Single Player bugs and crashes v39 plus (SVN) - After the 20th of July 2019

I'm going to get a save off CHOO and then run till I repro it and save a full dump for anyone who wants to look at it.
 
I'm getting this error:
Code:
Traceback (most recent call last):
  File "BugEventManager", line 308, in _handleDefaultEvent
  File "CvEventManager", line 615, in onLoadGame
  File "CvEventManager", line 488, in gameStart
  File "CvAdvisorUtils", line 70, in resetNoLiberateCities
AttributeError: 'CvUnitInfo' object has no attribute 'getHasBuilding'
@KaTiON_PT: you sure you rebuilt the dll from the most up to date source files?
Alberts added the getHasBuilding method to unit infos in rev 10860-62
 
upload_2019-8-2_21-44-37.png

?
This on a cleanup after the commit failed with equally as many words. Anyone have a clue? This is new to me.

Not much I can do here til this is cleared up somehow.
 
ok, we r having the OLD "blacK" plots again, this hasnt happened around 10 years now or so, i think it is gamefont, but not sure again, its been along time, TB do u or DH remember this?? its only on the "fresh water" plots , l believe??
 

Attachments

  • fresh.JPG
    fresh.JPG
    276.8 KB · Views: 69
  • water.JPG
    water.JPG
    272.2 KB · Views: 40
ok, we r having the OLD "blacK" plots again, this hasnt happened around 10 years now or so, i think it is gamefont, but not sure again, its been along time, TB do u or DH remember this?? its only on the "fresh water" plots , l believe??

This might be a issue with the fpk's.
 
True... I'm just thinking that if the area data isn't storing the revealed plots then we'd have seen a major problem in that before now. It might be possible that this is the only place in the code that it would cause a crash and that it's a rare spot to hit, but I doubt it.

It could be that the save contains some kind of corruption from the problematic svn updates. Even if the game doesn't crash there could be some kind of memory corruption beeing caused by bad pointers. This could also happen in python because @Toffer90 fixed my crash problems by fixing python errors.
 
If there is some resources that couldn't be found or loaded from art or FPK I would expect there should be a log error correct?
 
If there is some resources that couldn't be found or loaded from art or FPK I would expect there should be a log error correct?

Resmgr.log and missing textures are displayed pink in game.

The Black Tiles problem comes up from time to time in all CivIV mods. Unpacking the art and starting the game without creating a fpk usally fixes it.
 
Left the game or VS running with a minidump loaded?
That was it! Thank you!


ok, we r having the OLD "blacK" plots again, this hasnt happened around 10 years now or so, i think it is gamefont, but not sure again, its been along time, TB do u or DH remember this?? its only on the "fresh water" plots , l believe??
@billw2015 THIS is what happens when you have too many FPKs. I KNEW there was a reason we backed down from having so many. This is it. We must be able to work with fewer FPKs and for the plans you have, I think that means we're going to be required to stop storing FPKs on the repo at all if I'm not mistaken.

@All: I'm going to bring this up elsewhere but I think this may mean that we no longer have a choice but to implement the FPK live mechanism and do it very soon - and part of that will be that FPKs must be packed much larger than they are so we aren't exceeding the limit of how many FPKs we can have. Obviously, different systems encounter this limitation at different points. Used to be around 8 FPKs was too many for most.

If you aren't keeping up with some of the more technical discussions on the FPK live thread I get it. Let me summarize:

We would be moving to a new storage method. To make the transition would be extremely simple. It would require taking these steps:
1) Delete the mod in the mods folder (you can save your usersettings folder somewhere if you like)
2) Check out a new repository for Caveman2Cosmos somewhere else in the same drive (usually C) as you store your program files folder where BeyondTheSword/Mods is, just make sure you set it up OUTSIDE the Firaxis (best even outside Program Files) directory.
3) Run a batch file that will be clearly labeled as your install file.

That's it.

You will end up with a Caveman2Cosmos mod folder in the mods folder in BeyondTheSword.

You will not generally mod in that active mods folder. However, ANY changes you make in your external Caveman2Cosmos folder will immediately be reflected in your actual game mod folder as well (with the exception of art files and code, which will get updated IF you use the batch file to load the mod with updates, that will also be clearly provided for you in the checkout.

Then any updates and commits would be done through the external folder.

That would be ALL you would HAVE to know. Period. I don't think that's too complicated a system and it will resolve a number of issues and pave the way for more improvements in team methodology.

Is that pretty much the sum total of it as you intend it as well @billw2015 ?


It could be that the save contains some kind of corruption from the problematic svn updates. Even if the game doesn't crash there could be some kind of memory corruption beeing caused by bad pointers. This could also happen in python because @Toffer90 fixed my crash problems by fixing python errors.
I agree. I'm about to load the save on a debug run and see what I can figure out about it.


If there is some resources that couldn't be found or loaded from art or FPK I would expect there should be a log error correct?
I'm not sure if the RESMGR log will catch the problem with too many FPKs - it COULD load everything fine but then have problems during the run with referencing the locations of the graphic files. Just sharing some theory here if you see some confusing results.
 
@billw2015 THIS is what happens when you have too many FPKs. I KNEW there was a reason we backed down from having so many. This is it. We must be able to work with fewer FPKs and for the plans you have, I think that means we're going to be required to stop storing FPKs on the repo at all if I'm not mistaken.
Just put fewer but bigger fpk's into the svn.
 
@billw2015 THIS is what happens when you have too many FPKs. I KNEW there was a reason we backed down from having so many. This is it. We must be able to work with fewer FPKs and for the plans you have, I think that means we're going to be required to stop storing FPKs on the repo at all if I'm not mistaken.
.
yeppers black tiles r gone with newly packed FPK's. so far.. thx. . . SO
 
Only svn matters at the moment.

The for the git repo unpacked art would be the best solution anyways.
Right so that brings to the foreground this matter:

If we are going to move to Git for the mod team and for modding commits and keep SVN as a reporting tool to release to the public (those who are already hooked up to us there) our more common weekly updates and debug fixes, we still need to somehow ensure that modding isn't being committed to the SVN at that point somehow and from there SVN becomes basically output only.

How do we actually MAKE this transition. Do we just say, after such and such date, no further modding commits are going to be accepted to the SVN, only updates from the Git side will be placed into the SVN and you can get your updates from the SVN but no non-authorized commits to SVN after that hard established date?
 
Back
Top Bottom