40 release!

[Osk21]

Am I correct that an updated SVN works as well as a potential 40 Release? Or am I better to download a brand new 40 download?

 

[Thunderbrd]

Am I correct that an updated SVN works as well as a potential 40 Release? Or am I better to download a brand new 40 download?

It would be the same thing, just not connected to the SVN for further updating afterward.

 

[Osk21]

It would be the same thing, just not connected to the SVN for further updating afterward.

So would I be ok if I started a new game, with the most recent SVN after a version 40 release?

 

[Thunderbrd]

So would I be ok if I started a new game, with the most recent SVN after a version 40 release?

yeah, I mean you could basically start one now and it would be pretty close to what the v40 release will be.

 

[Osk21]

yeah, I mean you could basically start one now and it would be pretty close to what the v40 release will be.

Great!!! I will start over. I am very interested in Researching ALL technologies and completely taking a Mastery Victory to all overall victories. I realize that this could over a year and I want to play on snail or eternity. and I love winning!!! lol not particularly interested in losing a year long game. What is your recommendation for level and speed etc. I welcome comments from the whole Caveman2Cosmos team. Thanks again for all your efforts.

 

[Osk21]

Great!!! I will start over. I am very interested in Researching ALL technologies and completely taking a Mastery Victory to all overall victories. I realize that this could over a year and I want to play on snail or eternity. and I love winning!!! lol not particularly interested in losing a year long game. What is your recommendation for level and speed etc. I welcome comments from the whole Caveman2Cosmos team. Thanks again for all your efforts.

I would love to be able to Settle the Moons and planets in the same game.

 

[raxo2222]

Someone was reporting bug with blacksmith event or something like that - there was python error, that indicated deeper problem with code.
It was introduced recently.
Also only some buildings and techs spelling was fixed, as those can be found in multiple game text files including modules especially Pepper's module.

@billw2015 you should check it out.
Also there are memory issues, that were detected by coverity.

 

[billw2015]

Also only some buildings and techs spelling was fixed, as those can be found in multiple game text files including modules especially Pepper's module.

If someone wants to help then it would go faster! It will take me another week or so if I only do that, and I'm really not down for only doing text corrections for a week, it is quite boring!

 

[JosEPh_II]

If someone wants to help then it would go faster! It will take me another week or so if I only do that, and I'm really not down for only doing text corrections for a week, it is quite boring!

Agree your talents are needed elsewhere.

Now have GithubDesktop installed on new comp. I had Dev bat run but I did not realize you actually needed to start a game for the FPKs to build. So when I did the MakeFinalRelease it did not like it. lol I have it figured out now. Maybe....

 

[billw2015]

Someone was reporting bug with blacksmith event or something like that - there was python error, that indicated deeper problem with code.

So I put in a check in the Python code for the error condition. I'm not sure if it indicates a deeper problem in the code, or just a mistake in the Python event handler itself. However the fix is reasonable if no other symptoms are evident I think.

Also there are memory issues, that were detected by coverity.

I might knock some of those out today actually.

I had Dev bat run but I did not realize you actually needed to start a game for the FPKs to build. So when I did the MakeFinalRelease it did not like it.

Not sure I understand what you mean. MakeFinalRelease should not depend on anything, especially anything to do with FPKs, it should always work by itself. First time setup on new computer for using FinalRelease should just be sync, DevSetup.bat, MakeFinalRelease.bat. And from then on you just launch the game and it will update FPKs and DLL automatically when changes happen either by you or from updates from git.

 

[Toffer90]

So I put in a check in the Python code for the error condition. I'm not sure if it indicates a deeper problem in the code, or just a mistake in the Python event handler itself. However the fix is reasonable if no other symptoms are evident I think.

The other symptoms may be that quests cannot expire for any players ever, would be relatively difficult for players to notice if the expire code isn't working as it should, especially if they complete the quests before they expire as that omits the expire code in any case. Might be reported as "I completed a quest, then all the AI's completed the same quest one by one, I thought only one player could complete a quest". I don't know if the message log actually tells you when an AI player completes a quest.
Might be that it only informs you about your own quests, tells you that it has expired when another player completes it, but if it doesn't expire then the player would be none the wiser.

Firaxis was quite diligent at adding in such "valid input checks" where needed in python methods, but they never had a check for the player in quest event trigger points.
This means that C2C has made a change that result in the dll making unnecessary quest expire checks for null-players, or that the dll is incapable if reporting the player that an event should expire for when passing the task on to python.
The question is if this error can happen in new games, or if it only affects old games due to a specific technical change.

I will change your change to the blacksmith expire event to match the change I made for expire alternate energy quest event some months ago; for consistency reasons and to not hide potential deeper problems completely.

