step-bmi-model.py

"""Step through a DiffusionModel run using only BMI calls."""

import numpy as np

from diffusion import BmiDiffusionModel

CONFIG_FILE = "config.yaml"


print("Create an instance of BmiDiffusionModel.")
m = BmiDiffusionModel()
print(m.get_component_name())

print("Initialize the model.")
m.initialize(CONFIG_FILE)
print(CONFIG_FILE)

print("List the model's exchange items.")
print("Number of input vars:", m.get_input_item_count())
for var in m.get_input_var_names():
    print(f" - {var}")
print("Number of output vars:", m.get_output_item_count())
for var in m.get_output_var_names():
    print(f" - {var}")

var_name = m.get_output_var_names()[0]
print(f"Info for variable {var_name}")
print(" - variable type:", m.get_var_type(var_name))
print(" - units:", m.get_var_units(var_name))
print(" - itemsize:", m.get_var_itemsize(var_name))
print(" - nbytes:", m.get_var_nbytes(var_name))
print(" - location:", m.get_var_location(var_name))

print("Get grid info.")
grid_id = m.get_var_grid(var_name)
print(" - grid id:", grid_id)
print(" - grid type:", m.get_grid_type(grid_id))
grid_rank = m.get_grid_rank(grid_id)
print(" - rank:", grid_rank)
grid_size = m.get_grid_size(grid_id)
print(" - size:", grid_size)
grid_shape = np.empty(grid_rank, dtype=np.int32)
try:
    m.get_grid_shape(grid_id, grid_shape)
except RuntimeError:
    print(" - shape: n/a")
else:
    print(" - shape:", grid_shape)
grid_spacing = np.empty(grid_rank, dtype=np.float64)
try:
    m.get_grid_spacing(grid_id, grid_spacing)
except RuntimeError:
    print(" - spacing: n/a")
else:
    print(" - spacing:", grid_spacing)
grid_origin = np.empty(grid_rank, dtype=np.float64)
try:
    m.get_grid_origin(grid_id, grid_origin)
except RuntimeError:
    print(" - origin: n/a")
else:
    print(" - origin:", grid_origin)

print("Get time information from the model.")
print(" - start time:", m.get_start_time())
print(" - end time:", m.get_end_time())
print(" - current time:", m.get_current_time())
print(" - time step:", m.get_time_step())
print(" - time units:", m.get_time_units())

print(f"Get initial values of {var_name}...")
val = np.empty(grid_size, dtype=m.get_var_type(var_name))
m.get_value(var_name, val)
print(f" - values at time {m.get_current_time()}, plus sum:")
print(val, val.sum())

print("Advance model by a single time step.")
m.update()
print(" - new time:", m.get_current_time())

print("Get new values of {}...".format(var_name))
val = np.empty(grid_size, dtype=m.get_var_type(var_name))
m.get_value(var_name, val)
print(f" - values at time {m.get_current_time()}, plus sum:")
print(val, val.sum())

print("Advance model to a later time.")
while m.get_current_time() < 10 * m.get_time_step():
    m.update()
print(" - new time:", m.get_current_time())

print("Get new values of {}...".format(var_name))
val = np.empty(grid_size, dtype=m.get_var_type(var_name))
m.get_value(var_name, val)
print(f" - values at time {m.get_current_time()}, plus sum:")
print(val, val.sum())

print("Finalize the model.")
m.finalize()
print("Done.")