Using C++20 in QC #2018
Using C++20 in QC #2018
Conversation
It did not really need c++20, but had a related todo comment. Initially, I assumed that using the concrete type as an argument would prevent us from using it as const, ref, const ref, but actually c++ converts it properly.
Applied only to simple cases, but we could probably rewrite more with additional time investment.
|
Rebasing an old PR with some attempts at C++20... |
| } | ||
|
|
||
| /// Produces the most constrained Activity which will match all the provided | ||
| template <typename Iter, typename Accessor = const Activity& (*)(const Activity&)> | ||
| Activity strictestMatchingActivity( | ||
| Iter begin, Iter end, Accessor accessor = [](const Activity& a) -> const Activity& { return a; }) |
There was a problem hiding this comment.
why was accessor removed?
There was a problem hiding this comment.
The accessor was added in the first place because we did not have a way to easily pass some kind of iterator of all the activities if they are members bigger structures. std::ranges allow for that, as it is demostrated in the test, so we can remove this and expect a RangeOfActivities instead.
| template <class... Args, class Enable = std::enable_if_t<(... && std::is_convertible_v<Args, DataSourceType>)>> | ||
| bool isOneOf(Args... dataSourceType) const | ||
|
|
||
| template <class... DataSourceType> |
There was a problem hiding this comment.
shouldn't there be any concept similarly to computeNumericalActivityField ?
There was a problem hiding this comment.
Why? this method is a member of DataSourceSpec and expects DataSourceType only, but it can accept one or more of those.
Do you mean just saying something like requires T == DataSourceType)?
There was a problem hiding this comment.
the "todo" comment was probably my lack of awareness, that it could be done simpler.
There was a problem hiding this comment.
well, there was a structure that checks whether the template args can be converted to the enum, so I thought that you want to keep the functionality
There was a problem hiding this comment.
no, not really. i think i had thought that one needs is_convertible to make it work also with const DataSourceType or const DataSourceType&, which apparently is not needed in the end.
Now mostly as a demonstrator, since we need O2 & ROOT to move first.
I can confirm though that the QC library compiles fine and passes the tests in o2-qc-test-core