Marko Lindqvist wrote:
> On 25 April 2013 02:40, Emmet Hikory wrote:
> > Looking at the UTYF_BOMBARDER nativity issue, I notice that GNA patches
> > 1850 (unreachableprotects) and 1851 (capturer/capturable) seem to have
> > surrounded the prior bombarder code, with the effect that Unreachable
> > units do not protect other units from being captured or bombarded,
> > regardless of the setting.
> >
>  After so long time I can't say how long they had been in development
> before being first submitted (timestamps in patches indicate that "final"
> versions were created just before submitting, but that's the case with
> almost all my patches). Given my usual workflow it's entirely possible that
> despite them having consequtive patch numbers, they have been developet
> separately even without realizing they could affect each other.

    Heh.  At least it's recent enough that the patch author is active on 
the mailing list :) Given the uncertainty, I'll consider the relation 
between capturable and unreachableprotects to be entirely coincidental. 
I want to look at the implications a bit more, but will probably submit 
a patch moving the capturer/capturable check inside the 
unreachableprotectsr+alliance guard (we don't really want to be capturing 
allied freight, etc.).

>  Addition of "unreachableprotects" setting certainly was meant to keep old
> hardcoded behavior when enabled for compatibility of old rulesets.

    I've just read through the ML traffic about PR#8455, but suspect 
I'll have to do that again (gmane threading doesn't work so well on deep 
archives) to determine the intent of the commit (as the intent of the 
original patch changed significantly during patch review).  I suspect that
I'll end up wanting to rewrite bombardment entirely, rather than just trying
to fix the nativity quickly, as there are so many deep assumptions buried
in the logic and functional composition.  Depending on the outcome of all
of this, it may or may not be moved inside the unreachableprotects guard.


Freeciv-dev mailing list

Reply via email to