Please describe the feature you have in mind and explain what the current shortcomings are?
Implicit layer save paths aren't 'collected' by the current publishing logic to be included - because for most studios adopting USD with Solaris the first time these so called "implicit layers" were not intentionally created to become their own USD layers, but just a byproduct of the default values of e.g. a SOP import.
In essence, they are not intentionally generated and should actually not have been separate layers.
How would you imagine the implementation of the feature?
Add a validator that recognizes all the implicit layers the USD ROP would write out and report them.
The validator could potentially even have e.g. repair actions like "Set explicit save path" or "Flatten layer" if we could somehow auto-detect where it is in the graph, etc. but that may be complicated with custom HDAs
Are there any labels you wish to add?
[X] I have added the relevant labels to the enhancement request.
Describe alternatives you've considered:
We could also just take along implicit layers for the ride in the publishing but 99% of the time they just aren't intentional.
Additional context:
The issue or confusion comes up every now and then, e.g. in this issue here
Also note that this would only be published currently if a checkbox is manually disabled on the USD ROP that Houdini has enabled by default:
Is there an existing issue for this?
Please describe the feature you have in mind and explain what the current shortcomings are?
Implicit layer save paths aren't 'collected' by the current publishing logic to be included - because for most studios adopting USD with Solaris the first time these so called "implicit layers" were not intentionally created to become their own USD layers, but just a byproduct of the default values of e.g. a SOP import.
In essence, they are not intentionally generated and should actually not have been separate layers.
How would you imagine the implementation of the feature?
Add a validator that recognizes all the implicit layers the USD ROP would write out and report them.
The validator could potentially even have e.g. repair actions like "Set explicit save path" or "Flatten layer" if we could somehow auto-detect where it is in the graph, etc. but that may be complicated with custom HDAs
Are there any labels you wish to add?
Describe alternatives you've considered:
We could also just take along implicit layers for the ride in the publishing but 99% of the time they just aren't intentional.
Additional context:
The issue or confusion comes up every now and then, e.g. in this issue here
Also note that this would only be published currently if a checkbox is manually disabled on the USD ROP that Houdini has enabled by default: