Bash shell Decimal to Binary base 2 conversion
Asked Answered
E

9

47

I'm looking for an easy way in Bash to convert a decimal number into a binary number. I have variables that need to be converted:

$ip1 $ip2 $ip3 $ip4

Is there a simple method to do this without looking at every individual number?

I would prefer not to have to write a lot of code.

Ethyne answered 23/4, 2012 at 10:7 Comment(0)
O
87

You can use bc as:

echo "obase=2;$ip1" | bc

See it

Odense answered 23/4, 2012 at 10:9 Comment(9)
At the moment i have echo "obase=2;$ip1" | bc echo "obase=2;$ip2" | bc echo "obase=2;$ip3" | bc echo "obase=2;$ip4" | bc and the echo each line is outputing a newline. is there a way to have it output tip to tailEthyne
@DanielDC: You can do: echo "obase=2;10;20;30" | bc | tr -d '\n'Odense
@DanielDC: Add a | tr -d '\n' at the end.Odense
lol i totally forgot the | tr -d '\n' thanks heaps for your help man appreciate itEthyne
Great solution; a slightly more compact (and marginally better-performing) version is: bc <<<"obase=2;$ip1".Arterialize
This is a great solution. The explanation is that bc is a calculator which accepts input from stdin. In this case, the input is the string to set the base of the output to 2 (presumable the default input is base 10), and then just echo the number in the "ip1" variable. The <<< is just a quick way to pass a string or variable to somethings stdin (ie. it is equivelent to echo "obase=2;$ip1" | bc (but presumably preferred since it eliminates the call to echo). A more straightforward example of bc adding two and two is bc <<<"2+2".Kitchens
In case you're looking for the reversal process, ie binary to decimal, here you go echo $((2#$@)) where $@ is your binary number.Swear
How do you specify the number of digits?Nombril
I asked that question here: How to make bc output a desired number of base-2 binary digitsNombril
G
60

Convert decimal to binary with bash builtin commands (range 0 to 255):

D2B=({0..1}{0..1}{0..1}{0..1}{0..1}{0..1}{0..1}{0..1})

echo ${D2B[7]}

00000111

echo ${D2B[85]}

01010101

echo ${D2B[127]}

01111111


To remove leading zeros, e.g. from ${D2B[7]}:

echo $((10#${D2B[7]}))

111


This creates an array with 00000000 00000001 00000010 ... 11111101 11111110 11111111 with bash‘s brace expansion. The position in array D2B represents its decimal value.

See also: Understanding code ({0..1}{0..1}{0..1}{0..1}{0..1}{0..1}{0..1}{0..1})

Gondolier answered 19/9, 2014 at 22:58 Comment(11)
Very nice pure bash solution! To generalize this for range 0 to 2^n-1, use eval D2B='('$(for ((i=0; i<$n; i++)); do printf '%s' "{0..1}"; done)')' (after setting n). The only drawback is that this method takes O(2^n) space (superpolynomial!). This quickly draws unwieldy.Divaricate
Incredible tight solution! Can you explain what happens? What does {0..1} ?Handling
The output of echo $D2B is 00000000 which was confusing me, as this does not look like an array but a simple string. Thank you for enlightening me!Handling
@domih: Here D2B is an array. I do not recommend using $D2B because it is intended to display strings. To access the first element of an array use ${D2B[0]}.Gondolier
simple example to help understand: a=({0..1}{2..3}); echo ${a[@]}. The echo statement prints a for all indices.Economics
This is really cool, but won't this blow up in memory because it allocates the entire array?Erastianism
@bjd2385: Memory consumption is kept within limits with this small array. This shows the complete array: echo {0..1}{0..1}{0..1}{0..1}{0..1}{0..1}{0..1}{0..1}Gondolier
@Gondolier but if I tried this option with a 4 byte arrangement, e.g. to represent an IP address in base 10 for a lookup...it seems to crash my terminal. Very cool though! There are other methods of accomplishing that though to get around this.Erastianism
@bjd2385: With more digits the array scales very quickly.Gondolier
I never known Brace Expansion can do multi-dimensional orthogonal. very impressive!!Grammalogue
the magic of Brace Expansion x Array. you are a bash artistGrammalogue
E
7

Decimal to binary conversion in Bash:

I'm using Ubuntu 14.04 to do this.

Convert the decimals 1 through 5 to binary.

el@apollo:~$ bc <<< "obase=2;1"
1
el@apollo:~$ bc <<< "obase=2;2"
10
el@apollo:~$ bc <<< "obase=2;3"
11
el@apollo:~$ bc <<< "obase=2;4"
100
el@apollo:~$ bc <<< "obase=2;5"
101

Bonus example:

el@apollo:~$ bc <<< "obase=2;1024"
10000000000

el@apollo:~$ bc <<< "obase=2;2^128"
100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
Enchase answered 20/6, 2014 at 20:38 Comment(0)
S
5

I like to use dc for this. It's very concise:

$ n=50; dc -e "$n 2op"
110010

The commands here are as follows:

  1. Push the number, n, on the stack, via shell expansion.
  2. Push 2 on the stack, then use o to pop the stack and use 2 as the output radix.
  3. p to print the top of the stack (which is just n), using the output radix set in step 2 (so print in binary).

