Byte to Byte Val (Broken)

-- Code starts at line 200 --

local byte = string.byte
local len = string.len
local sub = string.sub

local shift_6 = 2^6
local shift_12 = 2^12
local shift_18 = 2^18

local function utf8charbytes(s, i)
  -- argument defaults
  i = i or 1

  -- argument checking
  if type(s) ~= "string" then
    error("bad argument #1 to 'utf8charbytes' (string expected, got ".. type(s).. ")")
  end
  if type(i) ~= "number" then
    error("bad argument #2 to 'utf8charbytes' (number expected, got ".. type(i).. ")")
  end

  local c = byte(s, i)
  --print(c)
  -- determine bytes needed for character, based on RFC 3629
  -- validate byte 1
  if c >= 0 and c <= 127 then
    -- UTF8-1
    return 1

  elseif c >= 194 and c <= 223 then
    -- UTF8-2
    local c2 = byte(s, i + 1)

    if not c2 then
      error("UTF-8 string terminated early")
    end

    -- validate byte 2
    if c2 < 128 or c2 > 191 then
      error("Invalid UTF-8 character")
    end

    return 2

  elseif c >= 224 and c <= 239 then
    -- UTF8-3
    local c2 = byte(s, i + 1)
    local c3 = byte(s, i + 2)

    if not c2 or not c3 then
      error("UTF-8 string terminated early")
    end

    -- validate byte 2
    if c == 224 and (c2 < 160 or c2 > 191) then
      error("Invalid UTF-8 character")
    elseif c == 237 and (c2 < 128 or c2 > 159) then
      error("Invalid UTF-8 character")
    elseif c2 < 128 or c2 > 191 then
      error("Invalid UTF-8 character")
    end

    -- validate byte 3
    if c3 < 128 or c3 > 191 then
      error("Invalid UTF-8 character")
    end

    return 3

  elseif c >= 240 and c <= 244 then
    -- UTF8-4
    local c2 = byte(s, i + 1)
    local c3 = byte(s, i + 2)
    local c4 = byte(s, i + 3)

    if not c2 or not c3 or not c4 then
      error("UTF-8 string terminated early")
    end

    -- validate byte 2
    if c == 240 and (c2 < 144 or c2 > 191) then
      error("Invalid UTF-8 character2.1")
    elseif c == 244 and (c2 < 128 or c2 > 143) then
      error("Invalid UTF-8 character2.2")
    elseif c2 < 128 or c2 > 191 then
      error("Invalid UTF-8 character2.3")
    end

    -- validate byte 3
    if c3 < 128 or c3 > 191 then
      error("Invalid UTF-8 character3")
    end

    -- validate byte 4
    if c4 < 128 or c4 > 191 then
      error("Invalid UTF-8 character4")
    end

    return 4

  else
    error("Invalid UTF-8 character5")
  end
end


local function utf8sub(s, i, j)
  -- argument defaults
  j = j or -1

  local pos = 1
  local bytes = len(s)
  local length = 0

  -- only set l if i or j is negative
  local l = (i >= 0 and j >= 0) or utf8len(s)
  local startChar = (i >= 0) and i or l + i + 1
  local endChar   = (j >= 0) and j or l + j + 1

  -- can't have start before end!
  if startChar > endChar then
    return ""
  end

  -- byte offsets to pass to string.sub
  local startByte,endByte = 1,bytes

  while pos <= bytes do
    length = length + 1

    if length == startChar then
      startByte = pos
    end

    pos = pos + utf8charbytes(s, pos)

    if length == endChar then
      endByte = pos - 1
      break
    end
  end

  if startChar > length then startByte = bytes+1   end
  if endChar   < 1      then endByte   = 0         end

  return sub(s, startByte, endByte)
end

local function utf8unicode(str, i, j, byte_pos)
  i = i or 1
  j = j or i

  if i > j then return end

  local ch,bytes

  if byte_pos then
    bytes = utf8charbytes(str,byte_pos)
    ch  = sub(str,byte_pos,byte_pos-1+bytes)
  else
    ch,byte_pos = utf8sub(str,i,i), 0
    bytes       = #ch
  end

  local unicode

  if bytes == 1 then unicode = byte(ch) end
  if bytes == 2 then
    local byte0,byte1 = byte(ch,1,2)
    local code0,code1 = byte0-0xC0,byte1-0x80
    unicode = code0*shift_6 + code1
  end
  if bytes == 3 then
    local byte0,byte1,byte2 = byte(ch,1,3)
    local code0,code1,code2 = byte0-0xE0,byte1-0x80,byte2-0x80
    unicode = code0*shift_12 + code1*shift_6 + code2
  end
  if bytes == 4 then
    local byte0,byte1,byte2,byte3 = byte(ch,1,4)
    local code0,code1,code2,code3 = byte0-0xF0,byte1-0x80,byte2-0x80,byte3-0x80
    unicode = code0*shift_18 + code1*shift_12 + code2*shift_6 + code3
  end

  return unicode,bytes
end

local function hexy(IN)
    local B,K,O,I,D=16,"0123456789ABCDEF","",0
    while IN>0 do
      I=I+1
      IN,D=math.floor(IN/B),math.fmod(IN,B)+1
      O=string.sub(K,D,D)..O
    end
    return O
  end

-- Where the code starts --

for i=1,9 do
A,B = utf8unicode("!— ! ÿª #",i)
--   [byte |    hex val    | char size]
print(A.." | "..hexy(A).." | "..B)
end