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 Why is atoi dangerous and what should be used instead?

Parent

Why is atoi dangerous and what should be used instead?

+3
−0

According to Which functions in the C standard library must always be avoided?, the atoi family of functions is dangerous and should never be used for any purpose. The rationale given in the answer is this:

These have no error handling but invoke undefined behavior whenever errors occur. Completely superfluous functions that can be replaced with the strtol() family of functions. strtol(src, NULL, 10) is per definition 100% equivalent to atoi except with well-defined error handling. References: ISO 9899:2018 7.22.1.2, MISRA-C:2012 rule 21.7.

Where exactly is it said that the functions could invoke undefined behavior and how is the strtol (family) guaranteed to be equivalent?

History
Why does this post require attention from curators or moderators?
You might want to add some details to your flag.
Why should this post be closed?

0 comment threads

Post
+2
−0

The atoi family of functions should never be used for any purpose - they are broken by design.

The reason why can be found in the C standard C23 7.24.1:

The functions atof, atoi, atol, and atoll are not required to affect the value of the integer expression errno on an error. If the value of the result cannot be represented, the behavior is undefined.

Meaning that the functions do not necessary have any error handling at all - if you pass a string which consists of other things than digit characters, anything can happen.

Furthermore, the same chapter makes this guarantee (C23 7.24.2):

Except for the behavior on error, they are equivalent to

atoi: (int)strtol(nptr, nullptr, 10)
atol: strtol(nptr, nullptr, 10)
atoll: strtoll(nptr, nullptr, 10)

So the solution is to always use the strtol family of functions instead. They have error handling but are otherwise 100% equivalent to the atoi family, when passing base 10 = decimal as parameter.

(In fact some of the better standard libraries implements atoi as a mere wrapper macro around strtol.)

History
Why does this post require attention from curators or moderators?
You might want to add some details to your flag.

1 comment thread

atoi(3) is not broken by design. (3 comments)
atoi(3) is not broken by design.
alx‭ wrote about 15 hours ago · edited about 15 hours ago

atoi(3) is not broken by design. It actually has a good design. It was broken by history.

The standard could perfectly define the behavior on error, and it would be a decent API. They just decided to not define it.

If I were to define the behavior of atoi(3), I would make it equivalent to this:

int
atoi(const char *s)
{
    int  i, e;

    i = strtoi(s, NULL, 10, INT_MIN, INT_MAX, &e);
    errno = e;
    return i;
}

See also: https://man.netbsd.org/strtoi.3

Lundin‭ wrote about 14 hours ago · edited about 14 hours ago

alx‭ Yes they could define it but chose not to, and so the functions are broken. It's entirely a fault caused by the C standard and would have been easy to fix during C90 standardization, when they chose to pick up the function from the smelly swamp of sloppily designed, sloppily specified Unix crap. But this was never done. The C99 says the functions were kept around solely for backwards compatibility and for the reason that they might perform an itty bit better because of the lack of error handling. They have to justify the bad specification somehow...

Lundin‭ wrote about 14 hours ago

Btw errno error handling is another huge designer facepalm originating from that same smelly swamp, so it's not something to encourage or specify in when designing new functions. Someone asked "surely it isn't possible to make something even worse than Windows awful GetLastError" and then *nix/POSIX proved them wrong...