4 This document describes the coding style conventions used in Busybox. If you
5 add a new file to Busybox or are editing an existing file, please format your
6 code according to this style. If you are the maintainer of a file that does
7 not follow these guidelines, please -- at your own convenience -- modify the
8 file(s) you maintain to bring them into conformance with this style guide.
9 Please note that this is a low priority task.
11 To help you format the whitespace of your programs, an ".indent.pro" file is
12 included in the main Busybox source directory that contains option flags to
13 format code as per this style guide. This way you can run GNU indent on your
14 files by typing 'indent myfile.c myfile.h' and it will magically apply all the
15 right formatting rules to your file. Please _do_not_ run this on all the files
16 in the directory, just your own.
23 Here is the order in which code should be laid out in a file:
25 - commented program name and one-line description
26 - commented author name and email address(es)
27 - commented GPL boilerplate
28 - commented longer description / notes for the program (if needed)
29 - #includes of .h files with angle brackets (<>) around them
30 - #includes of .h files with quotes ("") around them
31 - #defines (if any, note the section below titled "Avoid the Preprocessor")
32 - const and global variables
33 - function declarations (if necessary)
34 - function implementations
38 Whitespace and Formatting
39 -------------------------
41 This is everybody's favorite flame topic so let's get it out of the way right
45 Tabs vs. Spaces in Line Indentation
46 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
48 The preference in Busybox is to indent lines with tabs. Do not indent lines
49 with spaces and do not indents lines using a mixture of tabs and spaces. (The
50 indentation style in the Apache and Postfix source does this sort of thing:
51 \s\s\s\sif (expr) {\n\tstmt; --ick.) The only exception to this rule is
52 multi-line comments that use an asterisk at the beginning of each line, i.e.:
55 /t * This is a block comment.
56 /t * Note that it has multiple lines
57 /t * and that the beginning of each line has a tab plus a space
58 /t * except for the opening '/*' line where the slash
59 /t * is used instead of a space.
62 Furthermore, The preference is that tabs be set to display at four spaces
63 wide, but the beauty of using only tabs (and not spaces) at the beginning of
64 lines is that you can set your editor to display tabs at *whatever* number of
65 spaces is desired and the code will still look fine.
71 Put spaces between terms and operators. Example:
75 for(i=0;i<num_items;i++){
79 for (i = 0; i < num_items; i++) {
81 While it extends the line a bit longer, the spaced version is more
82 readable. An allowable exception to this rule is the situation where
83 excluding the spacing makes it more obvious that we are dealing with a
84 single term (even if it is a compound term) such as:
86 if (str[idx] == '/' && str[idx-1] != '\\')
90 if ((argc-1) - (optind+1) > 0)
96 If an opening bracket starts a function, it should be on the
97 next line with no spacing before it. However, if a bracket follows an opening
98 control block, it should be on the same line with a single space (not a tab)
99 between it and the opening control block statement. Examples:
109 Don't do this either:
115 And for heaven's sake, don't do this:
130 Spacing around Parentheses
131 ~~~~~~~~~~~~~~~~~~~~~~~~~~
133 Put a space between C keywords and left parens, but not between
134 function names and the left paren that starts it's parameter list (whether it
135 is being declared or called). Examples:
140 for(i = 0; i < n; i++) {
145 for (i = 0; i < n; i++) {
147 But do functions like this:
149 static int my_func(int foo, char bar)
153 Also, don't put a space between the left paren and the first term, nor between
154 the last arg and the right paren.
159 strcmp( thisstr, thatstr )
164 strcmp(thisstr, thatstr)
170 Also, please "cuddle" your else statements by putting the else keyword on the
171 same line after the right bracket that closes an 'if' statement.
190 The exception to this rule is if you want to include a comment before the else
196 /* otherwise, we're just kidding ourselves, so re-frob the input */
204 Variable and Function Names
205 ---------------------------
207 Use the K&R style with names in all lower-case and underscores occasionally
208 used to separate words (e.g., "variable_name" and "numchars" are both
209 acceptable). Using underscores makes variable and function names more readable
210 because it looks like whitespace; using lower-case is easy on the eyes.
228 - Enums, macros, and constant variables should all be in upper-case with
229 words optionally seperatedy by underscores (i.e. FIFOTYPE, ISBLKDEV()).
231 - Nobody is going to get mad at you for using 'pvar' as the name of a
232 variable that is a pointer to 'var'.
