Replace jessilib::encoding with jessilib::text_encoding

src/bot/console/console.cpp

@@ -33,7 +33,7 @@ void console_input_loop() {
 	std::wstring input;
 	auto shutdown_future = get_shutdown_future();
 	while (shutdown_future.wait_for(std::chrono::milliseconds(10)) != std::future_status::ready) {
-		std::getline(std::wcin, input); // TODO: use a non-bloicking call and poll running periodically?
+		std::getline(std::wcin, input); // TODO: use a non-blocking call and poll running periodically?
 		jessibot::io::console_command_context context{ jessilib::string_cast<char8_t>(input) };
 		if (!command_manager::instance().execute_command(context)) {
 			text error_text{ u8"ERROR", text::property::bold, color{ 0xFF0000 }};

src/common/config.cpp

@@ -53,6 +53,11 @@ std::string config::format() const {
 	return m_format;
 }
 
+text_encoding config::encoding() const {
+	std::shared_lock<std::shared_mutex> guard{ m_mutex };
+	return m_encoding;
+}
+
 /** Modifiers */
 void config::set_data(const object& in_data) {
 	std::lock_guard<std::shared_mutex> guard{ m_mutex };
@@ -60,16 +65,17 @@ void config::set_data(const object& in_data) {
 }
 
 /** File I/O */
-void config::load(const std::filesystem::path& in_filename, const std::string& in_format) {
+void config::load(const std::filesystem::path& in_filename, const std::string& in_format, text_encoding in_encoding) {
 	jessilib_assert(!in_filename.empty());
 	std::lock_guard<std::shared_mutex> guard{ m_mutex };
 
 	// Determine format
 	m_filename = in_filename;
 	m_format = get_format(m_filename, in_format);
+	m_encoding = in_encoding;
 
 	// Load
-	m_data = read_object(m_filename, m_format);
+	m_data = read_object(m_filename, m_format, m_encoding);
 }
 
 void config::reload() {
@@ -92,20 +98,21 @@ void config::write() const {
 	}
 }
 
-void config::write(const std::filesystem::path& in_filename , const std::string& in_format) {
+void config::write(const std::filesystem::path& in_filename , const std::string& in_format, text_encoding in_encoding) {
 	jessilib_assert(!in_filename.empty());
 	std::lock_guard<std::shared_mutex> guard{ m_mutex };
 
 	// Setup
 	m_filename = in_filename;
 	m_format = get_format(m_filename, in_format);
+	m_encoding = in_encoding;
 
 	// Write
-	write_object(m_data, m_filename, m_format);
+	write_object(m_data, m_filename, m_format, m_encoding);
 }
 
 /** Static File I/O */
-object config::read_object(const std::filesystem::path& in_filename, const std::string& in_format) {
+object config::read_object(const std::filesystem::path& in_filename, const std::string& in_format, text_encoding in_encoding) {
 	// Open up file for reading
 	std::ifstream file{ in_filename, std::ios::in | std::ios::binary };
 	if (!file) {
@@ -114,10 +121,10 @@ object config::read_object(const std::filesystem::path& in_filename, const std::
 	}
 
 	// Deserialize1
-	return deserialize_object(file, get_format(in_filename, in_format));
+	return deserialize_object(file, get_format(in_filename, in_format), in_encoding);
 }
 
-void config::write_object(const object& in_object, const std::filesystem::path& in_filename, const std::string& in_format) {
+void config::write_object(const object& in_object, const std::filesystem::path& in_filename, const std::string& in_format, text_encoding in_encoding) {
 	// Open up file for writing
 	std::ofstream file{ in_filename, std::ios::out | std::ios::binary };
 	if (!file) {
@@ -126,7 +133,7 @@ void config::write_object(const object& in_object, const std::filesystem::path&
 	}
 
 	// Deserialize1
-	return serialize_object(file, in_object, get_format(in_filename, in_format));
+	return serialize_object(file, in_object, get_format(in_filename, in_format), in_encoding);
 }
 
 std::string config::get_format(const std::filesystem::path& in_filename, const std::string& in_format) {

src/common/parser/parser.cpp

@@ -17,12 +17,11 @@
  */
 
 #include "parser.hpp"
-#include "unicode.hpp"
 #include <istream>
 
 namespace jessilib {
 
-object parser::deserialize_bytes(std::istream& in_stream, encoding in_read_encoding) {
+object parser::deserialize_bytes(std::istream& in_stream, text_encoding in_read_encoding) {
 	std::vector<byte_type> data;
 
