GHCi as a Hex-Calculator interactive

The GHCi (REPL for Haskell) is a very useful interactive tool, it’s not only for debugging :)
This package “ghci-hexcalc” is an interactive hex-calculator using Haskell/GHCi. This is a simple and casual tool like Perl and Excel for our daily work.

Interactive oriented features:

• Short-named operators and functions
• Show values in hexadecimal format by default
• Suppress type annotation of numeric literals by type inference
• Postfix-notation available
• Highlight available

See also description on Hackage.

Contents

• Run
• Example of use
• Specification
• Appendix

Run

Bare GHCi:

$ ghci src/Data/GHex.hs

or

$ ghci -ghci-script example/example.ghci

Stack:

$ stack exec -- ghci  src/Data/GHex.hs

Example of use

Numeric literals by Hex type annotation

The value of Hex type is shown as hexadecimal format.

ghci> 1 :: Hex
0x0000_0000_0000_0001

ghci> 0xff :: Hex
0x0000_0000_0000_00ff

ghci> 0b1011 :: Hex
0x0000_0000_0000_000b

ghci> 16 + 3 :: Hex
0x0000_0000_0000_0013

Variables on GHCi

You could use variables of Haskell syntax on GHCi.

ghci> x = 255 :: Hex
ghci> x + 3
ghci> y = it       -- `it` is GHCi's variable. It stores the previous result.

Arithmetic operations

You could also use arithmetic operators in Hex type.

ghci> x + 3
ghci> (x * 256) -1
ghci> x + 2^10
ghci> neg x

Logical operations

Numeric literals applied to functions of this package are inferred as Hex type.

ghci> 0xff .& 6
ghci> 256 .| 16
ghci> 100 .^ 5
ghci> inv 255

Shift operations

ghci> 1 .<< 16
ghci> 256 .>> 1

Div and mod operations

ghci> 0xff0000 ./ 256
ghci> 0xfedc .% 256

Generate bit and byte with position

ghci> bit1 15
0x0000_0000_0000_8000

ghci> bits 7 4
0x0000_0000_0000_00f0

ghci> bitList [15, 14, 1]
0x0000_0000_0000_c002

ghci> byte1 2
0x0000_0000_00ff_0000

ghci> bytes 4 3
0x0000_00ff_ff00_0000

Extract and replace bits

ghci> gets 0xabcd 15 12
0x0000_0000_0000_000a

ghci> puts 0xabcd 15 12 7
0x0000_0000_0000_7bcd

Set and clear bits

ghci> sbits 0x1234 11 8
0x0000_0000_0000_1f34

ghci> cbits 0x1234 7 4
0x0000_0000_0000_1204

Get asserted bit positions and count bits

ghci> pos1 0x0081
[7,0]

ghci> pos0 $ inv 0x0100
[8]

ghci> count1 0b11001
3

Permute, split and merge

ghci> gather 0x12345678 0x0ff000f0
0x0000_0000_0000_0237

ghci> scatter 0x12345678 0xff00ff00 0xabcd
0x0000_0000_ab34_cd78

ghci> (3,0b101) .++ (2,0b11)
(5,0x0000_0000_0000_0017)

Predefined-constants

ghci> mega
0x0000_0000_0010_0000

ghci> giga
0x0000_0000_4000_0000

ghci> 4 * giga - 1
0x0000_0000_ffff_ffff

ghci> 2^32 ./ giga
0x0000_0000_0000_0004

Postfix-notation

The operator .@ is an operator for postfix notation. It’s the same as Data.Function.(&).

The following two are the same:

ghci> pos1 0xf0
[7,6,5,4]

ghci> 0xf0 .@pos1
[7,6,5,4]

Formatting for hex, bin, dec, Tera,Giga,Mega,Kilo, signed and floating

Formatting functions convert a Hex type value to a string type for each format.

ghci> 2^16 .@hex
"0x0000_0000_0001_0000"

ghci> 100 .@bin
"0b110_0100"

ghci> 100 .@bin16
"0b0000_0000_0110_0100"

ghci> giga .@dec
"1073741824"

ghci> bit 43 .@decT
"8"

ghci> 0xffffffffffffffff .@signed
"-1"

ghci> 0x3fc00000 .@float
"1.5"

Hilighting specified bits

The function color highlights specified bits. It inverts the color in the ANSI sequence for the specified bits.

