Let the candidate restore the starter code
Les mainteneurs répondent en général sous 1 jour
Évaluation
Cette issue n'a pas encore été évaluée.
Description
Who is affected
A candidate in a coding interview who has edited the editor into a state they want to start over from. The starter is the only template they have, and nothing on the interview page puts it back.
Observed
On main at 1ba82f9, with cargo run -- web and ?problem=chargeback-pair-match, I replaced the Python buffer with my own text, clicked JavaScript and then Python again, and the buffer still held my text. Ctrl+Z at that point did not bring the starter back either. The page has no control with a reset, restore or starter label: I listed every button and summary on the interview page and found none. codeByLanguage is seeded from problem.starterCode once (web/interview.js:422), and setLanguage only swaps buffers between tabs (web/interview.js:2209).
Desired outcome
The candidate can put the starter code of the language they are in back into the editor, without leaving the interview.
Present workaround
Select all, delete, and retype the signature and the class wrapper.
Scope
The editor toolbar and the editor buffer on the interview page, in web/interview.html and web/interview.js. Not the agent, the recording contract, the judges or the problem bank. I have no data on how often candidates want this, so this issue rests on the missing path, not on a measured need.
Approaches
- A button that replaces the buffer at once. Smallest change, but one misclick discards the candidate's work with no way back.
- The same button behind a confirmation. Safe, but adds a step, and a modal competes with the Meet presentation flow.
- A button that replaces the buffer through
document.execCommand("insertText"), as the editor's own programmatic edits already do (web/interview.js:817), so Ctrl+Z restores the previous buffer. No extra step, but undo depends on the browser accepting the command, and the existing fallback assigns.value, which drops the undo history.
I prefer 3 with the fallback taking the confirmation of 2, so the case where undo is unavailable is the only one that asks.
Compatibility
No new message type or recording kind is needed. A reset is a new code_update, and the recording's editor event is already a whole-buffer snapshot that replaces the previous one (docs/recording-contract.md). On the agent side, a code change that does not arrive with a language switch is already read as the candidate editing, and the comment in src/agent/events.rs covers a candidate clearing the editor and starting over.
Open question
Interview contract version 17 records Test only while the editor still holds the code the run executed. I have not checked what happens to a Test credit that was already recorded when the buffer is reset afterwards. I will measure that before the pull request and report it here.
Acceptance criteria
- A control on the interview page replaces the editor buffer of the current language with that language's starter. The buffers of the other languages are unchanged.
- A reset cannot silently destroy work: it is either undone with Ctrl+Z or confirmed first.
- The reset reaches the interviewer and the recording through the normal code publish.
- The control is unavailable whenever the editor is disabled, including after the interview ends.
- A browser test fails if the reset leaves the buffer unchanged or discards the other languages' buffers.
Related: #252 added the editor options and states that the editor preserves native undo and redo; approach 3 keeps that.
I will work on this and can open a pull request with the test first. Please assign it to me.
- Langage dominant
- Rust
- Étoiles
- 223
- Forks
- 148
- Merge moyen
- 1 j 10 h
- PR mergées (30 j)
- 227
Préparer son environnement
- Aucun Dockerfile ni fichier Docker Compose
- Propose un modèle de pull request
- Lire le guide de contribution
Par où commencer
- Lisez l'issue en entier, puis le guide de contribution du projet.
- Signalez en commentaire que vous la prenez — cela évite que deux personnes fassent le même travail.
- Forkez le dépôt et travaillez sur une branche.
- Ouvrez une pull request qui référence le numéro de l'issue.
Autres issues de sysprog21/codetrial
-
Difficulté 4/5 3-5 jours Accessibilité débutants 48/100
Les mainteneurs répondent en général sous 1 jour
-
CODETRIAL_DURATION_MIN does not preselect the interview length in the lobbyPeut-être pris @RWong04 l’a pris il y a 1 jour. Ouverte
sysprog21/codetrial#582 · 3 commentaires · 1 personne assignée ·
Les mainteneurs répondent en général sous 1 jour
-
Let candidates choose a report improvement as the next practice focusPeut-être pris @Hsieh-HC l’a pris il y a 1 jour. Ouverte
Difficulté 3/5 1-2 jours Accessibilité débutants 74/100
sysprog21/codetrial#580 · 3 commentaires · 1 personne assignée ·
Les mainteneurs répondent en général sous 1 jour
-
Example drawing leaves too little room to draw, even when enlargedPeut-être pris @Disesfgewu l’a pris il y a 1 jour. Ouverte
Difficulté 3/5 1-2 jours Accessibilité débutants 68/100
sysprog21/codetrial#579 · 2 commentaires · 1 personne assignée ·
Les mainteneurs répondent en général sous 1 jour
-
Mechanize UI color token enforcement checkPeut-être pris @ArthurArthurArthur0817 l’a pris il y a 1 jour. Ouverte
Difficulté 3/5 1-2 jours Accessibilité débutants 68/100
sysprog21/codetrial#578 · 1 commentaire · 1 personne assignée ·
Les mainteneurs répondent en général sous 1 jour
Toutes les issues de sysprog21/codetrial
Issues similaires
-
Difficulté 2/5 1-3 heures Accessibilité débutants 69/100
Les mainteneurs répondent en général sous 1 jour
-
A match arm above the largest i64 is compared as 0Peut-être pris Une pull request liée à cette issue est ouverte ou déjà fusionnée. Ouverte
Difficulté 2/5 1-3 heures Accessibilité débutants 86/100
Les mainteneurs répondent en général sous 1 jour
-
Difficulté 1/5 Moins d'une heure Accessibilité débutants 83/100
DioxusLabs/dioxus#6063 ·
Les mainteneurs répondent en général sous 1 jour
-
`nsh-core` fails to compile when taffy's default `calc` feature is enabled elsewhere in the graphOuverte
Difficulté 2/5 1-3 heures Accessibilité débutants 76/100
longbridge/nsh#966 ·
Les mainteneurs répondent en général sous 1 jour
-
Windows docs: pin a tested ConPTY version for Kitty graphics; 1.25.260930003 corrupts large uploadsOuverte
Difficulté 1/5 1-3 heures Accessibilité débutants 88/100
raphamorim/rio#2021 ·
Les mainteneurs répondent en général sous 1 jour