Result: done
Files: crates/nikaia-std/src/tools/emit_stmt.nika crates/nikaia-std/src/tools/lowered/emit_stmt.rs tests/compiler/tests/a_let_mut_tuple_marks_every_name.nika tests/compiler/tests/a_let_mut_tuple_marks_every_name.stdout
Commit: parser: let mut (a, b) marks every name mutable
Body: The parser and checker already accepted let mut (a, b). Both messages came from rustc, on the lowered Rust: emit_stmt_let wrote let mut (a, b) = …, which rustc refuses, and for mut (a, b) became for (mut a, mut b), so rustc gave unused_mut for the b the body never assigns. Now the tuple let writes #[allow(unused_mut)] let (mut a, mut b) = …, and a for/pausing while let with a mut head of two or more names gets the same allow. A single let mut x / for mut x the body never changes still warns. The emitter has no per-name assignment facts, so a tuple head where no name is assigned also stays silent.
Fixes: #1126
Pushed: local 40944720e2250a324be84c1679be9d83e2dbb143
Tests: the new a_let_mut_tuple_marks_every_name failed on the unfixed binary (rustc: “mut must be attached to each individual binding”) and passes now with –both-settings, in a scratch project with a_let_tuple_with_one_mut_name_is_not_supported_yet (still ok, NK1307). Hand probes: for mut (a, b) with the new binary prints 3 and gives no warning; let mut x never changed still warns. lower-std plus a fixpoint rebuild left only emit_stmt.rs changed. cargo fmt, clippy –all-targets and pre-push-check.sh are clean. Grepped it for let mut ( / for (mut : nothing found.
Gate: green
Deviation: none
Found: none
Needs: none
Changelog: Fixed: let mut (a, b) = … compiles and marks every name mutable, as NK1307’s help says; a mut tuple head no longer warns about a name the body leaves alone.
Owner: none
Disk freed: none
Rs2nika: none
Attribution:
Co-Authored-By: Claude Opus 5.5 noreply@anthropic.com
Claude-Session: https://claude.ai/code/session_01GFeMqxRbtUMiGVyDMxsvEq
