bug: support partial wiring of multi-output sources by gbin · Pull Request #883 · copper-project/copper-rs

gbin · 2026-03-02T19:14:03Z

Summary

The proc macro could not accomodate on multi output tasks a partial subscription to the set of output messages.

Related issues

Closes #

Changes

Ask the user to confirm all the unused connection with a line like:

(
            src: "src",
            dst: "__nc__",
            msg: "bool",
),

Testing

just fmt
just lint
just test
optional full just std-ci (if std/runtime paths are impacted)
optional full just nostd-ci (if embedded/no_std paths are impacted)
Other (please specify):

pro-tip: just with no parameters in the root defaults to just fmt, just lint, and just test.

Checklist

I have updated docs or examples where needed
I have added or updated tests where needed
I have considered platform impact (Linux/macOS/Windows/embedded)
I have considered config/logging changes (if applicable)
This change is not a breaking change (or I documented it below)

Breaking changes (if any)

Additional context

Copilot

Pull request overview

Adds runtime projection support so downstream tasks can subscribe to only a subset of a multi-output source’s messages, fixing partial wiring for multi-output tasks in the proc-macro-generated runtime.

Changes:

Introduces CuOutputGet and CuOutputProject to extract/project specific CuMsg<T> out of a source output pack (including tuples).
Updates derive macro generated execution to run the source into a “full output” buffer, then project into the connected downstream output slot.
Adds a compile-pass test + config validating partial subscription (i32 only) from a (i32, bool) source.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 5 comments.

File	Description
core/cu29_runtime/src/cutask.rs	Adds typed lookup + projection traits/functions enabling subset routing from multi-output sources.
core/cu29_derive/src/lib.rs	Changes generated execution to process into a full output pack and project into the connected output type.
core/cu29_derive/tests/config/multi_output_source_subset_valid.ron	Adds config wiring a multi-output source to a sink subscribing to only `i32`.
core/cu29_derive/tests/compile_pass/copper_runtime/multi_output_source_subset.rs	Adds compile-pass test verifying the partial subscription compiles.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-02T19:52:23Z

+macro_rules! impl_cu_output_get_tuple {
+    ($($idx:tt => $name:ident),+) => {
+        impl<T, $($name),+> CuOutputGet<T> for ($(CuMsg<$name>,)+)
+        where
+            T: CuMsgPayload + 'static,
+            $($name: CuMsgPayload + 'static),+
+        {
+            fn output_msg(&self) -> &CuMsg<T> {
+                $(
+                    if TypeId::of::<T>() == TypeId::of::<$name>() {
+                        let msg = &self.$idx;
+                        // SAFETY: TypeId equality proves T == $name.
+                        return unsafe { &*(msg as *const CuMsg<$name> as *const CuMsg<T>) };
+                    }
+                )+
+                panic!(
+                    "Requested output message type '{}' is not present in source output tuple",
+                    type_name::<T>()
+                );
+            }
+        }
+    };
+}
+
+impl_cu_output_get_tuple!(0 => T1, 1 => T2);
+impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3);
+impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4);
+impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4, 4 => T5);
+impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4, 4 => T5, 5 => T6);
+impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4, 4 => T5, 5 => T6, 6 => T7);
+impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4, 4 => T5, 5 => T6, 6 => T7, 7 => T8);
+impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4, 4 => T5, 5 => T6, 6 => T7, 7 => T8, 8 => T9);


The tuple CuOutputGet<T> impl is overly permissive: it makes (CuMsg<T1>, CuMsg<T2>, ...) implement CuOutputGet<T> for any T, so a missing type compiles and then panics at runtime. It also requires unsafe casting. Prefer generating per-element trait impls (e.g., implement CuOutputGet<T1> and CuOutputGet<T2> separately for the tuple, each returning &self.0 / &self.1), which removes TypeId/unsafe entirely and turns “type not present” into a compile-time error (and prevents ambiguous duplicate payload types via coherence).

Suggested change

macro_rules! impl_cu_output_get_tuple {

($($idx:tt => $name:ident),+) => {

impl<T, $($name),+> CuOutputGet<T> for ($(CuMsg<$name>,)+)

where

T: CuMsgPayload + 'static,

$($name: CuMsgPayload + 'static),+

{

fn output_msg(&self) -> &CuMsg<T> {

$(

if TypeId::of::<T>() == TypeId::of::<$name>() {

let msg = &self.$idx;

// SAFETY: TypeId equality proves T == $name.

return unsafe { &*(msg as *const CuMsg<$name> as *const CuMsg<T>) };

}

)+

panic!(

"Requested output message type '{}' is not present in source output tuple",

type_name::<T>()

);

}

}

};

}

impl_cu_output_get_tuple!(0 => T1, 1 => T2);

impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3);

impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4);

impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4, 4 => T5);

impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4, 4 => T5, 5 => T6);

impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4, 4 => T5, 5 => T6, 6 => T7);

impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4, 4 => T5, 5 => T6, 6 => T7, 7 => T8);

impl_cu_output_get_tuple!(0 => T1, 1 => T2, 2 => T3, 3 => T4, 4 => T5, 5 => T6, 6 => T7, 7 => T8, 8 => T9);

impl<T1, T2> CuOutputGet<T1> for (CuMsg<T1>, CuMsg<T2>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T1> {

