not a valid SHA1 hash string.',
+        )],
     )
     created_at = models.DateTimeField(editable=False)
     updated_at = models.DateTimeField()
@@ -182,14 +186,14 @@ def backup_file(self) -> None:
             self.save()
             return
 
-        md5 = hashlib.md5(response.content).hexdigest()
+        sha1 = hashlib.sha1(response.content).hexdigest()
         content_file = ContentFile(response.content)
 
         try:
-            file_backup = ScrapedPdf.objects.get(md5_hash=md5)
+            file_backup = ScrapedPdf.objects.get(sha1_hash=sha1)
         except ScrapedPdf.DoesNotExist:
-            file_backup = ScrapedPdf(md5_hash=md5)
-            file_backup.file.save(name=md5, content=content_file)
+            file_backup = ScrapedPdf(sha1_hash=sha1)
+            file_backup.file.save(name=sha1, content=content_file)
             file_backup.save()
 
         self.scraped_pdf = file_backup

examiner/tests/test_models.py

@@ -123,12 +123,12 @@ def test_file_backup(tmpdir, settings):
 
     # The downloaded file should be hashed and the result stored
     assert exam_url.dead_link is False
-    expected_md5_hash = 'adc7a2fa473be1b091f7324aa4067c8a'
+    expected_sha1_hash = '4dc828ea76ab618be6d72d135af13c40de3b9ce6'
     file_backup = exam_url.scraped_pdf
-    assert file_backup.md5_hash == expected_md5_hash
+    assert file_backup.sha1_hash == expected_sha1_hash
 
     # And the stored file should be named according to its hash
-    assert file_backup.file.name == 'examiner/FileBackup/' + expected_md5_hash
+    assert file_backup.file.name == 'examiner/FileBackup/' + expected_sha1_hash
 
     # The directory for file backups should now contain one file
     backup_directory = Path(settings.MEDIA_ROOT / 'examiner/FileBackup/')
@@ -238,9 +238,9 @@ def test_string_content():
     """FileBackup PDFs should be parsable."""
     pdf_path = Path(__file__).parent / 'data' / 'matmod_exam_des_2017.pdf'
     pdf_content = ContentFile(pdf_path.read_bytes())
-    md5 = 'a8c5b61d8e750db6e719937a251e93b9'
-    pdf_backup = ScrapedPdf(md5_hash=md5)
-    pdf_backup.file.save(md5, content=pdf_content)
+    sha1 = 'a8c5b61d8e750db6e719937a251e93b9'
+    pdf_backup = ScrapedPdf(sha1_hash=sha1)
+    pdf_backup.file.save(sha1, content=pdf_content)
     pdf_backup.read_text()
     pdf_backup.save()