sized-random-access-range

 

C++

Compiler support

Freestanding and hosted

Language

Standard library

Standard library headers

Named requirements

Feature test macros (C++20)

Language support library

Concepts library (C++20)

Diagnostics library

Memory management library

Metaprogramming library (C++11)

General utilities library

Containers library

Iterators library

Ranges library (C++20)

Algorithms library

Strings library

Text processing library

Numerics library

Date and time library

Input/output library

Filesystem library (C++17)

Concurrency support library (C++11)

Execution control library (C++26)

Technical specifications

Symbols index

External libraries

 

Ranges library

Range access begin cbegin      end cend rbegin crbegin      rend crend size ssize      data cdata reserve_hint (C++26)         empty

Range conversions std::from_range_t std::from_range (C++23)(C++23)         to (C++23)

Dangling iterator handling dangling borrowed_iterator_t borrowed_subrange_t

Range primitives
range_size_trange_difference_trange_value_t      elements_of (C++23)	iterator_tconst_iterator_tsentinel_tconst_sentinel_t (C++23)(C++23)	range_reference_trange_const_reference_trange_rvalue_reference_trange_common_reference_t (C++23)

Range concepts
range borrowed_range      common_range	sized_range viewable_range      view	input_range output_range forward_range	bidirectional_range      random_access_range contiguous_range	approximately_sized_range (C++26) constant_range (C++23) sized-random-access-range (C++26)

Views
view_interface	subrange

Range factories
empty_viewviews::empty	single_viewviews::single	basic_istream_viewviews::istream	iota_viewviews::iota views::indices (C++26)	repeat_viewviews::repeat (C++23)(C++23)

Range adaptors
views::all_tviews::all as_rvalue_viewviews::as_rvalue (C++23)(C++23)     filter_viewviews::filter transform_viewviews::transform take_viewviews::take take_while_viewviews::take_while common_viewviews::common views::counted as_input_viewviews::as_input (C++26)(C++26)	ref_view drop_viewviews::drop drop_while_viewviews::drop_while lazy_split_viewviews::lazy_split split_viewviews::split join_viewviews::join join_with_viewviews::join_with (C++23)(C++23) concat_viewviews::concat (C++26)(C++26) cache_latest_viewviews::cache_latest (C++26)(C++26)	owning_view reverse_viewviews::reverse as_const_viewviews::as_const (C++23)(C++23) elements_viewviews::elements keys_viewviews::keys values_viewviews::values enumerate_viewviews::enumerate (C++23)(C++23) zip_viewviews::zip (C++23)(C++23) zip_transform_viewviews::zip_transform (C++23)(C++23)	adjacent_viewviews::adjacent (C++23)(C++23) views::pairwise (C++23) adjacent_transform_viewviews::adjacent_transform (C++23)(C++23) views::pairwise_transform (C++23) chunk_viewviews::chunk (C++23)(C++23) slide_viewviews::slide (C++23)(C++23) chunk_by_viewviews::chunk_by (C++23)(C++23) stride_viewviews::stride (C++23)(C++23) cartesian_product_viewviews::cartesian_product (C++23)(C++23)

Range generators std::generator (C++23)

Range adaptor closure objects range_adaptor_closure (C++23)

Range adaptor objects

Helper items
copyable-box movable-box (until C++23)    (C++23)	simple-view non-propagating-cache

 

template< class T > concept /*sized-random-access-range*/ = ranges::random_access_range<T> && ranges::sized_range<T>;		(since C++26) (exposition only*)

The exposition-only concept sized-random-access-range specifies the requirements of a range type that knows its size in constant time and allows random access to its elements.

Notes

sized-random-access-range is used by parallel range algorithms to reject unbounded range argument. The reasons for that are:

• For efficient parallel implementation, the iteration space bounds need to be known. Otherwise, it is hard to apply the “divide and conquer” strategy for creating work for multiple execution threads.
• While serial range algorithms allow passing an “infinite” range like std::ranges::views::iota(0), it may result in an endless loop. It is hard to imagine usefulness of that in the case of parallel execution. Requiring data sequences to be bounded potentially prevents errors at run-time.
• Using explicitly bounded output ranges follows established practices of secure coding, which recommend or even require to always specify the size of the output in order to prevent out-of-range data modifications.
• If several input ranges or sequences are bounded, an algorithm should usually stop as soon as the end is reached for the shortest one. There are already precedents in the standard that an algorithm takes two sequences with potentially different input sizes and chooses the smaller size as the number of iterations it is going to make, such as std::ranges::transform and std::ranges::mismatch.
• As the output data sequences are bounded, they are no more assumed to have sufficient space for all the input. There is thus a possibility that some elements of the input sequences are not processed by an algorithm, and it makes sense for it to return iterators to the positions in these sequences where the processing stopped.