Mutually-exclusive augmentations which introduce same-named nodes

[lordvalumart]

I would like to augment a container in two mutually-exclusive ways, depending on its contents. I would like both augmentations to introduce a node with the same name but a different type in each case. pyang rejects this with error: there is already a child node to <...> at <...> with the name "..." defined at <...>.

Reproducer:

module my-module {
    yang-version 1.1;
    namespace "urn:example:my-module";
    prefix "my-module";

    container my-container {
        leaf my-integer {
            type uint32;
        }
    }

    augment "/my-container" {
        when 'my-integer = 0';

        leaf my-new-node {
            type string;
        }
    }

    augment "/my-container" {
        when 'my-integer != 0';

        leaf my-new-node {
            type uint32;
        }
    }
}

Can anyone say whether this is a bug in pyang, or does the YANG spec prohibit this?

Thanks!

[llhotka]

24. 7. 2024 v 14:24, lordvalumart @.***>:

I would like to augment a container in two mutually-exclusive ways, depending on its contents. I would like both augmentations to introduce a node with the same name but a different type in each case. pyang rejects this with error: there is already a child node to <...> at <...> with the name "..." defined at <...>.

YANG schema validators are unable to determine that the two XPath expressions are mutually exclusive.

Lada

PS. Strictly speaking, they aren’t: if „my-integer“ instance doesn’t exist, then both are false.

Reproducer: module my-module { yang-version 1.1; namespace "urn:example:my-module"; prefix "my-module";

container my-container { leaf my-integer { type uint32; } }

augment "/my-container" { when 'my-integer = 0';

leaf my-new-node { type string; } }

augment "/my-container" { when 'my-integer != 0';

leaf my-new-node { type uint32; } } }

Can anyone say whether this is a bug in pyang, or does the YANG spec prohibit this? Thanks! — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.***>

-- Ladislav Lhotka PGP Key ID: 0xB8F92B08A9F76C67

[lordvalumart]

Thanks, that makes sense. I guess then the message from pyang isn't quite accurate; if the augmentation contains a when clause pyang can't determine whether or not there is an error.