[reverse.iterators] Dissolve single-item subclauses. #1832
Conversation
|
@zygoloid : This merge went the wrong way: You merged "master" into my branch, it seems, rather than the other way round. |
|
Rebased. |
|
@jensmaurer This is github's "rebase someone else's PR" feature. Sadly it appears to be broken right now (it fails to actually re-check whether the PR is mergeable afterwards or something). |
source/iterators.tex
Outdated
|
|
||
| \indexlibrarymember{operator+}{reverse_iterator}% | ||
| \begin{itemdecl} | ||
| constexpr reverse_iterator operator+(difference_type n) const; |
There was a problem hiding this comment.
I find it a bit strange to call this an "observer", since it does not of itself provide any mechanism to actually observe the state of the iterator. (It seems a little odd to call operator*, operator->, and operator[] observers, rather than something like "element access", too).
I might be tempted to instead lump operator++, operator--, operator+, operator+=, operator-, operator-= under something like "reverse_iterator navigation", put these three under "reverse_iterator element access", and keep base as a single-item subclause (which I think makes sense, as it's the one and only part of the reverse_iterator interface that is not part of some Iterator concept).
There was a problem hiding this comment.
Sounds reasonable. Done.
|
This is worth calling out to @mclow as it will affect a small number of issues in the LWG issues list. I think the proposed change looks cleaner though. |
|
Thanks for the callout; I can fix the issues. |
zygoloid
added
the
needs rebase
|
Postponed to post-meeting mailing due to changes to stable names. |
jensmaurer
removed
the
needs rebase
|
Rebased. |
jensmaurer
added
the
decision-required
zygoloid
added
after-motions
|
To be merged after Jacksonville motions. |
| @@ -1315,8 +1313,6 @@ | |||
| \tcode{u.current}. | |||
| \end{itemdescr} | |||
|
|
|||
| \rSec4[reverse.iter.op=]{\tcode{reverse_iterator::operator=}} | |||
There was a problem hiding this comment.
We're missing an xrefdelta entry for this.
| \movedxref{reverse.iter.op+=}{reverse.iter.nav} | ||
| \movedxref{reverse.iter.op\dcr}{reverse.iter.nav} | ||
| \movedxref{reverse.iter.op-=}{reverse.iter.nav} | ||
| \movedxref{reverse.iter.op++}{reverse.iter.nav} |
| \movedxref{reverse.iter.op>=}{reverse.iter.cmp} | ||
| \movedxref{reverse.iter.op<=}{reverse.iter.cmp} | ||
|
|
||
| \movedxref{reverse.iter.star}{reverse.iter.elem} |
There was a problem hiding this comment.
This was called reverse.iter.op.star, not reverse.iter.star, for some reason.
Partially addresses #1429.