ghci> 0xff .@color (bits 7 4)
0b0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_1111_1111
                                                                        ^^^^

ghci> 0xffffffff .@color mega
0b0000_0000_0000_0000_0000_0000_0000_0000_1111_1111_1111_1111_1111_1111_1111_1111
                                                       ^

ghci> 0 .@color (bitList [54,53,4,3,2])
0b0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_0000_0000
             ^^                                                            ^ ^^

Input and convert

The function inputRawHexIO inputs a string and converts it to a Hex type.

ghci> inputRawHexIO
ff aa  (your input)
ghci> x = it
ghci> x
0x0000_0000_0000_ffaa

ghci> x <- inputRawHexIO
ff aa  (your input)
ghci> x
0x0000_0000_0000_ffaa

Floating conversion examples

ghci> float2hex 1.0
0x0000_0000_3f80_0000

ghci> hex2float 0x3fc00000
1.5

ghci> double2hex 1.0
0x3ff0_0000_0000_0000

ghci> hex2double 0x40091eb851eb851f
3.14

Combination examples

ghci> x .| bit1 18
ghci> (x .<< 4) .& 0xf0
ghci> bit1 33 ./ giga
ghci> 2 * mega .@dec
ghci> 4 * tera .@pos1
ghci> foldr1 (.|) [0xa, 0xcc, 0xd1b]
ghci> 0 .@color (tera .| giga .| mega .| kilo)

Using Data.Bits library

Hex type is deriving Data.Bits type. So you could use functions of Data.Bits.

ghci> x `testBit` 8
ghci> x `clearBit` 15
ghci> x .&. 0xff
ghci> x `rotateL` 4
ghci> countLeadingZeros x

Clear screen

ghci> cls

Simple help

Show simple usage:

ghci> usage

Listing APIs with :browse ghci command:

ghci> :browse
newtype Hex = Hex Word
(.&) :: Hex -> Hex -> Hex
(.|) :: Hex -> Hex -> Hex
(.^) :: Hex -> Hex -> Hex
inv :: Hex -> Hex
  :

When you run with ghci -haddock, you could also use :doc ghci command (ghc8.6 or later):

ghci> :doc bits
 Set bits from n1 to n2

 >>> bits 15 8
 0x0000_0000_0000_ff00

Specification

Please see also Hackage document in detail.

General

• Core type:
  • The core type of this package is the Hex type.
  • Hex type is implemented in unsigned Word.
  • Hex type is 64 bit length on x86_64.
• Operators and functions:
  • Operators in this package begin with .(dot), like .& and .|.
  • Most functions align bits to the LSB side.

Numeric literals by Hex type annotation

Operation	Description
:: Hex	Type annotation for basic type Hex
255 :: Hex	Decimal number literal
0xff :: Hex	Hexadecimal number literal
0b1101 :: Hex	binary number literal

Derived operations

Operation	Description
Many operations	Eq, Ord, Num, Enum, Real, Bounded, Integral, Bits and FiniteBits class available

Postfix operator

Operation	Description
.@	Postfix-notation operator

Arithmetic operations

Operation	Description
+, -, *, ^, …	Num, Real class available
neg x1	Negation. (inv x1 + 1)
signext x1 n1	Sign extention
x1 ./ x2	Integer division
x1 .% x2	Integer modulo

Logical operations

Operation	Description
x1 .& x2	Bitwise “and”
x1 .| x2	Bitwise “or”
x1 .^ x2	Bitwise “xor”
inv x1	Bitwise “not” (invert)

Shift operations

Operation	Description
x1 .<< n1	Logical left shift
x1 .>> n1	Logical right shift

Generate bit and byte with position

Operation	Description
bit1 n1	Set a bit
bits n1 n2	Set bits from n1 to n2
bitList [n1, n2, … nn]	Set bits with List
byte1 n1	Set a byte
bytes n1 n2	Set bytes from n1 to n2
mask n1	Set bits from 0 to n1

Extract and replace bits

Operation	Description
gets x1 n1 n2	Extract bits from n1 to n2
puts x1 n1 n2 x2	Replace bits from n1 to n2
getBit1 x1 n1	Extract bit at n1
getBits x1 n1 n2	Synonym to gets
getByte1 x1 n1	Extract bytes from n1 to 0
getBytes x1 n1 n2	Extract bytes from n1 to n2
putBit1 x1 n1 x2	Replace byte at n1
putBits x1 n1 n2 x2	Synonym to puts
putBytes x1 n1 n2 x2	Replace bytes from n1 to n2

Set and clear bits

Operation	Description
sbits x1 n1 n2	Set bits from n1 to n2 of x1
cbits x1 n1 n2	Clear bits from n1 to n2 of x1

