What would be the problem with 2 million ingredients? That this solution has no early return in case of 0? If you return any positive value, you have to look at all ingredients either way.

what if the recipe has 2 million ingredients? ok ok, I'm overthink, I think that this solution fits very will for the whole context of the problem

Congrats! The shortest solution.

super logic

Ashamed that I didn't realize it was this simple...

all occurrences

will this solution delete all the repetitions of the letters in text (e.g. "a" found 2,3.. n times) or delete only one occurence?

thank you

I can see some of your solution, I do not knwo if it is the most recent one. And yes, I can see a bug in it, which would cause the crash.
As a first step to diagnose the issue, I would recommend recreating the failing test case in your local IDE and debugging trough it. Use your IntelliJ, Eclipse, NetBeans, or whatever is what you use for coding on your computer, and run your solution with the input which Codewars says you can;t handle. Debug through your code, find the place where it crashes.

You can also visit CW Discord and #help-solve or #java channel.

Now fixed.

Approved.

Oh, never mind, I misread! Have changed the test logs in the fork to report test input, too.

Seems like that would duplicate the existing output from the test harness, no? The logs are already in the following form:

Just return the numbers: expected <solution return value> to equal <expected return value>.

Looks good to me. Just a remark: it's better practice to provide the input to user for failed tests (something like failed test with value = ..., expected ... but got ...).