templated input adapters

This commit is contained in:
Francois Chabot 2020-02-19 10:32:49 -05:00
parent 973c52dd4a
commit 617b3cf42e
7 changed files with 623 additions and 310 deletions

View file

@ -31,9 +31,10 @@ namespace detail
/*!
@brief deserialization of CBOR, MessagePack, and UBJSON values
*/
template<typename BasicJsonType, typename SAX = json_sax_dom_parser<BasicJsonType>>
template<typename BasicJsonType, typename SAX = json_sax_dom_parser<BasicJsonType>, typename InputAdapterType = input_adapter_protocol>
class binary_reader
{
using input_adapter_ptr_t = std::shared_ptr<InputAdapterType>;
using number_integer_t = typename BasicJsonType::number_integer_t;
using number_unsigned_t = typename BasicJsonType::number_unsigned_t;
using number_float_t = typename BasicJsonType::number_float_t;
@ -46,7 +47,7 @@ class binary_reader
@param[in] adapter input adapter to read from
*/
explicit binary_reader(input_adapter_t adapter) : ia(std::move(adapter))
explicit binary_reader(input_adapter_ptr_t adapter) : ia(std::move(adapter))
{
(void)detail::is_sax_static_asserts<SAX, BasicJsonType> {};
assert(ia);
@ -1965,7 +1966,7 @@ class binary_reader
private:
/// input adapter
input_adapter_t ia = nullptr;
input_adapter_ptr_t ia = nullptr;
/// the current character
int current = std::char_traits<char>::eof();