 	// Read entire stream into data
@@ -36,10 +35,10 @@ object parser::deserialize_bytes(std::istream& in_stream, encoding in_read_encod
 	return deserialize_bytes(bytes_view_type{ &data.front(), data.size() }, in_read_encoding);
 }
 
-void parser::serialize_bytes(std::ostream& in_stream, const object& in_object, encoding in_write_encoding) {
+void parser::serialize_bytes(std::ostream& in_stream, const object& in_object, text_encoding in_write_encoding) {
 	// TODO: replace this method
 	auto bytes = serialize_bytes(in_object, in_write_encoding);
-	in_stream << bytes;
+	in_stream.write(bytes.data(), bytes.size());
 }
 
 } // namespace jessilib

src/common/parsers/json.cpp

@@ -20,26 +20,26 @@
 
 namespace jessilib {
 
-object json_parser::deserialize_bytes(bytes_view_type in_data, encoding in_write_encoding) {
+object json_parser::deserialize_bytes(bytes_view_type in_data, text_encoding in_write_encoding) {
 	object result;
 
-	if (in_write_encoding == encoding::utf_8) {
+	if (in_write_encoding == text_encoding::utf_8) {
 		std::u8string_view data_view = jessilib::string_view_cast<char8_t>(in_data);
 		deserialize_json<char8_t, true>(result, data_view);
 	}
-	else if (in_write_encoding == encoding::utf_16) {
+	else if (in_write_encoding == text_encoding::utf_16) {
 		std::u16string_view data_view = jessilib::string_view_cast<char16_t>(in_data);
 		deserialize_json<char16_t, true>(result, data_view);
 	}
-	else if (in_write_encoding == encoding::utf_32) {
+	else if (in_write_encoding == text_encoding::utf_32) {
 		std::u32string_view data_view = jessilib::string_view_cast<char32_t>(in_data);
 		deserialize_json<char32_t, true>(result, data_view);
 	}
-	else if (in_write_encoding == encoding::wchar) {
+	else if (in_write_encoding == text_encoding::wchar) {
 		std::wstring_view data_view = jessilib::string_view_cast<wchar_t>(in_data);
 		deserialize_json<wchar_t, true>(result, data_view);
 	}
-	else if (in_write_encoding == encoding::multibyte) {
+	else if (in_write_encoding == text_encoding::multibyte) {
 		// TODO: support without copying... somehow
 		auto u8_data = mbstring_to_ustring<char8_t>(jessilib::string_view_cast<char>(in_data));
 		std::u8string_view data_view = u8_data.second;
@@ -49,15 +49,15 @@ object json_parser::deserialize_bytes(bytes_view_type in_data, encoding in_write
 	return result;
 }
 
-std::string json_parser::serialize_bytes(const object& in_object, encoding in_write_encoding) {
+std::string json_parser::serialize_bytes(const object& in_object, text_encoding in_write_encoding) {
 	switch (in_write_encoding) {
-		case encoding::utf_8:
+		case text_encoding::utf_8:
 			return serialize_impl<char8_t, char>(in_object);
-		case encoding::utf_16:
+		case text_encoding::utf_16:
 			return serialize_impl<char16_t, char>(in_object);
-		case encoding::utf_32:
+		case text_encoding::utf_32:
 			return serialize_impl<char16_t, char>(in_object);
-		case encoding::wchar:
+		case text_encoding::wchar:
 			return serialize_impl<char16_t, char>(in_object);
 		default:
 			break;

src/common/serialize.cpp

@@ -53,8 +53,8 @@ object deserialize_object(std::u8string_view in_data, const std::string& in_form
 	return get_parser(in_format)->deserialize(in_data);
 }
 
-object deserialize_object(std::istream& in_stream, const std::string& in_format) {
+object deserialize_object(std::istream& in_stream, const std::string& in_format, text_encoding in_encoding) {
-	return get_parser(in_format)->deserialize_bytes(in_stream, encoding::utf_8);
+	return get_parser(in_format)->deserialize_bytes(in_stream, in_encoding);
 }
 
 /** Serialization */
@@ -62,8 +62,10 @@ std::u8string serialize_object(const object& in_object, const std::string& in_fo
 	return get_parser(in_format)->serialize<char8_t>(in_object);
 }
 
-void serialize_object(std::ostream& in_stream, const object& in_object, const std::string& in_format) {
+void serialize_object(std::ostream& in_stream, const object& in_object, const std::string& in_format, text_encoding in_encoding) {
-	get_parser(in_format)->serialize_bytes(in_stream, in_object, encoding::utf_8);
+	in_object.get<object::string_view_type>(object::string_view_type{});
+
+	get_parser(in_format)->serialize_bytes(in_stream, in_object, in_encoding);
 }
 
 } // namespace jessilib

src/include/jessilib/config.hpp

@@ -24,6 +24,7 @@
 #include <filesystem>
 #include <shared_mutex>
 #include "object.hpp"
+#include "text_encoding.hpp"
 
 namespace jessilib {
 
@@ -41,19 +42,20 @@ public:
 	object data() const;
 	std::filesystem::path filename() const;
 	std::string format() const;
+	text_encoding encoding() const;
 
 	/** Modifiers */
 	void set_data(const object& in_data);
 
 	/** File I/O */
-	void load(const std::filesystem::path& in_filename, const std::string& in_format = {});
+	void load(const std::filesystem::path& in_filename, const std::string& in_format = {}, text_encoding in_encoding = text_encoding::utf_8);
 	void reload();
 	void write() const;
-	void write(const std::filesystem::path& in_filename , const std::string& in_format = {});
+	void write(const std::filesystem::path& in_filename, const std::string& in_format = {}, text_encoding in_encoding = text_encoding::utf_8);
 
 	/** Static File I/O */
-	static object read_object(const std::filesystem::path& in_filename, const std::string& in_format = {});
+	static object read_object(const std::filesystem::path& in_filename, const std::string& in_format = {}, text_encoding in_encoding = text_encoding::utf_8);
-	static void write_object(const object& in_object, const std::filesystem::path& in_filename, const std::string& in_format = {});
+	static void write_object(const object& in_object, const std::filesystem::path& in_filename, const std::string& in_format = {}, text_encoding in_encoding = text_encoding::utf_8);
 
 	/** Static helpers */
 	static std::string get_format(const std::filesystem::path& in_filename, const std::string& in_format = {});
@@ -62,6 +64,7 @@ private:
 	mutable std::shared_mutex m_mutex;
 	object m_data;
 	std::string m_format;
+	text_encoding m_encoding;
 	std::filesystem::path m_filename;
 };
 

src/include/jessilib/object.hpp

@@ -252,12 +252,24 @@ public:
 	T get(DefaultT&& in_default_value) const {
 		const string_type* result = std::get_if<string_type>(&m_value);
 		if (result != nullptr) {
-			return *result;
+			return T{ *result };
 		}
 
 		return { in_default_value.begin(), in_default_value.end() };
 	}
 
+	// TODO: support other basic_string_view types
+	template<typename T, typename DefaultT = T,
+		typename std::enable_if<std::is_same<T, string_view_type>::value && std::is_same<typename std::decay<DefaultT>::type, string_view_type>::value>::type* = nullptr>
+	T get(DefaultT&& in_default_value) const {
+		const string_type* result = std::get_if<string_type>(&m_value);
+		if (result != nullptr) {
+			return T{ *result };
+		}
+
+		return in_default_value;
+	}
+
 	/** arrays */
 
 	// reference getter (array_type)

src/include/jessilib/parser.hpp

@@ -20,7 +20,7 @@
 
 #include <memory>
 #include "object.hpp"
-#include "unicode_base.hpp"
+#include "text_encoding.hpp"
 #include "impl/parser_manager.hpp"
 
 namespace jessilib {
@@ -40,15 +40,15 @@ public:
 	 * @param in_stream Stream to deserialize object from
 	 * @return A valid (possibly null) object
 	 */
-	virtual object deserialize_bytes(std::istream& in_stream, encoding in_read_encoding);
+	virtual object deserialize_bytes(std::istream& in_stream, text_encoding in_read_encoding);
-	virtual object deserialize_bytes(bytes_view_type in_data, encoding in_read_encoding) = 0;
+	virtual object deserialize_bytes(bytes_view_type in_data, text_encoding in_read_encoding) = 0;
-	virtual void serialize_bytes(std::ostream& in_stream, const object& in_object, encoding in_write_encoding);
+	virtual void serialize_bytes(std::ostream& in_stream, const object& in_object, text_encoding in_write_encoding);
-	virtual std::string serialize_bytes(const object& in_object, encoding in_write_encoding) = 0;
+	virtual std::string serialize_bytes(const object& in_object, text_encoding in_write_encoding) = 0;
 
 	template<typename CharT>
 	object deserialize(std::basic_string_view<CharT> in_text) {
 		bytes_view_type byte_view{ reinterpret_cast<const byte_type*>(in_text.data()), in_text.size() * sizeof(CharT) };
-		return deserialize_bytes(byte_view, default_encoding_info<CharT>::text_encoding);
+		return deserialize_bytes(byte_view, default_encoding_info<CharT>::encoding);
 	}
 
 	// Perhaps this could be condensed down to a simple method such that: serialize(out_variant, in_object, encoding)?

src/include/jessilib/parsers/json.hpp

@@ -30,8 +30,8 @@ namespace jessilib {
 class json_parser : public parser {
 public:
 	/** deserialize/serialize overrides */
-	object deserialize_bytes(bytes_view_type in_data, encoding in_write_encoding) override;
+	object deserialize_bytes(bytes_view_type in_data, text_encoding in_write_encoding) override;
-	std::string serialize_bytes(const object& in_object, encoding in_write_encoding) override;
+	std::string serialize_bytes(const object& in_object, text_encoding in_write_encoding) override;
 
 	std::u8string serialize_u8(const object& in_object) override { return serialize_impl<char8_t>(in_object); }
 	std::u16string serialize_u16(const object& in_object) override { return serialize_impl<char16_t>(in_object); }

src/include/jessilib/serialize.hpp

@@ -20,6 +20,7 @@
 
 #include <istream>
 #include "object.hpp"
+#include "text_encoding.hpp"
 
 namespace jessilib {
 
@@ -34,10 +35,10 @@ object deserialize_object(std::u16string_view in_data, const std::string& in_for
 object deserialize_object(std::u32string_view in_data, const std::string& in_format);
 object deserialize_object(const std::vector<char8_t>& in_data, const std::string& in_format);
 //object deserialize_object(std::u8string_view in_data, const std::string& in_format);
-object deserialize_object(std::istream& in_stream, const std::string& in_format); // TODO: add encoding param
+object deserialize_object(std::istream& in_stream, const std::string& in_format, text_encoding in_encoding = text_encoding::unknown);
 
 /** Serialization */
 std::u8string serialize_object(const object& in_object, const std::string& in_format); // TODO: templatize?
-void serialize_object(std::ostream& in_stream, const object& in_object, const std::string& in_format); // TODO: add encoding param
+void serialize_object(std::ostream& in_stream, const object& in_object, const std::string& in_format, text_encoding in_encoding = text_encoding::unknown);
 
