[ILLink] Add interface implementing type to OverrideInformation and use where it makes sense

[jtschuster]

Fixes#98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

This also reduces allocations of OverrideInformation in a hello world from ~35mb to ~1.2mb.

[ghost]

Tagging subscribers to 'linkable-framework': @eerhardt, @vitek-karas, @LakshanF, @sbomer, @joperezr, @marek-safar
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Fixes #98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

Author:	jtschuster
Assignees:	-
Labels:	linkable-framework
Milestone:	-

[sbomer]

It's possible for a base to provide a static interface method, which seems like it might have broken the old logic. Could you add a testcase for this (using static interface methods, where the interface type isn't instantiated), and try to confirm that this change fixes it?

edit: addressed by your comment at #97487 (comment).

[sbomer]

LGTM aside from the last few comments, thanks!

I'm not sure how well this InterfaceImplementor concept will translate once you improve support for recursive interfaces, since the impl will no longer always be on the "Implementor" type. But looks good for the current support.

[jtschuster]

Fixes#98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

This also reduces allocations of OverrideInformation in a hello world from ~35mb to ~1.2mb.

[ghost]

Tagging subscribers to 'linkable-framework': @eerhardt, @vitek-karas, @LakshanF, @sbomer, @joperezr, @marek-safar
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Fixes #98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

Author:	jtschuster
Assignees:	-
Labels:	linkable-framework
Milestone:	-

[sbomer]

It's possible for a base to provide a static interface method, which seems like it might have broken the old logic. Could you add a testcase for this (using static interface methods, where the interface type isn't instantiated), and try to confirm that this change fixes it?

edit: addressed by your comment at #97487 (comment).

[sbomer]

LGTM aside from the last few comments, thanks!

I'm not sure how well this InterfaceImplementor concept will translate once you improve support for recursive interfaces, since the impl will no longer always be on the "Implementor" type. But looks good for the current support.

[jtschuster]

Fixes#98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

This also reduces allocations of OverrideInformation in a hello world from ~35mb to ~1.2mb.

[ghost]

Tagging subscribers to 'linkable-framework': @eerhardt, @vitek-karas, @LakshanF, @sbomer, @joperezr, @marek-safar
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Fixes #98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

Author:	jtschuster
Assignees:	-
Labels:	linkable-framework
Milestone:	-

[sbomer]

It's possible for a base to provide a static interface method, which seems like it might have broken the old logic. Could you add a testcase for this (using static interface methods, where the interface type isn't instantiated), and try to confirm that this change fixes it?

edit: addressed by your comment at #97487 (comment).

[sbomer]

LGTM aside from the last few comments, thanks!

I'm not sure how well this InterfaceImplementor concept will translate once you improve support for recursive interfaces, since the impl will no longer always be on the "Implementor" type. But looks good for the current support.

[jtschuster]

Fixes#98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

This also reduces allocations of OverrideInformation in a hello world from ~35mb to ~1.2mb.

[ghost]

Tagging subscribers to 'linkable-framework': @eerhardt, @vitek-karas, @LakshanF, @sbomer, @joperezr, @marek-safar
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Fixes #98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

Author:	jtschuster
Assignees:	-
Labels:	linkable-framework
Milestone:	-

[sbomer]

It's possible for a base to provide a static interface method, which seems like it might have broken the old logic. Could you add a testcase for this (using static interface methods, where the interface type isn't instantiated), and try to confirm that this change fixes it?

edit: addressed by your comment at #97487 (comment).

[sbomer]

LGTM aside from the last few comments, thanks!

I'm not sure how well this InterfaceImplementor concept will translate once you improve support for recursive interfaces, since the impl will no longer always be on the "Implementor" type. But looks good for the current support.

[jtschuster]

Fixes#98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

This also reduces allocations of OverrideInformation in a hello world from ~35mb to ~1.2mb.

