0xDEADBEEF is not tasty, but still 0xFEEDBEEF

Question

Inspired by "You had me at hello"

Introduction

For those who don't know, hexadecimal is a base system like decimal. Contrary to decimal however, it uses 16 (hence the name) digits instead of 10.

Here are the first few numbers while counting up in hexadecimal to maybe give you a feel of things, although you should probably google it for more info.

0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 1A, 1B, 1C, 1D, 1E, 1F, 20...

From now on I'll refer to hexadecimal as hex.

---

Challenge

Take a possibly infinite stream of bytes, and convert it to hex. This should be done as you read since you might get an infinite stream and keep on reading forever.

The twist, is that if a sequence of converted numbers equals the exact hex number DEADBEEF (3735928559 in decimal), you must output FEEDBEEF (4276993775 in decimal) instead.

If no DEADBEEF is found, print the string as it was

Inter-character matches work too, so 2D EA DB EE F0 works and should be converted to 2F EE DB EE F

---

Example

This might seem a bit vague, so here are 2 step-by-step examples.

Example 1

Input: Hello, World (bytes represented as string)

Step 1: Convert the input to ASCII: 072 101 108 108 111 044 032 087 111 114 108 100

Step 2: Convert the ASCII to hex: 48 65 6C 6C 6F 2C 20 57 6F 72 6C 64

Step 3: Print out everything until DEADBEEF is found. It wasn't found, so just print Hello, World again.

Example 2

Input: Hello,Þ­¾ï,,

Step 1: Convert to ASCII: 72 101 108 108 111 44 222 173 190 239 44 44

Step 2: Convert to hex: 48 65 6C 6C 6F 2C DE AD BE EF 2C 2C

Step 3: Is there DEADBEEF? Yep. so take everything up to it. 48 65 6C 6C 6F 2C

Step 4: Append FEEDBEEF to it. 48 65 6C 6C 6F 2C FE ED BE EF

Step 5: Convert back to a string and print it. Output is Hello,þí¾ï

---

Scoring

Try to make a program that does the above in the fewest bytes of source code as possible.

---

Remember!

Make sure your program works for finite and infinite streams. The stream can be STDIN, file, function argument (if you can pass streams as arguments), or your language's closest alternative. Output is to STDOUT or function returning (if you can return streams).

More test cases (contain invisible characters)

eeþí¾Þ­¾ïe (6565feedbedeadbeef65) -> eeþí¾þí¾ïe
Þ­¾ï (deadbeef)-> þí¾ï
Þ=êÛîôV -> Þ?îÛîôV

Answer 1

Japt, 34 30 bytes

-4 bytes from @Shagy

csG r`Ü%¼ef.+``fe‚¼ef` ò ®nG d

Try it online!

Answer 2

Python 3, 121 bytes

I=open(0).read
p=print
x=i=I(1)
while i:p(end=x[:-4]);x=x[-4:];x!='Þ­¾ï'or+p(end='þí¾ï');i=I(1);x+=i
p(end=x[:5])

Try it online!

Answer 3

Clean, 143 bytes

uses octal escapes here and on TIO to prevent UTF-8 mangling, but it does work with the actual values and so each escape sequence is counted as one byte

This one works with inter-character sequences.

import StdEnv
$l=:[a,'\352\333\356',b:t]|(toInt a)rem 16==13&&(toInt b)>>4==15=[a+'\002':['\356\333\356\015']]=[hd l: $(tl l)]
$['\336\255\276\357':_]=['\376\355\276\357']
$[h:t]=[h: $t]
$e=e

Try it online!

Clean, 38 bytes

uses octal escapes here and on TIO to prevent UTF-8 mangling, but it does work with the actual values and so each escape sequence is counted as one byte

This one does not work with inter-character sequences.

$['\336\352\276\357':_]=['\376\355\276\357']
$[h:t]=[h: $t]
$e=e

Try it online!

Defines the function $ :: [Char] -> [Char] taking and returning a list of characters.
Since Clean is lazily-evaluated, this is equivalent to a stream.

Answer 4

05AB1E, 30 bytes

ÇhJl’–ìÀÜ’¡Dнsg1›’ŽåÀܒ׫2ôHçJ

Try it online or verify a few more test cases.

Explanation:

Ç                               # Convert the (implicit) input-string to codepoint decimals
 h                              # Convert them to hexadecimal
  J                             # Join them together
   l                            # Convert to lowercase
    ’–ìÀÜ’                      # Push compressed string "deadbeef"
          ¡                     # Split the hexadecimal string by this "deadbeef"
           D                    # Duplicate
            н                   # Take the first item of the split-list
             s                  # Swap so the full list is at the top of the stack again
              g1›               # Check if its length is larger than 1
                 ’ŽåÀܒ׫       # If it is: concat "feedbeef" to the head-string
                         2ô     # Split the string into parts of size 2
                           H    # Convert hexadecimal back to decimal
                            ç   # Decimal back to characters
                             J  # Join the characters (and output implicitly)

See this 05AB1E tips of mine (section How to use the dictionary?) to understand why ’–ìÀÜ’ is "deadbeef" and ’ŽåÀÜ’ is "feedbeef".

Answer 5

C (gcc), 137 bytes

i=16;n;long long u;c;R(k){u=u<<4|c>>k&15,u^=u+'!RA'<<32?0:' @';}main(){for(;n+=!!~(c=getchar(n-=!i&&~putchar(u>>32)));i/=2)R(4),R(0);}

Try it online!