[Freeciv-Dev] Re: (PR#10061) assertion "punit->transported_by > 0" failed

["Marko Lindqvist" <marko.lindqvist@xxxxxxxxxxx>]

<URL: http://rt.freeciv.org/Ticket/Display.html?id=10061 >

Jason Short wrote:
> <URL: http://rt.freeciv.org/Ticket/Display.html?id=10061 >
> 
> Marko Lindqvist wrote:
> 
>><URL: http://rt.freeciv.org/Ticket/Display.html?id=10061 >
>>
>>  I managed to reproduce this with plain CVS server.
>>
>>set timeout=-1
>>create Hammurabi
>>set aifill=7
>>set randseed=1094968423
>>set seed=1819942531
>>
>>  Crashes somewhere between years 1500 and 1600.
> 
> 
> I made it through 1750 with this autogame.  Are you sure you're using 
> up-to-date CVS code to reproduce it?  Even a small change will cause 
> autogames to differ and make us unable to reproduce it.

  I know, that's why I mentioned even manually created player name (it 
determines nation for first player -> affects set of available nations 
when rand() selecting nations for others.)
  I eventually managed to reproduce this from later savegame too, I'll 
attach save as comment to RT. But it's possible that actual bug is 
already happened before saving and save itself is buggy.

   But if we know
> the exact date of the CVS version that is used to reproduce it, we can do a
> 
>    cvs up -D "Sun Sep 12 01:57:58 EDT 2004"

  That automatically generated version-string part would be nice now. 
Hopefully it eventually becomes reality in #4712.
  My 'master' CVS directory (one I update with CVS and from which diff + 
patch updates other directories) seems to be from "Sep 11 17:51 EEST". 
So that date & time is my best bet for now. I can look into this in more 
detail later.

> and then find the bug before fixing it in the HEAD version (assuming it 
> still applies).


  - Caz