View file

@ -45,14 +45,11 @@ struct input_adapter_protocol
virtual ~input_adapter_protocol() = default;
};
/// a type to simplify interfaces
using input_adapter_t = std::shared_ptr<input_adapter_protocol>;
/*!
Input adapter for stdio file access. This adapter read only 1 byte and do not use any
buffer. This adapter is a very low level adapter.
*/
class file_input_adapter : public input_adapter_protocol
class file_input_adapter final : public input_adapter_protocol
{
public:
JSON_HEDLEY_NON_NULL(2)
@ -87,7 +84,7 @@ characters following those used in parsing the JSON input. Clears the
std::istream flags; any input errors (e.g., EOF) will be detected by the first
subsequent call for input from the std::istream.
*/
class input_stream_adapter : public input_adapter_protocol
class input_stream_adapter final : public input_adapter_protocol
{
public:
~input_stream_adapter() override
@ -128,7 +125,7 @@ class input_stream_adapter : public input_adapter_protocol
};
/// input adapter for buffer input
class input_buffer_adapter : public input_adapter_protocol
class input_buffer_adapter final : public input_adapter_protocol
{
public:
input_buffer_adapter(const char* b, const std::size_t l) noexcept
@ -285,7 +282,7 @@ struct wide_string_input_helper<WideStringType, 2>
};
template<typename WideStringType>
class wide_string_input_adapter : public input_adapter_protocol
class wide_string_input_adapter final : public input_adapter_protocol
{
public:
explicit wide_string_input_adapter(const WideStringType& w) noexcept
@ -331,112 +328,164 @@ class wide_string_input_adapter : public input_adapter_protocol
std::size_t utf8_bytes_filled = 0;
};
class input_adapter
inline std::shared_ptr<file_input_adapter> input_adapter(std::FILE* file)
{
return std::make_shared<file_input_adapter>(file);
}
inline std::shared_ptr<input_stream_adapter> input_adapter(std::istream& stream)
{
return std::make_shared<input_stream_adapter>(stream);
}
inline std::shared_ptr<input_stream_adapter> input_adapter(std::istream&& stream)
{
return std::make_shared<input_stream_adapter>(stream);
}
template<typename CharT,
typename std::enable_if<
std::is_pointer<CharT>::value and
std::is_integral<typename std::remove_pointer<CharT>::type>::value and
sizeof(typename std::remove_pointer<CharT>::type) == 1,
int>::type = 0>
std::shared_ptr<input_buffer_adapter> input_adapter(CharT b, std::size_t l)
{
return std::make_shared<input_buffer_adapter>(reinterpret_cast<const char*>(b), l);
}
template<typename CharT,
typename std::enable_if<
std::is_pointer<CharT>::value and
std::is_integral<typename std::remove_pointer<CharT>::type>::value and
sizeof(typename std::remove_pointer<CharT>::type) == 1,
int>::type = 0>
std::shared_ptr<input_buffer_adapter> input_adapter(CharT b)
{
return input_adapter(reinterpret_cast<const char*>(b),
std::strlen(reinterpret_cast<const char*>(b)));
}
template<class IteratorType,
typename std::enable_if<
std::is_same<typename iterator_traits<IteratorType>::iterator_category, std::random_access_iterator_tag>::value,
int>::type = 0>
std::shared_ptr<input_buffer_adapter> input_adapter(IteratorType first, IteratorType last)
{
#ifndef NDEBUG
// assertion to check that the iterator range is indeed contiguous,
// see https://stackoverflow.com/a/35008842/266378 for more discussion
const auto is_contiguous = std::accumulate(
first, last, std::pair<bool, int>(true, 0),
[&first](std::pair<bool, int> res, decltype(*first) val)
{
res.first &= (val == *(std::next(std::addressof(*first), res.second++)));
return res;
}).first;
assert(is_contiguous);
#endif
// assertion to check that each element is 1 byte long
static_assert(
sizeof(typename iterator_traits<IteratorType>::value_type) == 1,
"each element in the iterator range must have the size of 1 byte");
const auto len = static_cast<size_t>(std::distance(first, last));
if (JSON_HEDLEY_LIKELY(len > 0))
{
// there is at least one element: use the address of first
return std::make_shared<input_buffer_adapter>(reinterpret_cast<const char*>(&(*first)), len);
}
else
{
// the address of first cannot be used: use nullptr
return std::make_shared<input_buffer_adapter>(nullptr, len);
}
}
inline std::shared_ptr<wide_string_input_adapter<std::wstring>> input_adapter(const std::wstring& ws)
{
return std::make_shared<wide_string_input_adapter<std::wstring>>(ws);
}
inline std::shared_ptr<wide_string_input_adapter<std::u16string>> input_adapter(const std::u16string& ws)
{
return std::make_shared<wide_string_input_adapter<std::u16string>>(ws);
}
inline std::shared_ptr<wide_string_input_adapter<std::u32string>> input_adapter(const std::u32string& ws)
{
return std::make_shared<wide_string_input_adapter<std::u32string>>(ws);
}
template<class ContiguousContainer, typename
std::enable_if<not std::is_pointer<ContiguousContainer>::value and