 } // namespace jessilib

src/include/jessilib/text_encoding.hpp

@@ -0,0 +1,224 @@
+/**
+ * Copyright (C) 2021 Jessica James.
+ *
+ * Permission to use, copy, modify, and/or distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
+ * SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION
+ * OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
+ * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ *
+ * Written by Jessica James <jessica.aj@outlook.com>
+ */
+
+#pragma once
+
+#include <bit>
+#include <string_view>
+
+namespace jessilib {
+
+static constexpr char32_t byte_order_mark_codepoint = 0xFEFF;
+
+static_assert(sizeof(wchar_t) == sizeof(char16_t) || sizeof(wchar_t) == sizeof(char32_t),
+	"Unexpected wchar_t size; neither char16 nor char32");
+
+enum class text_encoding {
+	utf_8 = 0,
+
+	utf_16_little = 1,
+	utf_16_big = 2,
+	utf_16_native = (std::endian::native == std::endian::little ? utf_16_little : utf_16_big),
+	utf_16_foreign = (std::endian::native == std::endian::little ? utf_16_big : utf_16_little),
+	utf_16 = utf_16_native, // Alias for native
+
+	utf_32_little = 3,
+	utf_32_big = 4,
+	utf_32_native = (std::endian::native == std::endian::little ? utf_32_little : utf_32_big),
+	utf_32_foreign = (std::endian::native == std::endian::little ? utf_32_big : utf_32_little),
+	utf_32 = utf_32_native, // Alias for native
+
+	wchar = 5, // essentially only really for std::wcout / std::wcout
+	multibyte = 6, // essentially only really for std::cout / std::cin
+	//wchar = (sizeof(wchar_t) == sizeof(char16_t) ? utf_16 : utf_32),
+
+	unknown
+};
+
+template<text_encoding EncodingV>
+struct encoding_info;
+
+template<>
+struct encoding_info<text_encoding::utf_8> {
+	using data_type = char8_t;
+	static constexpr bool is_little = false;
+	static constexpr bool is_big = false;
+	static constexpr bool is_native = true;
+	static constexpr bool is_foreign = !is_native;
+	static constexpr bool is_agnostic = true;
+	static constexpr size_t bom_byte_size = 3;
+	static constexpr text_encoding encoding = text_encoding::utf_8;
+};
+
+template<>
+struct encoding_info<text_encoding::utf_16_little> {
+	using data_type = char16_t;
+	static constexpr bool is_little = true;
+	static constexpr bool is_big = false;
+	static constexpr bool is_native = std::endian::native == std::endian::little;
+	static constexpr bool is_foreign = !is_native;
+	static constexpr bool is_agnostic = false;
+	static constexpr size_t bom_byte_size = 2;
+	static constexpr text_encoding encoding = text_encoding::utf_16;
+};
+
+template<>
+struct encoding_info<text_encoding::utf_16_big> {
+	using data_type = char16_t;
+	static constexpr bool is_little = false;
+	static constexpr bool is_big = true;
+	static constexpr bool is_native = std::endian::native == std::endian::big;
+	static constexpr bool is_foreign = !is_native;
+	static constexpr bool is_agnostic = false;
+	static constexpr size_t bom_byte_size = 2;
+	static constexpr text_encoding encoding = text_encoding::utf_16;
+};
+
+template<>
+struct encoding_info<text_encoding::utf_32_little> {
+	using data_type = char32_t;
+	static constexpr bool is_little = true;
+	static constexpr bool is_big = false;
+	static constexpr bool is_native = std::endian::native == std::endian::little;
+	static constexpr bool is_foreign = !is_native;
+	static constexpr bool is_agnostic = false;
+	static constexpr size_t bom_byte_size = 4;
+	static constexpr text_encoding encoding = text_encoding::utf_32;
+};
+
+template<>
+struct encoding_info<text_encoding::utf_32_big> {
+	using data_type = char32_t;
