transcode_view & transcode_iterator

The transcode_view class provides a one-by-one view of the stored range’s code units as another encoding’s code units. Dereferencing the iterators returns a single code_unit value corresponding to the desired encoding’s type.

The range-based classes are excellent ways to walk over units of information in a low-memory environment, as they only store the minimum amount of data necessary to perform their operations on the fly. This reduces the speed but is fine for one-at-a-time encoding operations. To decode eagerly and in bulk, see the transcode functions. The paired transcode_iterator class does the bulk of the work and stores all of the information. It is paired with an empty, blank sentinel value so as to decrease the cost of iteration.

template<typename _FromEncoding, typename _ToEncoding = utf8_t, typename _Range = __txt_detail::__default_char_view_t<code_unit_t<_FromEncoding>>, typename _FromErrorHandler = default_handler_t, typename _ToErrorHandler = default_handler_t, typename _FromState = decode_state_t<_FromEncoding>, typename _ToState = encode_state_t<_ToEncoding>> class transcode_view : public view_base

A transcoding iterator that takes an input of code units and provides an output over the code units of the desired _ToEncoding after converting from the _FromEncoding in a fashion that will never produce a ztd::text::encoding_error::insufficient_output error.

Remark

This type produces proxies as their reference type, and are only readable, not writable iterators. The type will also try many different shortcuts for decoding the input and encoding the intermediates, respectively, including invoking a few customization points for either decode_one. or encode_one . It may also call transcode_one to bypass having to do the round-trip through two encodings, which an encoding pair that a developer is interested in can use to do the conversion more quickly. The view presents code units one at a time, regardless of how many code units are output by one decode operation. This means if, for example, one (1) UTF-16 code unit becomes two (2) UTF-8 code units, it will present each code unit one at a time. If you are looking to explicitly know each collection of characters, you will have to use lower-level interfaces.

Template Parameters

_FromEncoding – The encoding to read the underlying range of code points as.
_ToEncoding – The encoding to read the underlying range of code points as.
_Range – The range of input that will be fed into the _FromEncoding’s decode operation.
_FromErrorHandler – The error handler for any decode-step failures.
_ToErrorHandler – The error handler for any encode-step failures.
_FromState – The state type to use for the decode operations to intermediate code points.
_ToState – The state type to use for the encode operations to intermediate code points.

Public Types

using iterator = transcode_iterator<_FromEncoding, _ToEncoding, _Range, _FromErrorHandler, _ToErrorHandler, _FromState, _ToState>: The iterator type for this view.

using sentinel = transcode_sentinel_t: The sentinel type for this view.

using range_type = _Range : The underlying range type.

using from_encoding_type = _FromEncoding : The encoding type used for decoding to intermediate code point storage.

using to_encoding_type = _ToEncoding : The encoding type used for encoding to the final code units storage.

using from_error_handler_type = _FromErrorHandler : The error handler when a decode operation fails.

using to_error_handler_type = _ToErrorHandler : The error handler when an encode operation fails.

using from_state_type = _FromState : The state type used for decode operations.

using to_state_type = _ToState : The state type used for encode operations.

Public Functions

inline constexpr transcode_view(range_type __range) noexcept