std::is_base_of<std::random_access_iterator_tag, typename iterator_traits<decltype(std::begin(std::declval<ContiguousContainer const>()))>::iterator_category>::value,
int>::type = 0>
std::shared_ptr<input_buffer_adapter> input_adapter(const ContiguousContainer& c)
{
return input_adapter(std::begin(c), std::end(c));
}
template<class T, std::size_t N>
std::shared_ptr<input_buffer_adapter> input_adapter(T (&array)[N])
{
return input_adapter(std::begin(array), std::end(array));
}
// This class only handles inputs of input_buffer_adapter type.
// It's required so that expressions like {ptr, len} can be implicitely casted
// to the correct adapter.
class span_input_adapter
{
public:
// native support
JSON_HEDLEY_NON_NULL(2)
input_adapter(std::FILE* file)
: ia(std::make_shared<file_input_adapter>(file)) {}
/// input adapter for input stream
input_adapter(std::istream& i)
: ia(std::make_shared<input_stream_adapter>(i)) {}
/// input adapter for input stream
input_adapter(std::istream&& i)
: ia(std::make_shared<input_stream_adapter>(i)) {}
input_adapter(const std::wstring& ws)
: ia(std::make_shared<wide_string_input_adapter<std::wstring>>(ws)) {}
input_adapter(const std::u16string& ws)
: ia(std::make_shared<wide_string_input_adapter<std::u16string>>(ws)) {}
input_adapter(const std::u32string& ws)
: ia(std::make_shared<wide_string_input_adapter<std::u32string>>(ws)) {}
/// input adapter for buffer
template<typename CharT,
typename std::enable_if<
std::is_pointer<CharT>::value and
std::is_integral<typename std::remove_pointer<CharT>::type>::value and
sizeof(typename std::remove_pointer<CharT>::type) == 1,
int>::type = 0>
input_adapter(CharT b, std::size_t l)
span_input_adapter(CharT b, std::size_t l)
: ia(std::make_shared<input_buffer_adapter>(reinterpret_cast<const char*>(b), l)) {}
// derived support
/// input adapter for string literal
template<typename CharT,
typename std::enable_if<
std::is_pointer<CharT>::value and
std::is_integral<typename std::remove_pointer<CharT>::type>::value and
sizeof(typename std::remove_pointer<CharT>::type) == 1,
int>::type = 0>
input_adapter(CharT b)
: input_adapter(reinterpret_cast<const char*>(b),
std::strlen(reinterpret_cast<const char*>(b))) {}
span_input_adapter(CharT b)
: span_input_adapter(reinterpret_cast<const char*>(b),
std::strlen(reinterpret_cast<const char*>(b))) {}
/// input adapter for iterator range with contiguous storage
template<class IteratorType,
typename std::enable_if<
std::is_same<typename iterator_traits<IteratorType>::iterator_category, std::random_access_iterator_tag>::value,
int>::type = 0>
input_adapter(IteratorType first, IteratorType last)
{
#ifndef NDEBUG
// assertion to check that the iterator range is indeed contiguous,
// see https://stackoverflow.com/a/35008842/266378 for more discussion
const auto is_contiguous = std::accumulate(
first, last, std::pair<bool, int>(true, 0),
[&first](std::pair<bool, int> res, decltype(*first) val)
{
res.first &= (val == *(std::next(std::addressof(*first), res.second++)));
return res;
}).first;
assert(is_contiguous);
#endif
span_input_adapter(IteratorType first, IteratorType last)
: ia(input_adapter(first, last)) {}
// assertion to check that each element is 1 byte long
static_assert(
sizeof(typename iterator_traits<IteratorType>::value_type) == 1,
"each element in the iterator range must have the size of 1 byte");
const auto len = static_cast<size_t>(std::distance(first, last));
if (JSON_HEDLEY_LIKELY(len > 0))
{
// there is at least one element: use the address of first
ia = std::make_shared<input_buffer_adapter>(reinterpret_cast<const char*>(&(*first)), len);
}
else
{
// the address of first cannot be used: use nullptr
ia = std::make_shared<input_buffer_adapter>(nullptr, len);
}
}
/// input adapter for array
template<class T, std::size_t N>
input_adapter(T (&array)[N])
: input_adapter(std::begin(array), std::end(array)) {}
span_input_adapter(T (&array)[N])
: span_input_adapter(std::begin(array), std::end(array)) {}
/// input adapter for contiguous container
template<class ContiguousContainer, typename
std::enable_if<not std::is_pointer<ContiguousContainer>::value and
std::is_base_of<std::random_access_iterator_tag, typename iterator_traits<decltype(std::begin(std::declval<ContiguousContainer const>()))>::iterator_category>::value,
int>::type = 0>
input_adapter(const ContiguousContainer& c)
: input_adapter(std::begin(c), std::end(c)) {}
span_input_adapter(const ContiguousContainer& c)
: span_input_adapter(std::begin(c), std::end(c)) {}
operator input_adapter_t()
std::shared_ptr<input_buffer_adapter> get()
{
return ia;
}
private:
/// the actual adapter
input_adapter_t ia = nullptr;
std::shared_ptr<input_buffer_adapter> ia = nullptr;
};
} // namespace detail
} // namespace nlohmann

