763946 - [shipping] sign-off view skips pushes that are older than the fall-back sign-off

Status: RESOLVED FIXED

(Webtools Graveyard :: Elmo, defect)

Description

[Axel Hecht [:Pike]]

Just came across this on sk and fx15, the fallback sign-off has a newer push-date than the pushes on the next appversion.

Resolution is to only cut-off at push-date, if we're not falling back.

Also a regression from data1.1, bug 650816

Comment 1

[Axel Hecht [:Pike]]

Attachment: don't cut at current_so.push.push_date if there's a fallback

This one was easy, just needed to pass in the fallback data, too, and only cut at push_date if there's no fallback, i.e., the current signoff is on the current tree flow, and thus, you're really showing the pushes in question.

Comment 2

[Peter Bengtsson [:peterbe]]

Comment on attachment 632281 [details] [diff] [review]
don't cut at current_so.push.push_date if there's a fallback

Review of attachment 632281 [details] [diff] [review]:
-----------------------------------------------------------------

::: apps/shipping/api.py
@@ +294,5 @@
>      if Action.REJECTED in flags and len(initial_diff) < 2:
>          initial_diff.append(action4id[flags[Action.REJECTED]].signoff_id)
> +    # if we're having a sign-off on this appversion, i.e no fallback,
> +    # show only new pushes
> +    if current_so is not None and fallback is None:

Why not just:

`if current_so and not fallback:`

Comment 3

[[github robot]]

Commit pushed to develop at https://github.com/mozilla/elmo

https://github.com/mozilla/elmo/commit/be1330e18e628133c62e9d76f5aa284804616f22
bug 763946, don't cut pushes to new ones if we're having a fallback, r=peterbe

Comment 4

[Axel Hecht [:Pike]]

FIXED and on production.

To copy from irc, I tend to like reading is None and is not None, I also like the specific difference against other boolean false values.