According to me, it is zero but there seems to be bit confusion here
I have tested it with gcc compiler and it gives me zero as output. I know that in C++, size of an empty class is 1. Let me know if I am missing anything here.
+26
A:
A struct cannot be empty in C because the syntax forbids it. Furthermore, there is a semantic constraint that makes behavior undefined if a struct has no named member:
struct-or-union-specifier:
struct-or-union identifieropt { struct-declaration-list }
struct-or-union identifier
struct-or-union:
struct
union
struct-declaration-list:
struct-declaration
struct-declaration-list struct-declaration
struct-declaration:
specifier-qualifier-list struct-declarator-list ;
/* type-specifier or qualifier required here! */
specifier-qualifier-list:
type-specifier specifier-qualifier-listopt
type-qualifier specifier-qualifier-listopt
struct-declarator-list:
struct-declarator
struct-declarator-list , struct-declarator
struct-declarator:
declarator
declaratoropt : constant-expression
If you write
struct identifier { };
It will give you a diagnostic message, because you violate syntactic rules. If you write
struct identifier { int : 0; };
Then you have a non-empty struct with no named members, thus making behavior undefined, and not requiring a diagnostic:
If the struct-declaration-list contains no named members, the behavior is undefined.
Notice that the following is disallowed because a flexible array member cannot be the first member:
struct identifier { type ident[]; };
Johannes Schaub - litb
2009-10-26 18:28:25
but following code works perfectly with gcc compiler and it prints zero ` struct abc { }; printf("size of empty struct %d\n", sizeof(struct abc));'
vinit dhatrak
2009-10-26 18:35:40
compiling with `-ansi -pedantic` gives "main.c:2: warning: struct has no members"
Johannes Schaub - litb
2009-10-26 18:40:47
Empty structs are a GCC extension.
Michael Burr
2009-10-26 18:41:17
@litb gr8 input !!!
vinit dhatrak
2009-10-26 18:45:36
I don't believe the C standard says a diagnostic is required if you violate a rule contained only in the BNF. For the most part, a diagnostic is required only if a "shall" or "shall not", contained with a "constraints" clause, is violated. In C++, the situation is slightly different (it specifies that "The set of *diagnosable rules* consists of all syntactic and semantic rules in this International Standard..."
Jerry Coffin
2009-10-26 18:50:02
C99 at least says "A conforming implementation shall produce at least one diagnostic message (identified in an implementation-defined manner) if a preprocessing translation unit or translation unit contains a violation of any syntax rule or constraint, ..."
Johannes Schaub - litb
2009-10-26 18:51:28
Congrats on maxing your rep out on one answer... :P
GMan
2009-10-26 18:55:20
"With the gcc compiler" being the operative words; this is a gcc-specific extension (of questionable value IMO). Visual Studio rightly chokes on it.
John Bode
2009-10-26 19:03:39
Oh, thanks so much. I appreciate that you enjoy the rep festival with me.
Johannes Schaub - litb
2009-10-26 19:05:45
+3
A:
In C99: "If the struct-declaration-list contains no named members, the behavior is undefined."
The syntax doesn't really allow it anyway, though I don't see anything that says a diagnostic is required, which puts it pretty much back in the "undefined behavior" camp.
Jerry Coffin
2009-10-26 18:32:11
"No named members" refers to the specific situation with unnamed bitfleld (see litb's reply above). A plain empty struct (no members at all) is a constraint violation, not UB.
AndreyT
2009-10-26 18:48:22
I've just reread it, and I see no such constraint. I'm looking at §6.7.2.1. The constraints I see are in paragraphs 2 through 4. Paragraph 2 says the only allowed incomplete type is a flexible array. Paragraph 3 says a bitfield can have a width of zero and no name, or a name and a width between 1 and the size of the underlying object. Paragraph 4 says the underlying type of a bitfield shall be _Bool, unsigned, or int (or something else that's implementation defined).
Jerry Coffin
2009-10-26 19:01:43
It's a syntax rule violation, which demands a diagnostic message.
Johannes Schaub - litb
2009-10-26 19:11:25
If it's a syntax rule violation, then it certainly requires a diagnostic -- but I can't see anything that says the BNF forms part of the syntax rules. For the most part, they're treated as non-normative (e.g. Appendix A contains the whole BNF, but is specifically labeled as "informative").
Jerry Coffin
2009-10-26 19:20:34
@Jerry, the rules within "Syntax" sections (like 6.7.2.1/1 in C99) are syntax rules of course. Annex A only summarizes all rules that are given already in clause 6, thus it's informative.
Johannes Schaub - litb
2009-10-26 19:57:04
"Of course" usually translates to having taken something for granted without any real support -- and this seems to be no exception. Don't get me wrong -- I'm not saying you're wrong (exactly), only that this is the sort of thing it should explicitly specify, not take for granted.
Jerry Coffin
2009-10-26 20:16:22
+2
A:
The C grammar doesn't allow the contents of a struct to be empty - there has to be at least an unnamed bitfield or a named member (as far as the grammar is concerned - I'm not sure if a struct that contains only an unnamed bitfield is otherwise valid).
Support for empty structs in C are an extension in GCC.
In C++ and empty struct/class member-specification is explicitly permitted, but the size is defined to be 1 - unless as part of the empty base optimization the compiler is allowed to make an empty base class take no space in the derived class.
Michael Burr
2009-10-26 18:33:15