Have you checked all the edge cases? Remember, rank 0 is not a rank.

The original text of the extract is somewhere in a_scandal_in_bohemia (not always at beginning).

a_scandal_in_bohemia is the full text in english. It is preloaded.
extract is a random extract from a_scandal_in_bohemia but it has been encrypted.

Getting the original text for the second line isn't trivial: it is a big part of the puzzle. Keep thinking! :)

note: the extract in the sample tests is 100% of the text. The extracts in the full tests are c.80% of the text starting at different points to make things harder.

To print the whole text: print(a_scandal_in_bohenia).
To print the random extract: print(extract) (or whatever is the argument of your function)

try printing the input

I guess the short answer is that your solution doesn't work - some random test cases are generated and you don't pass all of them, but because they're generated randomly, you don't necessarily always fail on the same exact test case.

I just ran my code. There are 103 test cases - same number every time. Does your code actually pass all tests and have you submitted?