Small change to c_unpack function

[traversc]

No description provided.

[eddelbuettel]

Wow. That is impressive. I slept on a PR for over six years. Ouch.

Do you recall why raw was better? Did you have some message not fitting into a plain C++ vectors?

[traversc]

Wow I had forgotten about this too!

The reason was to avoid an unnecessary copy to std::vector.

[eddelbuettel]

So from eyeball test still valid?

As an aside, trying to update at CRAN (to get rid of stale 'do not set C++11 unless needed' the package published there has) I ran into R CMD check nagging over 'do use compare class(x) == value but use inherits()' so I pushed a commit. If you have a second, glance at it.

Otherwise I think the package is now ... mostly unused but will look into the merge (with a big fat mea culpa for taking over half a decade, shame on me) and an update at CRAN just to have a clean(er) sheet. I presume that is ok with you?

[traversc]

The commit looks good to me, and yes I think using RawVector over std::vector still makes sense.

[eddelbuettel]

I'll put a ChangeLog in for you but date it to today. maybe with a note that is was from way-back-when.