Comments on µContest

It looks like the activation mail doesn't work.
I haven't received anything and there's no option to resend the email :\

Edit: They activated me manually now. I also received the activation email 9 hours after registering at 06:33 CEST.

lol you don't loose time. Ya it seems OVH's servers were busy

Now I have another problem
Everytime I submit an answer, it only returns an empty page. http://www.microcontest.com/interface.php states that it should return "0" or the points gained.
I submit the answers via POST to http://www.microcontest.com/contests/$levelid/validation.php with PHPSESSID as cookie, which I got from the request for the variables, and the solution as POST, where the name of the var is, what the level site says. In level 1 for example "s", so my POST is s=123.

Any ideas?

Edit:

Ok, I got it now.

The solution submission needs the 60gp cookie. This cookie is given when requesting the variables - But ONLY, if there is already a PHPSESSID set.
That is of course totally incompatible with the doc @ http://www.microcontest.com/interface.php, which says that the PHPSESSID is generated upon requesting the variables, and that you need to use it for the solution submission.

But without PHPSESSID => no 60gp cookie => no solution submission..

So what I do now is:

1) Let the program login to the site with username, sha1 via the login form. Save the PHPSESSID
2) Request the variables using PHPSESSID in cookie & username+sha1 in POST, save 60gp cookie
3) Submit solution, using PHPSESSID & 60gp

:S

Cita de noother

Julio 13, 2011 - 18:38:49

This cookie is given when requesting the variables - But ONLY, if there is already a PHPSESSID set.

I think this is false actually. In fact, our lib doesn't send any PHPSESSID to contest.php when the POST is performed. Moreover, contest.php sends back the PHPSESSID AND the cookie 60gp, i checked with wireshark.

You're right. It was my fault.

I didn't get the cookie because my http-function didn't parse the header correctly. The page has Set-Cookie 2 times, so it replaced the first with the second cookie. (I somehow had in mind, that cookies are sent in 1 line, sepereated by ';', just like it is when you sent them to a host)

It also wasn't clear to me, that this qp60 cookie actually exists at the beginning. The doc says "(Cookie, PHPSESSID)". Looks like I interpreted that wrong by assuming it want's to say 'You need to send the Cookie "PHPSESSID"'.

Thank you..

But still, I only get a blank validation-page, if the submitted solution is wrong or the timelimit has been exceeded. Not a '0' as it is stated in the interface.php.
Also it would be nice, if the page would return some more status codes. Like:

>0 Solution correct, points
0 Solution wrong
-1 Challenge already solved
-2 Timelimit exceeded
-3 Request ignored, because of protocol violation (no qp60-cookie for example)

In some cases, it's very hard to figure out what's wrong at the moment :\

greets

Yes you are totally right, these kind of error codes are planned in the next version of the protocol.
But it is still strange that you don't get a correct validation page. I've never got the problem, and with my tests with wireshark, I actually got '0', because I have already validated the contest.
If you want an example : Example. (I remplaced 0 by 3 at the end for the example but I actually got 0).
I will add it to the doc page.

Hi, I think the score point is not correct.
I have solved 66% and the score of microcontest is 18613.
18613*0.66=8107 and I have 4859 pts.
Am I wrong?

Yes you are. Score doesn't work this way here.

Cita de thefinder

Febrero 13, 2014 - 13:16:45

Hi, I think the score point is not correct.
I have solved 66% and the score of microcontest is 18613.
18613*0.66=8107 and I have 4859 pts.
Am I wrong?

Yes I think you are.
wechall scoring is not equal. It is set so that the jump from 1% to 2% is worth much less points than the jump from 99% to 100% - Exponential scoring.
This forces users to try harder challenges and complete (or nearly complete) whole sites.
It also means that a user who signs up to 50 sites and gets 10% on each will not be regarded as highly rank as someone with say 3 sites of 100%.

Wechall exponential scoring is a good thing, but only when the site itself isn't already exponentially scoring. A µContest challenge can score from 1 to 20 points based upon number of users who solved it. So if you solved 50% of the challenges but mainly the easy ones, you score only like 12%. Which is fair regarding difficulty, but adding another layer of exponential scoring on wechall is quite a bit drastic.