Follow-up Comment #4, bug #22876 (project freeciv):
> If it's too hard to fix run-time propagation properly on 2.4 or
> 2.5 [...] then I think reverting to the previous buggy
> load-time propagation is a reasonable fallback position.
Agreed.
> (We can do that while keeping the speedup of patch #5357,
> right?)
It might need some change. In the example with tech A [root_req=B], tech C
[root_req=D] and tech E [requires A and C], player_invention_recheable()
currently (after patch #5357 applied) tests if B and D are known. With
reverting partially the patch, it will still test if both are known, but
root_req of E will be only B xor D. I suspect it could have some side
effects.
_______________________________________________________
Reply to this item at:
<http://gna.org/bugs/?22876>
_______________________________________________
Message posté via/par Gna!
http://gna.org/
_______________________________________________
Freeciv-dev mailing list
[email protected]
https://mail.gna.org/listinfo/freeciv-dev
