Fix header files

[DimitriPapadopoulos]

json.h: <inttypes.h> → <stdint.h>
int64_t is defined in <stdint.h>, no need for the full <inttypes.h>

json.h: <stdlib.h> → <stddef.h>
<stddef.h> is enough to define size_t

json.c: <stdlib.h>
Now that <stdlib.h> has been removed from json.h, add it to json.c for malloc()/calloc() and free()

[DimitriPapadopoulos]

Note that this patch does not change the status of the library regarding ANSI C vs. C89 compatibility:

• The additional <stddef.h> header is ANSI C.
• The ANSI C header <stdlib.h> is just moved around, from json.h to json.c.
• Neither <inttypes.h> nor <stdint.h> are ANSI C. Both are C99 and both are included only if json_int_t is not already defined, and additionally <inttypes.h> includes <stdint.h> anyway:

#ifndef json_int_t
   #ifndef _MSC_VER
      #include <inttypes.h>

[LB--]

Thanks! After doing some more thinking and research, perhaps it would be better to use a progressive-enhancement approach where we check the presence and value of __STDC_VERSION__ to know what is safe to use. I think also we should stick to long/long long/int_fast64_t since int64_t and co are optional and not guaranteed to exist. Though, that might be better suited to a different commit, as there are more places than just here that use int64_t. The user can also override the default by defining json_int_t, which I hope to add proper documentation for later. Thoughts?

Also, would you like your name added to the AUTHORS file? You may add it yourself or I will add it for you, or you can opt-out, whichever you prefer.

[DimitriPapadopoulos]

I agree:

• check __STDC_VERSION__ to introduce C99 headers for example,
• avoid int64, but on the other hand int64 is used only if json_int_t is not defined.

However, I also agree the above ideas are orthogonal to the current patch, which is about using the proper header files without changing the status with respect to ANSI C or int64. I would rather see this pull request merged before working on the above ideas.

[LB--]

without changing the status

Good point, I'll go ahead and merge this PR in that case. Yes or no to AUTHORS file?

[DimitriPapadopoulos]

I've submitted another PR for AUTHORS.