[ghost]

Tagging subscribers to 'linkable-framework': @eerhardt, @vitek-karas, @LakshanF, @sbomer, @joperezr, @marek-safar
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Fixes #98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

Author:	jtschuster
Assignees:	-
Labels:	linkable-framework
Milestone:	-

[sbomer]

It's possible for a base to provide a static interface method, which seems like it might have broken the old logic. Could you add a testcase for this (using static interface methods, where the interface type isn't instantiated), and try to confirm that this change fixes it?

edit: addressed by your comment at #97487 (comment).

[sbomer]

LGTM aside from the last few comments, thanks!

I'm not sure how well this InterfaceImplementor concept will translate once you improve support for recursive interfaces, since the impl will no longer always be on the "Implementor" type. But looks good for the current support.

[jtschuster]

Fixes#98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

This also reduces allocations of OverrideInformation in a hello world from ~35mb to ~1.2mb.

[ghost]

Tagging subscribers to 'linkable-framework': @eerhardt, @vitek-karas, @LakshanF, @sbomer, @joperezr, @marek-safar
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Fixes #98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

Author:	jtschuster
Assignees:	-
Labels:	linkable-framework
Milestone:	-

[sbomer]

It's possible for a base to provide a static interface method, which seems like it might have broken the old logic. Could you add a testcase for this (using static interface methods, where the interface type isn't instantiated), and try to confirm that this change fixes it?

edit: addressed by your comment at #97487 (comment).

[sbomer]

LGTM aside from the last few comments, thanks!

I'm not sure how well this InterfaceImplementor concept will translate once you improve support for recursive interfaces, since the impl will no longer always be on the "Implementor" type. But looks good for the current support.

[jtschuster]

Fixes#98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

This also reduces allocations of OverrideInformation in a hello world from ~35mb to ~1.2mb.

[ghost]

Tagging subscribers to 'linkable-framework': @eerhardt, @vitek-karas, @LakshanF, @sbomer, @joperezr, @marek-safar
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Fixes #98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

Author:	jtschuster
Assignees:	-
Labels:	linkable-framework
Milestone:	-

[sbomer]

It's possible for a base to provide a static interface method, which seems like it might have broken the old logic. Could you add a testcase for this (using static interface methods, where the interface type isn't instantiated), and try to confirm that this change fixes it?

edit: addressed by your comment at #97487 (comment).

[sbomer]

LGTM aside from the last few comments, thanks!

I'm not sure how well this InterfaceImplementor concept will translate once you improve support for recursive interfaces, since the impl will no longer always be on the "Implementor" type. But looks good for the current support.

[jtschuster]

Fixes#98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

This also reduces allocations of OverrideInformation in a hello world from ~35mb to ~1.2mb.

[ghost]

Tagging subscribers to 'linkable-framework': @eerhardt, @vitek-karas, @LakshanF, @sbomer, @joperezr, @marek-safar
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Fixes #98255

Adds the type that implements the interface to the OverrideInformation class to be able to properly handle cases where the base provides the interface method or the type relies on a default interface method. Adds the InterfaceImplementor class that is basically a tuple of (TypeDefinition typeWithInterface, InterfaceImplementation ImplementedInterface) to do this. Also cleans some unnecessary code in OverrideInformation and ensures that OverrideInformation will have an InterfaceImplementor if the base method is an interface method.

Author:	jtschuster
Assignees:	-
Labels:	linkable-framework
Milestone:	-

[sbomer]

It's possible for a base to provide a static interface method, which seems like it might have broken the old logic. Could you add a testcase for this (using static interface methods, where the interface type isn't instantiated), and try to confirm that this change fixes it?

edit: addressed by your comment at #97487 (comment).

[sbomer]

LGTM aside from the last few comments, thanks!

I'm not sure how well this InterfaceImplementor concept will translate once you improve support for recursive interfaces, since the impl will no longer always be on the "Implementor" type. But looks good for the current support.