[Community puzzle] Next car license plate?

This topic is about the puzzle Next car license plate ?.

Feel free to send your feedback or ask some help here!


Someone at CG’s going to say that sort of thing only happens to me, but… something’s wildly wrong in the way that new community puzzle’s displayed.


(for even wilder fun, check out the dark theme)


The puzzle is fine in edit mode in the puzzle editor, but mostly the same in display mode. (and never fine in the IDE, be it spawned from edit or display mode)

(if you think it’s my browser, or the fact I forgot to clear my cache before posting, or anything like that, just PM me, that’ll be enough noise)

It looks like that to me too (Ubuntu 16.04 with i3wm, tested in latest versions of Chromium and Firefox).
I opened two or three other Community Puzzles, these are ok.

One of my own (pending) puzzles is now in this state as well. Might be something in the lines of “puzzle edited after such-or-such date”.

Same happened to my puzzle:

Workaround: edit, then save again. Which I just did for yours.

(for the record, that didn’t work yesterday)

What a weird bug. Thank you.

Sorry! in fact, i just fixed this bug some hours ago.
Indeed, now you just have to save again the contribution to fix it!

1 Like

Don’t get that error. But there is, I believe, a mistake on the cover page.
AB-001-AA, AA-002-AB, AA-003-AB, …, AA-999-AB,
Those last three mentioned on the fourth row are a repeat of the last three on the second row. So… yeah. A little confusing

Umm… Yeah. You’re right. Fixed it. Thanks, Shelby.

  • danBhentschel

The example lines in the puzzle intro shows:
ZY-001-ZZ, ZY-002-ZZ, ZY-003-ZZ, …, ZY-999-ZZ,
ZZ-001-ZZ, ZZ-002-ZZ, ZZ-003-ZZ, …, ZZ-999-ZZ.

I think it should have been:
ZY-001-ZZ, ZY-002-ZZ, ZY-003-ZZ, …, ZY-999-ZZ,
ZZ-001-AA, ZZ-002-AA, ZZ-003-AA, …, ZZ-999-ZZ.

If the input is:

Output should be:

I passed the “Very big” test in the IDE unit, but the validator test has rejected it. What’s the expected behaviour on very big tests?

Thanks for reporting. The problem setter’s code as well as mine confirm that, I’ve edited the statement.

nothing special, just a slightly bigger number - about 10% larger.

It seems that removing the modulo-26 (ordinal wrap) operation on the leftmost letter fixed the solution, which means that license plates can go over ZO-OOO-OO? I really have no idea why that’s the case, but it seems that single change made the tests accept my answer.
Or maybe we can blame javascript’s floating point numbers, haha!

I don’t get it, ZO-OOO-OO isn’t valid (the middle part are numbers, not letters), but it goes above ZO-001-AA.
However the number comfortably fits in a 32bit signed int, so it is far away from JavaScript rounding errors.

I have some trouble with this puzzle. Passing all tests except the big one:
my code answers : JQ-027-YY instead of JQ-027-XY… I don’t understand how an error can appear just for 1 letter and not for others

For info, letter I and O are not present on FR plates… (too much confusing with 1 and Q)
Could be an optional challenge… :slight_smile:

O, I and U are excluded of the french plates, they can be confused with 0, 1 and V. Moreover, SS is removed too because of the nazi organisation.

1 Like

All my tests validate and works. however only get 85% completed as 1.000.000 mill I rejected :frowning:

1 Like

Same for me, stuck at 85% because of it. My guess is the validator lands on an edge case that none of the tests do not, e.g. ZZ->AA transition or 999->001. But it’s always tough to debug those, naturally.

*Edited: Solved it by iterating plate by plate. Best hint I can give is consider the edge case where “carrying the 1” trickles all the way from the middle to the right to the left side.