David Chamont

For a later session, about smartPointers exercice

1

I find disturbing that the first step (problem1) requires the use of the array flavor of std::make_unique, which has not been covered in the course.

Atomic exercise DOES NOT race on my machine

6

@hageboeck @sponce @bernhardmgruber I suspect the `-O2` compiling option (with g++ 11.3. ) make that `for (int i = 0; i < 100; ++i) { a++; }` is some howreplaced...

- We should declare as a requirement the knowledge of object-functions. - I wonder how the idea of "manahattan distance" mix well with the idea of complexes. Perhaps we should...

For a later session : about the templates exercise

1

I dislike a little the final question (ordering complex of complex), because I suspect such "complex of complex" is a mathematical non-sense. Isn´t it ?

Exercise code/operators, about the solution

5

@sponce @bernhardmgruber @hageboeck I generally recommand to make any unary constructor explicitly. Yet, in this example, how dangerous is it to let the compiler transform an int into some Fraction...

Present C++ notebooks

1

If needed...

Have a progressive stl exercise ?

1

I do not know what would be the right solution, but it is a problem that the stl exercise cannot be compiled until every hole is completed.

Extract from global README a dedicated README-contributors

1

As a reminder.

Use std::vector in smartPointers and stl exercises ?

8

It is my strong feeling that the use of raw C arrays makes the syntax uselessly complex, and divert the learners from what we want to demonstrate. It would rather...

Any plan to support C++20 ?

9

There was some hope that after the recent cling 1.0 release, we may get C++20 in xeus-cling. Optimistic ?