[freeciv-ai] Re: better easy mode

[Raimar Falke <rf13@xxxxxxxxxxxxxxxxx>]

On Thu, Oct 31, 2002 at 01:23:00PM +0000, Per I. Mathisen wrote:
> On Thu, 31 Oct 2002, Raimar Falke wrote:
> > > This is only a sanity check since it should already have this role. (The
> > > current AI has a problem with forgetting roles.) In my massive ai tree
> > > this problem has been fixed and the above line has been replaced by
> > >
> > >     assert(punit->ai.ai_role == AIUNIT_ESCORT);
> >
> > From my short reading there my be a problem with bodyguard assignment.
> 
> No, I used to have a LOG_ERROR there in massive ai patch for a while. It
> never happened.
> 
> > > Your patch will now be assimilated into my tree...
> >
> > Bad. It should instead be applied.
> 
> You mean you really want the duplication of ai_unit_new_role() ??
> 
> That is crazy.
> 
> > > ...and then I'll try to [back]port these changes as a separate patch.
> 
> See the attached patch. You wanted the massive ai patch split up, well,
> here is one piece of it... please read & comment.

My patch is there to remove a bug. I can and I think also everybody
else prove that it does what it intends. It may be ugly to have two
wrappers but if I don't know what the comment "/* do not use
ai_unit_new_role() */" I better play save than sorry for a bug fixing
patch.

You change behaviour in a lot of places in your patch. You can do this
but than it isn't a bug fix patch anymore. And IMHO there is no good
reason to merge the adding of the wrapper with the ai_fuzzy or the
Gilligan's Island change. These changes are no bad per se but they
should be in a seperate patch and seperate commit. Same for the #ifdef
DEBUG change.

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
 "The very concept of PNP is a lovely dream that simply does not translate to
  reality. The confusion of manually doing stuff is nothing compared to the
  confusion of computers trying to do stuff and getting it wrong, which they
  gleefully do with great enthusiasm." 
    -- Jinx Tigr in the SDM