Quuxplusone added inline comments.
================
Comment at: llvm/include/llvm/ADT/IterableTraits.h:17-31
+namespace detail {
+
+template <typename I> auto is_forward_iterator(...) -> std::false_type;
+
+template <typename I>
+auto is_forward_iterator(int) -> decltype(
+ std::next(std::declval<I>()),
----------------
Instead of all this, I would recommend
```
template<class It>
using is_input_iterator = std::is_base_of<std::input_iterator_tag, typename
std::iterator_traits<It>::iterator_category>;
template<class R>
using iterator_type_of = std::decay_t<decltype(std::begin(std::declval<R>()))>;
template<class R, class IsInputIterable = std::true_type>
struct is_range_iterable : std::false_type {};
template<class R>
struct is_range_iterable<R, typename
is_input_iterator<iterator_type_of<R>>::type> : std::true_type {};
```
(Notice that we're supposed to be checking for //input// iterability, not
//forward// iterability. But also, the right way to see if something is an
input iterator is not to check an ad-hoc bunch of operators like `*` and `!=`,
but simply to check its STL iterator category.)
However, in real life, I personally would reject this patch, because I think
"saving a couple of calls to begin/end at the call site" is not worth the peril
of letting people accidentally write e.g.
```
SmallVector<SomeKindaString> v("hello world");
assert(v.size() == 12); // elements are "h"s, "e"s, "l"s, etc.
```
================
Comment at: llvm/include/llvm/ADT/SmallVector.h:1191-1194
+ template <typename Iterable>
+ explicit SmallVector(
+ Iterable &&Range,
+ std::enable_if_t<llvm::is_range_iterable<Iterable>::value, bool> = true)
----------------
Watch out: this greedy constructor template will be used for non-const copy
construction, e.g.
SmallVector<int> x;
SmallVector<int> y = x; // calls
SmallVector<int>::SmallVector<SmallVector<int>&>(SmallVector<int>&, bool),
// which is a better match than SmallVector<int>::SmallVector(const
SmallVector<int>&)
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D102760/new/
https://reviews.llvm.org/D102760
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
