Package `mdtopx.io.malla`
Triangulated mesh (TIN) formats. Read into scene.meshes, one MeshEntity per mesh, containing mesh (vertices and triangles). Follow the common interface: info, read, write.
| Module | Format | Reads | Writes | Options |
|---|---|---|---|---|
mdt |
Native MDTopX MDT (.mdt) |
✔ | — | — |
stl |
Stereolithography (.stl) |
✔ | ✔ | write: ascii (False) |
obj |
Wavefront OBJ (.obj), writes textures |
✔ | ✔ | write: local_origin (True), texture_size (4096) |
Native MDTopX MDT
The mdt module reads MDTopX digital terrain models, .mdt files from any version: 3, used by MDTop, through the current version 9. Version 9 is version 8 with integrated textures appended. It reads only geometry:
.mdt contents |
In the scene |
|---|---|
| Terrain triangulation | scene.meshes[0] (id = 0) |
| Objects such as buildings, trees or bridges, each with its triangulation | scene.meshes[1:] (id = 1, 2…), in file order |
| Each mesh's triangle code | Mesh layer_code |
| Breakline codes | scene.layers, one per unique code |
| Each triangle's constrained sides | Triangle constraint1, constraint2, constraint3: index in scene.layers of the breakline code for sides a-b, b-c and c-a (-1 = none) |
| Model boundaries | scene.polylines |
Triangle indices a, b, c start at 0 and reference their own mesh's vertices. Deleted triangles are not loaded.
The reference system, contour settings, point LiDAR attributes, textures and associated database (.dbf) are not read.
This format declares no extension, so an
IoHostcannot find it. Always read it withmdt.read().
Example
from collections import Counter
from mdtopx.io.malla import mdt
escena = mdt.read(r"C:\datos\Triangulación de CAMPUS_SUR.mdt")
print("Mallas:", len(escena.meshes), "| límites:", len(escena.polylines))
print("Objetos por código:", Counter(m.layer_code for m in escena.meshes[1:]))
codigos = [capa.name for capa in escena.layers]
print("Códigos de ruptura:", codigos)
# Terrain. Lists are copied on each access: store them once in variables.
terreno = escena.meshes[0]
vertices = terreno.mesh.vertices
triangulos = terreno.mesh.triangles
print(f"Terreno: {len(vertices)} puntos, {len(triangulos)} triángulos")
zs = [v.z for v in vertices]
print(f"Cotas: {min(zs):.3f} a {max(zs):.3f}")
# Breakline sides by code
rupturas = Counter()
for t in triangulos:
for r in (t.constraint1, t.constraint2, t.constraint3):
if r >= 0:
rupturas[codigos[r]] += 1
print("Lados de ruptura:", dict(rupturas))
# Model boundaries
for limite in escena.polylines[:3]:
print(f"Límite {limite.layer_code}: {len(limite.vertices)} vértices, "
f"cerrado={limite.closed}, perímetro={limite.perimeter:.2f} m")
To display progress when reading a large file, pass a function in progress:
escena = mdt.read(ruta, progress=lambda f: print(f"\r{f:.0%}", end="") or True)
STL
The stl module automatically detects and reads binary or ASCII STL files. It writes binary, or ASCII with ascii=True, exporting all scene meshes.
Save .mdt terrain as STL:
import mdtopx
from mdtopx.io.malla import mdt, stl
escena = mdt.read(r"C:\datos\Terreno.mdt")
solo_terreno = mdtopx.Scene()
solo_terreno.meshes = [escena.meshes[0]]
stl.write(r"C:\datos\Terreno.stl", solo_terreno)
Wavefront OBJ
The obj module reads an OBJ mesh: vertices and faces, splitting polygons into triangles, ignoring textures, normals and materials, and merging groups into one mesh. It writes scene meshes with georeferenced plan images from scene.images as textures: an .obj, an .mtl and a PNG per texture tile. This matches MDTopX's Export OBJ with textures:
- Images are split into tiles no larger than
texture_sizepixels; triangles crossing tile boundaries are split there. Where images overlap, the last inscene.imagestakes precedence except where transparent (has_alpha). - With
local_origin=True, vertices are written relative to the mesh minimum rounded to the meter, recorded in the.objline# origen X Y Zand<nombre>_origen.txt. Reading an OBJ with this origin adds it back. - Z points upward.
Export .mdt terrain with an orthophoto texture:
import mdtopx
from mdtopx.io.malla import mdt, obj
from mdtopx.io.image import tiff
escena = mdtopx.Scene()
escena.meshes = [mdt.read(r"C:\datos\Terreno.mdt").meshes[0]]
escena.images = tiff.read(r"C:\datos\orto.tif").images # with georeferencing stored in the GeoTIFF
obj.write(r"C:\datos\Terreno.obj", escena)