Indian Hill C Style and Coding Standards as amended for U of T Zoology UNIX- L.W. Cannon R.A. Elliott L.W. Kirchhoff J.H. Miller J.M. Milner R.W. Mitze E.P. Schan N.O. Whittington Bell Labs Henry Spencer Zoology Computer Systems University of Toronto ABSTRACT This document is an annotated (by the last author) version of the original paper of the same title. It describes a set of coding standards and recommendations which are local standards for officially-supported UNIX programs. The scope is coding style, not functional organization. April 18, 1990 _________________________ - UNIX is a trademark of Bell Laboratories. 1. Introduction This document is a result of a committee formed at Indian Hill to establish a common set of coding standards and recommendations for the Indian Hill community. The scope of this work is the coding style, not the functional organization of programs. The standards in this document are not specific to ESS programming only. [ In fact, they're pretty good general standards. ``To be clear is professional; not to be clear is unprofessional.'' - Sir Ernest Gowers. This document is presented unadulterated; U of T variations, comments, exceptions, etc. are presented in footnotes. ] { Now, U of T variations are in []'s, while NetHack variations are in {}'s. Otherwise it's just about impossible to read on-line. } We have tried to combine previous work [1,6] on C style into a uniform set of standards that should be appropriate for any project using C. [ Of necessity, these standards cannot cover all situations. Experience and informed judgement count for much. Inexperienced programmers who encounter unusual situations should consult 1) code written by experienced C programmers following these rules, or 2) experienced C programmers. ] { This document applies to the the ``core'' code files and should be at least considered for port-specific code files, although in that case norms for the port's system may also be considered. A couple notes on general layout before getting to the specific layout rules. Each indenting level should be 4 positions, and tabs ( ^I's ) should be multiples of 8 positions. Occasionally tabs may be used for each indenting level, provided this does not cause line wrapping. } 2. File Organization A file consists of various sections that should be separated by several blank lines. Although there is no max- imum length requirement for source files, files with more than about 1500 lines are cumbersome to deal with. The edi- tor may not have enough temp space to edit the file, compi- lations will go slower, etc. Since most of us use 300 baud terminals, entire rows of asterisks, for example, should be discouraged. [ This is not a problem at U of T, or most other sensible places, but rows of asterisks are still annoying. ] Also lines longer than 80 columns are not handled well by all terminals and should be avoided if pos- sible. [ Excessively long lines which result from deep indenting are often a symptom of poorly-organized code. ] { For NetHack, lines should be limited to 79 characters (barring long strings) since even 80 is a nuisance in some situations. A long string may be moved left from its natural indentation to avoid line wrapping. } The suggested order of sections for a file is as fol- lows: 1. Any header file includes should be the first thing in the file. { NetHack files should have a copyright/ license section before any of the others mentioned here. } 2. Immediately after the includes should be a prologue that tells what is in that file. A description of the purpose of the objects in the files (whether they be functions, external data declarations or definitions, or something else) is more useful than a list of the object names. [ A common variation, in both Bell code and ours, is to reverse the order of sections 1 and 2. This is an acceptable practice. ] 3. Any typedefs and defines that apply to the file as a whole are next. 4. Next come the global (external) data declarations. If a set of defines applies to a particular piece of glo- bal data (such as a flags word), the defines should be immediately after the data declaration. [ Such defines should be indented to put the defines one level deeper than the first keyword of the declaration to which they apply. ] { For NetHack, the defines should not be indented if there are a large number of them. They may even go in a separate file if the declaration tells where to find them. } 5. The functions come last. [ They should be in some sort of meaningful order. Top-down is generally better than bottom-up, and a ``breadth-first'' approach (functions on a similar level of abstraction together) is preferred over depth-first (functions defined as soon as possible after their calls). Considerable judgement is called for here. If defining large numbers of essentially-independent utility functions, consider alphabetical order. ] 2.1. File Naming Conventions UNIX requires certain suffix conventions for names of files to be processed by the cc command [5]. [ In addition