note description: "Eiffel implementations for Unicode encoding conversion." legal: "See notice at end of class." status: "See notice at end of class." date: "$Date: 2020-05-20 11:14:50 +0000 (Wed, 20 May 2020) $" revision: "$Revision: 104299 $" class UNICODE_CONVERSION create {ENCODING} default_create feature {NONE} -- Initialization default_create -- Process instances of classes with no creation clause. -- (Default: do nothing.) -- (from ANY) do end feature {ENCODING} -- Access last_converted_stream: STRING_8 -- Stream prepresentation of last converted string. -- (from ENCODING_I) require -- from ENCODING_I last_conversion_successful: last_conversion_successful do check from_precondition: attached last_converted_string as l_last then if last_was_wide_string then Result := string_16_to_stream (l_last.as_string_32) else Result := string_general_to_stream (l_last) end end ensure -- from ENCODING_I last_converted_stream_not_void: Result /= Void end last_converted_string: detachable READABLE_STRING_GENERAL -- Last converted string. -- (from ENCODING_I) last_converted_string_8: STRING_8 -- (from ENCODING_I) require -- from ENCODING_I last_conversion_successful: last_conversion_successful not_wide: not last_was_wide_string do check attached last_converted_string as l_string then Result := string_general_to_stream (l_string) end ensure -- from ENCODING_I last_converted_stream_not_void: Result /= Void end feature -- Access generating_type: TYPE [detachable UNICODE_CONVERSION] -- Type of current object -- (type of which it is a direct instance) -- (from ANY) external "built_in" ensure -- from ANY generating_type_not_void: Result /= Void end generator: STRING_8 -- Name of current object's generating class -- (base class of the type of which it is a direct instance) -- (from ANY) external "built_in" ensure -- from ANY generator_not_void: Result /= Void generator_not_empty: not Result.is_empty end feature -- Comparison frozen deep_equal (a: detachable ANY; b: like arg #1): BOOLEAN -- Are a and b either both void -- or attached to isomorphic object structures? -- (from ANY) do if a = Void then Result := b = Void else Result := b /= Void and then a.is_deep_equal (b) end ensure -- from ANY instance_free: class shallow_implies_deep: standard_equal (a, b) implies Result both_or_none_void: (a = Void) implies (Result = (b = Void)) same_type: (Result and (a /= Void)) implies (b /= Void and then a.same_type (b)) symmetric: Result implies deep_equal (b, a) end frozen equal (a: detachable ANY; b: like arg #1): BOOLEAN -- Are a and b either both void or attached -- to objects considered equal? -- (from ANY) do if a = Void then Result := b = Void else Result := b /= Void and then a.is_equal (b) end ensure -- from ANY instance_free: class definition: Result = (a = Void and b = Void) or else ((a /= Void and b /= Void) and then a.is_equal (b)) end frozen is_deep_equal alias "≡≡≡" (other: UNICODE_CONVERSION): BOOLEAN -- Are Current and other attached to isomorphic object structures? -- (from ANY) require -- from ANY other_not_void: other /= Void external "built_in" ensure -- from ANY shallow_implies_deep: standard_is_equal (other) implies Result same_type: Result implies same_type (other) symmetric: Result implies other.is_deep_equal (Current) end is_equal (other: UNICODE_CONVERSION): BOOLEAN -- Is other attached to an object considered -- equal to current object? -- (from ANY) require -- from ANY other_not_void: other /= Void external "built_in" ensure -- from ANY symmetric: Result implies other ~ Current consistent: standard_is_equal (other) implies Result end frozen standard_equal (a: detachable ANY; b: like arg #1): BOOLEAN -- Are a and b either both void or attached to -- field-by-field identical objects of the same type? -- Always uses default object comparison criterion. -- (from ANY) do if a = Void then Result := b = Void else Result := b /= Void and then a.standard_is_equal (b) end ensure -- from ANY instance_free: class definition: Result = (a = Void and b = Void) or