From Compsci.ca Wiki
The brainfuck language is an esoteric programming language noted for its extreme minimalism. It was designed to challenge and amuse programmers, and is not suitable for practical use. Its name has been variously bowdlerized, as in brainf*ck or brainfsck. The name of the language is generally not capitalized, although it is a proper noun.
Urban Müller created brainfuck in 1993 with the intention of designing a language which could be implemented with the smallest possible compiler , inspired by the 1024 byte compiler for the FALSE programming language. Several brainfuck compilers have been made smaller than 200 bytes. The classic distribution is Müller's version 2, containing a compiler for the Amiga, an interpreter, example programs, and a readme document.
The language consists of eight commands, listed below. A brainfuck program is a sequence of these commands, possibly interspersed with other characters (which are ignored). The commands are executed sequentially, except as noted below.
The brainfuck language uses a simple machine model consisting, besides the program, of an array of 30,000 byte cells initialized to zero; a movable pointer into the array (initialized to point to the leftmost byte of the array); and two streams of bytes for input and output (most often connected to a keyboard and a monitor respectively, and using the ASCII character encoding).
The eight language commands, each consisting of a single character, are the following:
|increment the pointer (to point to the next cell to the right).|
|decrement the pointer (to point to the next cell to the left).|
|increment (increase by one) the byte at the pointer.|
|decrement (decrease by one) the byte at the pointer.|
|output the value of the byte at the pointer.|
|accept one byte of input, storing its value in the byte at the pointer.|
|jump forward to the command after the corresponding |
|jump back to the command after the corresponding |
] command may instead be translated as an unconditional jump to the corresponding
[ command, or vice versa; programs will behave the same but will run more slowly.)
Brainfuck programs can be translated into C using the following substitutions, assuming
ptr is of type
unsigned char* and has been initialized to point to an array of zeroed bytes:
|brainfuck command||C equivalent|
As the name suggests, brainfuck programs tend to be difficult to comprehend. Partly this is because any mildly complex task requires a long sequence of commands; partly it is because the program's text gives no direct indications of the program's state. These, as well as brainfuck's inefficiency and its limited input/output capabilities, are some of the reasons it is not used for serious programming. Nonetheless, like any Turing-complete language, brainfuck is theoretically capable of computing any computable function or simulating any other computational model, if given an unlimited memory store. A variety of brainfuck programs have been written.
Brainfuck's formal "parent language"
Except for its two I/O commands, brainfuck is a minor variation of the formal programming language P prime prime (P′′) created by Corrado Böhm in 1964. (In fact, using six symbols equivalent to the respective brainfuck commands +, -, <, >, [, ], Böhm provided an explicit program for each of the basic functions that together serve to compute any computable function. So in a very real sense, the first "brainfuck" programs appear in Böhm's 1964 paper – and they were programs sufficient to prove Turing-completeness.)
The following program prints "Hello World!" and a newline to the screen:
++++++++++ [>+++++++>++++++++++>+++>+<<<<-] The initial loop to set up useful values in the array >++. Print 'H' >+. Print 'e' +++++++. Print 'l' . Print 'l' +++. Print 'o' >++. Print ' ' <<+++++++++++++++. Print 'W' >. Print 'o' +++. Print 'r' ------. Print 'l' --------. Print 'd' >+. Print '!' >. Print newline
For readability, this code has been spread across many lines and comments have been added. Brainfuck ignores all characters but
+-<>,. so no special syntax for comments is needed. The code could just as well have been written as:
The first line initialises
a = 10 by simply incrementing ten times from 0. The loop from line 2 effectively sets the initial values for the array: a = 70 (close to 72, the ASCII code for the character 'H'),
a = 100 (close to 101 or 'e'),
a = 30 (close to 32, the code for space) and
a = 10 (newline). The loop works by multiplying the value of
10, by 7, 10, 3, and 1, saving the results in other cells. After the loop is finished, a is zero.
>++. then moves the pointer to
a which holds
70, adds two to it (producing 72 which is the ASCII character code of a capital H), and outputs it.
The next line moves the array pointer to
a and adds one to it, producing
101, a lower-case 'e', which is then output.
As 'l' happens to be the seventh letter after 'e', to output 'll' we add another seven (
a and output the result twice.
'o' is the third letter after 'l', so we increment
a three more times and output the result.
The rest of the program goes on in the same way. For the space and capital letters, different array cells are selected and incremented or decremented as needed.
A simple program fragment which sets the current location to 0, by iteratively decrementing until it is equal to 0.
