[tuple.creation] Simplify the introductory notation, and then use notation that matches the introduction. #1251
Conversation
|
If you like this general direction, I could rephrase |
|
@jwakely: I'd appreciate your feedback. |
|
Seems like an improvement to me. |
|
@tkoeppe: We have merge conflicts here. |
|
Yes, I think this is an improvement. I'm not keen on putting "where indexing is zero-based" in parentheses though, I preferred the original form. |
|
@jwakely: Yes, I agree, I'll unparenthesize that. |
|
@jwakely: Done. |
| @@ -2047,7 +2040,7 @@ | |||
| \indexlibrary{\idxcode{tuple_cat}} | |||
| \begin{itemdecl} | |||
| template <class... Tuples> | |||
| constexpr tuple<@\placeholder{CTypes}@...> tuple_cat(Tuples&&... tpls); | |||
| constexpr tuple<CTypes...> tuple_cat(Tuples&&... tpls); | |||
| \end{itemdecl} | |||
|
|
|||
| \begin{itemdescr} | |||
There was a problem hiding this comment.
The following description uses italic teletype T_i for elements of Tuples, which seems inconsistent with the changes made above.
There was a problem hiding this comment.
@zygoloid: I hadn't revised tuple_cat at all, and that was still the original formatting. I've now added a complete review of this description to the PR.
| parameter in the function parameter pack \tcode{tpls}, where all indexing is | ||
| zero-based. | ||
|
|
||
| \pnum | ||
| \requires For all $i$, $\tcode{U}_i$ shall be the type | ||
| $\cv_i$ \tcode{tuple<$Args_i...$>}, where $\cv_i$ is the (possibly empty) $i^{th}$ | ||
| cv-qualifier-seq and $Args_i$ is the parameter pack representing the element | ||
| types in $\tcode{U}_i$. Let ${A_{ik}}$ be the ${k_i}^{th}$ type in $Args_i$. For all |
There was a problem hiding this comment.
Please note that "k_i^th type" doesn't make sense - it should be "k^th type".
There was a problem hiding this comment.
While I agree, I'm also happier leaving that alone for now as you have done. I think using k_i is merely confusing rather than wrong.
| @@ -2047,39 +2040,44 @@ | |||
| \indexlibrary{\idxcode{tuple_cat}} | |||
| \begin{itemdecl} | |||
| template <class... Tuples> | |||
| constexpr tuple<@\placeholder{CTypes}@...> tuple_cat(Tuples&&... tpls); | |||
There was a problem hiding this comment.
CTypes is still italicized in the synopsis. (Likewise for make_tuple.)
There was a problem hiding this comment.
Ahh, yes; I change the synopsis to match all the changes in this section.
source/utilities.tex
Outdated
| \tcode{is_constructible_v<$A_{ik}$, $cv_i$ $A_{ik}$\&> == true}, otherwise | ||
| \tcode{is_constructible_v<$A_{ik}$, $cv_i A_{ik}$\&\&> == true}. | ||
| $\cv_i$ \tcode{tuple<$\tcode{Args}_i$...>}, where $\cv_i$ is the (possibly empty) $i^\text{th}$ | ||
| cv-qualifier-seq and $\tcode{Args}_i$ is the parameter pack representing the element |
There was a problem hiding this comment.
\grammarterm for cv-qualifier-seq, please.
There was a problem hiding this comment.
... but that doesn't make sense, since there is no such grammar element here. Perhaps it means that $\cv_i$ are the cv-qualifiers of $\tcode{T}_i$, but I'm not really sure. Let's make the formatting fix to italicize the grammar term cv-qualifier-seq here and file an LWG issue asking for this description to be revisited.
There was a problem hiding this comment.
OK, that's already applied then.
source/utilities.tex
Outdated
| $\tcode{A}_{ik}$ the following requirements shall be satisfied: | ||
| \begin{itemize} | ||
| \item If $\tcode{T}_i$ is deduced as an lvalue reference type, then | ||
| \tcode{is_constructible_v<$A_{ik}$, $cv_i\;A_{ik}$\&> == true}, otherwise |
There was a problem hiding this comment.
Missing tcode for A here and on the line below.
…ation that matches the introduction.
Here is a slight simplification of [tuple.creation]. I bugged me that the introduction was defining things that were only tangentially useful. I changed it to introduce the concept of indexing into a pack only once, and I used the "variable-pack" notation (e.g.
TTypes,VTypes) in all the functions in this subclause.I will work on
tuple_catseparately.Before:
After: