I was recently trying to research my chances of qualifying for ACM ICPC by looking at the latest regionals I could find problems for, ACPC 2015, which can be found at the ACM-ICPC Live Archive. I had first solved a few of the problems on the Hackerrank instance before I found the official one, but when I tried submitting my code that solved the Hackerrank instance of them it would give WA on the ICPC OJ.
Also it seems there are no successful submissions whatsoever to the whole contest on the whole OJ which seems suspicious. Does anyone know if the ICPC OJ is known for having wrong test cases? Or is output formatting more strict or something?
As a reference to test it if anyone wants to try to solve the very easiest problem (it's div 2 problem A level) they can try here.
My code for solving it was this. And I really can't see how it should be wrong, and also as previously mentioned the Hackerrank instance accepted it.
If anyone could help enlighten me that would be awesome.
As you can see here:
https://icpcarchive.ecs.baylor.edu/index.php?option=com_onlinejudge&Itemid=8&category=677
The problems have red ticks by their side. I don't remember where I saw this, but red tags means the problem doesn't have checker (so you will get always WA). Green one means ok problem and yellow means special checker (so you should go for those ones for practicing).
As you, I discovered that the worst way (by getting tons of WA in problems with red ticks).
Ahhh, that makes sense. Thank you! I guess I'll just have to trust that Hackerrank has correct checkers, or trust that I can find all edge cases myself then :).