Using size_t limits max number to 2^32.

[FreeBear-nc]

When dealing with SD cards we could be asked to print a uint64 (for example, from SD.totalGytes()).
This attempt to handle the very large numbers that might be expected from an SD card.

Yes, there is a bit of ugliness from recasting to a uint32 before calling the snprintf function, but this avoids implicit casts where data could be truncated.

[marsman7]

filesize = filesize / (D_FILE_SIZE_DIVIDER/100);
With this code you loose resolution. In Example

original

#define D_FILE_SIZE_DIVIDER 1024

filesize = 16.777.216;
filesize *= 100;  // 1.677.721.600
filesize = filesize / D_FILE_SIZE_DIVIDER;  // 1.638.400

your code

#define D_FILE_SIZE_DIVIDER 1024

filesize = 16.777.216;
filesize = filesize / (D_FILE_SIZE_DIVIDER/100);  // 1.677.721

[FreeBear-nc]

Noted.
Was trying to avoid an overflow when multiplying by 100 which the current code will do at (approx) 2^28. But with a 2^64 limit, it is going to be a very, very long time before we see file sizes in the Exbi range (that is 2^60).

[fvanroie]

With uint64_t you can still do the *100 before dividing by D_FILE_SIZE_DIVIDER and keep the resolution.
It will indeed only pose a problem with unintelligible numbers.

[FreeBear-nc]

Reworked the code to use an array of suffixes and reduce the number of snprintf_P statements. This means if TiB or others are needed, they just need to be added to the array.

[fvanroie]

Thanks, I did a bit more refactoring to make the code even more universal, and added Terabytes. cc1d6d9
It should also sidestep the overflow issue entirely now.