-is-this-fft-'s blog

"Files '21.in' and '50.in' contain equal tests after well-formatting"
By -is-this-fft-, history, 4 years ago, In English

Hi.

Sometimes I use Codeforces groups for private trainings. On occassion, this means importing (via Polygon) problems from some other contests (that have public test data and are free to be used like that).

But when importing test cases "from the files" or "from the archive", this happens a lot:
(By the way I have no idea why it happens so frequently. My best guess is sometimes that OI-style contests have to repeat test cases in different subtasks. But I have also seen this with ICPC-style.)

Anyway, my question is:

  • Why does Polygon even care if there are multiple equal tests? It doesn't seem to care if there are two equal tests from generators.
  • Why does it only show one error at a time, and a seemingly random pair at that (It doesn't seem like it is the "lexicographically smallest" pair)?
  • Can I somehow turn this validation off? There are some (probably more important) checks that you can turn off.

(I guess I should be smarter and write a local script to remove the duplicates).

  • Vote: I like it
  • +32
  • Vote: I do not like it

»
4 years ago, # |
  Vote: I like it +21 Vote: I do not like it

It doesn't seem to care if there are two equal tests from generators.

no, cares, when you press commit

→ Reply
  • »
    »
    4 years ago, # ^ |
      Vote: I like it -17 Vote: I do not like it

    I think it cares if the generator command is the same, but it doesn't care if the output is the same.

    → Reply
»
4 years ago, # |
← Rev. 2   Vote: I like it +8 Vote: I do not like it

There is a similar warning for Polygon-generated tests. Thus an error when uploading seems a bit extreme to me.

Why it happens so frequently: perhaps authors don't care enough, and/or have no ready means to automatically check for duplicates. I know we had duplicate tests in a recent contest, and I'm not proud of that. However, at the point it was discovered, it was more convenient overall to just use the tests as they were than to remove duplicates. This of course hints that the processes may be improved, but I digress.

→ Reply
  • »
    »
    4 years ago, # ^ |
      Vote: I like it +9 Vote: I do not like it

    I know we had duplicate tests in a recent contest, and I'm not proud of that.

    You brought shame on your whole family.

    → Reply
    • »
      »
      »
      4 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      Hah, yeah, kind of.

      Seriously though, duplicate tests are in almost all cases not a problem in itself, but rather an indication that the problem just might be not polished. Definitely useful for a coordinator as a hint to take a closer look.

      → Reply