to the suffix conventions given here, it is conventional to use `Makefile' (not `makefile') for the control file for make and `README' for a summary of the contents of a directory or directory tree. ] The following suffixes are required: o C source file names must end in .c o Assembler source file names must end in .s In addition the following conventions are universally followed: o Relocatable object file names end in .o o Include header file names end in .h or .d [ .h is preferred. An alternate convention that may be preferable in multi-language environments is to use the same suffix as an ordinary source file but with two periods instead of one (e.g. ``foo..c''). ] o Ldp specification file names end in .b [ No idea what this is. ] o Yacc source file names end in .y o Lex source file names end in .l 3. Header Files Header files are files that are included in other files prior to compilation by the C preprocessor. Some are defined at the system level like stdio.h which must be included by any program using the standard I/O library. Header files are also used to contain data declarations and defines that are needed by more than one program. [ Don't use absolute pathnames for header files. Use the construction for getting them from a standard place, or define them relative to the current directory. The -I option of the C compiler is the best way to handle extensive private libraries of header files; it permits reorganizing the directory structure without having to alter source files. ] Header files should be functionally organized, i.e., declarations for separate subsystems should be in separate header files. Also, if a set of declarations is likely to change when code is ported from one machine to another, those declarations should be in a separate header file. Header files should not be nested. Some objects like typedefs and initialized data definitions cannot be seen twice by the compiler in one compilation. On non-UNIX sys- tems this is also true of uninitialized declarations without the extern keyword. [ It should be noted that declaring variables in a header file is often a poor idea. Frequently it is a symptom of poor partitioning of code between files.] This can happen if include files are nested and will cause the compilation to fail. { NetHack header files should use the #ifndef HEADER_H/#define HEADER_H/contents/#endif idiom to bracket their contents, and can then be nested as desired. } 4. External Declarations External declarations should begin in column 1. Each declaration should be on a separate line. A comment describing the role of the object being declared should be included, with the exception that a list of defined con- stants do not need comments if the constant names are suffi- cient documentation. The comments should be tabbed so that they line up underneath each other. [ So should the constant names and their defined values. ] Use the tab character (CTRL I if your terminal doesn't have a separate key) rather than blanks. For structure and union template declarations, each element should be alone on a line with a comment describing it. { Closely related elements may be grouped together on a line if a single comment can easily cover them. } The opening brace ( { ) should be on the same line as the structure tag, and the closing brace should be alone on a line in column 1, i.e. struct boat { int wllength; /* water line length in feet */ int type; /* see below */ long sarea; /* sail area in square feet */ }; /* * defines for boat.type * [ These defines are better put right after the * declaration of type, within the struct declaration, * with enough tabs after # to indent define one level * more than the structure member declarations. ] */ #define KETCH 1 #define YAWL 2 #define SLOOP 3 #define SQRIG 4 #define MOTOR 5 If an external variable is initialized the equal sign should not be omitted. [ Any variable whose initial value is important should be explicitly initialized, or at the very least should be commented to indicate that C's default initialization to 0 is being relied on. The empty initializer, ``{}'', should never be used. Structure initializations should be fully parenthesized with braces. Constants used to initialize longs should be explicitly long. ] int x = 1; char *msg = "message"; struct boat winner = { 40, /* water line length */ YAWL, 600 /* sail area */ }; [ In any file which is part of a larger whole rather than a self-contained program, maximum use should be made of the static keyword to make functions and variables local to single files. Variables in particular should be accessible from other files only when there is a clear need that cannot be filled in another way. Such usages should be commented to make it clear that another file's variables are being used; the comment should name the other file. ] 5. Comments Comments that describe data structures, algorithms, etc., should be in block comment form with the opening /* in column one, a * in column 2 before each line of comment text, and the closing */ in columns 2-3. [ Some automated program-analysis packages use a different character in column 2 as a marker for lines with specific items of information. In particular, a line with a `-' here in a comment preceding a function is sometimes assumed to be a one-line summary of the function's purpose. ] /* * Here is a block comment. * The comment text should be tabbed over19 * and the opening /* and closing star-slash * should be alone on a line. * [ A common practice in both Bell and local code is * to use a space rather than a tab after the *. This * is acceptable. ] */ Note that grep ^.