I need to store 3D points in a database (SQLite using SQLAlchemy). Each 3D point set would contain ~1000 points and I have up to hundreds of point sets. I have been using JSON to convert each of the point X, Y, Z coordinates to a large string then storing the 3 strings in a single table.
from db.base import Base
from sqlalchemy import Column, Integer, String
class PointData(Base):
''' Raw point data, cannot store array in sqlite '''
#JSON strings
x_data = Column(String(), nullable=True)
y_data = Column(String(), nullable=True)
z_data = Column(String(), nullable=True)
The other option would be to create a table for each point in the point set - and I would end up with ~1000 tables. Which is a better method for database overall size and performance - large single tables or many more small tables?
Aucun commentaire:
Enregistrer un commentaire