# filters.splitter¶

The splitter filter breaks a point cloud into square tiles of a size that you choose. The origin of the tiles is chosen arbitrarily unless specified as an option.

The splitter takes a single PointView as its input and creates a PointView for each tile as its output.

Splitting is usually applied to data read from files (which produce one large stream of points) before the points are written to a database (which prefer data segmented into smaller blocks).

Default Embedded Stage

This stage is enabled by default

## Example¶

{
"pipeline":[
"input.las",
{
"type":"filters.splitter",
"length":"100",
"origin_x":"638900.0",
"origin_y":"835500.0"
},
{
"type":"writers.pgpointcloud",
"connection":"dbname='lidar' user='user'"
}
]
}


## Options¶

length
Length of the sides of the tiles that are created to hold points. [Default: 1000]
origin_x
X Origin of the tiles. [Default: none (chosen arbitarily)]
origin_y
Y Origin of the tiles. [Default: none (chosen arbitarily)]
buffer
Amount of overlap to include in each tile. This buffer is added onto length in both the x and the y direction. [Default: 0.0]