Complete.Org: Mailing Lists: Archives: freeciv-dev: June 2005:
[Freeciv-Dev] Re: (PR#13342) pubserver 2.0 crash in found_new_tech
Home

[Freeciv-Dev] Re: (PR#13342) pubserver 2.0 crash in found_new_tech

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
Subject: [Freeciv-Dev] Re: (PR#13342) pubserver 2.0 crash in found_new_tech
From: "Jason Short" <jdorje@xxxxxxxxxxxxxxxxxxxxx>
Date: Mon, 27 Jun 2005 15:10:28 -0700
Reply-to: bugs@xxxxxxxxxxx

<URL: http://bugs.freeciv.org/Ticket/Display.html?id=13342 >

Mateusz Stefek wrote:
> <URL: http://bugs.freeciv.org/Ticket/Display.html?id=13342 >
> 
>>[jdorje - Mon Jun 27 19:34:06 2005]:
>>
>>I reverted the 13198 patch in S2_0.
>>
>>This should (once the pubservers rebuild) prevent future crashes of this
>>type.  However it means the (relatively minor) 13198 bug is back.
>>
>>I'm not sure if the 13198 patch should be reverted in the dev branch
>>also, or if we should just look for a better fix.
> 
> The problems with those crashes were that after choose_random_tech()
> player.research.researching was still A_UNSET.
> As I can see this is the main reason to call that function in S2_0
> 
> I think you should commit the 13198 back into S2_0 and change
> choose_random_tech() to be similar to it's HEAD version:
> 
> choose_random_tech()
> {
>   do {
>   choose(random_tech)
>   } while(player researching A_UNSET)
> }

That would work but it would give somewhat wrong behavior - maybe better
or worse than the current somewhat wrong behavior, I don't know.

Current somewhat wrong behavior: when server chooses new tech
automatically on end-of-turn, bulbs are lost.

New somewhat wrong behavior: Above bug is fixed.  But when a hut is
investigated the new tech is given (for free) properly, but some other
techs may also be given (and paid for) incorrectly.

-jason





[Prev in Thread] Current Thread [Next in Thread]