else ((a /= Void and b /= Void) and then a.standard_is_equal (b)) end frozen standard_is_equal alias "≜" (other: UNICODE_CONVERSION): BOOLEAN -- Is other attached to an object of the same type -- as current object, and field-by-field identical to it? -- (from ANY) require -- from ANY other_not_void: other /= Void external "built_in" ensure -- from ANY same_type: Result implies same_type (other) symmetric: Result implies other.standard_is_equal (Current) end feature {ENCODING} -- Status report is_code_page_convertable (a_from_code_page, a_to_code_page: READABLE_STRING_8): BOOLEAN obsolete "Use is_code_page_convertible [2020-04-22]" -- Is a_from_code_page convertible to a_to_code_page. -- (from ENCODING_I) do Result := is_code_page_convertible (a_from_code_page, a_to_code_page) end last_conversion_successful: BOOLEAN -- Was last conversion successful? -- (from ENCODING_I) last_was_wide_string: BOOLEAN -- Last conversion result was wide string? -- (from ENCODING_I) feature -- Status report conforms_to (other: ANY): BOOLEAN -- Does type of current object conform to type -- of other (as per Eiffel: The Language, chapter 13)? -- (from ANY) require -- from ANY other_not_void: other /= Void external "built_in" end same_type (other: ANY): BOOLEAN -- Is type of current object identical to type of other? -- (from ANY) require -- from ANY other_not_void: other /= Void external "built_in" ensure -- from ANY definition: Result = (conforms_to (other) and other.conforms_to (Current)) end feature -- Conversion convert_to (a_from_code_page: READABLE_STRING_8; a_from_string: READABLE_STRING_GENERAL; a_to_code_page: READABLE_STRING_8) -- Convert between Unicode encodings. require -- from ENCODING_I a_from_code_page_valid: is_code_page_valid (a_from_code_page) a_to_code_page_valid: is_code_page_valid (a_to_code_page) code_page_convertible: is_code_page_convertible (a_from_code_page, a_to_code_page) a_from_string_not_void: a_from_string /= Void do reset if a_from_code_page.is_case_insensitive_equal (a_to_code_page) then last_conversion_successful := True if a_from_string.is_string_8 then last_converted_string := a_from_string.to_string_8 else last_converted_string := a_from_string.as_string_32 end else if a_from_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf8) then last_converted_string := utf8_to_utf32 (if a_from_string.is_valid_as_string_8 then a_from_string.to_string_8 else {UTF_CONVERTER}.utf_32_string_to_utf_8_string_8 (a_from_string) end) last_conversion_successful := True elseif a_from_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf32) then if a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf8) then last_converted_string := utf32_to_utf8 (a_from_string.as_string_32) last_conversion_successful := True elseif a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf16) then last_converted_string := utf32_to_utf16 (a_from_string.as_string_32) last_was_wide_string := True last_conversion_successful := True end elseif a_from_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf16) then if a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf32) then last_converted_string := utf16_to_utf32 (a_from_string.as_string_32) last_conversion_successful := True end end end ensure -- from ENCODING_I success_implies_not_void: last_conversion_successful implies last_converted_stream /= Void success_implies_not_void: last_conversion_successful implies last_converted_string /= Void end multi_byte_to_pointer (a_string: READABLE_STRING_8): MANAGED_POINTER -- Managed pointer of a_string. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_string_not_void: a_string /= Void do Result := (create {C_STRING}.make (a_string)).managed_data ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void end pointer_to_multi_byte (a_multi_string: POINTER; a_count: INTEGER_32): STRING_8 -- STRING_8 read from a_multi_string. