obj_handler.py

import re

import numpy as np

def extract_data(file):
    """
    @brief: This function takes in a .obj file and returns a list containing
            the indexes of the vertexes that make up the face, and a dict
            that contains the index of the vertex, along with its coordinates

    @param: .obj file

    @ret  : verticies (the dict containing coordinates)
    @ret  : faces (the list containing the vertexes' indexes)
    """

    verticies = np.empty((1, 3))
    faces = []

    # Read more about how waveform (.obj) files are structured to understand
    # how this code exactly works, but shortly:
    #   * If the line starts with a "v", then that's a vertex and what follows is
    #     its X, Y, Z coordinates (foats)
    #   * If the line starts with a "f", then that's a face and what follows is
    #     the list of verticies to be connected to create a face
    #     (formatted a bit strangely though, I recommend checking an example)

    for line in file.readlines():
        if line[0:2] == "v ":
            pt = np.array([[float(x) for x in re.findall("[-+]?[.]?[\d]+(?:,\d\d\d)*[\.]?\d*(?:[eE][-+]?\d+)?", line)]])
            verticies = np.concatenate([verticies, pt])
        elif line[0:2] == "f ":
            faces.append([int(vertex.split("/")[0]) for vertex in line[2:-2].split(' ')])
    return verticies, faces