Casey Carter

@tomaszkam, I also suggest you compile `-DNDEBUG` to disable the debug checking in cmcstl2.

> They are (disguised) function objects, so my code will work as written in C++20. (We're a little circumspect in the current draft about whether the algorithms in `std::ranges` are...

> Putting -NDEBUG with -O2 does not help: Is that with `itinerary.cpp` from #182? I can't explain the differences we're seeing from the same command line and almost precisely the...

> I changed ... to something that is guaranteed to work with standard I've made a similar change to #182 (I stored the lambda in a named variable outside the...

> Being able to pass functions to functions is fundamental and very useful. Being able to pass function templates and overload sets to functions would be useful as well. I...

The (almost totally undocumented) design for output cursors is that you either: * Don't implement `next`, indicating that your output cursor is position-less/write-once. These iterators usually do all of their...

> Why is it called `post_increment()` instead of `next(int)`? Because we don't have to differentiate prefix and postfix versions of the single name `++` - we can use different names...

> `insert_iterator` (broken in the TS: updates to the stored iterator are lost), `ostreambuf_iterator` (also broken in the TS: updates to the value of `fail()` can be lost) Which is...

> IMO we should defer all basic_iterator issues until after Kona. I'd like to see the open bugs in the Ranges TS addressed first. Yes, apologies to Tom. I had...

I *think* this is another instance of [GCC PR 78715](https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78715), but I'll keep this issue open until the bug is resolved so I can make sure.