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_multi_string_not_default: a_multi_string /= default_pointer a_count_non_negative: a_count >= 0 local i: INTEGER_32 l_managed_pointer: MANAGED_POINTER do create l_managed_pointer.share_from_pointer (a_multi_string, a_count) create Result.make (a_count) from i := 0 until i >= a_count loop Result.append_code (l_managed_pointer.read_natural_8 (i).to_natural_32) i := i + 1 end ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void end pointer_to_string_32 (a_w_string: POINTER; a_count: INTEGER_32): STRING_32 -- STRING_32 read from a_w_string of a_count bytes. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_w_string_not_default: a_w_string /= default_pointer a_count_non_negative: a_count >= 0 local i: INTEGER_32 l_managed_pointer: MANAGED_POINTER l_size: INTEGER_32 do create l_managed_pointer.share_from_pointer (a_w_string, a_count) l_size := a_count // 4 create Result.make (l_size) from i := 0 until i >= l_size loop if i * 4 <= a_count then Result.append_code (l_managed_pointer.read_natural_32 (i * 4)) end i := i + 1 end ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void end pointer_to_wide_string (a_w_string: POINTER; a_count: INTEGER_32): STRING_32 -- STRING_32 read from a_w_string of a_count bytes. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_w_string_not_default: a_w_string /= default_pointer a_count_non_negative: a_count >= 0 local i: INTEGER_32 l_managed_pointer: MANAGED_POINTER l_size: INTEGER_32 do create l_managed_pointer.share_from_pointer (a_w_string, a_count) l_size := (a_count + 1) // 2 create Result.make (l_size) from i := 0 until i >= l_size loop if i * 2 <= a_count then Result.append_code (l_managed_pointer.read_natural_16 (i * 2).to_natural_32) end i := i + 1 end ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void end string_16_to_stream (a_string: STRING_32): STRING_8 -- We use a_string as 2 bytes encoding string, the first two bytes are not used. -- in the endianness of the current platform. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_string_not_void: a_string /= Void local l_managed_pointer: MANAGED_POINTER i, l_count: INTEGER_32 do l_managed_pointer := wide_string_to_pointer (a_string) create Result.make (l_managed_pointer.count) from i := 0 l_count := l_managed_pointer.count - 2 until i = l_count loop Result.append_character (l_managed_pointer.read_natural_8 (i).to_character_8) i := i + 1 end ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void valid_count: Result.count = a_string.count * 2 end string_32_to_multi_byte (a_string: STRING_32): STRING_8 -- Byte stream of a_string in endianness of the current platform. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_string_not_void: a_string /= Void local i: INTEGER_32 l_code: NATURAL_32 l_count: INTEGER_32 l_is_little_endian: BOOLEAN do l_count := a_string.count if l_count > 0 then create Result.make (l_count * 4) from i := 1 l_is_little_endian := Is_little_endian until i > l_count loop l_code := a_string.code (i) if l_is_little_endian then Result.append_code (l_code & 255); Result.append_code (l_code & 65280 |>> 8); Result.append_code (l_code & 16711680 |>> 16); Result.append_code (l_code & 4278190080 |>> 24) else Result.append_code (l_code & 4278190080 |>> 24); Result.append_code (l_code & 16711680 |>> 16); Result.append_code (l_code & 65280 |>> 8); Result.append_code (l_code & 255) end i := i + 1 end else create Result.make_empty end ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void end string_8_to_wide_string (a_w_string: STRING_8): STRING_32 -- Interpret a_w_string as a sequence of 2-byte characters into a STRING_32 -- in endianness of the current platform. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_w_string_not_void: a_w_string /= Void local i: INTEGER_32 l_size, l_count: INTEGER_32 l_is_little_endian: BOOLEAN l_code: NATURAL_32 do l_count := a_w_string.count l_size := (l_count + 1) // 2 l_is_little_endian := Is_little_endian create Result.make (l_size) from i := 1 until i > l_count loop if i + 1 <= l_count then if l_is_little_endian then l_code := a_w_string.code (i) | (a_w_string.code (i + 1) |<< 8) else l_code := (a_w_string.code (i) |<< 8) | a_w_string.code (i + 1) end; Result.append_code (l_code) end i := i + 1 end ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void end string_general_to_stream (a_string: READABLE_STRING_GENERAL): STRING_8 -- Streamize a_string. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_string_not_void: a_string /= Void do if a_string.is_string_8 then Result := a_string.to_string_8 else Result := string_32_to_multi_byte (a_string.as_string_32) end ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void end wide_string_to_pointer (a_string: READABLE_STRING_32): MANAGED_POINTER -- Managed pointer of a_string which is taken as -- 16bits string. High 16bits of characters of a_string are discarded. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_string_not_void: a_string /= Void local i, nb: INTEGER_32 do nb := a_string.count create Result.make ((nb + 1) * 2) from i := 0 until i = nb loop Result.put_natural_16 (a_string.code (i + 1).to_natural_16, i * 2) i := i + 1 end; Result.put_natural_16 (0, i * 2) ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void end feature -- Duplication frozen clone (other: detachable ANY): like other obsolete "Use `twin' instead. [2017-05-31]" -- Void if other is void; otherwise new object -- equal to other -- -- For non-void other, clone calls copy; -- to change copying/cloning semantics, redefine copy. -- (from ANY) do if other /= Void then Result := other.twin end ensure -- from ANY instance_free: class equal: Result ~ other end copy (other: UNICODE_CONVERSION) -- Update current object using fields of object attached -- to other, so as to yield equal objects. -- (from ANY) require -- from ANY other_not_void: other /= Void type_identity: same_type (other) external "built_in" ensure -- from ANY is_equal: Current ~ other end frozen deep_clone (other: detachable ANY): like other obsolete "Use `deep_twin' instead. [2017-05-31]" -- Void if other is void: otherwise, new object structure -- recursively duplicated from the one attached to other -- (from ANY) do if other /= Void then Result := other.deep_twin end ensure -- from ANY instance_free: class deep_equal: deep_equal (other, Result) end frozen deep_copy (other: UNICODE_CONVERSION) -- Effect equivalent to that of: -- copy (other . deep_twin) -- (from ANY) require -- from ANY other_not_void: other /= Void do copy (other.deep_twin) ensure -- from ANY deep_equal: deep_equal (Current, other) end frozen deep_twin: UNICODE_CONVERSION -- New object structure recursively duplicated from Current. -- (from ANY) external "built_in" ensure -- from ANY deep_twin_not_void: Result /= Void deep_equal: deep_equal (Current, Result) end frozen standard_clone (other: detachable ANY): like other obsolete "Use `standard_twin' instead. [2017-05-31]" -- Void if other is void; otherwise new object -- field-by-field identical to other. -- Always uses default copying semantics. -- (from ANY) do if other /= Void then Result := other.standard_twin end ensure -- from ANY instance_free: class equal: standard_equal (Result, other) end frozen standard_copy (other: UNICODE_CONVERSION) -- Copy every field of other onto corresponding field -- of current object. -- (from ANY) require -- from ANY other_not_void: other /= Void type_identity: same_type (other) external "built_in" ensure -- from ANY is_standard_equal: standard_is_equal (other) end frozen standard_twin: UNICODE_CONVERSION -- New object field-by-field identical to other. -- Always uses default copying semantics. -- (from ANY) external "built_in" ensure -- from ANY standard_twin_not_void: Result /= Void equal: standard_equal (Result, Current) end frozen twin: UNICODE_CONVERSION -- New object equal to Current -- twin calls copy; to change copying/twinning semantics, redefine copy. -- (from ANY) external "built_in" ensure -- from ANY twin_not_void: Result /= Void is_equal: Result ~ Current end feature -- Basic operations frozen as_attached: attached UNICODE_CONVERSION obsolete "Remove calls to this feature. [2017-05-31]" -- Attached