Created
November 28, 2023 17:04
-
-
Save kddnewton/089d23d49adb5551792293fdb5bf64a0 to your computer and use it in GitHub Desktop.
Maybe encoding bug?
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# frozen_string_literal: true | |
class Context < BasicObject | |
def method_missing(name, *) = :identifier | |
def self.const_missing(name) = :constant | |
end | |
$context = Context.new | |
def evaluate(encoding, codepoint) | |
character = codepoint.chr(encoding) | |
if character.match?(/[[:lower:]]/) | |
type = $context.instance_eval("# encoding: #{encoding.name}\n#{character}") | |
puts "#{encoding.name} 0x#{codepoint.to_s(16)}" if type != :identifier | |
elsif character.match?(/[[:upper:]]/) | |
type = $context.instance_eval("# encoding: #{encoding.name}\n#{character}") | |
puts "#{encoding.name} 0x#{codepoint.to_s(16)}" if type != :constant | |
end | |
end | |
encodings_7bits = [ | |
Encoding::US_ASCII | |
] | |
encodings_7bits.each do |encoding| | |
(0..0x7F).each do |codepoint| | |
evaluate(encoding, codepoint) | |
end | |
end | |
encodings_1byte = [ | |
Encoding::ASCII_8BIT, | |
Encoding::ISO8859_1, | |
Encoding::ISO8859_2, | |
Encoding::ISO8859_3, | |
Encoding::ISO8859_4, | |
Encoding::ISO8859_5, | |
Encoding::ISO8859_6, | |
Encoding::ISO8859_7, | |
Encoding::ISO8859_8, | |
Encoding::ISO8859_9, | |
Encoding::ISO8859_10, | |
Encoding::ISO8859_11, | |
Encoding::ISO8859_13, | |
Encoding::ISO8859_14, | |
Encoding::ISO8859_15, | |
Encoding::ISO8859_16, | |
Encoding::CP878, | |
Encoding::KOI8_U, | |
Encoding::WINDOWS_1250, | |
Encoding::WINDOWS_1251, | |
Encoding::WINDOWS_1252, | |
Encoding::WINDOWS_1253, | |
Encoding::WINDOWS_1254, | |
Encoding::WINDOWS_1257, | |
Encoding::CP437, | |
Encoding::CP720, | |
Encoding::CP737, | |
Encoding::CP775, | |
Encoding::IBM850, | |
Encoding::IBM852, | |
Encoding::CP852, | |
Encoding::IBM855, | |
Encoding::CP855, | |
Encoding::CP857, | |
Encoding::CP860, | |
Encoding::CP861, | |
Encoding::CP862, | |
Encoding::CP863, | |
Encoding::CP864, | |
Encoding::CP865, | |
Encoding::CP866, | |
Encoding::CP869, | |
Encoding::CP1258, | |
Encoding::GB1988, | |
Encoding::MACCENTEURO, | |
Encoding::MACCROATIAN, | |
Encoding::MACCYRILLIC, | |
Encoding::MACGREEK, | |
Encoding::MACICELAND, | |
Encoding::MACROMAN, | |
Encoding::MACROMANIA, | |
Encoding::MACTHAI, | |
Encoding::MACTURKISH, | |
Encoding::MACUKRAINE, | |
Encoding::Windows_1256, | |
Encoding::WINDOWS_1255, | |
Encoding::TIS_620, | |
Encoding::Windows_874 | |
] | |
encodings_1byte.each do |encoding| | |
(0..0xFF).each do |codepoint| | |
evaluate(encoding, codepoint) | |
end | |
end | |
encodings_2byte = [ | |
Encoding::BIG5, | |
Encoding::BIG5_HKSCS_2008, | |
Encoding::BIG5_UAO, | |
Encoding::EUCKR, | |
Encoding::EUCTW, | |
Encoding::SHIFT_JIS, | |
Encoding::SJIS_DOCOMO, | |
Encoding::SJIS_KDDI, | |
Encoding::SJIS_SOFTBANK, | |
Encoding::WINDOWS_31J, | |
Encoding::CP949, | |
Encoding::GB18030, | |
Encoding::CP936, | |
Encoding::CP950, | |
Encoding::CP951, | |
Encoding::EUCCN, | |
Encoding::GB12345, | |
Encoding::MACJAPANESE | |
] | |
encodings_2byte.each do |encoding| | |
(0..0xFFFF).each do |codepoint| | |
evaluate(encoding, codepoint) | |
rescue RangeError | |
end | |
end | |
encodings_3byte = [ | |
Encoding::EUCJP, | |
Encoding::EUC_JP_MS, | |
Encoding::EUC_JISX0213, | |
Encoding::CP51932 | |
] | |
encodings_3byte.each do |encoding| | |
(0..0xFFFFFF).each do |codepoint| | |
evaluate(encoding, codepoint) | |
rescue RangeError | |
end | |
end | |
encodings_unicode = [ | |
Encoding::UTF_8, | |
Encoding::UTF8_MAC, | |
Encoding::UTF8_DoCoMo, | |
Encoding::UTF8_KDDI, | |
Encoding::UTF8_SoftBank, | |
Encoding::CP65001, | |
Encoding::CESU_8 | |
] | |
encodings_unicode.each do |encoding| | |
(0...0xD800).each do |codepoint| | |
evaluate(encoding, codepoint) | |
end | |
(0xE000..0x10FFFF).each do |codepoint| | |
evaluate(encoding, codepoint) | |
end | |
end | |
encodings_4byte = [ | |
Encoding::EMACS_MULE, | |
Encoding::STATELESS_ISO_2022_JP, | |
Encoding::STATELESS_ISO_2022_JP_KDDI, | |
] | |
encodings_4byte.each do |encoding| | |
(0..0xFFFFFFFF).each do |codepoint| | |
evaluate(encoding, codepoint) | |
rescue RangeError | |
end | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment