creditbest.GetScore() crashes foldit

Case number:845829-997237
Opened by:pauldunn
Opened on:Tuesday, March 11, 2014 - 21:24
Last modified:Friday, March 8, 2019 - 15:58

creditbest.GetScore function will crash a client (both main and devprev) if the track doesn't have a credited best yet. (possibly related to

Load un-evo'd player's share in new track.
run creditbest.GetScore()

Alternate procedure:
Load puzzle that has unmet conditions from the start like [DEVPREV] Conditions
run creditbest.GetScore()

(Tue, 03/11/2014 - 21:24  |  5 comments)

NickyCGS's picture
User offline. Last seen 7 years 21 weeks ago. Offline
Joined: 06/27/2013
Groups: Repro-men

Thanks for the feedback Pauldunn! Reproduced this (and the linked feedback) issue has been sent to developers.

LociOiling's picture
User offline. Last seen 3 hours 26 min ago. Offline
Joined: 12/27/2012

Still an issue on both main and devprev.

More specifically, both creditbest.GetScore () and creditbest.GetEnergyScore () crash the client when working on an unevolved solution.

Similarly, creditbest.GetSegmentEnergyScore () and creditbest.GetSegmentSubscore () crash the client for an unevolved solution. Interestingly (?), when these functions are called with a bad argument, such as segment index 999 for the first argument, the error is detected and no crash occurs.

A crash also occurs for creditbest.AreConditionsMet ().

No debug.txt is generated for any of the crashes.

The functions creditbest.Restore() and creditbest.GetExplorationMultiplier do not crash. Sadly, creditbest.GetScore still crashes after a call to creditbest.Restore ().

The other poses -- current, recentbest, and absolutebest -- do not have this problem.

Using pcall to invoke GetScore and the other crashing functions does not help, nor does running the entire recipe inside an xpcall.

An "unhandled exception" traceback appears in log.txt (see attached).

Joined: 03/05/2015
Groups: Gargleblasters

Crossposted from 1299 feedback... crash on running LUA recipe Banded Worm Pairs
(didn't know about the scriptlog, will be attaching these regularly now)
Working in main, the client which on my mac shows as "game_dynamic" in the finder bar.

Joined: 03/05/2015
Groups: Gargleblasters

Recipe crashes are the most common for me so it is "helpful" to know that this issue is still outstanding.

Wipf's picture
User offline. Last seen 2 years 16 weeks ago. Offline
Joined: 10/06/2018
Groups: Go Science

Are the developers working on this or will we have to live with this bug?

Basically the game client crashes instead of throwing a LUA error when a recipe tries to access the creditbest solution but the current undo track does not have one, because it never had a solution which statisfied all conditions to get credit.
I think that should not be so hard to fix, but this is open for years now. Are the developers overloaded?


Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, Boehringer Ingelheim, RosettaCommons