version of Current. -- (Can be used during transitional period to convert -- non-void-safe classes to void-safe ones.) -- (from ANY) do Result := Current end frozen default: detachable UNICODE_CONVERSION -- Default value of object's type -- (from ANY) do end frozen default_pointer: POINTER -- Default value of type POINTER -- (Avoid the need to write p.default for -- some p of type POINTER.) -- (from ANY) do ensure -- from ANY instance_free: class end default_rescue -- Process exception for routines with no Rescue clause. -- (Default: do nothing.) -- (from ANY) do end frozen do_nothing -- Execute a null action. -- (from ANY) do ensure -- from ANY instance_free: class end feature {NONE} -- Implementation Unicode_encodings: STRING_TABLE [READABLE_STRING_8] -- Supported Unicode encodings. once create Result.make (8); Result.put ({CODE_PAGE_CONSTANTS}.utf7, {CODE_PAGE_CONSTANTS}.utf7); Result.put ({CODE_PAGE_CONSTANTS}.utf8, {CODE_PAGE_CONSTANTS}.utf8); Result.put ({CODE_PAGE_CONSTANTS}.utf16, {CODE_PAGE_CONSTANTS}.utf16); Result.put ({CODE_PAGE_CONSTANTS}.utf16_le, {CODE_PAGE_CONSTANTS}.utf16_le); Result.put ({CODE_PAGE_CONSTANTS}.utf32, {CODE_PAGE_CONSTANTS}.utf32); Result.put ({CODE_PAGE_CONSTANTS}.utf32_le, {CODE_PAGE_CONSTANTS}.utf32_le); Result.put ({CODE_PAGE_CONSTANTS}.utf16_be, {CODE_PAGE_CONSTANTS}.utf16_be); Result.put ({CODE_PAGE_CONSTANTS}.utf32_be, {CODE_PAGE_CONSTANTS}.utf32_be) end feature -- Endian Is_little_endian: BOOLEAN -- Is this system little endian? -- (from ENCODING_HELPER) once Result := {PLATFORM}.is_little_endian ensure -- from ENCODING_HELPER instance_free: class end string_16_switch_endian (a_str: STRING_32): STRING_32 -- Switch endian of a_str for low bits. -- High bits are cleaned. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_str_not_void: a_str /= Void local l_code: NATURAL_32 i, l_count: INTEGER_32 do l_count := a_str.count create Result.make (l_count) from i := 1 until i > l_count loop l_code := a_str.code (i); Result.append_code (l_code & 255 |<< 8 & 65280 + l_code & 65280 |>> 8 & 255) i := i + 1 end ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void end string_32_switch_endian (a_str: STRING_32): STRING_32 -- Switch endian of a_str for both high and low bits. -- (from ENCODING_HELPER) require -- from ENCODING_HELPER a_str_not_void: a_str /= Void local l_code: NATURAL_32 i, l_count: INTEGER_32 do l_count := a_str.count create Result.make (l_count) from i := 1 until i > l_count loop l_code := a_str.code (i); Result.append_code (l_code & 255 |<< 24 & 4278190080 + l_code & 65280 |<< 8 + l_code & 16711680 |>> 8 + l_code & 4278190080 |>> 24 & 255) i := i + 1 end ensure -- from ENCODING_HELPER instance_free: class result_not_void: Result /= Void end feature -- Explicit Conversion append_code_point_to_utf8 (a_code: NATURAL_32; a_string: STRING_8) -- Append a Unicode code point a_code to a UTF-8 stream. require a_string_not_void: a_string /= Void a_code_is_valid: a_code >= 0 and then a_code <= 1114111 do {UTF_CONVERTER}.utf_32_code_into_utf_8_string_8 (a_code, a_string) ensure a_string_appended: (a_code <= 127 implies a_string.count = old a_string.count + 1) and ((a_code > 127 and a_code <= 2047) implies a_string.count = old a_string.count + 2) and ((a_code > 2047 and a_code <= 65535) implies a_string.count = old a_string.count + 3) and ((a_code > 65535 and a_code <= 1114111) implies a_string.count = old a_string.count + 4) end read_character_from_utf8 (a_position: INTEGER_32; a_read_bytes: detachable INTEGER_32_REF; a_string: READABLE_STRING_8): CHARACTER_32 -- Read a Unicode character from UTF-8 string. -- a_string is in UTF-8. -- a_position is the starting byte point of a character. -- a_read_bytes is the number of bytes read. require a_string_not_void: a_string /= Void a_position_in_range: a_position > 0 and a_position <= a_string.count a_position_valid: a_string.code (a_position).to_natural_8 <= 127 or (a_string.code (a_position).to_natural_8 & 224) = 192 or (a_string.code (a_position).to_natural_8 & 240) = 224 or (a_string.code (a_position).to_natural_8 & 248) = 240 or (a_string.code (a_position).to_natural_8 & 252) = 248 or (a_string.code (a_position).to_natural_8 & 254) = 252 local l_pos: INTEGER_32 l_nat8: NATURAL_8 l_code: NATURAL_32 do l_pos := a_position l_nat8 := a_string.code (l_pos).to_natural_8 if l_nat8 <= 127 then Result := l_nat8.to_character_32 elseif (l_nat8 & 224) = 192 then l_code := (l_nat8 & 31).to_natural_32 |<< 6 l_pos := l_pos + 1 l_nat8 := a_string.code (l_pos).to_natural_8 l_code := l_code | (l_nat8 & 63).to_natural_32 Result := l_code.to_character_32 elseif (l_nat8 & 240) = 224 then l_code := (l_nat8 & 15).to_natural_32 |<< 12 l_nat8 := a_string.code (l_pos + 1).to_natural_8 l_code := l_code | ((l_nat8 & 63).to_natural_32 |<< 6) l_nat8 := a_string.code (l_pos + 2).to_natural_8 l_code := l_code | (l_nat8 & 63).to_natural_32 Result := l_code.to_character_32 l_pos := l_pos + 2 elseif (l_nat8 & 248) = 240 then l_code := (l_nat8 & 7).to_natural_32 |<< 18 l_nat8 := a_string.code (l_pos + 1).to_natural_8 l_code := l_code | ((l_nat8 & 63).to_natural_32 |<< 12) l_nat8 := a_string.code (l_pos + 2).to_natural_8 l_code := l_code | ((l_nat8 & 63).to_natural_32 |<< 6) l_nat8 := a_string.code (l_pos + 3).to_natural_8 l_code := l_code | (l_nat8 & 63).to_natural_32 Result := l_code.to_character_32 l_pos := l_pos + 3 elseif (l_nat8 & 252) = 248 then Result := ' '.to_character_32 l_pos := l_pos + 4 else Result := ' '.to_character_32 l_pos := l_pos + 5 end if a_read_bytes /= Void then a_read_bytes.set_item (l_pos - a_position + 1) end end utf16_to_utf32 (a_str: READABLE_STRING_32): STRING_32 -- Convert utf16 to utf32. require a_str_not_void: a_str /= Void local i, l_count: INTEGER_32 l_code: NATURAL_32 l_temp: NATURAL_32 l_lower: NATURAL_32 do l_count := a_str.count create Result.make (l_count) from i := 1 until i > l_count loop l_code := a_str.code (i) i := i + 1 l_temp := l_code & 65535 if i <= l_count then l_lower := a_str.code (i) & 65535 end if l_temp >= 55296 and then l_temp <= 56319 and then i <= l_count and then l_lower >= 56320 and then l_lower <= 57343 then Result.append_code ((l_temp & 1023) |<< 10 + l_lower & 1023 + 65536) i := i + 1 else Result.append_code (l_temp) end end ensure instance_free: class result_not_void: Result /= Void end utf32_to_utf16 (a_str: READABLE_STRING_32): STRING_32 -- Convert utf32 to utf16 without data lose. require a_str_not_void: a_str /= Void local l_code: NATURAL_32 i, l_count: INTEGER_32 do create Result.make (a_str.count * 2) from i := 1 l_count := a_str.count until i > l_count loop l_code := a_str.code (i) & 1048575 if l_code > 65535 then l_code := l_code - 65536; Result.append_code (l_code |>> 10 | 55296); Result.append_code (l_code & 1023 | 56320) else Result.append_code (l_code) end i := i + 1 end ensure instance_free: class result_not_void: Result /= Void end utf32_to_utf8 (a_string: READABLE_STRING_32): STRING_8 -- Convert UTF32 to UTF8. require a_string_not_void: a_string /= Void do Result := {UTF_CONVERTER}.utf_32_string_to_utf_8_string_8 (a_string) ensure result_not_void: Result /= Void end utf8_to_utf32 (a_string: READABLE_STRING_8): STRING_32 -- UTF8 to UTF32 conversion, Eiffel implementation. require a_string_not_void: a_string /= Void do Result := {UTF_CONVERTER}.utf_8_string_8_to_string_32 (a_string) ensure result_not_void: Result /= Void end feature -- Output Io: STD_FILES -- Handle to standard file setup -- (from ANY) once create Result; Result.set_output_default ensure -- from ANY instance_free: class io_not_void: Result /= Void end out: STRING_8 -- New string containing terse printable representation -- of current object -- (from ANY) do Result := tagged_out ensure -- from ANY out_not_void: Result /= Void end print (o: detachable ANY) -- Write terse external representation of o -- on standard output. -- (from ANY) local s: READABLE_STRING_8 do if attached o then s := o.out if attached {READABLE_STRING_32} s as s32 then Io.put_string_32 (s32) elseif attached {READABLE_STRING_8} s