bool | uses Hamcrest matchers to provide a clean way

Plain old natural numbers will not do the trick, because you can't calculate the natural number length of an infinite list in finite time. However, lazy natural numbers can do it.

The boolean conversion operator for std::basic_istream is explicit. This means that instances of the type will not implicitly become a bool but can be converted to one explicitly, for instance by typing bool(infile).

Explicit boolean conversion operators are considered for conditional statements, i.e. the expression parts of if, while etc. More info about contextual conversions here.

However, a return statement will not consider the explicit conversion operators or constructors. So you have to explicitly convert that to a boolean for a return.

This is a null-parameter check syntax being introduced in C# 11.

The proposal is here, and the PR doing a first roll-out to the runtime is here.

The syntax:

Maybe I'm wrong, but it seems that last part:

It looks like GCC's naïveté about store-forwarding stalls is hurting its auto-vectorization strategy here. See also Store forwarding by example for some practical benchmarks on Intel with hardware performance counters, and What are the costs of failed store-to-load forwarding on x86? Also Agner Fog's x86 optimization guides.

(gcc -O3 enables -ftree-vectorize and a few other options not included by -O2, e.g. if-conversion to branchless cmov, which is another way -O3 can hurt with data patterns GCC didn't expect. By comparison, Clang enables auto-vectorization even at -O2, although some of its optimizations are still only on at -O3.)

It's doing 64-bit loads (and branching to store or not) on pairs of ints. This means, if we swapped the last iteration, this load comes half from that store, half from fresh memory, so we get a store-forwarding stall after every swap. But bubble sort often has long chains of swapping every iteration as an element bubbles far, so this is really bad.

(Bubble sort is bad in general, especially if implemented naively without keeping the previous iteration's second element around in a register. It can be interesting to analyze the asm details of exactly why it sucks, so it is fair enough for wanting to try.)

Anyway, this is pretty clearly an anti-optimization you should report on GCC Bugzilla with the "missed-optimization" keyword. Scalar loads are cheap, and store-forwarding stalls are costly. (Can modern x86 implementations store-forward from more than one prior store? no, nor can microarchitectures other than in-order Atom efficiently load when it partially overlaps with one previous store, and partially from data that has to come from the L1d cache.)

Even better would be to keep buf[x+1] in a register and use it as buf[x] in the next iteration, avoiding a store and load. (Like good hand-written asm bubble sort examples, a few of which exist on Stack Overflow.)

If it wasn't for the store-forwarding stalls (which AFAIK GCC doesn't know about in its cost model), this strategy might be about break-even. SSE 4.1 for a branchless pmind / pmaxd comparator might be interesting, but that would mean always storing and the C source doesn't do that.

If this strategy of double-width load had any merit, it would be better implemented with pure integer on a 64-bit machine like x86-64, where you can operate on just the low 32 bits with garbage (or valuable data) in the upper half. E.g.,

It's all slightly mysterious. gcc behaves the same as clang.

The standard has this to say (emphasis mine):

Absent default member initializers, if any non-static data member of a union has a non-trivial default constructor, copy constructor, move constructor, copy assignment operator, move assignment operator, or destructor, the corresponding member function of the union must be user-provided or it will be implicitly deleted for the union.

But I think the wording is a bit wooly here. I think what they actually mean is that you must provide an initialiser for the member that has (in your example) a non-trivial constructor, like so:

If I understand correctly, the elements of Lam -> Lam would not be all set-theoretic functions (which are bigger than Lam by Cantor's theorem), but only continuous functions.

Yes, that's a common answer to the apparent paradox.

More in general, types are types, with their logical theory (type theory), while sets are sets (working as in set theory).

One is tempted to interpret types as sets, i.e. to constructs models of the lambda calculus within sets.

I suggest you read the milestone paper "Polymorphism is not set theoretic" by Reynolds. This shows that is you have "forall" types as in System F, and you interpret functions as sets in the "obvious" way, you get a contradiction, by a similar argument you made (Cantor's, roughly). Indeed, Reynolds constructs T ~= (T -> Bool) -> Bool.

Reynolds suggests indeed, one of the option is to interpret function types as a "subset" of functions on sets, e.g. continuous functions as you mentioned.

If so, what is the topology that defines this continuity

Scott topology.

and why are Haskell terms of type Lam -> Lam continuous for it ?

Roughly, because all the basic operations (lambda abstractions, applications, ...) are continuous, so you get a composition of continuous operators which has to be continuous.

Note that defining a proper formal semantics of Haskell is a daunting task. Even focusing on System F only, which is much simpler in comparison, models are already rather complex.

You could start from the simply-typed lambda calculus and learn about their categorical models: Cartesian closed categories (CCCs). Working in, say, the category of Scott-continuous functions is a special case of CCC. The CCC general structure makes you understand that, after you have a basic structure (essentially Currying and little else) you can use that to create a full-model, interpreting any (typed) lambda term.

The __enter__ method should return the context object. with ... as ... uses the return value of __enter__ to determine what object to give you. Since your __enter__ returns nothing, it implicitly returns None, so test is None.

From this blog post by Eric Niebler, whose range-V3 library heavily influenced C++20 ranges

... custom view types can inher[i]t from view_interface to get a host of useful member functions, like .front(), .back(), .empty(), .size(), .operator[], and even an explicit conversion to bool so that view types can be used in if statements