[ENG-6459][ENG-6637][ENG-6638] Add Create new preprint version button

[futa-ikeda]

• Ticket: ENG-6459 ENG-6637 ENG-6638
• Feature flag: n/a

Todo

• Add Withdrawn preprint detail test

Purpose

• Add a new "Create new version" button to the preprint detail page

Summary of Changes

• Add a new canCreateNewVersion attr to preprint model
  • Only admins can create a new version
  • Only preprints that is the latest version can create a new version
  • Only preprints that are were published at some point (whether they are withdrawn or not, doesn't matter)
• Update tombstone page for preprint versions

Screenshot(s)

• Accepted
  

• Rejected (for pre-moderation providers)
  

• Withdrawn
  

Side Effects

QA Notes

[brianjgeiger]

A lot of TODOs in those tests, but I presume they'll come with later tickets.

[brianjgeiger]

Well those aren't helpful test failure messages

153) [Chrome 131.0] Exam Partition 3 - Acceptance | preprints | detail: Accepted preprint detail page
     global failure: Error

     http://localhost:7357/assets/vendor-bd608453ad69303c53031e33554ea087.js:263465

155) [Chrome 131.0] Exam Partition 3 - Acceptance | preprints | detail: Pre-mod: Rejected preprint detail page
     global failure: Error

     http://localhost:7357/assets/vendor-bd608453ad69303c53031e33554ea087.js:263465

[cslzchen]

Ah, I didn't know withdrawal can have pending and rejected states. I need to check BE to see if we are handling this properly.

[cslzchen]

@futa-ikeda for pre-moderation, "accepted" is correct but for post moderation, it should be "pending" or " accepted". I am not 100% sure how frontend maps backend wording.

See: https://openscience.atlassian.net/browse/ENG-6467. I think FE can probably just use the is_lateset_version without caring about the review state?

[futa-ikeda]

Hmmm, interesting... I see what you are saying about some of the reviews_states for pre-mod vs. post-mod lining up to be the same state (ie: for pre-mod, "reviews_state": "pending" => "is_published": false, but for post-moderation, either "reviews_state": "pending" or "accepted" => "is_published": true). I'm not sure if we can fully ignore the reviews_state flag here though, as we shouldn't be allowing a new version for "pre-moderation and rejected" and "pre-moderation and pending moderator approval", right?

[cslzchen]

Yep, as discussed, let's rely on date_published instead of is_published. There will be a BE fix to make sure is_latest_version checks date_published instead of is_published.

[futa-ikeda]

Well those aren't helpful test failure messages

153) [Chrome 131.0] Exam Partition 3 - Acceptance | preprints | detail: Accepted preprint detail page
     global failure: Error

     http://localhost:7357/assets/vendor-bd608453ad69303c53031e33554ea087.js:263465

155) [Chrome 131.0] Exam Partition 3 - Acceptance | preprints | detail: Pre-mod: Rejected preprint detail page
     global failure: Error

     http://localhost:7357/assets/vendor-bd608453ad69303c53031e33554ea087.js:263465

Turns out it was the citations mirage endpoint failing!