Trajectory (inherits CellularSpace)
Type that defines an ordered selection over a CellularSpace. It inherits CellularSpace; therefore it is possible to apply all functions of such type to a Trajectory. For instance, calling forEachCell() also traverses Trajectories.
Arguments
- build: A boolean value indicating whether the Trajectory should be computed when created. The default value is true.
- greater: A function (Cell, Cell)->boolean to sort the Trajectory. Such function must return true if the first Cell has priority over the second one. When using this argument, Trajectory compares each pair of Cells to establish an execution order to be used by forEachCell(). As default, the Trajectory will not be ordered and so forEachCell() will run in the order the Cells were pushed into the CellularSpace. See greaterByAttribute() for predefined options for this argument.
- random: A boolean value indicating that the Trajectory must be shuffled. The Trajectory will be shuffled every time one calls rebuild() or when the Trajectory is an action of an Event. This argument cannot be combined with argument greater.
- select: A function (Cell)->boolean indicating whether an Cell of the CellularSpace should belong to the Trajectory. If this function returns anything but false or nil for a given Cell, it will be added to the Trajectory. If this argument is missing, all Cells will be included in the Trajectory.
- target: The CellularSpace over which the Trajectory will take place.
Attributes
Some attributes of Trajectory have internal semantics. They can be used as read-only by the modeler.
- cells: A vector of Cells pointed by the Trajectory.
- cObj_: A pointer to a C++ representation of the Trajectory. Never use this object.
- greater: The last function used to sort the Trajectory.
- parent: The CellularSpace where the Trajectory takes place.
- select: The last function used to filter the Trajectory.
Usage
cell = Cell{
cover = Random{"forest", "deforested"},
dist = Random{min = 0, max = 50}
}
cs = CellularSpace{
xdim = 10,
instance = cell
}
traj = Trajectory{
target = cs,
select = function(c)
return c.cover == "forest"
end,
greater = function(c, d)
return c.dist < d.dist
end
}
traj = Trajectory{
target = cs,
greater = function(c, d)
return c.dist < d.dist
end
}
traj = Trajectory{
target = cs,
build = false
}Functions
| add | Add a new Cell to the Trajectory. |
| clear | Remove all Cells from the Trajectory. |
| clone | Return a copy of the Trajectory. |
| filter | Apply the filter over the Cells of the Trajectory. |
| get | Return a Cell from the Trajectory given its x and y locations. |
| randomize | Randomize the Cells of the Trajectory. |
| rebuild | Rebuild the Trajectory. |
| save | Save a subset from the target CellularSpace into a file. |
| sort | Sort the current CellularSpace subset. |
| # | Retrieve the number of Cells in the Trajectory. |
add
Add a new Cell to the Trajectory. It will be added to the end of the list of Cells.
Arguments
- #1: A Cell.
Usage
cs = CellularSpace{
xdim = 10
}
traj = Trajectory{
target = cs,
select = function(c)
return c.x > 3
end
}
traj:add(cs:get(1, 1))clear
Remove all Cells from the Trajectory.
Usage
cs = CellularSpace{
xdim = 10
}
traj = Trajectory{
target = cs
}
traj:clear()
print(#traj)clone
Return a copy of the Trajectory. It has the same parent, select, greater and Cells. Any change in the cloned Trajectory will not affect the original one.
Usage
cell = Cell{
cover = Random{"forest", "deforested"}
}
cs = CellularSpace{
xdim = 10,
instance = cell
}
traj = Trajectory{
target = cs,
select = function(c)
return c.cover == "forest"
end
}
copy = traj:clone()
print(#copy)
print(#traj)filter
Apply the filter over the Cells of the Trajectory. Cells that belong to the CellularSpace but do not belong to the Trajectory are ignored. This way, this function creates a subset over the subset of the CellularSpace.
Usage
cell = Cell{
dist = Random{min = 0, max = 50},
increase = function(self)
self.dist = self.dist + 2
end
}
cs = CellularSpace{
xdim = 10,
instance = cell
}
traj = Trajectory{target = cs, select = function(c)
return c.dist > 20
end}
traj:increase()
traj:filter()get
Return a Cell from the Trajectory given its x and y locations. If the Cell does not belong to the Trajectory then it will return nil.
Arguments
- #1: The x location.
- #2: The y location.
Usage
cs = CellularSpace{xdim = 10}
traj = Trajectory{target = cs}
traj:get(1, 1)randomize
Randomize the Cells of the Trajectory. It will change the traversing order used by forEachCell().
Usage
cs = CellularSpace{xdim = 10}
traj = Trajectory{target = cs}
traj:randomize()rebuild
Rebuild the Trajectory. It works as if the Trajectory was declared again with the same arguments.
Usage
cell = Cell{
dist = Random{min = 0, max = 50}
}
cs = CellularSpace{
xdim = 10,
instance = cell
}
traj = Trajectory{
target = cs,
select = function(cell) return cell.dist < 20 end,
greater = function(c1, c2) return c1.dist < c2.dist end
}
print(#traj)
forEachCell(cs, function(cell)
cell.dist = cell.dist + 10
end)
traj:rebuild()
print(#traj)save
Save a subset from the target CellularSpace into a file.
Arguments
- #1: A File which can be a .shp or .geojson extension.
- #2: A vector with the names of the attributes to be saved. If attrs is nil, all attributes will be saved.
Usage
cs = CellularSpace{
file = filePath("test/sampa.shp", "gis")
}
t = Trajectory{
target = cs,
select = function(cell)
return cell.ID % 2 == 0
end
}
t:save("odd.shp")sort
Sort the current CellularSpace subset. It updates the traversing order of the Trajectory.
Usage
cell = Cell{
dist = Random{min = 0, max = 50},
increase = function(self)
self.dist = self.dist + Random{min = 0, max = 3}:sample()
end
}
cs = CellularSpace{
xdim = 10,
instance = cell
}
traj = Trajectory{target = cs, greater = function(c, d)
return c.dist < d.dist
end}
traj:increase()
traj:sort()#
Retrieve the number of Cells in the Trajectory.
Usage
cs = CellularSpace{
xdim = 10
}
traj = Trajectory{
target = cs
}
print(#traj)