\* will catch all block comments in the file. In some cases, block comments inside a function are appropriate, and they should be tabbed over to the same tab setting as the code that they describe. Short comments may appear on a single line indented over to the tab setting of the code that follows. if (argc > 1) { /* Get input file from command line. */ if (freopen(argv[1], "r", stdin) == (FILE *)0) error("can't open %s\n", argv[1]); } Very short comments may appear on the same line as the code they describe, but should be tabbed over far enough to separate them from the statements. If more than one short comment appears in a block of code they should all be tabbed to the same tab setting. if (a == 2) return(TRUE); /* special case */ else return(isprime(a)); /* works only for odd a */ 6. Function Declarations Each function should be preceded by a block comment prologue that gives the name and a short description of what the function does. [ Discussion of non-trivial design decisions is also appropriate, but avoid duplicating infor- mation that is present in (and clear from) the code. It's too easy for such redundant information to get out of date.] { For NetHack, even the block comment can be simplified when the entire function is considered trivial.} If the function returns a value, the type of the value returned should be alone on a line in column 1 (do not default to int). If the function does not return a value then it should not be given a return type. { Since this is what ``void'' was invented for, use it. } If the value returned requires a long explanation, it should be given in the prologue; otherwise it can be on the same line as the return type, tabbed over. The function name and formal parameters should be alone on a line beginning in column 1. Each parameter should be declared (do not default to int), with a comment on a single line. { These parameter declarations should begin in column 1; tabbing them over is acceptable but not preferred. } The opening brace of the function body should also be alone on a line beginning in column 1. The function name, argument declaration list, and opening brace should be separated by a blank line. [ Nei- ther Bell nor local code has ever included these separating blank lines, and it is not clear that they add anything useful. Leave them out. ] { Unless deemed desirable in a port-specific file where all compilers for the port support ANSI C, all function declarations must be ``old-style''. } All local declarations and code within the function body should be tabbed over at least one tab. If the function uses any external variables, these should have their own declarations in the function body using the extern keyword. If the external variable is an array the array bounds must be repeated in the extern declaration. There should also be extern declarations for all functions called by a given function. This is particu- larly beneficial to someone picking up code written by another. If a function returns a value of type other than int, it is required by the compiler that such functions be declared before they are used. Having the extern declara- tion in the calling function's declarations section avoids all such problems. [ These rules tend to produce a lot of clutter. Both Bell and local practice frequently omits extern declarations for static variables and functions. This is permitted. Omission of declarations for standard library routines is also permissible, although if they are declared it is better to declare them within the functions that use them rather than globally. ] { All external NetHack functions should be declared in extern.h, widely- used system functions in system.h, and most widely-used global NetHack variables in decl.h. Explicit extern declar- ations elsewhere should limited to things not widely-used. } In general each variable declaration should be on a separate line with a comment describing the role played by the variable in the function. If the variable is external or a parameter of type pointer which is changed by the func- tion, that should be noted in the comment. All such com- ments for parameters and local variables should be tabbed so that they line up underneath each other. The declarations should be separated from the function's statements by a blank line. A local variable should not be redeclared in nested blocks. [ In fact, avoid any local declarations that over- ride declarations at higher levels. ] Even though this is valid C, the potential confusion is enough that lint will complain about it when given the -h option. 