[container.requirements.general]/Table 71 LWG 3352: Remove the requirement for strong_equality #3517
Conversation
…strong_equality Missed by #3475.
| & | ||
| \tcode{strong_ordering} & | ||
| & | ||
| \expects \tcode{X::iterator} meets the random access iterator requirements & |
There was a problem hiding this comment.
\expects seems wrong to me here: this requires all container iterators to implement <=> despite that it would be UB to evaluate i <=> j when i and j are iterators for non-random-access containers. Given that we have concepts that detect comparability based on the validity of expressions, it's not good design to have unusable expressions be well-formed.
I think we should handle this bit of cleanup via LWG issue rather than editorially.
There was a problem hiding this comment.
I think you are right. And there might 2 kinds of solution IMO, which is preferred?
- use
\mandates(or\constraints?) instead of\expects; - make a new requirement called "random access container" and move
i <=> jto its own table.
There was a problem hiding this comment.
Please be so kind and leave the LWG issue number in a comment here.
There was a problem hiding this comment.
I've submitted this issue to LWGChair, I'll reply back with an issue number once it's assigned.
jensmaurer
changed the title
jensmaurer
added
the
not-editorial
Missed by #3475.