Tim Taylor

I assume a pre-existing typo as was previously `long LDOUBLE`. Not familiar enough with the code base to check the intention there.

@yaccos Great - thank you for the bench marking! I appreciate the time you have taken to look at this even if it seems there is no completely satisfactory answer...

That would be better but the generic nature of Ops still give me pause. It's somewhat blurring a lint for `all.equal()` usage. If the metadata gets updated to `expect_equal()` I...

^Ignore the first linked issue above. I somehow replied on the wrong issue. I've confirmed locally that the second one is a manifestation of the issue described.

FWIW - my preference would be 2. I think consistency with R is useful (it was the discrepancy that led me here and it seems to have come up a...

> Thanks both - it sounds reasonable to me. Would you mind asking about this on R-devel to get a definite answer? Done. Just waiting for it to go through...

Simon's reponse https://stat.ethz.ch/pipermail/r-package-devel/2025q2/011629.html, snippet below: > R has a configuration option --enable-long-double which governs whether R should use algorithms which can benefit from extended precision where applicable. However, that is...

Not that I can think off. The first I've definitely done by accident whilst refactoring so would be nice to catch.

I'm a little confused (not unusual). What does the `cnd_class` argument now do that helps if it just replaces the class? I think the docs may need updating. Should we...

I can raise this as a new issue if preferred (perhaps also in testthat) but it is somewhat related ... Ignoring the 'type' discussion for the moment, should `cnd_entrace()` document...