add a hash of letter filenames to the dvla zip file name

if we partially retry a day, we would create new zip files, containing
different letters (if some were processed succesfully). We need these
files to have different filenames to earlier zip files so that we can
avoid overwriting log data in zips_sent.

Hashing the filename means that we'll only overwrite if it was the same
file containing the same content.

app/celery/letters_pdf_tasks.py

@@ -2,6 +2,8 @@ import io
 import math
 from datetime import datetime
 from uuid import UUID
+from hashlib import sha512
+from base64 import urlsafe_b64encode
 
 from PyPDF2.utils import PdfReadError
 from botocore.exceptions import ClientError as BotoClientError
@@ -115,14 +117,24 @@ def collate_letter_pdfs_for_day(date=None):
         # since it is triggered mid afternoon.
         date = datetime.utcnow().strftime("%Y-%m-%d")
 
-    letter_pdfs = s3.get_s3_bucket_objects(
-        current_app.config['LETTERS_PDF_BUCKET_NAME'],
-        subfolder=date
+    letter_pdfs = sorted(
+        s3.get_s3_bucket_objects(
+            current_app.config['LETTERS_PDF_BUCKET_NAME'],
+            subfolder=date
+        ),
+        key=lambda letter: letter['Key']
     )
     for i, letters in enumerate(group_letters(letter_pdfs)):
-        # eg NOTIFY.2018-12-31.001.ZIP
-        dvla_filename = 'NOTIFY.{date}.{num:03}.ZIP'.format(date=date, num=i + 1)
         filenames = [letter['Key'] for letter in letters]
+
+        hash = urlsafe_b64encode(sha512(''.join(filenames).encode()).digest())[:20].decode()
+        # eg NOTIFY.2018-12-31.001.Wjrui5nAvObjPd-3GEL-.ZIP
+        dvla_filename = 'NOTIFY.{date}.{num:03}.{hash}.ZIP'.format(
+            date=date,
+            num=i + 1,
+            hash=hash
+        )
+
         current_app.logger.info(
             'Calling task zip-and-send-letter-pdfs for {} pdfs to upload {} with total size {:,} bytes'.format(
                 len(filenames),