Issues

Messaging in multithreaded/concurrent platforms

#11634

· notramo opened

on Sep 23, 2026

Missing-field error for a where-clause parser row points inside the generic helper instead of at the call that picks the record type

Bug

#11633

· rtfeldman opened

on Sep 23, 2026

Derived parser/encoder closes an inferred tag union mid-inference, so type checking depends on expression order

Bug

#11632

· rtfeldman opened

on Sep 23, 2026

`roc build --opt=dev` on a warm cache links a cached procedure to the wrong constant after an edit, while `--no-cache` and `--opt=speed` are correct

#11627

· niclas-ahden opened

on Sep 23, 2026

A release-safe `roc build` panics in `lambda_solved` (`mergeTags`) when two functions `?` one generic function at different payloads of the same tag

#11626

· niclas-ahden opened

on Sep 23, 2026

`roc check` rejects a call to a method whose annotation has a `_` hole when that method dispatches to another type's method that calls a function defined further down (the callee's `a` is not instantiated)

#11625

· niclas-ahden opened

on Sep 23, 2026

`roc check` reuses a method's `..others` across two calls when the method dispatches to a sibling that calls a helper below the type and its signature has a `_` hole (follow-up to #11587)

#11624

· niclas-ahden opened

on Sep 23, 2026

`roc check` reports a `?` error union as used as `[]` (release-safe: "reached unreachable code") when a method whose body is below the type is dispatched after it

#11622

· niclas-ahden opened

on Sep 23, 2026

`roc check` hangs (release-safe: "reached unreachable code") on an unannotated recursive function that uses `?`, called from another unannotated function

#11621

· niclas-ahden opened

on Sep 23, 2026

Platform requires cannot name a type from the platform's imports after independent canonicalization

#11619

· lukewilliamboswell opened

on Sep 23, 2026

`var` reassignments inside a `for` used as a match-arm expression are dropped

Bug

#11618

· lukewilliamboswell opened

on Sep 23, 2026

Key layout-only builtin specializations by representation instead of type

Task

#11616

· rtfeldman opened

on Sep 23, 2026

Navigation Menu

Uh oh!

All issues

Messaging in multithreaded/concurrent platforms

Missing-field error for a where-clause parser row points inside the generic helper instead of at the call that picks the record type

Derived parser/encoder closes an inferred tag union mid-inference, so type checking depends on expression order

`roc build --opt=dev` on a warm cache links a cached procedure to the wrong constant after an edit, while `--no-cache` and `--opt=speed` are correct

A release-safe `roc build` panics in `lambda_solved` (`mergeTags`) when two functions `?` one generic function at different payloads of the same tag

`roc check` rejects a call to a method whose annotation has a `_` hole when that method dispatches to another type's method that calls a function defined further down (the callee's `a` is not instantiated)

`roc check` reuses a method's `..others` across two calls when the method dispatches to a sibling that calls a helper below the type and its signature has a `_` hole (follow-up to #11587)

`roc check` reports a `?` error union as used as `[]` (release-safe: "reached unreachable code") when a method whose body is below the type is dispatched after it

`roc check` hangs (release-safe: "reached unreachable code") on an unannotated recursive function that uses `?`, called from another unannotated function

Platform requires cannot name a type from the platform's imports after independent canonicalization

`var` reassignments inside a `for` used as a match-arm expression are dropped

Key layout-only builtin specializations by representation instead of type

Uh oh!

Search results