It seems that solutions are accepted that find the following files: foopng and foo.pngbar when the first argument is png. That is wrong and should be checked.

This solution returns wrong results! If you have a file foopng and your first argument is png this file foopng is found, but it should not! The dot must be escaped: grep "\.$1"

And if you have a file bar.pngGGG the file is found, but it should not. So check for end of line: grep "\.$1$"

NOT best practices!

Neither solutions with echo, or echo -e nor printf work. I also tried pipeline it to sed 's%$%\r\n%', to no avail.

Why does printf not work. I had to use echo. Why?

I get a Segmentation fault when I use this solution. The size of result is never defined. How could that have worked?

For python I didn’t use numpy, just plain python. But I have not yet passed the final test (error at an edge case).

Well, I think if you read a bit at Wikipedia, it isn’t to difficult: https://en.wikipedia.org/wiki/Integer_(computer_science)

https://en.wikipedia.org/wiki/Two's_complement

https://en.wikipedia.org/wiki/Endianness

Maybe this is clever if you are a slow typer and reader or have a very limited space in memory or disk, but for me it seems like code obfuscation. Never mind, just my opinion.

Well, it is an issue, because the test cases don’t detect wrong results. How is this just a suggestion? It is wrong by the definition given in the description.

Can’t anybody else fix the kata or the test cases?

Add a test case for args (1, 96, 3612) which should result in 9 and NOT in 7 (as it is accepted now). This was already said by jolaf.

This is still not solved. The args (1, 96, 3612) are still accepted if the result is 7. But it should only be accepted if the result is 9! Please add a test case for this.

Solution is wrong for the cases (2211,0) and (11,0). Test case should be added.

This is the first solution I have found, that is fast and works with all the test and edge cases (i.e. negative numbers, lots of 0 and 1 as arguments).