Raw Robj does not inherit the "raw" class

[JosiahParry]

Raw Robj does not inherit the "raw" class

[yutannihilation]

"raw" is an implicit class. extendr's inherits() simply checks the class attribute, while R's inherits() considers implicit class. ?class says:

If the object does not have a class attribute, it has an implicit class, notably "matrix", "array", "function" or "numeric" or the result of typeof(x)

This is the C version of R's ihnerits(), but this isn't exposed as an API. So, there's no easy way to mimic the R's inherits().

https://github.com/wch/r-source/blob/ec91103ee93b1d4e9f46a37c2dece3761192d9aa/src/main/objects.c#L1013

If the behavior is confusing, extendr might be better to consider removing it. But, I guess what you really want is is_raw() in this case, isn't it?

[JosiahParry]

Good to know! Thank you for sharing that. I suppose is_raw() is what I could be using here. I didn't know about it. In the use case I opted for Raw::try_from() and matched on the results which is sufficient.

I think the inherits() is still useful so I don't think removal should be necessary. I've made a PR #689 to clarify the behavior of inherits() in the docstring.

[yutannihilation]

Can this be closed?

[CGMossa]

Can we please hold off on this? I want to propose a PR based on this.

…

On Sat, 20 Jan 2024, 03.13 Hiroaki Yutani, ***@***.***> wrote: Can this be closed? — Reply to this email directly, view it on GitHub <#687 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAIDVSH33YHZNINAHNP5G53YPMR4RAVCNFSM6AAAAABB6OITDGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMBRGU4TINJWGA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[yutannihilation]

No problem (anyway I don't have permission to close, you know). I was just wondering if this was forgotten to be closed when #689 got merged.