On Wed, Jul 25, 2007 at 03:35:46PM +1200, Ralph Versteegen wrote:
> On 7/25/07, James Paige <[EMAIL PROTECTED]> wrote:
> > On Tue, Jul 24, 2007 at 09:32:29PM +1200, Ralph Versteegen wrote:
> > > On 7/24/07, [EMAIL PROTECTED]
> > > <[EMAIL PROTECTED]> wrote:
> > > > james
> > > > 2007-07-23 09:49:20 -0700 (Mon, 23 Jul 2007)
> > > > 300
> > > > Spell learning fixes. Resolves bug 410
> > > > * Learned spells are properly eported after battle (missing HeroDef 
> > > > conversion)
> > > > * Fixed unreported bug that prevented spells from being learned if the 
> > > > levelup bug bitset was on
> > > > * Fixed broken spell-forgetting when experience has been lowered with 
> > > > plotscripting
> > > > ---
> > > > U   wip/bmod.bas
> > > > U   wip/bmodsubs.bas
> > >
> > > Besides cleanup, I don't see anything that actually changed
> > > (especially to do with reporting after battles). Were the offsets into
> > > buffer() wrong? Were the calls to loadattackdata and redbadbinstring
> > > not equivalent to readattackname?
> >
> > buffer was empty (or rather, it contained old data unelated to heroes)
> > Most of the code had been updated to use a HeroDef instance rather than
> > data in buffer, but half the code for spell learning and all the code
> > for spell forgetting had been missed.
> >
> > As for the no-learning-when-level-up-bug-enabled, that one was a much
> > older bug caused by the fact that when the level-up bug is enabled it
> > would load equipment data into buffer, clobbering the hero data that was
> > needed later for learning.
> >
> 
> But, hang on! Fenrir reported he'd first seen the bug on the 2nd,
> while Mike didn't introduce the HeroDef UDT until the 19th. I guess
> you fixed the actual bugs Fenrir reported in other commits, and this
> was just fixes to bugs introduced since then?
> 
> TMC

Yeah, Actually, The code was a real mess since the dawn of time. I 
actually fixed the endless loop part in an earlier commit (1205). The 
incomplete application of the HeroDef to that code was only part 
of this bug.

The *real* bug here is that back in 1998 I had no concept of the 
importance of separating model and view, so I wrote some very wonky 
end-of-battle code :)

---
James Paige
_______________________________________________
Ohrrpgce mailing list
ohrrpgce@lists.motherhamster.org
http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org

Reply via email to