6.1. Examples /* * skyblue() * * Determine if the sky is blue. */ int /* TRUE or FALSE */ skyblue() { extern int hour; if (hour < MORNING || hour > EVENING) return(FALSE); /* black */ else return(TRUE); /* blue */ } /* * tail(nodep) * * Find the last element in the linked list * pointed to by nodep and return a pointer to it. */ NODE * /* pointer to tail of list */ tail(nodep) NODE *nodep; /* pointer to head of list */ { register NODE *np; /* current pointer advances to NULL */ register NODE *lp; /* last pointer follows np */ np = lp = nodep; while ((np = np->next) != (NODE*) 0) lp = np; return(lp); } 7. Compound Statements Compound statements are statements that contain lists of statements enclosed in braces. The enclosed list should be tabbed over one more than the tab position of the com- pound statement itself. The opening left brace should be at the end of the line beginning the compound statement and the closing right brace should be alone on a line, tabbed under the beginning of the compound statement. Note that the left brace beginning a function body is the only occurrence of a left brace which is alone on a line. { The case and default keywords may be indented further within a switch statement.} 7.1. Examples if (expr) { statement; statement; } if (expr) { statement; statement; } else { statement; statement; } Note that the right brace before the else and the right brace before the while of a do-while statement (below) are the only places where a right braces appears that is not alone on a line. for (i = 0; i < MAX; i++) { statement; statement; } while (expr) { statement; statement; } do { statement; statement; } while (expr); switch (expr) { case ABC: case DEF: statement; break; case XYZ: statement; break; default: statement; break24; } [ The last break is, strictly speaking, unnecessary, but it is required nonetheless because it prevents a fall-through error if another case is added later after the last one. ] Note that when multiple case labels are used, they are placed on separate lines. The fall through feature of the C switch statement should rarely if ever be used when code is executed before falling through to the next one. If this is done it must be commented for future maintenance. { Falling though is used more widely in NetHack, but it should still be commented if the code is split as described above. } if (strcmp(reply, "yes") == EQUAL) { statements for yes ... } else if (strcmp(reply, "no") == EQUAL) { statements for no ... } else if (strcmp(reply, "maybe") == EQUAL) { statements for maybe ... } else { statements for none of the above ... } The last example is a generalized switch statement and the tabbing reflects the switch between exactly one of several alternatives rather than a nesting of statements. 8. Expressions 8.1. Operators The old versions of equal-ops =+, =-, =*, etc. should not be used. The preferred use is +=, -=, *=, etc. All binary operators except . and -> should be separated from their operands by blanks. [ Some judgement is called for in the case of complex expressions, which may be clearer if the ``inner'' operators are not surrounded by spaces and the ``outer'' ones are. ] In addition, keywords that are followed by expressions in parentheses should be separated from the left parenthesis by a blank. [ Sizeof is an exception, see the discussion of function calls. Less logically, so is return. ] Blanks should also appear after commas in argument lists to help separate the arguments visually. { Dropping the spaces after keywords and between arguments is acceptable, especially if it avoids line wrapping. } On the other hand, macros with arguments and function calls should not have a blank between the name and the left parenthesis. In particular, the C preprocessor requires the left parenthesis to be immediately after the macro name or else the argument list will not be recognized. Unary operators should not be separated from their single operand. Since C has some unexpected precedence rules, all expressions involving mixed operators should be fully parenthesized. Examples a += c + d; a = (a + b) / (c * d); strp->field = str.fl - ((x & MASK) >> DISP); while (*d++ = *s++) ; /* EMPTY BODY */ 8.2. Naming Conventions Individual projects will no doubt have their own naming conventions. There are some general rules however. o An initial underscore should not be used for any user- created names. [ Trailing underscores should be avoided too. ] UNIX uses it for names that the user should not have to know (like the standard I/O library). [ This convention is reserved for system purposes. If you must have your own private identi- fiers, begin them with a capital letter identifying the package to which they belong. ] o Macro names, typedef names, and define names should be all in CAPS. { NetHack typedef names tend to be lowercase. } o Variable names, structure tag names, and function names should be in lower