thiemowmde added a comment.

This proposal changes the code path from "Entity > Fingerprint > TermType > 
Language" to "Entity > Language > TermType". Which is, as far as I can see, the 
code path that's way, way more common in our code base.

The serialization should, in my opinion, stay as it is. Which is "Entity > 
TermType > Language", right? Why change it, if it isn't broken? We didn't 
changed it when "Fingerprint" was introduced, right? So why change it now?

The question how this will affect the (de)serializers is, indeed, a very good 
one. I think it won't be a problem. We will replace the current 
`FingerprintSerializer` with something like a `EntityTermSetSerializer` which 
will do more or less the same. Most important difference: If we ever want to 
(for example) deserialize labels without deserializing descriptions and 
aliases, this will still be possible, but different. Instead of returning a set 
of Term objects for all labels, this deserializer will return a set of 
EntityTerm objects where the description and aliases part is either null or 
somehow marked as "not deserialized" (could use `false` as a special value for 
that, for example). This will require more memory (for the two `false` 
properties).


TASK DETAIL
  https://phabricator.wikimedia.org/T87237

REPLY HANDLER ACTIONS
  Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign 
<username>.

EMAIL PREFERENCES
  https://phabricator.wikimedia.org/settings/panel/emailpreferences/

To: thiemowmde
Cc: adrianheine, Lydia_Pintscher, JeroenDeDauw, Aklapper, thiemowmde, 
Wikidata-bugs, aude



_______________________________________________
Wikidata-bugs mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs

Reply via email to