If you want padding:

$ n=50; pad_size=16; printf "%0${pad_size}d\n" $(dc -e "$n 2op")
0000000000110010
Sigurd answered 11/5, 2022 at 13:42 Comment(0)
B
3

General method for converting an integer number into another representation with another base (but base<=10 because of using digits 0..9 for representation, only):

function convertIntvalToBase () # (Val Base)
{
   val=$1
   base=$2
   result=""
   while [ $val -ne 0 ] ; do
        result=$(( $val % $base ))$result #residual is next digit
        val=$(( $val / $base ))
   done
   echo -n $result
}

e.g.

convertIntvalToBase $ip1 2     # converts $ip1 into binary representation
Borrow answered 4/11, 2014 at 16:5 Comment(0)
S
3

Defined as a function in bash:

# to Binary:
toBinary(){
    local n bit
    for (( n=$1 ; n>0 ; n >>= 1 )); do  bit="$(( n&1 ))$bit"; done
    printf "%s\n" "$bit"
}
Shallow answered 19/9, 2021 at 23:13 Comment(3)
Any reason for using printf "%s\n" "$bit" rather than just echo $bit?Sigurd
Many reasons. Please read Why is printf better than echo?.Shallow
Ew, yeah that's compelling enough I guess.Sigurd
H
1

Decimal to Binary using only Bash

Any integer number can be converted ti binary using it::

touch dec2bin.bash && chmod +x "$_" && vim "$_"

And, then copy paste the following:

#!/bin/bash
num=$1;
dec2bin()
{
    op=2; ## Since we're converting to binary
    quo=$(( $num/ $op)); ## quotient
    rem=$(( $num% $op)); ## remainder
    array=(); ## array for putting remainder inside array
    array+=("$rem"); ## array expansion
        until [[ $quo -eq 0 ]]; do
            num=$quo; ## looping to get all remainder, untill the remainder is 0
            quo=$(( $num / $op));
            rem=$(( $num % $op));
            array+="$rem"; ## array expansion
        done
    binary=$(echo "${array[@]}" | rev); ## reversing array
    printf "$binary\n"; ## print array
}
main()
{
[[ -n ${num//[0-9]/} ]] &&
    { printf "$num is not an integer bruv!\n"; return 1;
    } || { dec2bin $num; }
}
main;

For example:

./dec2bin.bash $var
110100100

Integer must be added!!

./dec2bin.bash 420.py
420.py is not an integer bruv!

Also, another way using python: Much slower

python -c "print(bin(420))"
0b110100100

Hexadecimal to Binary using only Bash

Similarly, hexadecimal to binary, as follows using only bash:

#!/usr/local/bin/bash ## For Darwin :( higher bash :)
#!/bin/bash ## Linux :)
hex=$1;
hex2bin()
{
    op=2; num=$((16#$hex));
    quo=$(( $num/ $op));
    rem=$(( $num% $op));
    array=();
    array+=("$rem");
        until [[ $quo -eq 0 ]]; do
            num=$quo;
            quo=$(( $num / $op));
            rem=$(( $num % $op));
            array+="$rem";
        done
    binary=$(echo "${array[@]}" | rev);
    printf "Binary of $1 is: $binary\n";
}
main()
{
[[ -n ${hex//[0-9,A-F,a-f]/} ]] &&
    { printf "$hex is not a hexa decimal number bruv!\n"; return 1;
    } || { hex2bin $hex; }
}
main;

For example:

./hex2bin.bash 1aF
Binary of 1aF is: 110101111

Hex must be passed:

./hex2bin.bash XyZ
XyZ is not a hexa decimal number bruv!
His answered 30/4, 2019 at 11:37 Comment(0)
N
1

To make @codaddict's answer a little more pretty, use this to prefix the output with 0b for "binary":

printf "0b%s\n" "$(echo "obase=2; $((num1 + num2))" | bc)"

Example:

num1=2#1111  # binary  1111 (decimal 15)
num2=2#11111 # binary 11111 (decimal 31)
printf "0b%s\n" "$(echo "obase=2; $((num1 + num2))" | bc)"

Output:

0b101110

This is decimal 46.

For details on the input base-2 formatting in bash, such as 2#1111 above, see the very end of my answer here: How to use all bash operators, and arithmetic expansion, in bash.

To have at least 8 digits in the output, use:

printf "0b%08d\n" $(echo "obase=2; $((num1 + num2))" | bc)

Source: David Rankin, in an answer to my question here.

Nombril answered 22/3, 2022 at 7:27 Comment(0)
L
0
toBin ()
{
    printf "%08d\n" $(dc -e "$1 2op")
}


$ toBin 37
00100101
Ladanum answered 2/1, 2023 at 16:10 Comment(0)

© 2022 - 2024 — McMap. All rights reserved.