Rule 8-5-2: what about strings?

[jocob]

struct Test
{
char str1[20];
char str2[20];
};

struct Test t =
{
"Hello",
"World"
};

LDRA flags this with 397 S (Array initialisation has insufficient items.) violation of 8-5-2 (Braces shall be used to indicate and match the structure in the non-zero initialization of arrays and structures.)

Looking at the details of the rule in the guidelines, I see why: "The non-zero initialization of arrays or structures requires an explicit initializer for each element."

It seems there should be an exception for initializing a character array with the string syntax, since:
a. the unused characters are implicitly null character 0
b. there is no MISRA compliant way to explicitly add null characters to a "string"

I think MISRA C 2004 has a similar problem, but I haven't looked at 2012 yet.

[dg1980]

Hi,

what´s the output of LDRA if you use this ugly alternative?

Code:

struct Test t =
{
  {'H', 'e', 'l', 'l', 'o', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0'},
  {'W', 'o', 'r', 'l', 'd', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0', '\0'}
};

[misra cpp]

You are correct that the example initializing using "Hello" etc. is a violation of 8-5-2 (and dg1980's example padded with '\0's is compliant). However, the first example would be compliant under MISRA C:2012 - it has the exception for strings.

The next issue of MISRA C++ aims to remove unnecessary differences between MISRA C and C++, and this is likely to be a rule that will be modified for that reason - though they may still not be identical because there is a semantic difference between C and C++ for a statement like
char str[4] = "abcd";
(its legal in C but a constraint error in C++)


As a matter of etiquette, we don't encourage discussion of particular tool's behaviour on the Bulleting Board - good or bad