case. [ It is best to avoid names that differ only in case, like foo and FOO. The potential for confusion is considerable. ] { Some systems do not distinguish case in function names, so especially avoid using both foo() and Foo(). } Some macros (such as getchar and putchar) are in lower case since they may also exist as functions. Care is needed when interchanging macros and functions since functions pass their parameters by value whereas macros pass their arguments by name substitution. [ This differ- ence also means that carefree use of macros requires care when they are defined. Remember that complex expressions can be used as parameters, and operator- precedence problems can arise unless all occurrences of parameters in the definition have parentheses around them. There is little that can be done about the problems caused by side effects in parameters except to avoid side effects in expressions (a good idea anyway).] 8.3. Constants Numerical constants should not be coded directly. [ At the very least, any directly-coded numerical constant must have a comment explaining the derivation of the value. ] { There are, however, a number of NetHack constants that have no particular derivation, being numbers picked out of the air for a particular piece of code, often indicating the probability of something. As long as these constants do not occur outside the limited bit of code, don't bother forcing a name on something that doesn't want one. } The define feature of the C preprocessor should be used to assign a meaningful name. This will also make it easier to admin- ister large programs since the constant value can be changed uniformly by changing only the define. The enumeration data type is the preferred way to handle situations where a variable takes on only a discrete set of values, since additional type checking is available through lint. { However, some older compilers cannot handle enumeration types, so it's back to defined constants. } There are some cases where the constants 0 and 1 may appear as themselves instead of as defines. For example if a for loop indexes through an array, then for (i = 0; i < ARYBOUND; i++) is reasonable, as is fptr = fopen(filename, "r"); if (fptr == (FILE*)0) error("can't open %s\n", filename); In the last example, although the defined constant NULL is available as part of the most versions of the standard I/O library's header file, stdio.h, its type is uncertain and it requires considerable thought to decide when an uncast NULL is safe for all allowed (and disallowed, but occurring!) definitions of NULL and all degrees of prototyping. Since a cast is sometimes necessary even with NULL, just explicitly cast 0 to the pointer type in each use. { The wording of this paragraph has changed from the original to suit the needs of NetHack. } { The rest of this document is not binding on NetHack. One added point, though -- #ifdef/#else/#endif sets should have a single space inserted between the `#' and the first letter for each ifdef nesting level (not counting OVLx, or HEADER_H in header files. #else and #endif lines should be commented with a reminder of their condition, unless the conditional section is very short. } 9. Portability The advantages of portable code are well known. This section gives some guidelines for writing portable code, where the definition of portable is taken to mean that a source file contains portable code if it can be compiled and executed on different machines with the only source change being the inclusion of possibly different header files. The header files will contain defines and typedefs that may vary from machine to machine. Reference [1] contains useful information on both style and portability. Many of the recommendations in this document originated in [1]. The following is a list of pitfalls to be avoided and recommen- dations to be considered when designing portable code: o First, one must recognize that some things are inherently non-portable. Examples are code to deal with particular hardware registers such as the program status word, and code that is designed to support a particular piece of hardware such as an assembler or I/O driver. Even in these cases there are many rou- tines and data organizations that can be made machine independent. It is suggested that source file be organized so that the machine-independent code and the machine-dependent code are in separate files. Then if the program is to be moved to a new machine, it is a much easier task to determine what needs to be changed. [ If you #ifdef dependencies, make sure that if no machine is specified, the result is a syntax error, not a default machine! ] It is also possible that code in the machine-independent files may have uses in other programs as well. o Pay attention to word sizes. The following sizes apply to basic types in C for the machines that will be used most at IH: [ The 3B is a Bell Labs machine. The VAX, not shown in the table, is similar to the 3B in these respects. The 68000 resembles either the pdp11 or the 