However, we need a JIRA for platform too, so that the removal of empty segments is done at the level of AbstractBlockAsyncRenderer#createId() or even beyond (simplest) and then later move to a HASH for ex.
We already have XWIKI-23439 - while I describe a different cause (and possible solution) there, I believe it's the right issue as the described bug is exactly what we would be fixing.
At this point, I have no idea if the use of a cache is really an issue in practice or not. I still believe it could be an issue, but I'm not convinced that this is causing real issues in practice - so I would rather suggest to create a separate improvement issue for that.
This message was sent by Atlassian Jira (v9.3.0#930000-sha1:287aeb6)
If image attachments aren't displayed, see this article.
