Package `mdtopx.io`
The mdtopx.io package provides access to MDTopX's file formats. Each format is a plugin, a DLL, with its own Python module in a subpackage according to data type:
| Subpackage | Data type |
|---|---|
mdtopx.io.pointcloud |
Point clouds |
mdtopx.io.vector |
Drawing |
mdtopx.io.raster |
Elevation grids (DEM) |
mdtopx.io.malla |
Triangulated meshes |
mdtopx.io.image |
Color images |
mdtopx.io.model |
BIM models |
These are the same plugins loaded by MDTopX, so files are read identically from Python and the program.
Common format interface
All format modules share this interface:
| Function | Action |
|---|---|
info() |
Returns a FormatInfo: name, extensions, can_read, can_write and category. |
read(path, *, scene=None, progress=None, …) |
Reads the file into a Scene. Creates a new scene if scene is omitted; otherwise adds contents to the existing one. Returns the scene. |
write(path, scene, *, progress=None, …) |
Writes the scene to the file. |
Read-only formats have no write; write-only formats have no read. Some formats accept specific keyword options, such as decimals, documented on each subpackage's page.
read and write raise IOError on failure and FileNotFoundError if the format plugin is missing.
from mdtopx.io.pointcloud import las
from mdtopx.io.vector import dxf
# Read two files into the same scene
escena = las.read(r"C:\datos\nube.laz")
dxf.read(r"C:\datos\ejes.dxf", scene=escena)
print(len(escena.clouds), "nubes y", len(escena.polylines), "polilíneas")
Package functions
| Function | Action |
|---|---|
load_format(name) |
Loads plugin plugin_<name> and returns its Format (info, read, write). Each plugin loads once per process. Used internally by format modules. |
plugin_dir() |
Plugin search folder: MDTOPX_PLUGINS environment variable, or mdtopx/io/plugins if unset. |
host(folder=None) |
Returns an IoHost with all plugins in the folder loaded. |
default_host() |
Shared IoHost with all package plugins, created on first use. |
divide(input_path, params, tile_name, *, progress=None) |
Splits a point cloud file, described below. |
Read a file without knowing its format
An IoHost selects the plugin by file extension and, when several formats share an extension, by contents. read and write return a Status instead of raising an exception:
import mdtopx
from mdtopx.io import default_host
host = default_host()
for f in host.formats():
print(f.name, f.extensions)
escena = mdtopx.Scene()
if host.read(r"C:\datos\terreno.grdasc", escena) == mdtopx.Status.Ok:
print(escena.grids[0].grid.rows, "filas")
Some formats cannot be opened this way because they declare no extension. For example, read native MDTopX MDT through its module.
Split point clouds
divide splits a point cloud file into several files without loading the entire cloud into memory, reading and writing in chunks. params is a DivideParams; its mode (DivideMode) selects the criterion:
| Mode | Criterion |
|---|---|
Sequential |
Point count (max_points_per_tile). |
Grid |
Grid sheets (origin_x, origin_y, sheet_width, sheet_height). |
Altitude |
Elevation ranges (z_min, vertical_step). |
GpsInterval |
GPS time intervals (gps_start, gps_interval). |
Surfaces |
Surfaces (surfaces, a list of DivideSurface). |
Pk |
Chainage along an axis (axis, pk_step, pk_initial). |
tile_name receives a TileInfo (index and, in Surfaces and Pk modes, pk_start and pk_end) and returns the tile's path. Its extension determines the output format, which can differ from the input format.
Both the input file and tiles must use a format supporting chunked reading and writing: LAS, LAZ or E57. For other formats,
dividereturns a result withstatusset toInvalidData.
import mdtopx
from mdtopx.io import divide
params = mdtopx.DivideParams()
params.mode = mdtopx.DivideMode.Grid
params.origin_x, params.origin_y = 440000, 4470000
params.sheet_width = params.sheet_height = 500
resultado = divide(r"C:\datos\vuelo.laz", params,
lambda info: rf"C:\datos\hojas\hoja_{info.index:04d}.laz")
print(resultado.tiles, "hojas,", resultado.points, "puntos")