Fix a few small bugs in submdspan #6621
Conversation
| @@ -21741,11 +21741,11 @@ | |||
| \begin{itemdescr} | |||
| \pnum | |||
| \constraints | |||
| \tcode{sizeof...(slices)} equals \tcode{Extents::rank()}. | |||
| \tcode{sizeof...(slices)} equals \tcode{decltype(src)::rank()}. | |||
There was a problem hiding this comment.
decltype(src) is const extents<IndexType, Extents...>&and reference types don't have members, so this is ill-formed.src.rank()would work here, as wouldextents<IndexType, Extents...>::rank()`. (I think the latter is more conventional for the Standard Library, though I have no data to back that up nor strong feelings either way.)
| \hphantom{0 } \le \tcode{extents().extent($k$)}$ | ||
| $0 \le \tcode{\exposid{first_}<index_type, $k$>(slices...)}$ | ||
| $\le \tcode{\exposid{last_}<$k$>(extents(), slices...)}$ | ||
| $\le \tcode{extents().extent($k$)}$ |
There was a problem hiding this comment.
FWIW, I didn't read this before as comparing each expression to 0, but as chained comparisons as you intend. In either case, you this line is now overlong and needs to be broken manually (hence the CI failure).
There was a problem hiding this comment.
Is there a reason you're changing the linebreaking here? Is there non-whitespace change here that you want to make?
|
Please get in touch and we can figure this out. |
This fixes a couple bugs in
subdmspan. Some of the wording for constraints/mandates/preconditions ofsubmdspan_extentswas straight copied fromsubmdspan_mappingbut the names of the things it refers to are not quite right (srcis actually the equivalent ofextents()anddecltype(src)the equivalent ofExtents). Also fixes the last precondition insubmdspan_mappingwhich should be essentially the same as insubmdspan_extentsi.e. a strict ordering of the values, not comparing to 0 each of them.