[associative.reqmts.general] Fix confusing local use of 'r' #5138
Conversation
jensmaurer
changed the title
| @@ -1710,21 +1710,24 @@ | |||
| \tcode{c} denotes a possibly \keyword{const} value of type \tcode{X::key_compare}; | |||
| \item | |||
| \tcode{kl} is a value such that \tcode{a} is partitioned\iref{alg.sorting} | |||
| with respect to \tcode{c(r, kl)}, with \tcode{r} the key value of \tcode{e} | |||
There was a problem hiding this comment.
I appreciate the heavy maths jargon here, and even though it resonates on a personal level, I think it's maybe a bit unnecessary in a general text. What do you think of changing this to the more normal "where r is the key value of e and e is in [or "an element of"] a"?
There was a problem hiding this comment.
Well, the existing text already has the phrase; we're just replacing "r" with "x" and making sure the phrase is duplicated everywhere "x" is used.
The later bullets already contain a "with" phrase, so continuing with "where" is a bit inconsistent.
There was a problem hiding this comment.
(If we want to improve clarity here, adding "all" would probably also be helpful. But maybe that's implied by "partitioned".)
There was a problem hiding this comment.
It's the "consistency with a bad idea" thing... OK, let's make the change as is, and maybe we can rephrase everything afterwards.
There was a problem hiding this comment.
(Note to selves: revisit this.)
|
@jwakely , @StephanTLavavej , please have a look. |
There was a problem hiding this comment.
As I noted in #5134 it would be a smaller edit to change the first r which is the one that introduced the problem (the cases being changed were there first, since C++14).
But this is also OK. We'd need some of these changes to ensure r is defined for the bullets following (8.19) anyway.
|
@tkoeppe, this seems ready. |
Fixes #5134