234 Note: The Busybox codebase is very much a mixture of code gathered from a
235 variety of sources. This explains why the current codebase contains such a
236 hodge-podge of different naming styles (Java, Pascal, K&R, just-plain-weird,
237 etc.). The K&R guideline explained above should therefore be used on new files
238 that are added to the repository. Furthermore, the maintainer of an existing
239 file that uses alternate naming conventions should -- at his own convenience
240 -- convert those names over to K&R style; converting variable names is a very
241 low priority task. Perhaps in the future we will include some magical Perl
242 script that can go through and convert variable names, left as an exercise for
245 For the time being, if you want to do a search-and-replace of a variable name
246 in different files, do the following in the busybox directory:
248 $ perl -pi -e 's/\bOldVar\b/new_var/g' *.[ch]
252 Avoid The Preprocessor
253 ----------------------
255 At best, the preprocessor is a necessary evil, helping us account for platform
256 and architecture differences. Using the preprocessor unnecessarily is just
263 Use 'const <type> var' for declaring constants.
269 Do this instead, when the variable is in a header file and will be used in
270 several source files:
274 Or do this when the variable is used only in a single source file:
276 static const int var = 80;
278 Declaring variables as '[static] const' gives variables an actual type and
279 makes the compiler do type checking for you; the preprocessor does _no_ type
280 checking whatsoever, making it much more error prone. Declaring variables with
281 '[static] const' also makes debugging programs much easier since the value of
282 the variable can be easily queried and displayed.
288 Use 'static inline' instead of a macro.
292 #define mini_func(param1, param2) (param1 << param2)
296 static inline int mini_func(int param1, param2)
298 return (param1 << param2);
301 Static inline functions are greatly preferred over macros. They provide type
302 safety, have no length limitations, no formatting limitations, and under gcc
303 they are as cheap as macros. Besides, really long macros with backslashes at
304 the end of each line are ugly as sin.
310 Code cluttered with ifdefs is difficult to read and maintain. Don't do it.
311 Instead, put your ifdefs in a header, and conditionally define 'static inline'
312 functions, (or *maybe* macros), which are used in the code.
316 ret = my_func(bar, baz);
319 #ifdef BB_FEATURE_FUNKY
320 maybe_do_funky_stuff(bar, baz);
327 #ifdef BB_FEATURE_FUNKY
328 static inline void maybe_do_funky_stuff (int bar, int baz)
330 /* lotsa code in here */
333 static inline void maybe_do_funky_stuff (int bar, int baz) {}
336 (in the .c source file)
338 ret = my_func(bar, baz);
341 maybe_do_funky_stuff(bar, baz);
343 The great thing about this approach is that the compiler will optimize away
344 the "no-op" case (the empty function) when the feature is turned off.
346 Note also the use of the word 'maybe' in the function name to indicate
347 conditional execution.
354 Strings in C can get a little thorny. Here's some guidelines for dealing with
355 strings in Busybox. (There is surely more that could be added to this
362 Put all help/usage messages in usage.c. Put other strings in messages.c.
363 Putting these strings into their own file is a calculated decision designed to
364 confine spelling errors to a single place and aid internationalization
365 efforts, if needed. (Side Note: we might want to use a single file - maybe
366 called 'strings.c' - instead of two, food for thought).
369 Testing String Equivalence
370 ~~~~~~~~~~~~~~~~~~~~~~~~~~
372 There's a right way and a wrong way to test for sting equivalence with
377 if (!strcmp(string, "foo")) {
382 if (strcmp(string, "foo") == 0){
385 The use of the "equals" (==) operator in the latter example makes it much more
386 obvious that you are testing for equivalence. The former example with the
387 "not" (!) operator makes it look like you are testing for an error. In a more
388 perfect world, we would have a streq() function in the string library, but
389 that ain't the world we're living in.
392 Avoid Dangerous String Functions
393 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
395 Unfortunately, the way C handles strings makes them prone to overruns when
396 certain library functions are (mis)used. The following table offers a summary
397 of some of the more notorious troublemakers:
399 function overflows preferred
400 ----------------------------------------
401 strcpy dest string strncpy
402 strcat dest string strncat
403 gets string it gets fgets
404 getwd buf string getcwd
405 [v]sprintf str buffer [v]snprintf
406 realpath path buffer use with pathconf
407 [vf]scanf its arguments just avoid it
410 The above is by no means a complete list. Be careful out there.
414 Avoid Big Static Buffers
415 ------------------------
417 First, some background to put this discussion in context: Static buffers look
420 /* in a .c file outside any functions */
421 static char *buffer[BUFSIZ]; /* happily used by any function in this file,
424 The problem with these is that any time any busybox app is run, you pay a
425 memory penalty for this buffer, even if the applet that uses said buffer is
426 not run. This can be fixed, thusly:
432 strcpy(buffer, lotsa_chars); /* happily uses global *buffer */
436 buffer = xmalloc(sizeof(char)*BUFSIZ);
439 However, this approach trades bss segment for text segment. Rather than
440 mallocing the buffers (and thus growing the text size), buffers can be
441 declared on the stack in the *_main() function and made available globally by
442 assigning them to a global pointer thusly:
444 static char *pbuffer;
448 strcpy(pbuffer, lotsa_chars); /* happily uses global *pbuffer */
452 char *buffer[BUFSIZ]; /* declared locally, on stack */
453 pbuffer = buffer; /* but available globally */
456 This last approach has some advantages (low code size, space not used until
457 it's needed), but can be a problem in some low resource machines that have
458 very limited stack space (e.g., uCLinux).
