Re: [PATCH 2/2] mingw: avoid const warning

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Am 06.06.2014 21:13, schrieb Junio C Hamano:
> Karsten Blees <karsten.blees@xxxxxxxxx> writes:
> 
>> Am 29.05.2014 12:47, schrieb Stepan Kasal:
>>> Fix const warnings in http-fetch.c and remote-curl.c main() where is
>>> argv declared as const.
>>>
>>> The fix should work for all future declarations of main, no matter
>>> whether the second parameter's type is "char**", "const char**", or
>>> "char *[]".
>>
>> I'm 100% in favor of a solution that doesn't restrict main to non-const char**! Thanks.
>>
>>> Signed-off-by: Stepan Kasal <kasal@xxxxxx>
>>> ---
>>>  compat/mingw.h | 5 +++--
>>>  1 file changed, 3 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/compat/mingw.h b/compat/mingw.h
>>> index 15f0c9d..8745d19 100644
>>> --- a/compat/mingw.h
>>> +++ b/compat/mingw.h
>>> @@ -369,10 +369,11 @@ extern CRITICAL_SECTION pinfo_cs;
>>>  void mingw_startup();
>>>  #define main(c,v) dummy_decl_mingw_main(); \
>>>  static int mingw_main(c,v); \
>>> -int main(int argc, char **argv) \
>>> +int main(c, char **main_argv_not_used) \
>>>  { \
>>> +	typedef v, **argv_type; \
>>>  	mingw_startup(); \
>>> -	return mingw_main(__argc, __argv); \
>>> +	return mingw_main(__argc, (argv_type)__argv); \
>>>  } \
>>>  static int mingw_main(c,v)
>>>  
>>
>> I have to admit I had trouble understanding what 'typedef v,
>> **arv_type;' does (looks invalid at first glance), and why you
>> would need main_argv_not_used instead of just main(c,v).
>>
>> So, I'd like to award +10 points for cleverness, but -10 for
>> obscurity ;-) Probably deserves a comment or an explanation in the
>> commit message.
> 
> Agreed.  The "typedef" one is a cute hack.
> 
> I am wondering why the solution is not a more obvious "drop const
> that is not ANSI C", though.  I only have a ready-access to N1570
> draft but in it I find:
> 

Actually, that was the original solution ($gmane/247535). I just complained because it was slightly different from what we had in msysgit for quite some time [1], causing merge conflicts.

I guess compilers probably won't complain if you declare argv const, even if the standard is more strict. After all, you aren't supposed to modify argv.

> 
>> A simpler solution that works with all definitions of main() is to
>> cast to void* (tell the compiler all responsibility is on
>> us).
> 
> Can you cast away the constness that way, though?
> 

Not 'away'. This passes a non-const value to a const parameter, which is typically not a problem. Its just 'char**' to 'const char**' that produces the warning, because 'const char' hides behind a non-const pointer, see [2]. 'void*' to 'const char**' works, though.

[1] https://github.com/msysgit/git/commit/6949537a
[2] http://c-faq.com/ansi/constmismatch.html

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html




[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]