Restructure chapters and teach ownership as a spiral#31
Merged
Conversation
Reorders the course to the target layout and splits the single ownership chapter into a spiral: a short seed in the strings chapter, a Moves & Copy chapter right after strings, a Borrowing & references chapter right after functions, and a late Memory & ownership chapter that consolidates the model and adds a gentle lifetimes intuition. Also moves the ? operator up to sit right after Result and keeps it to a single error type per exercise (the multi-error file exercise is now a parse-only sum). The Box<dyn Error> material moves to the env-file parser, where reading a file and parsing it genuinely mixes error types. Smart pointers and the password validator are now bonus chapters at the end. Other changes: - Strings intro gets the 30-second ownership rules and points at the new chapters. - Downstream chapters (functions, vectors, hashmaps, tuples, result, conditionals) updated so ownership/borrowing read as already-taught instead of "later", and stale numeric chapter references are now name-based. - Dropped two banned tics (the "lightbulb moment" line in functions and the "magpie" opener in the password validator). - Appendix gains a short "why this course is built this way" page with the comparison table and a future-directions note. - Renamed chapter dirs in both examples/ and solutions/, regenerated every main.rs, and updated the deliberately-broken list in check-examples.sh. Note: renaming chapter directories changes the DB exercise_name stems, so progress on the dev/restructure database resets. Production is on v1, so it's unaffected. Local CI is green: fmt, clippy, tests, check-examples, check-solutions (75/75), typos, lychee.
Add a comment showing the format! one-liner and what it expands to (a fresh String buffer via write_fmt), contrasted with the in-place push_str version.
Rewrites submissions.exercise_name from the current main chapter names to the new restructured layout (prefix shifts, the ownership split into moves/borrowing, the ? operator move, and the sum_numbers_in_file rename). Preserves all rows (no DELETE), idempotent, validated against a scratch SQLite DB. Runs automatically at server startup via sqlx.
Mapped every key in analysis/course.db (58 participants, 1359 rows) straight to its final restructured home, including the per-chapter step renumberings from the content PRs that the snapshot predates. Exact-match UPDATEs only, so it's collision-free and order-independent. Validated against a copy of the snapshot: all 73 surviving distinct keys land on real exercise files; the only unmapped rows (~180) are on exercises that were removed (count_items, countdown, fallback, frequent_words, char_classes, advisor, status, records, number_to_string) and are left in place, not deleted.
The build context excludes .git, so build.rs couldn't read git info in the image and the footer showed 'unknown'. build.rs now prefers GIT_BRANCH / GIT_HASH from the environment (and Coolify's auto-injected SOURCE_COMMIT for the short hash), falling back to git, then 'unknown'. The Dockerfile declares these as build args and exports them as env before the build. Verified the override lands in the binary.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This reorders the course to the layout we planned and changes how ownership is taught.
Instead of one big ownership chapter, ownership is now a spiral. The strings chapter plants the 30-second version (one owner, dropped at end of scope, borrow without taking). A new Moves & Copy chapter comes right after strings, where the first move actually bites. A new Borrowing & references chapter comes right after functions, where you start wanting to pass a value without giving it up. The old ownership chapter becomes a late Memory & ownership chapter that ties the three rules together and adds a gentle lifetimes intuition (no syntax, just the shape of the error). Its exercises were donated to the two early chapters, so nothing is lost.
The ? operator moves up to sit right after Result, where it belongs. To keep it focused, every exercise there now uses a single error type (the old multi-error file exercise became a parse-only sum). The Box material moves to the env-file parser, which is the first place that genuinely mixes file I/O and parse errors. Smart pointers and the password validator are now bonus chapters at the end.
Along the way I fixed the cross-references that the reorder broke (ownership and borrowing now read as already-taught rather than "later"), turned the remaining numeric "chapter N" mentions into names, dropped two banned tics (the "lightbulb moment" line and the "magpie" opener), and added a short appendix page explaining why the course is built this way.
On existing students' progress: migration 010 remaps submissions.exercise_name to the new layout. I built it from the real production snapshot (analysis/course.db, 58 participants, 1359 rows) rather than guessing, so it maps each actual key straight to its final home, including the per-chapter step renumberings the snapshot predates. I validated it against a copy of that snapshot: all 73 surviving distinct keys land on real exercise files. About 180 rows stay put because they belong to exercises that were removed entirely (count_items, countdown, fallback, frequent_words, char_classes, advisor, status, records, number_to_string) and have nowhere to map; nothing is deleted, so no data is lost. The migration runs automatically at server startup. So students keep their progress for every exercise that still exists.
Local CI is green: fmt, clippy, tests, check-examples, check-solutions (75/75), typos, and lychee. Worth eyeballing the TOC, progress count, and the next-chapter button once it's deployed, since the two bonus chapters are now interspersed in the ordering.