&self.0

}

}

impl<T1, T2> CuOutputGet<T2> for (CuMsg<T1>, CuMsg<T2>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T2> {

&self.1

}

}

impl<T1, T2, T3> CuOutputGet<T1> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T1> {

&self.0

}

}

impl<T1, T2, T3> CuOutputGet<T2> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T2> {

&self.1

}

}

impl<T1, T2, T3> CuOutputGet<T3> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T3> {

&self.2

}

}

impl<T1, T2, T3, T4> CuOutputGet<T1> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T1> {

&self.0

}

}

impl<T1, T2, T3, T4> CuOutputGet<T2> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T2> {

&self.1

}

}

impl<T1, T2, T3, T4> CuOutputGet<T3> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T3> {

&self.2

}

}

impl<T1, T2, T3, T4> CuOutputGet<T4> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T4> {

&self.3

}

}

impl<T1, T2, T3, T4, T5> CuOutputGet<T1> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T1> {

&self.0

}

}

impl<T1, T2, T3, T4, T5> CuOutputGet<T2> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T2> {

&self.1

}

}

impl<T1, T2, T3, T4, T5> CuOutputGet<T3> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T3> {

&self.2

}

}

impl<T1, T2, T3, T4, T5> CuOutputGet<T4> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T4> {

&self.3

}

}

impl<T1, T2, T3, T4, T5> CuOutputGet<T5> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T5> {

&self.4

}

}

impl<T1, T2, T3, T4, T5, T6> CuOutputGet<T1> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T1> {

&self.0

}

}

impl<T1, T2, T3, T4, T5, T6> CuOutputGet<T2> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T2> {

&self.1

}

}

impl<T1, T2, T3, T4, T5, T6> CuOutputGet<T3> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T3> {

&self.2

}

}

impl<T1, T2, T3, T4, T5, T6> CuOutputGet<T4> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T4> {

&self.3

}

}

impl<T1, T2, T3, T4, T5, T6> CuOutputGet<T5> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T5> {

&self.4

}

}

impl<T1, T2, T3, T4, T5, T6> CuOutputGet<T6> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T6> {

&self.5

}

}

impl<T1, T2, T3, T4, T5, T6, T7> CuOutputGet<T1> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T1> {

&self.0

}

}

impl<T1, T2, T3, T4, T5, T6, T7> CuOutputGet<T2> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T2> {

&self.1

}

}

impl<T1, T2, T3, T4, T5, T6, T7> CuOutputGet<T3> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T3> {

&self.2

}

}

impl<T1, T2, T3, T4, T5, T6, T7> CuOutputGet<T4> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T4> {

&self.3

}

}

impl<T1, T2, T3, T4, T5, T6, T7> CuOutputGet<T5> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T5> {

&self.4

}

}

impl<T1, T2, T3, T4, T5, T6, T7> CuOutputGet<T6> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T6> {

&self.5

}

}

impl<T1, T2, T3, T4, T5, T6, T7> CuOutputGet<T7> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T7> {

&self.6

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8> CuOutputGet<T1> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T1> {

&self.0

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8> CuOutputGet<T2> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T2> {

&self.1

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8> CuOutputGet<T3> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T3> {

&self.2

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8> CuOutputGet<T4> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T4> {

&self.3

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8> CuOutputGet<T5> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T5> {

&self.4

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8> CuOutputGet<T6> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T6> {

&self.5

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8> CuOutputGet<T7> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T7> {

&self.6

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8> CuOutputGet<T8> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T8> {

&self.7

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8, T9> CuOutputGet<T1> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>, CuMsg<T9>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

T9: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T1> {

&self.0

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8, T9> CuOutputGet<T2> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>, CuMsg<T9>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

T9: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T2> {

&self.1

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8, T9> CuOutputGet<T3> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>, CuMsg<T9>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

T9: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T3> {

&self.2

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8, T9> CuOutputGet<T4> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>, CuMsg<T9>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

T9: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T4> {

&self.3

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8, T9> CuOutputGet<T5> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>, CuMsg<T9>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

T9: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T5> {

&self.4

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8, T9> CuOutputGet<T6> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>, CuMsg<T9>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

T9: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T6> {

&self.5

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8, T9> CuOutputGet<T7> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>, CuMsg<T9>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

T9: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T7> {

&self.6

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8, T9> CuOutputGet<T8> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>, CuMsg<T9>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

T9: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T8> {

&self.7

}

}

impl<T1, T2, T3, T4, T5, T6, T7, T8, T9> CuOutputGet<T9> for (CuMsg<T1>, CuMsg<T2>, CuMsg<T3>, CuMsg<T4>, CuMsg<T5>, CuMsg<T6>, CuMsg<T7>, CuMsg<T8>, CuMsg<T9>)

where

T1: CuMsgPayload + 'static,

T2: CuMsgPayload + 'static,

T3: CuMsgPayload + 'static,

T4: CuMsgPayload + 'static,

T5: CuMsgPayload + 'static,

T6: CuMsgPayload + 'static,

T7: CuMsgPayload + 'static,

T8: CuMsgPayload + 'static,

T9: CuMsgPayload + 'static,

{

fn output_msg(&self) -> &CuMsg<T9> {

&self.8

}

}

That suggestion, as written, is not implementable on stable Rust.