A continuous loop that takes text input from the keyboard and echoes it to the screen (similar to the Unix cat program). Note that this assumes the cell is set to 0 when a '
,' command is executed after the end of input (sometimes called end-of-file or "EOF"); implementations vary on this point. For implementations that set the cell to -1 on EOF, or leave the cell's value unchanged, this program would be written "
,+[-.,+]" or "
Moving the pointer
A version of the last one that also saves all the input in the array for future use, by moving the pointer each time.
This adds the current location (destructively, it is left at zero) to the next location.
Conditional loop statements
This will take lowercase input from the keyboard and make it uppercase. To exit, press the enter key.
First, we input the first character using the
, and immediately subtract 10 from it. (Most, but not all, brainfuck implementations use 10 as the code for a newline character.) If the user hits enter, the loop command (
[) will jump past the end of the loop, because we will have set the first byte to zero. If the character input was not a 10, we boldly assume it was a lowercase letter, and enter the loop, wherein we subtract another 22 from it, for a total of 32, which is the difference between an ASCII lowercase letter and the corresponding uppercase letter.
Next we output it. Now we input the next character, and again subtract 10. If this character was a linefeed, we exit the loop; otherwise, we go back to the start of the loop, subtract another 22, output, and so on. When we exit the loop, the program terminates, as there are no more commands.
Copying a byte
In the following section, [n] denotes the nth byte in the array: , ,  and so on.
Brainfuck does not include an operation for copying bytes. This must be done with the looping construct and arithmetical operators. Moving a byte is simple enough; moving the value of  to  can be done as follows:
However, this resets the value of  to 0. We can restore the value of  after copying by taking advantage of the ability to copy a value to two places at once. To copy the value of  to both  and  is simple:
We can take advantage of this to restore the value of . Therefore, we can nondestructively copy  to  (using  as scratch space) as follows:
This program adds two single-digit numbers and displays the result correctly if it too has only one digit:
The first number is input in , and 48 is subtracted from it to correct it (the ASCII codes for the digits 0-9 are 48-57). This is done by putting a 6 in  and using a loop to subtract 8 from  that many times. (This is a common method of adding or subtracting large numbers.) Next, the second number is input in .
The next loop
[<+>-] does the real work, moving the second number onto the first, adding them together and zeroing . Each time through, it adds one to  and subtracts one from ; so by the time  is zeroed, as many have been added to  as have been removed from . Now a return is input in . (We're not error-checking the input at all.)
Then the pointer is moved back to the , which is then output. ( is now a + (b + 48), since we didn't correct b; which is identical to (a + b) + 48, which is what we want.) Now the pointer is moved to , which holds the return that was input; it is now output, and we're done.
Apparently, some implementations prefer this variant which does not use linefeeds at all:
,>,>++++++++[<------<------>>-] <<[>[>+>+<<-]>>[<<+>>-]<<<-] >>>++++++[<++++++++>-],<.>.
Like the previous, but does multiplication, not addition.
The first number is input in , the second number is input in , and both numbers are corrected by having 48 subtracted.
Now we enter the main multiplication loop. The basic idea is that each time through it we subtract one from  and add  to the running total kept in . In particular: the first inner loop moves  onto both  and , while zeroing . (This is the basic way to duplicate a number.) The next inner loop moves  back into , zeroing . Then one is subtracted from , and the outer loop is ended. On exiting this loop,  is zero,  still has the second number in it, and  has the product of the two numbers. (Had we cared about keeping the first number, we could have added one to  each time through the outer loop, then moved the value from  back to  afterward.)
Now we add 48 to the product, input a return in , output the ASCIIfied product, and then output the return we just stored.
This example accepts two numbers from the user, divides them, and displays the truncated quotient (ie integer division).
The following code prompts the user for two numbers; the dividend is stored in memory location (0) and the divisor is stored in (1) The code then enters the main loop. With each iteration of the loop, the code subtracts the divisor from the dividend. If the difference is greater than zero, the cell holding the quotient is incremented, and the process repeated until the dividend reaches zero. The result is
,>,>++++++[-<--------<-------->>] Store 2 numbers from keyboard in (0) and (1); and subtract 48 from each <<[ This is the main loop, which continues until the dividend in (0) is zero >[->+>+<<] Destructively copy the divisor from (1) to (2) and (3); setting (1) to zero >[-<<- Subtract the divisor in (2) from the dividend in (0); the difference is stored in (0) and (2) is cleared [>]>>>[<[>>>-<<<[-]]>>]<<] If the dividend in (0) is zero; exit the loop >>>+ Add one to the quotient in (5) <<[-<<+>>] Destructively copy the divisor in (3) to (2) <<<] Move the stack pointer to (0) and go back to the start of the main loop >[-]>>>>[-<<<<<+>>>>>] Destructively copy the quotient in (5) to (0) (not necessary; but cleaner) <<<<++++++[-<++++++++>]<. Add 48 and print result
Partly because Urban Müller did not write a thorough language specification, the many subsequent brainfuck interpreters and compilers have come to use slightly different dialects of brainfuck.
