So someone here had the job to merge the current version of our software into another older repo.
(That older repo has some features for future versions, that should be only in there ... don´t ask ... not my decision)

Took him long enough. But he forgot to check one thing while merging: The encoding of the files.

And we´re german...

ALL umlauts, and all other special chars in the code were replaced with this: �
No global replace, because all chars were replaced with the same char.
(Why the fuck do we have special chars in the code in the first place???)

So to not need him to start all over again I compiled a list of common german words containing umlauts and did a global search and replace.

I think I got 90% of the errors like that.
Now he´s going to correct the rest of the errors.

Fuck the comments in the code though.
Just a waste of time...

During a code review I was doing the dev wrote a function that takes a tweet, iterates over each letter, if the letter is not in a specified unicode range the letter is removed from the sentence. So now you have a sentence with missing letters... Wtf