460 A macro is declared in busybox.h that implements compile-time selection
461 between xmalloc() and stack creation, so you can code the line in question as
463 RESERVE_BB_BUFFER(buffer, BUFSIZ);
465 and the right thing will happen, based on your configuration.
469 Miscellaneous Coding Guidelines
470 -------------------------------
472 The following are important items that don't fit into any of the above
476 Model Busybox Applets After GNU Counterparts
477 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
479 When in doubt about the proper behavior of a Busybox program (output,
480 formatting, options, etc.), model it after the equivalent GNU program.
481 Doesn't matter how that program behaves on some other flavor of *NIX; doesn't
482 matter what the POSIX standard says or doesn't say, just model Busybox
483 programs after their GNU counterparts and nobody has to get hurt.
485 The only time we deviate from emulating the GNU behavior is when:
487 - We are deliberately not supporting a feature (such as a command line
489 - Emulating the GNU behavior is prohibitively expensive (lots more code
490 would be required, lots more memory would be used, etc.)
491 - The difference is minor or cosmetic
493 A note on the 'cosmetic' case: Output differences might be considered
494 cosmetic, but if the output is significant enough to break other scripts that
495 use the output, it should really be fixed.
501 If a const variable is used only in a single source file, put it in the source
502 file and not in a header file. Likewise, if a const variable is used in only
503 one function, do not make it global to the file. Instead, declare it inside
504 the function body. Bottom line: Make a conscious effort to limit declarations
505 to the smallest scope possible.
507 Inside applet files, all functions should be declared static so as to keep the
508 global name space clean. The only exception to this rule is the "applet_main"
509 function which must be declared extern.
511 If you write a function that performs a task that could be useful outside the
512 immediate file, turn it into a general-purpose function with no ties to any
513 applet and put it in the utility.c file instead.
516 Brackets Are Your Friends
517 ~~~~~~~~~~~~~~~~~~~~~~~~~
519 Please use brackets on all if and else statements, even if it is only one
537 The "bracketless" approach is error prone because someday you might add a line
546 And the resulting behavior of your program would totally bewilder you. (Don't
547 laugh, it happens to us all.) Remember folks, this is C, not Python.
550 Function Declarations
551 ~~~~~~~~~~~~~~~~~~~~~
553 Do not use old-style function declarations that declare variable types between
554 the parameter list and opening bracket. Example:
558 int foo(parm1, parm2)
566 int foo(char parm1, float parm2)
570 The only time you would ever need to use the old declaration syntax is to
571 support ancient, antediluvian compilers. To our good fortune, we have access
572 to more modern compilers and the old declaration syntax is neither necessary
576 Emphasizing Logical Blocks
577 ~~~~~~~~~~~~~~~~~~~~~~~~~~
579 Organization and readability are improved by putting extra newlines around
580 blocks of code that perform a single task. These are typically blocks that
581 begin with a C keyword, but not always.
583 Furthermore, you should put a single comment (not necessarily one line, just
584 one comment) before the block, rather than commenting each and every line.
585 There is an optimal ammount of commenting that a program can have; you can
586 comment too much as well as too little.
588 A picture is really worth a thousand words here, so here is an example that
589 illustrates emphasizing logical blocks:
591 while (line = get_line_from_file(fp)) {
593 /* eat the newline, if any */
594 if (line[strlen(line)-1] == '\n') {
595 line[strlen(line)-1] = '\0';
598 /* ignore blank lines */
599 if (strlen(file_to_act_on) == 0) {
603 /* if the search string is in this line, print it,
604 * unless we were told to be quiet */
605 if (strstr(line, search) && !be_quiet) {
614 Processing Options with getopt
615 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
617 If your applet needs to process command-line switches, please use getopt() to
618 do so. Numerous examples can be seen in many of the existing applets, but
619 basically it boils down to two things: at the top of the .c file, have this
620 line in the midst of your #includes:
624 And a code block similar to the following near the top of your applet_main()
627 while ((opt = getopt(argc, argv, "abc")) > 0) {
639 show_usage(); /* in utility.c */
643 If your applet takes no options (such as 'init'), there should be a line
644 somewhere in the file reads:
646 /* no options, no getopt */
648 That way, when people go grepping to see which applets need to be converted to
649 use getopt, they won't get false positives.
651 Additional Note: Do not use the getopt_long library function and do not try to
652 hand-roll your own long option parsing. Busybox applets should only support
653 short options, plus explanations and examples in usage.h.