3B, depending on the particular compiler. ] type pdp11 3B IBM ________________________ char 8 8 8 short 16 16 16 int 16 32 32 long 32 32 32 In general if the word size is important, short or long should be used to get 16 or 32 bit items on any of the above machines. [ Any unsigned type other than plain unsigned int should be typedefed, as such types are highly compiler-dependent. This is also true of long and short types other than long int and short int. Large programs should have a central header file which supplies typedefs for commonly-used width-sensitive types, to make it easier to change them and to aid in finding width-sensitive code. ] If a simple loop counter is being used where either 16 or 32 bits will do, then use int, since it will get the most efficient (natural) unit for the current machine. [ Beware of making assumptions about the size of pointers. They are not always the same size as int. Nor are all pointers always the same size, or freely intercon- vertible. Pointer-to-character is a particular trouble spot on machines which do not address to the byte. ] o Word size also affects shifts and masks. The code x &= 0177770 will clear only the three rightmost bits of an int on a PDP11. On a 3B it will also clear the entire upper halfword. Use x &= ~07 instead which works properly on all machines. [ The or operator ( | ) does not have these problems, nor do bitfields (which, unfortunately, are not very portable due to defective compilers). ] o Code that takes advantage of the two's complement representation of numbers on most machines should not be used. Optimizations that replace arithmetic opera- tions with equivalent shifting operations are particu- larly suspect. You should weigh the time savings with the potential for obscure and difficult bugs when your code is moved, say, from a 3B to a 1A. o Watch out for signed characters. On the PDP-11, char- acters are sign extended when used in expressions, which is not the case on any other machine. In partic- ular, getchar is an integer-valued function (or macro) since the value of EOF for the standard I/O library is -1, which is not possible for a character on the 3B or IBM. [ Actually, this is not quite the real reason why getchar returns int, but the comment is valid: code which assumes either that characters are signed or that they are unsigned is unportable. It is best to completely avoid using char to hold numbers. Manip- ulation of characters as if they were numbers is also often unportable. ] o The PDP-11 is unique among processors on which C exists in that the bytes are numbered from right to left within a word. All other machines (3B, IBM, Interdata 8/32, Honeywell) number the bytes from left to right. [ Actually, there are some more right-to-left machines now, but the comments still apply. ] Hence any code that depends on the left-right orientation of bits in a word deserves special scrutiny. Bitfields within structure members will only be portable so long as two separate fields are never concatenated and treated as a unit. [1,3] [ The same applies to variables in general. Alignment considerations and loader peculiarities make it very rash to assume that two consecutively-declared variables are together in memory, or that a variable of one type is aligned appropriately to be used as another type. ] o Do not default the boolean test for non-zero, i.e. if (f() != FAIL) is better than if (f()) even though FAIL may have the value 0 which is con- sidered to mean false by C. [ A particularly notorious case is using strcmp to test for string equality, where the result should never ever be defaulted. The preferred approach is to define a macro STREQ: #define STREQ(a, b) (strcmp((a), (b)) == 0) ] This will help you out later when somebody decides that a failure return should be -1 instead of 0. [ An exception is commonly made for predicates, which are functions which meet the following restrictions: o Has no other purpose than to return true or false. o Returns 0 for false, 1 for true, nothing else. o Is named so that the meaning of (say) a `true' return is absolutely obvious. Call a predicate isvalid or valid, not checkvalid. ] o Be suspicious of numeric values appearing in the code. Even simple values like 0 or 1 could be better expressed using defines like FALSE and TRUE (see previ- ous item). [ Actually, YES and NO often read better. ] Any other constants appearing in a program would be better expressed as a defined constant. This makes it easier to change and also easier to read. o Become familiar with existing library functions and defines. [ But not too familiar. The internal details of library facilities, as opposed to their external interfaces, are subject to change without warning. They are also often quite unportable. ] You should not be writing your own string compare routine, or making your own defines for system structures. [ Or, especially, writing your own code to control terminals. Use the termcap package. ] Not only does this waste your time, but it prevents your program