Last active
December 16, 2015 17:39
-
-
Save cristobal/5471683 to your computer and use it in GitHub Desktop.
UCS2 encode a string for sending via AT modem using `%04x` hex format.
This file contains hidden or 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
| local function ucs2_encode(str) | |
| local cc = {} | |
| local bv = {1, 2, 4, 8, 16, 32, 64, 128} | |
| local fmt = "%04x" | |
| -- bit_set taken from http://ricilake.blogspot.no/2007/10/iterating-bits-in-lua.html | |
| local function bit_set(x, p) | |
| return x % (p + p) >= p | |
| end | |
| -- get bit_val | |
| -- x the number | |
| -- n the max bit bit are enumerated 1..8 instead of 0..7 lua world | |
| -- i the start bit pos | |
| -- o offset shift up/down | |
| local function bit_val(x, n, i, o) | |
| local v = 0 | |
| if not i then | |
| i = 1 | |
| end | |
| if not o then | |
| o = 0 | |
| end | |
| while i <= n do | |
| if bit_set(x, bv[i]) == true then | |
| v = v + bv[i + o] | |
| end | |
| i = i + 1 | |
| end | |
| return v | |
| end | |
| -- encode string | |
| local i = 1 | |
| local n = str:len() | |
| local c, c2, o, v | |
| while i <= n do | |
| -- read 1st byte, if uses one byte we got the char code | |
| -- otherwise read 2nd byte and compute new char code | |
| c = str:byte(i) | |
| o = 1 | |
| -- @see http://en.wikipedia.org/wiki/UTF-8 for UTF8-2 | |
| if c > 127 then | |
| -- read 2nd lower byte | |
| -- compute new char code value | |
| c2 = str:byte(i + 1) | |
| c = bit_val(c, 5, 3, -2) + | |
| bit_val(c, 2, 1, 6) + | |
| bit_val(c2, 6) | |
| o = 2 | |
| end | |
| table.insert(cc, string.format(fmt, c)) | |
| i = i + o | |
| end | |
| return string.upper(table.concat(cc, "")) | |
| end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment