[basic.scope.class] Reinstate a qualification that was lost in 0e26279b88c3b8b0a09babdeec8418d383f07419. #3312
Conversation
Without the introductory sentence that was deleted by that commit, we need to say explicitly that we are talking about a declaration _in a class_.
|
Why we don't need similar fix in the previous paragraph? |
|
The next paragraph is just an example, isn't it? |
|
The previous, not the next. The third paragraph also speaks about declarations whose scope extends to or past the end of a class definition. (Also the word "definition" is missing at the end). |
|
Ah, sorry, I see. (I had somehow misread the original message as "fourth and fifth", sorry!) So for the previous paragraph, we can say "hides a declaration of the same name in that class"? The "definition" was also absent from the original introductory paragraph, do you think it matters? |
It was originally saying "fourht and fifth", this was my mistake, sorry.
Let me redirect this to @zygoloid |
|
@languagelawyer For the third paragraph, do you have an example in mind? I think the slightly broader wording we have here is equivalent to the more-restrictive version. In fact, paragraph 3 seems to be entirely redundant, since it follows from [basic.lookup.unqual]p1 and p8. It's also unlikely to be worth spending much time on editorial improvements here, since http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2019/p1787r2.html is replacing all this wording and is likely to land next meeting (or maybe the one after, depending on how fast CWG gets through the NB comments). |
|
@zygoloid No, I don't have an example. I doubt there is an example even for the fourth paragraph. |
Without the introductory sentence that was deleted by that commit, we need to say explicitly that we are talking about a declaration in a class.
See also discussion on #1137.