+	static constexpr bool is_little = false;
+	static constexpr bool is_big = true;
+	static constexpr bool is_native = std::endian::native == std::endian::big;
+	static constexpr bool is_foreign = !is_native;
+	static constexpr bool is_agnostic = false;
+	static constexpr size_t bom_byte_size = 4;
+	static constexpr text_encoding encoding = text_encoding::utf_32;
+};
+
+template<>
+struct encoding_info<text_encoding::wchar> {
+	using data_type = wchar_t;
+	static constexpr bool is_little = std::endian::native == std::endian::little;
+	static constexpr bool is_big = std::endian::native == std::endian::big;
+	static constexpr bool is_native = true;
+	static constexpr bool is_foreign = !is_native;
+	static constexpr bool is_agnostic = true; // Not truly agnostic, but wchar_t is for system-local use only... probably.
+	static constexpr size_t bom_byte_size = 0; // Not supporting for wchar at this time
+	static constexpr text_encoding encoding = text_encoding::wchar;
+};
+
+template<>
+struct encoding_info<text_encoding::multibyte> {
+	using data_type = char;
+	static constexpr bool is_little = false;
+	static constexpr bool is_big = false;
+	static constexpr bool is_native = true;
+	static constexpr bool is_foreign = !is_native;
+	static constexpr bool is_agnostic = true;
+	static constexpr size_t bom_byte_size = 0;
+	static constexpr text_encoding encoding = text_encoding::multibyte;
+};
+
+template<typename CharT>
+struct default_encoding_info;
+
+template<>
+struct default_encoding_info<char8_t> : public encoding_info<text_encoding::utf_8> {
+};
+
+template<>
+struct default_encoding_info<char16_t> : public encoding_info<text_encoding::utf_16> {
+};
+
+template<>
+struct default_encoding_info<char32_t> : public encoding_info<text_encoding::utf_32> {
+};
+
+template<>
+struct default_encoding_info<wchar_t> : public encoding_info<text_encoding::wchar> {
+};
+
+/** Unicode byte-order markers */
+
+enum class bom_encoding {
+	utf_8 = 0,
+
+	utf_16_little = 1,
+	utf_16_big = 2,
+	utf_16_native = (std::endian::native == std::endian::little ? utf_16_little : utf_16_big),
+	utf_16 = utf_16_native,
+
+	utf_32_little = 4,
+	utf_32_big = 5,
+	utf_32_native = (std::endian::native == std::endian::little ? utf_32_little : utf_32_big),
+	utf_32 = utf_32_native,
+
+	unknown = 0xFF
+};
+
+// If this results in a non-native encoding, the swaps have to be done passing to decode_character
+constexpr bom_encoding peek_bom(std::string_view in_data) {
+	if (in_data.size() < 2) {
+		// Not enough space for any BOM
+		return bom_encoding::unknown;
+	}
+
+	// Try UTF-16 BE
+	if (in_data[0] == '\xFE' && in_data[1] == '\xFF') {
+		return bom_encoding::utf_16_big;
+	}
+
+	// Try UTF-16 LE
+	if (in_data[0] == '\xFF' && in_data[1] == '\xFE') {
+		// Check UTF-32 LE
+		if (in_data.size() >= 4
+			&& in_data[2] == 0 && in_data[3] == 0) {
+			return bom_encoding::utf_32_little;
+		}
+
+		return bom_encoding::utf_16_little;
+	}
+
+	if (in_data.size() < 3) {
+		// Not enough space for any other BOMs
+		return bom_encoding::unknown;
+	}
+
+	// Try UTF-8
+	if (in_data[0] == '\xEF' && in_data[1] == '\xBB' && in_data[2] == '\xBF') {
+		return bom_encoding::utf_8;
+	}
+
+	if (in_data.size() < 4) {
+		// Not enough space for any other BOMs
+		return bom_encoding::unknown;
+	}
+
+	// Try UTF-32 BE
+	if (in_data[0] == 0 && in_data[1] == 0 && in_data[2] == '\xFE' && in_data[3] == '\xFF') {
+		return bom_encoding::utf_32_big;
+	}
+
+	// No matches; return unknown (mostly likely encoded as UTF-8)
+	return bom_encoding::unknown;
+}
+
+} // namespace jessilib