In the classic distribution, the cells are 8-bit bytes, and this is still the most common size. However, to read non-textual data, a brainfuck program may need to distinguish an end-of-file condition from any possible byte value; thus 16-bit cells have also been used. Some implementations have used 32-bit cells, 64-bit cells, or bignum cells with practically unlimited range, but programs that use this extra range are likely to be slow, since storing or using cell values generally takes time proportional to the values stored or used.
In all these variants, the
. commands still read and write data in bytes. In most of them, the cells wrap around, i.e. incrementing a cell which holds its maximal value (with the
+ command) will bring it to its minimal value and vice versa. The exceptions are implementations which are distant from the underlying hardware, implementations that use bignums, and implementations that try to enforce portability.
Fortunately, it is usually easy to write brainfuck programs that do not ever cause integer wraparound or overflow. Such programs thus do not depend heavily on cell size. Generally this means avoiding increment of +255 (unsigned char wraparound); or avoiding overstepping the boundaries of [-128, +127] (signed char wraparound). For more details on integer wraparound, see the Integer overflow article.
In the classic distribution, the array has 30,000 cells, and the pointer begins at the leftmost cell. Any brainfuck implementation should thus provide at least that many cells, but surprisingly many implementations provide fewer. Even more cells are needed to store things like the millionth Fibonacci number, and the easiest way to make the language Turing-complete is to make the array unlimited on the right.
A few implementations extend the array to the left as well; this is an uncommon feature, and therefore portable brainfuck programs do not depend on it.
When the pointer moves outside the bounds of the array, some implementations will give an error message, some will try to extend the array dynamically, some will not notice and will produce unpredictable behavior, and a few will move the pointer to the opposite end of the array. Some tradeoffs are involved: expanding the array dynamically to the right is the most user-friendly approach and is good for memory-hungry programs, but it carries a speed penalty. If a fixed-size array is used it is helpful to make it very large, or better yet let the user set the size. Giving an error message for bounds violations is very useful for debugging but even that carries a speed penalty unless it can be handled by the operating system's memory protections.
Different operating systems (and sometimes different programming environments) use subtly different versions of ASCII. The most important difference is in the code used for the end of a line of text. MS-DOS and Microsoft Windows use a CRLF, i.e. a 13 followed by a 10, in most contexts. UNIX and its descendants, including Linux and Mac OS X, use just 10, and older Macs use just 13. It would be unfortunate if brainfuck programs had to be rewritten for different operating systems. Happily, a unified standard is easy to find. Urban Müller's compiler and his example programs use 10, on both input and output; so do a large majority of existing brainfuck programs; and 10 is also more convenient to use than CRLF. Thus, brainfuck implementations should make sure that brainfuck programs that assume newline=10 will run properly; many do so, but some do not.
This assumption is also consistent with most of the world's sample code for C and other languages, in that they use '\n', or 10, for their newlines. On systems that use CRLF line endings, the C standard library transparently remaps "\n" to "\r\n" on output and "\r\n" to "\n" on input for streams not opened in binary mode.
The behavior of the
, command when an end-of-file condition has been encountered varies. Some implementations set the cell at the pointer to 0, some set it to the C constant EOF (in practice this is usually -1), some leave the cell's value unchanged. There is no real consensus; arguments for the three behaviors are as follows.
Setting the cell to 0 avoids the use of negative numbers, and makes it marginally more concise to write a loop that reads characters until EOF occurs. This is a language extension devised by Panu Kalliokoski.
Setting the cell to -1 allows EOF to be distinguished from any byte value (if the cells are larger than bytes), which is necessary for reading non-textual data; also, it is the behavior of the C translation of
, given in Müller's readme file. However, it is not obvious that those C translations are to be taken as normative.
Leaving the cell's value unchanged is the behavior of Urban Müller's brainfuck compiler. This behavior can easily coexist with either of the others; for instance, a program that assumes EOF=0 can set the cell to 0 before each
, command, and will then work correctly on implementations that do either EOF=0 or EOF="no change". It is so easy to accommodate the "no change" behavior that any brainfuck programmer interested in portability should do so.
Many people have modified brainfuck in order to produce their own languages, often by adding commands, occasionally by removing them. Many of the resulting languages are included in the brainfuck derivatives list on the Esoteric Languages wiki. However, there are also unnamed minor variants, formed possibly as a result of inattention, of which some of the more common are:
- forbidding, rather than ignoring, any non-command characters in brainfuck programs
- introducing a comment marker which comments out the rest of the line
- various alterations of the loop semantics, sometimes destroying Turing completeness
- requiring a special character to mark the end of the program