Eliminate MIN_TILE_NUM

[Robert Leach]

So I was just looking at the new "MIN_TILE_NUM" related behavior (which is stored as a double). I noticed that if I try to zoom down to a single cell, it zooms all the way in and then bounces back out to the minimum.

First, I think we should remove MIN_TILE_NUM all together, as per the decision in the discussion in issue #65. Based on feedback that Anastasia was gathering, we decided for consistency's sake to always zoom completely down to the selection.

Second, if we would like to keep code around to possibly use a minimum zoom level, I suggest we use my code, since it stops the zoom before it zooms too far. Besides that, it already uses an int.

Please take a look @TreeView3Dev. I am going to set it to 1.0 right now, but I think removing it should be on our TODO. I will add a TODO tag above it.

[Christopher Keil]

Yes I agree it should be removed if there's no minimum number of tiles that we set. Also, do we really need 2 different variables that store the current number of tiles? (numVisible, tileNumVisible)

And shouldn't the variable that stores the current number of all tiles definitely be an integer? The calculations which currently assume for it to be a double would then need to be adjusted (e.g. using a cast)

[Robert Leach]

To address your question @TreeView3Dev, yes, we need the two variables because tileNumVisible changes at different times. It was preventing me from basing the window-rezoom upon window size change on tileNumVisible. If we can remove tileNumVisible and use numVisible instead, that would be OK, but if we do that, we cannot set numVisible in places where tileNumVisible is being set because it will mess up the window resize updates.

[Robert Leach]

There's a bit more to this issue than simply removing TILE_NUM_VISIBLE. I think that keeping the minimum zoom level code I implemented might be worthwhile for future features. Thus, I've moved this to alpha03.

[Robert Leach]

Removing milestone: Coding Convention Adherence (automated comment)