Communities

Writing
Writing
Codidact Meta
Codidact Meta
The Great Outdoors
The Great Outdoors
Photography & Video
Photography & Video
Scientific Speculation
Scientific Speculation
Cooking
Cooking
Electrical Engineering
Electrical Engineering
Judaism
Judaism
Languages & Linguistics
Languages & Linguistics
Software Development
Software Development
Mathematics
Mathematics
Christianity
Christianity
Code Golf
Code Golf
Music
Music
Physics
Physics
Linux Systems
Linux Systems
Power Users
Power Users
Tabletop RPGs
Tabletop RPGs
Community Proposals
Community Proposals
tag:snake search within a tag
answers:0 unanswered questions
user:xxxx search by author id
score:0.5 posts with 0.5+ score
"snake oil" exact phrase
votes:4 posts with 4+ votes
created:<1w created < 1 week ago
post_type:xxxx type of post
Search help
Notifications
Mark all as read See all your notifications »
Q&A

Welcome to Software Development on Codidact!

Will you help us build our independent community of developers helping developers? We're small and trying to grow. We welcome questions about all aspects of software development, from design to code to QA and more. Got questions? Got answers? Got code you'd like someone to review? Please join us.

Comments on Macro to verify that an expression has no side effects at compile time

Post

Macro to verify that an expression has no side effects at compile time

+2
−0

I'd like to implement a macro strnul():

#define strnul(s)  (s + strlen(s))

But that expands and evaluates s twice. I'd like to avoid that.

I could do (using GNU extensions)

#define strnul(s)   \
({                  \
    auto  s_ = s;   \
    s_ + strlen(s_);\
})

However, this adds some other problem: a local variable named s_. This local variable must have a name different from whatever users of this function may pass as argument, as otherwise we'd have auto s_ = s_;, which can't work.

But I have reasons to pass an argument called s_, so I'd like to find a way to entirely avoid the variable. If possible, an alternative would be to do

#define strnul(s)                         \
({                                        \
    static_assert(!has_side_effects(s));  \
    s + strlen(s);                        \
})

Is it possible to somehow do this in C? Even if the solution would require using compiler extensions, I'd be interested in finding a solution.

History

2 comment threads

Array length expressions might help soon (5 comments)
That's what functions are for (3 comments)
That's what functions are for
Lundin‭ wrote 8 months ago

The proper way to avoid this seems to be using a function. Why can't you do that? Otherwise the only trick I can think of is something like s += strlen(s) where compound assignment guarantees that s is only evaluated once. But since s isn't necessarily an lvalue in this case, or even a read/write location that should be updated, that probably won't work.

alx‭ wrote 8 months ago · edited 8 months ago

The problem with the function is that you need a cast within the function (as it gets a const char * but returns a char *). I'd like to have zero casts in a program. I think I prefer having to manually keep the names unique (by adding more trailing underscores) and avoid the function.

With functions, it'd have to be

#define QVoidptrof(p)  typeof(1?(p):(void*){})
#define QCharptrof(s)  typeof                                         \
(                                                                     \
	_Generic((QVoidptrof(s)){},                                   \
		const void *: (const char *){},                       \
		void *:       (char *){}                              \
	)                                                             \
)

#define strnul(s)  ((void)0, (QCharptrof(s)){strnul(s)})
char *
(strnul)(const char *s)
{
    (char *) s + strlen(s);
}
Lundin‭ wrote 8 months ago

alx‭ You don't need a cast, you can just pick char my_func() or const char my_func_const() with _Generic based on what was passed to the macro, much like QVoid/QChar will have to be implemented in C23.