Get asserted bit positions and count bits

Operation	Description
pos1 x1	Get bit positions asserted with 1
pos0 x1	Get bit positions asserted with 0
range x1	Get upper and lower boundaries
count1 x1	Count bit-1
count0 x1	Count bit-0

Permute

Operation	Description
bitrev x1	Reverse bits
byterev x1	Reverse bytes
gather x1 x2	Gather bits from x1 by x2
scatter x1 x2 x3	Scatter bits from x3 to x1 by x2

Split and merge

Operation	Description
splitBits x1	Split bits to List
splitBytes x1	Split bytes to List
mergeBits [x1, x2, .. xn]	Merge bits from List
mergeBytes [x1, x2, .. xn]	Merge bytes from List
splitPairs [n1, .. nn] x1	Split bits to pair of (length,Hex)
mergePairs [(n1,x1),..]	Merge bits from pair of (length,Hex)
(n1,x1) .++ (n2,x2)	Concatinate pairs of (length,Hex)

Predefined-constants

Constant	Description
exa	2^60 (It’s not 10^18)
peta	2^50 (It’s not 10^15)
tera	2^40 (It’s not 10^12)
giga	2^30 (It’s not 10^9)
mega	2^20 (It’s not 10^6)
kilo	2^10 (It’s not 10^3)
zero	0
one	1
all0	0x0
all1	inv all0
hexBitSize	64 on x86_64. Thus size of Word
hexBitSeq	[hexBitSize-1, hexBitSize-2, .. 0]

Formatting for hex, bin, dec, Tera,Giga,Mega,Kilo, signed and floating

Operation	Description
.@hex	Show in hexadecimal string
.@hex8	Show in hexadecimal string of 8bit
.@hex16	Show in hexadecimal string of 16bit
.@hex32	Show in hexadecimal string of 32bit
.@hex64	Show in hexadecimal string of 64bit
.@hexN n1	Show in hexadecimal string of n1 bit
.@bin	Show in binary string
.@bin8	Show in binary string of 8bit
.@bin16	Show in binary string of 16bit
.@bin32	Show in binary string of 32bit
.@bin64	Show in binary string of 64bit
.@binN n1	Show in binary string of n1 bit
.@dec	Show in decimal string
.@decE	Show in decimal of Exa unit
.@decP	Show in decimal of Peta unit
.@decT	Show in decimal of Tera unit
.@decG	Show in decimal of Giga unit
.@decM	Show in decimal of Mega unit
.@decK	Show in decimal of Kilo unit
.@signed	Show in singed decimal with Word
.@float	Show in float string
.@double	Show in double string
.@hexSized	Show in hexadecimal string of (len,Hex)
.@binSized	Show in binary string of (len,Hex)

Hilighting and pretty-print

Operation	Description
color x1 x2	Highlight bit of x1 specified with x2
ppr fun x1	Print x1 applied with fun

Input and convert

Operation	Description
inputRawHexIO	Input string and convert to Hex type

Floating convert

Operation	Description
float2hex	Convert Float to Hex type
hex2float	Convert Hex to Float type
double2hex	Convert Double to Hex type
hex2double	Convert Hex to Double type

Miscellaneous

Operation	Description
cls	Clear screen by ANSI sequence
usage	Show simple help

Appendix

GHC language extention for numeric literals

When -XBinaryLiterals extention enabled, you can use binary literals on GHC and GHCi, like 0b1101.

When -XNumericUnderscores extention enabled, you can use underscores in numeric literals on GHC and GHCi, like 0xff_ff. -XNumericUnderscores extension is available GHC 8.6 or later.

GHC language extensions can be described in ~/.ghci or ./ghci file:

:set -XBinaryLiterals
:set -XNumericUnderscores

GHC language extensions can also be specified as an option when starting GHC and GHCi:

$ ghci -XBinaryLiterals -XNumericUnderscores

Shell alias

It is useful to set the alias of the shell:

alias ghex="(cd $XXX/ghci-hexcalc; ghci -ghci-script example/example.ghci)"

Expression evaluation mode of GHC

You can also run in one shot mode (a expression evaluation mod) by ghc -e:

$ ghc src/Data/GHex.hs -e '4 * giga'
0x0000_0004_0000_0000

Default declaration on GHCi

With the default declaration, integer literals are inferred as Hex types.

ghci> default (Hex)
ghci> 255
0x0000_0000_0000_00ff