#!/usr/bin/env python
# coding: utf-8

# # Measuring X as a function of time
# 
# Sometimes we'd like to measure something as a function of elapsed wall clock time. QCoDeS provides a convenient default way of doing such a measurement, namely by using the `ElapsedTimeParameter`.
# 
# The main utility of having a default way of measuring time is the uniformity in data of different experiments.

# In[1]:


from pathlib import Path

import numpy as np

from qcodes.dataset import (
    Measurement,
    initialise_or_create_database_at,
    load_or_create_experiment,
    plot_dataset,
)
from qcodes.parameters import ElapsedTimeParameter, Parameter


# ### Prepatory footwork: setup database and experiment

# In[2]:


initialise_or_create_database_at(
    Path.cwd().parent / "example_output" / "x_as_a_function_of_time.db"
)
load_or_create_experiment("tutorial", "no_sample")


# ## The measurement itself
# 
# We'll measure some Brownian motion. We set up a parameter for the noise.

# In[3]:


noise = Parameter(
    "noise", label="Position", unit="m", get_cmd=lambda: np.random.randn()
)
time = ElapsedTimeParameter("time")


# In[4]:


meas = Measurement()
meas.register_parameter(time)
meas.register_parameter(noise, setpoints=[time])


# In[5]:


with meas.run() as datasaver:
    pos = 0
    time.reset_clock()
    for _ in range(100):
        pos += noise()
        now = time()
        datasaver.add_result((noise, pos), (time, now))

dataset = datasaver.dataset


# In[6]:


axs, cbs = plot_dataset(dataset)


# In[ ]: