An added non-ASCII character case breaks compiled runs

Aperta Adatta ai principianti
  • Qualcosa non va?
#690 0 commenti 0 reazioni 0 assegnatari Vedi su GitHub

I maintainer di solito rispondono entro 1 giorno

Nessuno ha ancora preso questa issue.

Valutazione

Difficoltà
2/5
Tempo stimato
1-3 ore
Idoneità per principianti
75/100
Tipo di issue
Bug
Chiarezza
Specificata chiaramente
Stato di attività
Attiva
Stack tecnologico
c, cpp, java, javascript
Ambito
testing-qa, tooling

Direzione di ricerca

Inizia con charLiteral in web/compiler-explorer.js e candidateTypeMatches in web/runners.js per ricostruire come vengono generati e accettati i casi di caratteri. Trova i test che coprono questi percorsi e aggiungi casi per é e un emoji. Il lavoro è completato quando l’aggiunta di uno dei due caratteri non ASCII viene rifiutata con un messaggio che lo nomina e i test passano.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

Observed behavior

An added character case whose cell is a non-ASCII character, such as é or an emoji, is written raw between quotes by charLiteral. C and C++ never hand it over as typed, and Java does only within the Basic Multilingual Plane:

  • C is built with clang on Compiler Explorer, and clang 21 (-std=c17) refuses the literal, so the run reports a compile failure.
  • C++ is built with gcc there, and g++ 15 (-std=c++20) reads 'é' as a multi-character constant, warns, and the solution receives -87 instead of the character. A correct solution can fail the case silently.
  • Java runs on JDK 25 there, and javac 25 receives é correctly (233) but refuses an emoji ("character literal contains more than one UTF-16 code unit"). gino1203 reported an earlier javac that accepts it and passes only the high surrogate, 55357.

Python and JavaScript are not affected.

Expected behavior

A character case either reaches every compiled language as typed or is refused when the case is added, with a message that names the character, as an integer outside 32 bits already is.

Reproduction

At main after #683, with the checkout's web/compiler-explorer.js:

const { nativeLiteral } = await import("./web/compiler-explorer.js");
nativeLiteral("é", "character", "cpp"); // "'é'", raw UTF-8 between quotes

Compiling char c = <that literal>; printf("%d", c); gives -87 with g++ 15 and refuses to compile with clang++ 21; the same in C. Checked locally with gcc 15, clang 21 and Java 25.

Initial analysis
  • Path: charLiteral in web/compiler-explorer.js escapes quotes, backslashes and, after #683, U+0000 to U+001F; every other character is written as is.
  • Reach: candidateTypeMatches in web/runners.js accepts any single code point as a character.
  • Domain: the six character judges (copper-pad-continuity, letter-grid-tracer, letter-panel-tracing, puzzle-grid-save-check, rooftop-panel-footprint, sealed-air-pocket-backfill) take letters, digits, ., 0 or 1, all ASCII. A C or C++ char cannot hold a character outside ASCII as one value, so refusing a non-ASCII character when the case is added is simpler than encoding it.
Action items
  1. Refuse a non-ASCII character case when it is added, naming the character.
  2. Add a test that fails on main for é and an emoji.

Raised by gino1203 in a review of #683.

Lingua principale
Rust
Stelle
228
Fork
150
Merge medio
1g 3h
PR unite (30g)
226

Preparare l'ambiente

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di sysprog21/codetrial

Tutte le issue di sysprog21/codetrial

Issue simili

Altre issue su Rust

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.