searching including duplicates filters on the status of the 'duplicate' rather than the 'duplicate-of' bug

If I ask to search including duplicates then for a bug marked as a duplicate the status of the bug marked as duplicate is used to determine whether to include it in the results.

This makes it useless to search (a) including duplicate bugs but (b) not including every possible status. For example:

https://edge.launchpad.net/ubuntu/+bugs?field.searchtext=autopkgtest&orderby=datecreated&search=Search&field.status%3Alist=NEW&field.status%3Alist=INCOMPLETE_WITH_RESPONSE&field.status%3Alist=INCOMPLETE_WITHOUT_RESPONSE&field.status%3Alist=INVALID&field.status%3Alist=WONTFIX&field.status%3Alist=CONFIRMED&field.status%3Alist=TRIAGED&field.status%3Alist=INPROGRESS&field.status%3Alist=FIXCOMMITTED&assignee_option=any&field.assignee=&field.bug_reporter=&field.bug_contact=&field.bug_commenter=&field.subscriber=&field.component-empty-marker=1&field.status_upstream-empty-marker=1&field.omit_dupes.used=&field.has_patch.used=&field.tag=&field.has_cve.used=&field.has_no_package.used=
 includes 127191. 127191 is a duplicate of 125231 which is Fix Released, so this bug ought not to have been listed. It is included because 127191 itself is New but this latter state is (a) not useful informative or in any way related to reality and (b) not possible to modify.

As a workaround to make my autopkgtest bug triage at all reasonable I will have to 1. unduplicate the bug; 2. set it to Fix Released; 3. duplicate it again. This is obviously not a good workflow.

The correct fix is to use the status of the duplicate target when searching. Of course a better fix would be to rework the model as requested in 52613. See also bug 3796.