src/include/jessilib/unicode_base.hpp

@@ -169,67 +169,6 @@ struct unicode_traits<wchar_t> : std::true_type {
 template<typename CharT>
 using encode_buffer_type = CharT[unicode_traits<CharT>::max_units_per_codepoint];
 
-// enum representing the character encodings I intend to support
-enum class encoding {
-	utf_8, // The most common and arguably superior encoding for files and networking protocols not in straight ASCII
-	utf_16,
-	utf_32,
-	wchar, // essentially only really for std::wcout / std::wcout
-	multibyte // essentially only really for std::cout / std::cin
-};
-
-template<encoding EncodingV>
-struct encoding_info;
-
-template<>
-struct encoding_info<encoding::utf_8> {
-	using data_type = char8_t;
-	static constexpr encoding text_encoding = encoding::utf_8;
-};
-
-template<>
-struct encoding_info<encoding::utf_16> {
-	using data_type = char16_t;
-	static constexpr encoding text_encoding = encoding::utf_16;
-};
-
-template<>
-struct encoding_info<encoding::utf_32> {
-	using data_type = char32_t;
-	static constexpr encoding text_encoding = encoding::utf_32;
-};
-
-template<>
-struct encoding_info<encoding::wchar> {
-	using data_type = wchar_t;
-	static constexpr encoding text_encoding = encoding::wchar;
-};
-
-template<>
-struct encoding_info<encoding::multibyte> {
-	using data_type = char;
-	static constexpr encoding text_encoding = encoding::multibyte;
-};
-
-template<typename CharT>
-struct default_encoding_info;
-
-template<>
-struct default_encoding_info<char8_t> : public encoding_info<encoding::utf_8> {
-};
-
-template<>
-struct default_encoding_info<char16_t> : public encoding_info<encoding::utf_16> {
-};
-
-template<>
-struct default_encoding_info<char32_t> : public encoding_info<encoding::utf_32> {
-};
-
-template<>
-struct default_encoding_info<wchar_t> : public encoding_info<encoding::wchar> {
-};
-
 /** single-unit helper utilities */
 char32_t fold(char32_t in_codepoint); // Folds codepoint for case-insensitive checks (not for human output)
 constexpr int as_base(char32_t in_character, unsigned int base); // The value represented by in_character in terms of base if valid, -1 otherwise

src/test/parser.cpp

@@ -30,50 +30,54 @@ using namespace std::literals;
 class test_parser : public parser {
 public:
 	/** deserialize/serialize overrides */
-	object deserialize_bytes(bytes_view_type in_data, encoding in_write_encoding) override {
+	object deserialize_bytes(bytes_view_type in_data, text_encoding in_write_encoding) override {
 		std::u8string u8_string;
 
 		switch (in_write_encoding) {
-			case encoding::utf_8:
+			case text_encoding::utf_8:
 				u8_string = string_view_cast<char8_t>(in_data);
 				break;
-			case encoding::utf_16:
+			case text_encoding::utf_16:
 				u8_string = jessilib::string_cast<char8_t>(string_view_cast<char16_t>(in_data));
 				break;
-			case encoding::utf_32:
+			case text_encoding::utf_32:
 				u8_string = jessilib::string_cast<char8_t>(string_view_cast<char32_t>(in_data));
 				break;
-			case encoding::wchar:
+			case text_encoding::wchar:
 				u8_string = jessilib::string_cast<char8_t>(string_view_cast<wchar_t>(in_data));
 				break;
-			case encoding::multibyte:
+			case text_encoding::multibyte:
 				u8_string = mbstring_to_ustring<char8_t>(string_view_cast<char>(in_data)).second;
 				break;
+			default:
+				break;
 		}
 
 		return deserialize_impl(std::u8string_view{ u8_string });
 	}
 
-	std::string serialize_bytes(const object& in_object, encoding in_write_encoding) override {
+	std::string serialize_bytes(const object& in_object, text_encoding in_write_encoding) override {
 		std::u8string u8_serialized = serialize_impl(in_object);
 
 		switch (in_write_encoding) {
-			case encoding::utf_8:
+			case text_encoding::utf_8:
 				return { u8_serialized.begin(), u8_serialized.end() };
-			case encoding::utf_16: {
+			case text_encoding::utf_16: {
 				auto casted = string_cast<char16_t>(u8_serialized);
 				return { reinterpret_cast<const char*>(casted.data()), casted.size() * sizeof(char16_t) };
 			}
-			case encoding::utf_32: {
+			case text_encoding::utf_32: {
 				auto casted = string_cast<char32_t>(u8_serialized);
 				return { reinterpret_cast<const char*>(casted.data()), casted.size() * sizeof(char32_t) };
 			}
-			case encoding::wchar: {
+			case text_encoding::wchar: {
 				auto casted = string_cast<wchar_t>(u8_serialized);
 				return { reinterpret_cast<const char*>(casted.data()), casted.size() * sizeof(wchar_t) };
 			}
-			case encoding::multibyte:
+			case text_encoding::multibyte:
 				return ustring_to_mbstring(u8_serialized).second;
+			default:
+				break;
 		}
 
 		return {};