Files
coursebank/CHANGELOG.md
T
alexmandClaude 7af1cd5b7e
Pipeline / check (pull_request) Successful in 13m38s
Pipeline / docs (pull_request) Skipped
Pipeline / nightly (pull_request) Skipped
Pipeline / release (pull_request) Skipped
fix(lecture): leave prerequisites off a lecture's pages
A target taught in one lecture can name a later lecture that builds on
it, and every page printed whatever named the lecture, so L4.2 opened
with five objectives from L3.2, L3.5, and L4.1. A lecture's objectives,
readings, and page now show an entry only when the lecture lists its
objective in teaches, or when no earlier lecture covers it. The
lectures: link is unchanged, so assembly, coverage, and the reports'
review lists still read it.

Reading prose that cites an objective or target from another lecture
now renders as its number on that lecture's page (L3.5's T 1) rather
than the target's full text. The trailing (T ...) list stays this
page's own.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-16 13:00:24 -05:00

4.4 KiB

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to calendar versioning (YY.MM.PATCH); see README.md.

[Unreleased]

Added

  • First public release.
  • Student reports add bonus questions and curves to the score and show how the total adds up, so it matches Gradescope and Canvas. Class means, bands, and grade distributions use the same total.
  • A zero-point Gradescope question with no option columns is recorded as a curve.
  • ingest warns when a question dropped as removed still gives every student points on the platform.
  • Schema 2.1, which is additive: a course that validated under 2.0 validates with no new errors. Rerun coursebank schema to pick up the new and changed JSON Schemas, including the new glossary.schema.json.
    • Four registries beside objectives and targets: terms (term-), equations (eq-), values (val-), and misconceptions (mis-), each defined once in an objective file, a new glossary.yaml, or course.yaml. Ids are unique across every registry.
    • Targets list the entries they turn on (terms, equations, values, misconceptions), carry a student-facing explanation, and keep the item-writing judgments in authoring.
    • Objectives and lectures gain a description, lectures gain materials, readings gain minutes, and bank options gain misconception_id.
    • policy gains levels (the course's names for levels 1 to 5), exam_conditions, blueprints, and target_text_forbid. No command reads exam_conditions or blueprints yet.
    • Brace references such as {term-gap-penalty} in student-facing prose render as links, values, quoted beliefs, or target numbers.
    • authoring, notes_private, evidence, conflicts, and status never reach a student page, report, or export.
  • validate checks the registries: id prefixes, ids unique across registries, the entries each target lists, equation symbols, values, citation keys, brace references, that every regular expression compiles, and each option's misconception_id.
  • Fourteen lint rules in a glossary family, from glossary-unused to glossary-proposed, and a note severity for the two that report conventions rather than problems.
  • lecture page ID writes a whole lecture page: description, objectives and targets with explanations and named ceilings, new and earlier terms, equations, numbers to know, common mistakes, readings with minutes, and materials. lecture glossary writes the course glossary page.
  • Student reports show what meeting each missed target looks like, the correction for a registry misconception, and group misses that share one.
  • migrate glossary --terms terms.yaml --values values.yaml brings 1.0 side files into glossary.yaml, reports what it renamed and dropped, and is a no-op when run again.

Changed

  • lint --min-severity defaults to note, and lint fails only on findings above a note.
  • Lecture pages (lecture objectives, lecture readings, lecture page) leave off prerequisites. An objective or target covered in an earlier lecture is no longer printed on a later lecture just because it lists that lecture in lectures:, unless the later lecture lists its objective in teaches. The lectures: link itself is unchanged, so assembly, coverage, and the reports' review lists still use it. Target numbers on those pages close up accordingly.
  • Reading prose that cites an objective or target from another lecture renders as its number on that lecture's page, such as L3.5's T 1, instead of the full text of the target.
  • The crate depends on regex, built without its optional features.

Fixed

  • A record with points: 0 on a bonus question no longer erases the points the platform awarded.
  • A curve whose position matches a recorded question number is no longer discarded at ingest.
  • Gradescope files named with the question's title, such as 39_Curve.csv, are now read. They used to be skipped without a word, which left titled questions and curves out of the totals. Any other CSV file in the directory now draws a warning, and two files for one question are refused. Re-ingest to pick this up.
  • The assessment JSON Schema now accepts dropped_as, dropped_before_printing, stem_digest, and learning_objectives on a placement, all of which the loader already read. Rerun coursebank schema to pick this up.