as s8 then Io.put_string (s8) else Io.put_string_32 (s.as_string_32) end end ensure -- from ANY instance_free: class end frozen tagged_out: STRING_8 -- New string containing terse printable representation -- of current object -- (from ANY) external "built_in" ensure -- from ANY tagged_out_not_void: Result /= Void end feature -- Platform Operating_environment: OPERATING_ENVIRONMENT -- Objects available from the operating system -- (from ANY) once create Result ensure -- from ANY instance_free: class operating_environment_not_void: Result /= Void end feature -- Query is_code_page_convertible (a_from_code_page, a_to_code_page: READABLE_STRING_8): BOOLEAN -- Is a_from_code_page convertible to a_to_code_page. do Result := a_from_code_page = {CODE_PAGE_CONSTANTS}.utf8 and then a_to_code_page = {CODE_PAGE_CONSTANTS}.utf32 if not Result then if a_from_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf8) then Result := a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf8) or else a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf32) elseif a_from_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf32) then Result := a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf32) or else a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf8) or else a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf16) elseif a_from_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf16) then Result := a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf16) or else a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf32) elseif a_from_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf7) then Result := a_to_code_page.is_case_insensitive_equal ({CODE_PAGE_CONSTANTS}.utf7) end end end is_code_page_valid (a_code_page: READABLE_STRING_8): BOOLEAN -- Is a_code_page valid? do if a_code_page /= Void and then not a_code_page.is_empty then Result := Unicode_encodings.has (a_code_page) end end is_valid_as_string_16 (a_string: READABLE_STRING_GENERAL): BOOLEAN -- Check high 16 bit of any char in a_string is zero. local i, nb: INTEGER_32 l_area: SPECIAL [CHARACTER_32] do if a_string /= Void then if a_string.is_string_32 then from nb := a_string.count Result := True l_area := a_string.as_string_32.area until i = nb or not Result loop Result := l_area [i].code <= 65535 i := i + 1 end else Result := True end end end is_valid_utf8 (a_string: READABLE_STRING_8): BOOLEAN -- Is a_string valid UTF-8 string? require a_string_not_void: a_string /= Void do Result := {UTF_CONVERTER}.is_valid_utf_8_string_8 (a_string) end Last_conversion_lost_data: BOOLEAN = False -- Has last conversion between two encodings lost data? feature {ENCODING} -- Reset reset -- Reset -- (from ENCODING_I) do last_converted_string := Void last_conversion_successful := False last_was_wide_string := False ensure -- from ENCODING_I last_converted_string_reset: last_converted_string = Void last_conversion_successful_reset: not last_conversion_successful end feature {NONE} -- Retrieval frozen internal_correct_mismatch -- Called from runtime to perform a proper dynamic dispatch on correct_mismatch -- from MISMATCH_CORRECTOR. -- (from ANY) local l_msg: STRING_32 l_exc: EXCEPTIONS do if attached {MISMATCH_CORRECTOR} Current as l_corrector then l_corrector.correct_mismatch else create l_msg.make_from_string ("Mismatch: ".as_string_32) create l_exc; l_msg.append (generating_type.name_32); l_exc.raise_retrieval_exception (l_msg) end end invariant -- from ANY reflexive_equality: standard_is_equal (Current) reflexive_conformance: conforms_to (Current) note library: "Encoding: Library of reusable components for Eiffel." copyright: "Copyright (c) 1984-2020, Eiffel Software and others" license: "Eiffel Forum License v2 (see http://www.eiffel.com/licensing/forum.txt)" source: "[ Eiffel Software 5949 Hollister Ave., Goleta, CA 93117 USA Telephone 805-685-1006, Fax 805-685-6869 Website http://www.eiffel.com Customer support http://support.eiffel.com ]" end -- class UNICODE_CONVERSION
Generated by ISE EiffelStudio