Spoiler expireAlternateEnergy :

def expireAlternativeEnergy1(argsList):

    CyPlayer = GC.getPlayer(argsList[1].ePlayer)

    # A player reported a 'NoneType' object has no attribute 'getBuildingClassCountWithUpgrades' exception in this function on SVN 11024.
    if not CyPlayer:
        print "[WARNING] CvRandonEventInterface.expireAlternativeEnergy1\n\tEVENTTRIGGER_ALTERNATIVE_ENERGY triggered for a non valid player (iPlayer not in range(51) == True)"
    else:
        if CyPlayer.getBuildingClassCountWithUpgrades(GC.getBuildingInfo(GC.getInfoTypeForString("BUILDING_GREAT_DAM")).getBuildingClassType()):
            return True
    return False

 

[raxo2222]

The other symptoms may be that quests cannot expire for any players ever, would be relatively difficult for players to notice if the expire code isn't working as it should, especially if they complete the quests before they expire as that omits the expire code in any case. Might be reported as "I completed a quest, then all the AI's completed the same quest one by one, I thought only one player could complete a quest". I don't know if the message log actually tells you when an AI player completes a quest.
Might be that it only informs you about your own quests, tells you that it has expired when another player completes it, but if it doesn't expire then the player would be none the wiser.

Firaxis was quite diligent at adding in such "valid input checks" where needed in python methods, but they never had a check for the player in quest event trigger points.
This means that C2C has made a change that result in the dll making unnecessary quest expire checks for null-players, or that the dll is incapable if reporting the player that an event should expire for when passing the task on to python.
The question is if this error can happen in new games, or if it only affects old games due to a specific technical change.

I will change your change to the blacksmith expire event to match the change I made for expire alternate energy quest event some months ago; for consistency reasons and to not hide potential deeper problems completely.

Spoiler expireAlternateEnergy :

def expireAlternativeEnergy1(argsList):

    CyPlayer = GC.getPlayer(argsList[1].ePlayer)

    # A player reported a 'NoneType' object has no attribute 'getBuildingClassCountWithUpgrades' exception in this function on SVN 11024.
    if not CyPlayer:
        print "[WARNING] CvRandonEventInterface.expireAlternativeEnergy1\n\tEVENTTRIGGER_ALTERNATIVE_ENERGY triggered for a non valid player (iPlayer not in range(51) == True)"
    else:
        if CyPlayer.getBuildingClassCountWithUpgrades(GC.getBuildingInfo(GC.getInfoTypeForString("BUILDING_GREAT_DAM")).getBuildingClassType()):
            return True
    return False

I got something like that for spreading corporations quest too.

 

[Toffer90]

I got something like that for spreading corporations quest too.

Yeah, it probably happens for all quest expire checks. Did this happen in a game that you started after 11024? That would tell us that it happens in new games too as this error was earliest reported by a player playing 11024.

But I'm adamant about adding this input check to all these quest expire methods, as I don't want to hide a potential deeper problem when the error message itself is completely benign, getting the error message has no consequences on the game.
It's like the assert one gets when running assert dll, and we don't want to hide away the assert messages when using the assert dll, the same way we don't want to hide away benign python errors that only shows up for players that have purposefully enabled them to show to actively help with debugging.

 

[raxo2222]

Yeah, it probably happens for all quest expire checks. Did this happen in a game that you started after 11024?

But I'm adamant about adding this input check to all these quest expire methods, as I don't want to hide a potential deeper problem when the error message itself is completely benign, getting the error message has no consequences on the game.
It's like the assert one gets when running assert dll, and we don't want to hide away the assert messages when using the assert dll, the same way we don't want to hide away benign python errors that only shows up for players that have purposefully enabled them to show to actively help with debugging.

I'm not sure - I loaded old save.

 

[Thunderbrd]

Ok, so are we ready for me to start packaging things up for release later today? And if I do, I assume that I'm taking it as basically a snapshot (copy) of the SVN version - without the svn related files - right?

 

[raxo2222]

Ok, so are we ready for me to start packaging things up for release later today? And if I do, I assume that I'm taking it as basically a snapshot (copy) of the SVN version - without the svn related files - right?

I'm checking resources - if there aren't any gaps in availability (those doesn't disrupt gameplay if building all buildings).
I found one thing:
Barrels, Copper Wares and Lead Wares are Enabled at Fermentation, Copper Working and Lead Working.
But you can actually make those wares (needed for barrels) at Metal Casting.
Can I move Lead Smith (makes Lead Wares) to Lead Working? It wouldn't need forge or foundry then though.

Then Barrel Maker could be unlocked at fermentation (currently needs Metal Casting too) - it needs Copper or Bronze or Lead Wares.

Spoiler :

I'm using my test mod but with unmodified buildings.

Buildings that needed barrels can use glass or pottery too.

Barrel Maker could have another OR prereq not just wares though.

 

[billw2015]

Also I am going to push the batch file + icons to SVN in a minute.

 

[Thunderbrd]

Can I move Lead Smith (makes Lead Wares) to Lead Working? It wouldn't need forge or foundry then though.

I suppose that makes sense. Lead can be worked by fingers when found if needbe. Doesn't quite require a forge exactly. And lead can be easily melted out of the stones, making it one of the first metals we worked, that and copper but I think it's even softer than copper so could make for a good exception to fix those other issues. And lead is otherwise such a double edged benefit so... yeah I agree with this. Not sure what others think here.

 

[billw2015]

And if I do, I assume that I'm taking it as basically a snapshot (copy) of the SVN version - without the svn related files - right?

Yeah, although we might want to repack the FPKs to get rid of the Patch.FPK. I will try and do that on my next commit as well.

 

[Thunderbrd]

Yeah, although we might want to repack the FPKs to get rid of the Patch.FPK. I will try and do that on my next commit as well.

OK, it will be later today that I'll be able to work on getting this packaged up. Hoping I do it right Anything else you think I should know?