I fully agree with Jacob. I thought that is why we have seperate
Terrain/Object/Airport folders in the first place...
On 03/17/2012 11:15 PM, Jacob Burbach wrote:
> On Sat, Mar 17, 2012 at 4:57 PM, Martin Spott<Martin.Spott@...> wrote:
>> Anders Gidenstam wrote:
>>> While the (old) new behaviour is as Martin expects it is not what most
>>> that has read Docs/README.scenery would expect (I'd think). However, I
>>> would not consider Docs/README.scenery a normative source (i.e. not how
>>> it should be but rather how it was) - but Martin's expectation (as far as
>>> I know about it) is also not the behaviour I'd like or want.
>> Well, to put it a bit harsh, if you ask every user, you'll probably
>> have to implement two dozend exceptions from the rule. This doesn't
>> make neither the developer happy nor the users, because it's getting
>> too confusing - and the same users will start complaining again ....
>> Let me explain it this way:
>> Nowadays we/you are confronted with a steadily growing number of
>> private Sceneries. There might be a scenario for France, one for Great
>> Britain, one for Germany and, because everybody needs to have their own
>> Scenery kingdom and these were the only ones left, one for BeNeLux and
>> one for Denmark. Because it's easier to maintain, the boundaries of
>> all these scenarios are aligned with full degrees. All these scenarios
>> are added to the scenery path.
>> Now think of an item which is situated somewhere in the German Bight in
>> a tile which doesn't have terrain, how often is every single windmill,
>> oil rig, buoy, light ship or whatever else going to be drawn in
>> FlightGear ? ;-)
>> Therefore I'd say the only really consistent way of dealing with the
>> scenery path is to stop searching the path whenever a terrain _or_ an
>> object tile is found in a scenery path item.
> Seems logical to me that objects and terrain should be considered
> separately. The loader should not stop looking for an object bucket
> just because it finds a terrain bucket...and vice versa. The loader
> knows what bucket it is looking for, so should just need to look
> through each path until it finds the first existing bucket of each
> type (or none) and load the data for that bucket/type as appropriate.
> Once it has found a particular type of bucket it no longer looks for
> it anymore, but continues looking for the other types until it finds
> one, or not. Seems simple enough to me, avoids any problem with things
> loading twice, and gives a very logical and expected behavior for
> This SF email is sponsosred by:
> Try Windows Azure free for 90 days Click Here
> Flightgear-devel mailing list