from taking advantage of any microcode assists or other means of improving performance of system routines. [ It also makes your code less readable, because the reader has to figure out whether you're doing something special in that reimplemented stuff to justify its existence. Furthermore, it's a fruitful source of bugs. ] o Use lint. It is a valuable tool for finding machine- dependent constructs as well as other inconsistencies or program bugs that pass the compiler. [ The use of lint on all programs is strongly recommended. It is difficult to eliminate complaints about functions whose return value is not used (in the current version of C, at least), but most other messages from lint really do indicate something wrong. The -h, -p, -a, -x, and -c options are worth learning. All of them will complain about some legitimate things, but they will also pick up many botches. Note that -p checks function-call type-consistency for only a subset of Unix library routines, so programs should be linted both with and without this option for best ``coverage''. ] 10. Lint Lint is a C program checker [2] that examines C source files to detect and report type incompatibilities, incon- sistencies between function definitions and calls, potential program bugs, etc. It is expected that projects will require programs to use lint as part of the official accep- tance procedure. [ Yes. ] In addition, work is going on in department 5521 to modify lint so that it will check for adherence to the standards in this document. It is still too early to say exactly which of the standards given here will be checked by lint. In some cases such as whether a comment is misleading or incorrect there is little hope of mechanical checking. In other cases such as checking that the opening brace of a function body is alone on a line in column 1, the test has already been added. [ Little of this is relevant at U of T. The version of lint that we have lacks these mods. ] Future bulletins will be used to announce new additions to lint as they occur. It should be noted that the best way to use lint is not as a barrier that must be overcome before official accep- tance of a program, but rather as a tool to use whenever major changes or additions to the code have been made. Lint can find obscure bugs and insure portability before problems occur. 11. Special Considerations This section contains some miscellaneous do's and don'ts. o Don't change syntax via macro substitution. It makes the program unintelligible to all but the perpetrator. o There is a time and a place for embedded assignment statements. [ The ++ and -- operators count as assign- ment statements. So, for many purposes, do functions with side effects. ] In some constructs there is no better way to accomplish the results without making the code bulkier and less readable. The while loop in section 8.1 is one example of an appropriate place. Another is the common code segment: while ((c = getchar()) != EOF) { process the character } Using embedded assignment statements to improve run- time performance is also possible. However, one should consider the tradeoff between increased speed and decreased maintainability that results when embedded assignments are used in artificial places. For exam- ple, the code: a = b + c; d = a + r; should not be replaced by d = (a = b + c) + r; even though the latter may save one cycle. Note that in the long run the time difference between the two will decrease as the optimizer gains maturity, while the difference in ease of maintenance will increase as the human memory of what's going on in the latter piece of code begins to fade. [ Note also that side effects within expressions can result in code whose semantics are compiler-dependent, since C's order of evaluation is explicitly undefined in most places. Compilers do differ. ] o There is also a time and place for the ternary ? : operator and the binary comma operator. The logical expression operand before the ? : should be parenthesized: (x >= 0) ? x : -x Nested ? : operators can be confusing and should be avoided if possible. There are some macros like getchar where they can be useful. The comma operator can also be useful in for statements to provide multi- ple initializations or incrementations. o Goto statements should be used sparingly as in any well-structured code. [ The continue statement is almost as bad. Break is less troublesome. ] The main place where they can be usefully employed is to break out of several levels of switch, for, and while nesting, e.g. for (...) for (...) { ... if (disaster) goto error; } ... error: clean up the mess [ The need to do such a thing may indicate that the inner constructs should be broken out into a separate function, with a success/failure return code. ] When a goto is necessary the accompanying label should be alone on a line and tabbed one tab position to the left of the associated code that follows. o This committee recommends that programmers not rely on automatic beautifiers for the following reasons. First, the