search

atomgalaxy / review-executor-sendrecv

A board for reviewing the executors proposal P0440
1 stars 0 forks source link

formulation of types in concepts #3

Open dietmarkuehl opened 4 years ago

dietmarkuehl commented 4 years ago

In multiple places the document refers to types using a phrase like the one below (this specific one is from 2.2.3.4):

... let E be a type such that decltype((e)) is E ...

I believe that actually says

... let E be decltype((e)) ...

At least, I don't see how any other type X can meet the requirement to also be decltype((e)). The currently used phrase causes me to suspect that there is something else going on and if that is the case it should be clarified.

ericniebler commented 4 years ago

I believe this bit of standardese boilerplate was first drafted by @CaseyCarter for the Ranges TS. Casey, do you recall why we wrote it that way?

CaseyCarter commented 4 years ago

I suspect this was an evolution of the phrase "Let E be an expression such that decltype((E)) is T" which we use in a few places in Ranges. :shrug: In any case, I agree with Dietmar that these can be simplified.

lewissbaker commented 4 years ago

This has also been raised in https://github.com/executors/executors/issues/465