It triggers coherence overlap (E0119) immediately, because these two impls overlap when T1 == T2:

impl<T1, T2> CuOutputGet for (CuMsg, CuMsg)

impl<T1, T2> CuOutputGet for (CuMsg, CuMsg)

Copilot · 2026-03-02T19:52:24Z

+        {
+            fn output_msg(&self) -> &CuMsg<T> {
+                $(
+                    if TypeId::of::<T>() == TypeId::of::<$name>() {


If this runtime lookup approach stays, consider fully qualifying TypeId/type_name as core::any::TypeId and core::any::type_name (or importing them explicitly) to avoid relying on ambient imports and to keep the code no_std-friendly.

Copilot · 2026-03-02T19:52:24Z

+                    "Requested output message type '{}' is not present in source output tuple",
+                    type_name::<T>()


If this runtime lookup approach stays, consider fully qualifying TypeId/type_name as core::any::TypeId and core::any::type_name (or importing them explicitly) to avoid relying on ambient imports and to keep the code no_std-friendly.

Copilot · 2026-03-02T19:52:24Z

                                step: CuComponentState::Process,
                                culistid: Some(clid),
                            });
+                            let mut __cu_full_output = Default::default();


The generated code now unconditionally constructs a fresh __cu_full_output via Default::default(). This introduces an implicit (and potentially macro-error-message-unfriendly) requirement that the task’s full Output type is Default, and it also adds per-call initialization overhead. Consider either (1) making the Default requirement explicit in the generated bounds / emitting a clearer compile-time error when it’s not satisfied, or (2) reusing an existing full-output buffer (if one exists in the runtime scaffolding) to avoid repeated initialization.

this is already explicit, all mesages are default constructible WTH?

Copilot · 2026-03-02T19:52:24Z

+                                #task_instance.process(&ctx, &mut __cu_full_output)
                            };
+                            cu29::cutask::project_output(&__cu_full_output, cumsg_output);


The generated code now unconditionally constructs a fresh __cu_full_output via Default::default(). This introduces an implicit (and potentially macro-error-message-unfriendly) requirement that the task’s full Output type is Default, and it also adds per-call initialization overhead. Consider either (1) making the Default requirement explicit in the generated bounds / emitting a clearer compile-time error when it’s not satisfied, or (2) reusing an existing full-output buffer (if one exists in the runtime scaffolding) to avoid repeated initialization.

gbin · 2026-03-02T20:26:01Z

ok this is a little bit fucked up this entire thing and can introduce some overhead.

The problem is that you cannot ignore any output from a multi output task because it is constructed from the connections.

if you don't connect them all the output is only a subset of the actual one.

I am thinking about helping that with a proper declaration of outputs or maybe just "not connected" entry in the task.

gbin · 2026-03-02T21:00:59Z

I completely reoverhalled this, the direction was BAD. no more overhead and added the notiong of __nc__ that you need to add to your connections to confirm you are dropping the output

makeecat · 2026-03-03T02:34:32Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 3b58d7491e

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

runtime: support partial wiring of multi-output sources

d8792f1

gbin added bug Something isn't working include in changelog labels Mar 2, 2026

fmt

9f770ec

gbin requested review from Copilot and makeecat March 2, 2026 19:43

Copilot started reviewing on behalf of gbin March 2, 2026 19:52 View session

Copilot AI reviewed Mar 2, 2026

View reviewed changes

Added __nc__ as a work around

20abf22

gbin added 4 commits March 2, 2026 15:02

Merge branch 'master' into gbin/core-multi-output-projection

e6c543d

fmt

8ad0011

clippy

3547371

order preservation for __nc__

3b58d74

chatgpt-codex-connector Bot reviewed Mar 3, 2026

View reviewed changes

Comment thread core/cu29_runtime/src/config.rs

makeecat approved these changes Mar 3, 2026

View reviewed changes

gbin added 2 commits March 3, 2026 08:17

fmt

5b1863c

fixes config serialization.

7f4b157

gbin merged commit b581c04 into master Mar 3, 2026
23 checks passed

gbin deleted the gbin/core-multi-output-projection branch March 3, 2026 14:44

		"Requested output message type '{}' is not present in source output tuple",
		type_name::<T>()

Conversation

gbin commented Mar 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Related issues

Changes

Testing

Checklist

Breaking changes (if any)

Additional context

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

gbin Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

gbin Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

gbin Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

gbin commented Mar 2, 2026

Uh oh!

gbin commented Mar 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

makeecat commented Mar 3, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

gbin commented Mar 2, 2026 •

edited

Loading

gbin commented Mar 2, 2026 •

edited

Loading