More comment, to-do

core/views/uploads.py

@@ -64,9 +64,13 @@ def logbookedit(request, year=None, slug=None):
     """Edit a logbook entry
     This is daft: we have the parsed identity of the person and we render it to text as 'fullname', to be re-parsed on re-importing.
     And there is no guarantee that this will be the same thing, esp. as aliases are used in the initial data input.
+    
     So we are losing all the cute aliases that have been used over the years by this export/re-import process. Bother.
     But they have already been lost in the Great Format Conversion of 2022-23 when everything was chnaged to use the same HTML parser.
     Which is a shame.
+    Fix is to add "alias_used" as a field in class PersonLogEntry, so that we can preserve
+    all those cute names. But it's rather a large manual effort (with some scripting) to recover the aliases from the original logbook
+    html files which are now only in the git history. Bother. Very sorry.
     """
     def clean_tu(tu):
         if tu =="":
@@ -81,8 +85,10 @@ def logbookedit(request, year=None, slug=None):
         """This gives each logbook entry a unique id based on the date+content, so the order of entries on a particular day
         does not matter. This is a change (August 2023) from previous process.
         Otherwise we could get 2023-07-20a and 2023-07-20b swapped on exporting and re-importing logbooks
-        because the database does not record precendence.
-        2 hex digits would seem adequate for each expo day, but we might get a collision..
+        because the database does not record precedence.
+        2 hex digits would seem adequate for each expo day, but we might get a collision.
+        The hash is based on the content after substitution of <p> so should be stable. Which means these ids
+        can be used elsewhere in the troggle system as permanent slugs.
         """
         sha.update(text.encode('utf-8'))
         return sha.hexdigest()[0:n]