it seem quite ok (seems to work like BitsPawn algorythm) but :
Crash when dubble clicking on "run filter" upper right during a wiggle all on puzzle 1054
Note: please let us sufficient time to test our "filter" recipes too (in addition to old recipes). i'd like to analyse the effect of the combination of our filter algorithms and this generic one (would it crash etc).
Script bug (unrecognized Lua command I think)
In Void Crusher on 1053 (no filter): error on calling current.GetExplorationMultiplier()
I hope old commands will be supported (or ignored), otherwise we'll loose existing recipes.
Crash after 6 hours work (Windows 7, sript Ebola Rebuild = many different actions & commands)
(started on line, then 5:30 hours off line)
Many loops OK, but it crashed when I came out from screen saving.
Here are several error messages in the log (+representative line number)
6789 etc:
core.scoring.methods: Error: don't recognize secondary structure character '_'
repeated
6987:
Error contacting server using URL: https://fold.it/comm/foldit_ops
(this is normal since I'm off line)
7006 etc:
***** STARTING THREAD ActionGlobalMinimizeBackbone
core.scoring.sc.ShapeComplementarityCalculator: Failed: No molecular dots for surface 0
protocols.filters.ShapeComplementarityFilter: Issue computing shape complementarity value - failing filter.
core.scoring.sc.ShapeComplementarityCalculator: Failed: No molecular dots for surface 0
protocols.filters.ShapeComplementarityFilter: Issue computing shape complementarity value - returning -1 instead.
Here are the latest lines:
*** STARTING THREAD ActionGlobalMinimize
delta_score: 0.034978
Playing sound: 4
*** ENDING THREAD ActionGlobalMinimize
Tool on_action_complete called
Tool on_action_complete called
***** STARTING THREAD ActionRepack
UNHANDLED EXCEPTION
1: library_main +5523953 bytes (no line)
2: library_main +7061406 bytes (no line)
3: library_main +402705 bytes (no line)
4: library_main +4230789 bytes (no line)
5: library_main +4231243 bytes (no line)
6: BaseThreadInitThunk +18 bytes (no line)
7: RtlInitializeExceptionChain +99 bytes (no line)
8: RtlInitializeExceptionChain +54 bytes (no line)
Latest line = 32072
I'll try to send you the log by mail.
I see the same issue with Loop Rebuild: GetExplorationMultiplier needs to be supported (or at least there should be a stub)
I accidentally removed one of the exploration lua hooks instead of just making it return 0. It will be put back in soon.
workaround for missing GetExplorationMultiplier…
add this code at the beginning of the failing recipe:
if current.GetExplorationMultiplier == nil then
current.GetExplorationMultiplier = function () return 0 end
end
There's been an update to the devprev release that restores GetExplorationMultiplier. There was no mention of the update on the front page blog or in the forum, so it's not clear if anything else got fixed.
No need to remove the code in my previous post, it won't hurt anything.
From jflat06's post on this thread, it sounds like GetExplorationMultiplier may be deprecated. Susume thinks any future exploration puzzles will be handled via filters. So it may be time to rip out GetExplorationMultiplier and the "normal" logic found in many recipes. I'll open a new feedback on that.
The update is listed as a hotfix in the comments to the devprev release.
WHERE can I find these filter switches in the original interface ?!
I do not see them. Nor do I see the original switch for the filters.
20150310-5abde18809-win_x86-devprev
I found them, they are in the drop down area where the filter scores are mentioned.
This is not documented at all !