VEC: Fix the inRange function

It now takes a PROPER square, as in ((xmin, ymin), (xmax, ymax))
instead of ((xmin, xmax), (ymin, ymax)) and also works
with negative values.

Because the meaning of the arguments has changed, we also
had to fix all uses of it.

Algebra/Vector.hs

@@ -24,12 +24,24 @@ data Alignment = CW
   deriving (Eq)
 
 
--- |Checks whether the Point is in a given dimension.
-inRange :: Square -- ^ the square: ((xmin, xmax), (ymin, ymax))
+-- |Convert two dimensions such as (xmin, xmax) and (ymin, ymax)
+-- to proper square coordinates, as in:
+-- ((xmin, ymin), (xmax, ymax))
+dimToSquare :: (Double, Double) -- ^ x dimension
+            -> (Double, Double) -- ^ y dimension
+            -> Square           -- ^ square describing those dimensions
+dimToSquare (x1, x2) (y1, y2) = ((x1, y1), (x2, y2))
+
+
+-- |Checks whether the Point is in a given Square.
+inRange :: Square -- ^ the square: ((xmin, ymin), (xmax, ymax))
         -> PT     -- ^ Coordinate
         -> Bool   -- ^ result
-inRange ((xmin, xmax), (ymin, ymax)) p =
-  x <= xmax && x >= xmin && y <= ymax && y >= ymin
+inRange ((xmin, ymin), (xmax, ymax)) p
+  = x >= min xmin xmax
+    && x <= max xmin xmax
+    && y >= min ymin ymax
+    && y <= max ymin ymax
   where
     (x, y) = unp2 p