which warnings ? i don't see any when i run the the Solution panel against the tests without those lines

also, the number of bytes in the array is length * sizeof *expected, not sizeof(char). (or length * sizeof(unsigned), but I prefer avoiding hardcoding). the memory size of an array is its length multiplied by the size of its element type.

what's the purpose of these lines ?

sub_str[rand() % str_len] = '\0';
stringer(0, NULL, exp_str);  

should have been raised as an issue. no UB here, just a mistake on my part, an == that should have been a != during comparison when either of the words was a null pointer. fixed.

• the index logic is not needed anymore
• the 3rd parameter of cr_assert_arr_eq() is the size in bytes of the array, not its length (as it calls memcmp() behind the scenes)

i added this fixed test to C.

• the testing code should not be in preloaded
• assertion should use cr_assert_arr_eq()

yes, the root author gets credited

added

approved

read the assertion message, not the line of code:

Invalid result for s = "", part1 = "ac", part2 = "ab"
Expected: false
Actual: true

size_t counts[36];
memset(counts, 0, 36 * sizeof(size_t));

can be just

size_t counts[36];
memset(counts, 0, sizeof counts);

or even

size_t counts[36] = {0};

size_t index = curr < 65 ? curr - 48 : curr > 90 ? curr - 87 : curr - 55;

this line could be made more readable by using character literals such as 'a' and '0'. or just set the buffer size to e.g. 128 and use direct indexing; this is nowhere close to a noticeable memory footprint

char rand_text[rand_len];

that does not account for the nul terminator, hence the crash.

also, please don't golf the Solution panel, it has to be maintainable; same remark for things like

size_t i = X - X;

approved

array[index] = (int)index;

this cast has no effect. array is already an array of int, so whatever you store inside is going to be implicitely cast to an int.

approved.