Deck generation from the ranked concept list #26

New issue

Open

opened 2026-09-26 13:58:36 +00:00 by belvedere · 0 comments

belvedere commented

2026-09-26 13:58:36 +00:00

Collaborator

Objective: Turn "the first 1000 words" into actual cards, honouring words the user already
knows.

Files:

Create: lib/first_thousand_words/study/deck_generator.ex
Modify: lib/first_thousand_words/study.ex (code interface)

Steps:

Select concepts that have a word in both the source and target language, ordered by
concept rank.
Take the first target_count (default 1000); skip concepts marked known by placement.
Bulk-insert cards with state: :new and due_at: nil.
Idempotent: rely on the (deck_id, concept_id, direction) unique identity.
Report how many cards were created and how many concepts were skipped for missing
translations — a silent shortfall is a bug.

Verify: generate a deck for a seeded pair; assert card count matches the coverage report.

Refs: depends on M2 coverage.

**Objective:** Turn "the first 1000 words" into actual cards, honouring words the user already knows. **Files:** - Create: `lib/first_thousand_words/study/deck_generator.ex` - Modify: `lib/first_thousand_words/study.ex` (code interface) **Steps:** 1. Select concepts that have a word in **both** the source and target language, ordered by concept `rank`. 2. Take the first `target_count` (default 1000); skip concepts marked known by placement. 3. Bulk-insert cards with `state: :new` and `due_at: nil`. 4. Idempotent: rely on the `(deck_id, concept_id, direction)` unique identity. 5. Report how many cards were created and how many concepts were skipped for missing translations — a silent shortfall is a bug. **Verify:** generate a deck for a seeded pair; assert card count matches the coverage report. **Refs:** depends on M2 coverage.

belvedere added this to the M3 — Spaced-repetition study engine milestone

2026-09-26 13:58:36 +00:00

belvedere added the

labels

2026-09-26 13:58:36 +00:00