View file

@ -22,19 +22,9 @@ namespace detail
// lexer //
///////////
/*!
@brief lexical analysis
This class organizes the lexical analysis during JSON deserialization.
*/
template<typename BasicJsonType>
class lexer
class lexer_base
{
using number_integer_t = typename BasicJsonType::number_integer_t;
using number_unsigned_t = typename BasicJsonType::number_unsigned_t;
using number_float_t = typename BasicJsonType::number_float_t;
using string_t = typename BasicJsonType::string_t;
public:
/// token types for the parser
enum class token_type
@ -75,9 +65,9 @@ class lexer
return "null literal";
case token_type::value_string:
return "string literal";
case lexer::token_type::value_unsigned:
case lexer::token_type::value_integer:
case lexer::token_type::value_float:
case token_type::value_unsigned:
case token_type::value_integer:
case token_type::value_float:
return "number literal";
case token_type::begin_array:
return "'['";
@ -103,15 +93,33 @@ class lexer
// LCOV_EXCL_STOP
}
}
};
/*!
@brief lexical analysis
explicit lexer(detail::input_adapter_t&& adapter)
This class organizes the lexical analysis during JSON deserialization.
*/
template<typename BasicJsonType, typename InputAdapterType = input_adapter_protocol>
class lexer : public lexer_base<BasicJsonType>
{
using input_adapter_ptr_t = std::shared_ptr<InputAdapterType>;
using number_integer_t = typename BasicJsonType::number_integer_t;
using number_unsigned_t = typename BasicJsonType::number_unsigned_t;
using number_float_t = typename BasicJsonType::number_float_t;
using string_t = typename BasicJsonType::string_t;
public:
using token_type = typename lexer_base<BasicJsonType>::token_type;
explicit lexer(input_adapter_ptr_t&& adapter)
: ia(std::move(adapter)), decimal_point_char(get_decimal_point()) {}
// delete because of pointer members
lexer(const lexer&) = delete;
lexer(lexer&&) = delete;
lexer(lexer&&) = default;
lexer& operator=(lexer&) = delete;
lexer& operator=(lexer&&) = delete;
lexer& operator=(lexer&&) = default;
~lexer() = default;
private:
@ -1480,7 +1488,7 @@ scan_number_done:
private:
/// input adapter
detail::input_adapter_t ia = nullptr;
input_adapter_ptr_t ia = nullptr;
/// the current character
std::char_traits<char>::int_type current = std::char_traits<char>::eof();

View file

@ -24,44 +24,46 @@ namespace detail
// parser //
////////////
enum class parse_event_t : uint8_t
{
/// the parser read `{` and started to process a JSON object
object_start,
/// the parser read `}` and finished processing a JSON object
object_end,
/// the parser read `[` and started to process a JSON array
array_start,
/// the parser read `]` and finished processing a JSON array
array_end,
/// the parser read a key of a value in an object
key,
/// the parser finished reading a JSON value
value
};
template<typename BasicJsonType>
using parser_callback_t =
std::function<bool(int depth, parse_event_t event, BasicJsonType& parsed)>;
/*!
@brief syntax analysis
This class implements a recursive descent parser.
*/
template<typename BasicJsonType>
template<typename BasicJsonType, typename InputAdapterType = input_adapter_protocol>
class parser
{
using input_adapter_ptr_t = std::shared_ptr<InputAdapterType>;
using number_integer_t = typename BasicJsonType::number_integer_t;
using number_unsigned_t = typename BasicJsonType::number_unsigned_t;
using number_float_t = typename BasicJsonType::number_float_t;
using string_t = typename BasicJsonType::string_t;
using lexer_t = lexer<BasicJsonType>;
using lexer_t = lexer<BasicJsonType, InputAdapterType>;
using token_type = typename lexer_t::token_type;
public:
enum class parse_event_t : uint8_t
{
/// the parser read `{` and started to process a JSON object
object_start,
/// the parser read `}` and finished processing a JSON object
object_end,
/// the parser read `[` and started to process a JSON array
array_start,
/// the parser read `]` and finished processing a JSON array
array_end,
/// the parser read a key of a value in an object
key,
/// the parser finished reading a JSON value
value
};
using parser_callback_t =
std::function<bool(int depth, parse_event_t event, BasicJsonType& parsed)>;
/// a parser reading from an input adapter
explicit parser(detail::input_adapter_t&& adapter,
const parser_callback_t cb = nullptr,
explicit parser(input_adapter_ptr_t&& adapter,
const parser_callback_t<BasicJsonType> cb = nullptr,
const bool allow_exceptions_ = true)
: callback(cb), m_lexer(std::move(adapter)), allow_exceptions(allow_exceptions_)
{
@ -486,7 +488,7 @@ class parser
private:
/// callback function
const parser_callback_t callback = nullptr;
const parser_callback_t<BasicJsonType> callback = nullptr;
/// the type of the last read token
token_type last_token = token_type::uninitialized;
/// the lexer