Leave answerless legacy input questions out of Perseus archives

[rtibblesbot]

Summary

• Perseus archive drops answerless legacy input questions (previously shipped as unsatisfiable numeric-input) from item files and exercise.json, with a publish warning — same rule as the QTI package.
• Without a QTI package, AssessmentMetaData takes item ids from the Perseus archive, so Kolibri never requests a dropped item.

References

Fixes #6210. Follows: #6209. Related: #6208.

Reviewer guidance

• Mastery n/m still count the dropped question, as the QTI path does; Kolibri's do_all reads the item ids, so it is unaffected.
• restore_channel ([QTI] Restoring a published channel loses or breaks its QTI exercises #6211, Restore QTI exercises when restoring a channel #6243) rebuilds items from exercise.json, so it no longer restores dropped questions; on unstable they come back for the author to fix.
• {"answer": "5", "correct": false} is now dropped, though Perseus rendered it as correct.
• Exercises only: unit Perseus archives keep these questions, since the unit's pre/post test lists every id.

AI usage

Claude Code implemented the fix and tests from a reviewed plan; verified with the full Python test suite and pre-commit.

🤖 Generated with Claude Code

Committed separately: no commit on this branch introduced the lines these changes touch, so they are a new commit rather than folded into the work they amend:

• contentcuration/contentcuration/tests/test_exportchannel.py
• contentcuration/contentcuration/utils/assessment/perseus.py

---

Written by rtibblesbot, an LLM-based coding agent.

🟡 Waiting for feedback Last updated: 2026-10-05 22:59 UTC

[rtibblesbot]

Outside this PR's scope — tracked under #5998:

• [QTI] A unit's published pre/post test lists an input question with no answer that its QTI package leaves out #6263 [QTI] A unit's published pre/post test lists an input question with no answer that its QTI package leaves out

---

@rtibblesbot's comments are generated by an LLM, and should be evaluated accordingly

How was this generated?

• Ran pre-flight CI checks (lint, format, tests) and verified all pass
• Rebased onto the target branch and resolved any conflicts
• Reorganized commit history into clean, logical commits
• Audited the diff to ensure only issue-relevant files are changed
• Built PR body from the repository's PR template with evidence blocks

[rtibblesbot]

Outside this PR's scope — tracked under #5998:

• Perseus export of a legacy input question drops JSON-number answers and accepts answers marked incorrect #6266 Perseus export of a legacy input question with a JSON-number answer has no accepted answer

---

@rtibblesbot's comments are generated by an LLM, and should be evaluated accordingly

How was this generated?

• Ran pre-flight CI checks (lint, format, tests) and verified all pass
• Rebased onto the target branch and resolved any conflicts
• Reorganized commit history into clean, logical commits
• Audited the diff to ensure only issue-relevant files are changed
• Built PR body from the repository's PR template with evidence blocks