Rework gigasecond to be slightly more challenging

[glennj]

I'm not terribly experienced in C. Have I implemented an OK way to handle strings without malloc/free?

Closes #924

[glennj]

Hmm I could be wrong but IIRC the intention of the gigasecond exercise is to be relatively simple.
Introducing time.h In any significant way may add needless complexity and could lead to different results depending on platform.

@wolf99 how does this submission meet your concerns?

[ryanplusplus]

Suggested change

	#define BUFFER_SIZE 20 // "YYYY-mm-dd HH:MM:SS"
	#define BUFFER_SIZE (strlen("YYYY-mm-dd HH:MM:SS") + 1)

[siebenschlaefer]

Alternatively: #define BUFFER_SIZE sizeof("YYYY-mm-dd HH:MM:SS")

[ryanplusplus]

Alternatively: #define BUFFER_SIZE sizeof("YYYY-mm-dd HH:MM:SS")

That's much better! I don't know why that never occurred to me.

[glennj]

Should it be a static size_t instead of a #define so it only needs to be calculated once?

[ryanplusplus]

It will be calculated at compile time even if it's a #define.

[ryanplusplus]

Can you bump up the difficulty for gigasecond in config.json?

[glennj]

I was totally Chesterton's Fencing with that construct_date function. Didn't understand why it's there, didn't complain about it ;)

[wolf99]

Thanks for this PR @glennj !