Ignore null revisions when checking if a non-admin can overwrite a redirect. They can only do so if the redirect has a single revision, and additional null revisions should not be counted.
Description
Status | Subtype | Assigned | Task | ||
---|---|---|---|---|---|
Open | Feature | None | T72239 Generate entries for deletions and restores in page histories | ||
Open | None | T211504 Ignore null revisions when checking if a non-admin can overwrite a redirect |
Event Timeline
So what's going to happen to these null revisions? Will they just get deleted? What happens if the move is reversed; should those deleted null revisions get automatically restored, or should a sysop have to come along and restore them?
Yes, they will just get deleted, along the original revision with the original redirect, This is what currently happens if there is only a single revision with the redirect and no intermediate null revisions.
There's no need to restore them, if you leave the redirect when renaming the page back to the original name.
Maybe it really is time to just get rid of null revisions and instead intersperse log entries with revisions in action=history. (See T217370.) That would eliminate the need for this task. The fact that we're willing to just delete null revisions and not restore them (in contrast to how it is with redirects, where if a move is reversed, the redirect is usually put back in place) shows they're pretty expendable.