main person who benefits from good program style is the programmer himself. This is especially true in the early design of handwritten algorithms or pseudo-code. Automatic beautifiers can only be applied to complete, syntactically correct programs and hence are not available when the need for attention to white space and indentation is greatest. It is also felt that programmers can do a better job of making clear the complete visual layout of a function or file, with the normal attention to detail of a careful program- mer. [ In other words, some of the visual layout is dictated by intent rather than syntax. Beautifiers cannot read minds. ] Sloppy programmers should learn to be careful programmers instead of relying on a beau- tifier to make their code readable. Finally, it is felt that since beautifiers are non-trivial programs that must parse the source, the burden of maintaining them in the face of the continuing evolution of C is not worth the benefits gained by such a program. 12. Project Dependent Standards Individual projects may wish to establish additional standards beyond those given here. The following issues are some of those that should be adddressed by each project pro- gram administration group. o What additional naming conventions should be followed? In particular, systematic prefix conventions for func- tional grouping of global data and also for structure or union member names can be useful. o What kind of include file organization is appropriate for the project's particular data hierarchy? o What procedures should be established for reviewing lint complaints? A tolerance level needs to be esta- blished in concert with the lint options to prevent unimportant complaints from hiding complaints about real bugs or inconsistencies. o If a project establishes its own archive libraries, it should plan on supplying a lint library file [2] to the system administrators. This will allow lint to check for compatible use of library functions. 13. Conclusion A set of standards has been presented for C programming style. One of the most important points is the proper use of white space and comments so that the structure of the program is evident from the layout of the code. Another good idea to keep in mind when writing code is that it is likely that you or someone else will be asked to modify it or make it run on a different machine sometime in the future. As with any standard, it must be followed if it is to be useful. The Indian Hill version of lint will enforce those standards that are amenable to automatic checking. If you have trouble following any of these standards don't just ignore them. Programmers at Indian Hill should bring their problems to the Software Development System Group (Lee Kirchhoff, contact) in department 5522. Programmers outside Indian Hill should contact the Processor Application Group (Layne Cannon, contact) in department 5512. [ At U of T Zoology, it's Henry Spencer in 336B. ] References [1] B.A. Tague, "C Language Portability", Sept 22, 1977. This document issued by department 8234 contains three memos by R.C. Haight, A.L. Glasser, and T.L. Lyon deal- ing with style and portability. [2] S.C. Johnson, "Lint, a C Program Checker", Technical Memorandum, 77-1273-14, September 16, 1977. [3] R.W. Mitze, "The 3B/PDP-11 Swabbing Problem", Memoran- dum for File, 1273-770907.01MF, September 14, 1977. [4] R.A. Elliott and D.C. Pfeffer, "3B Processor Common Diagnostic Standards- Version 1", Memorandum for File, 5514-780330.01MF, March 30, 1978. [5] R.W. Mitze, "An Overview of C Compilation of UNIX User Processes on the 3B", Memorandum for File, 5521- 780329.02MF, March 29, 1978. [6] B.W. Kernighan and D.M. Ritchie, The C Programming Language, Prentice-Hall 1978. /* * The C Style Summary Sheet Block comment, * by Henry Spencer, U of T Zoology describes file. */ #include Headers; don't nest. typedef int SEQNO; /* ... */ Global definitions. #define STREQ(a, b) (strcmp((a), (b)) == 0) static char *foo = (char *)0; /* ... */ Global declarations. struct bar { Static whenever poss. SEQNO alpha; /* ... */ # define NOSEQNO 0 int beta; /* ... */ Don't assume 16 bits. }; /* * Many unnecessary braces, to show where. Functions. */ static int /* what is returned */ Don't default int. bletch(a) int a; /* ... */ Don't default int. { int bar; /* ... */ extern int errno; /* ..., changed here */ extern char *index(); if (foobar() != FAIL) { if (!isvalid()) { return(OK); errno = ERANGE; } } else { x = &y + z->field; while (x == (y & MASK)) { } f += (x >= 0) ? x : -x; } for (i = 0; i < BOUND; i++) { /* lint -h[p]cax. */ do { } /* Avoid nesting ?: */ } while (index(a, b) != (char*)0); if (STREQ(x, "foo")) { x |= 07; /* 07 is... */ switch (...) { } else if (STREQ(x, "bar")) { case ABC: x &= ~077; /* 077 is... */ case DEF: } else if (STREQ(x, "ugh")) { printf("...", a, b); /* Avoid gotos */ break; } else { case XYZ: /* and continues. */ x = y; } /* FALLTHROUGH */ default: while ((c = getc()) != EOF) /* Limit imbedded =s. */ ; /* NULLBODY */ break; } }