-- Taken from GHC: compiler/utils/Encoding.hs
import Data.Char
import System.Environment (getArgs)
-- -----------------------------------------------------------------------------
-- The Z-encoding
{-
This is the main name-encoding and decoding function. It encodes any
string into a string that is acceptable as a C name. This is done
right before we emit a symbol name into the compiled C or asm code.
Z-encoding of strings is cached in the FastString interface, so we
never encode the same string more than once.
The basic encoding scheme is this.
* Tuples (,,,) are coded as Z3T
* Alphabetic characters (upper and lower) and digits
all translate to themselves;
except 'Z', which translates to 'ZZ'
and 'z', which translates to 'zz'
We need both so that we can preserve the variable/tycon distinction
* Most other printable characters translate to 'zx' or 'Zx' for some
alphabetic character x
* The others translate as 'znnnU' where 'nnn' is the decimal number
of the character
Before After
--------------------------
Trak Trak
foo_wib foozuwib
> zg
>1 zg1
foo# foozh
foo## foozhzh
foo##1 foozhzh1
fooZ fooZZ
:+ ZCzp
() Z0T 0-tuple
(,,,,) Z5T 5-tuple
(# #) Z1H unboxed 1-tuple (note the space)
(#,,,,#) Z5H unboxed 5-tuple
(NB: There is no Z1T nor Z0H.)
-}
type UserString = String -- As the user typed it
type EncodedString = String -- Encoded form
zDecodeString :: EncodedString -> UserString
zDecodeString [] = []
zDecodeString ('Z' : d : rest)
| isDigit d = decode_tuple d rest
| otherwise = decode_upper d : zDecodeString rest
zDecodeString ('z' : d : rest)
| isDigit d = decode_num_esc d rest
| otherwise = decode_lower d : zDecodeString rest
zDecodeString (c : rest) = c : zDecodeString rest
decode_upper, decode_lower :: Char -> Char
decode_upper 'L' = '('
decode_upper 'R' = ')'
decode_upper 'M' = '['
decode_upper 'N' = ']'
decode_upper 'C' = ':'
decode_upper 'Z' = 'Z'
decode_upper ch = {-pprTrace "decode_upper" (char ch)-} ch
decode_lower 'z' = 'z'
decode_lower 'a' = '&'
decode_lower 'b' = '|'
decode_lower 'c' = '^'
decode_lower 'd' = '$'
decode_lower 'e' = '='
decode_lower 'g' = '>'
decode_lower 'h' = '#'
decode_lower 'i' = '.'
decode_lower 'l' = '<'
decode_lower 'm' = '-'
decode_lower 'n' = '!'
decode_lower 'p' = '+'
decode_lower 'q' = '\''
decode_lower 'r' = '\\'
decode_lower 's' = '/'
decode_lower 't' = '*'
decode_lower 'u' = '_'
decode_lower 'v' = '%'
decode_lower ch = {-pprTrace "decode_lower" (char ch)-} ch
-- Characters not having a specific code are coded as z224U (in hex)
decode_num_esc :: Char -> EncodedString -> UserString
decode_num_esc d
= go (digitToInt d)
where
go n (c : rest) | isHexDigit c = go (16*n + digitToInt c) rest
go n ('U' : rest) = chr n : zDecodeString rest
go n other = error ("decode_num_esc: " ++ show n ++ ' ':other)
decode_tuple :: Char -> EncodedString -> UserString
decode_tuple d
= go (digitToInt d)
where
-- NB. recurse back to zDecodeString after decoding the tuple, because
-- the tuple might be embedded in a longer name.
go n (c : rest) | isDigit c = go (10*n + digitToInt c) rest
go 0 ('T':rest) = "()" ++ zDecodeString rest
go n ('T':rest) = '(' : replicate (n-1) ',' ++ ")" ++ zDecodeString rest
go 1 ('H':rest) = "(# #)" ++ zDecodeString rest
go n ('H':rest) = '(' : '#' : replicate (n-1) ',' ++ "#)" ++ zDecodeString rest
go n other = error ("decode_tuple: " ++ show n ++ ' ':other)
{-
Tuples are encoded as
Z3T or Z3H
for 3-tuples or unboxed 3-tuples respectively. No other encoding starts
Z<digit>
* "(# #)" is the tycon for an unboxed 1-tuple (not 0-tuple)
There are no unboxed 0-tuples.
* "()" is the tycon for a boxed 0-tuple.
There are no boxed 1-tuples.
-}
main :: IO ()
main = do
args <- getArgs
putStrLn . unwords $ map zDecodeString args