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

# # Basic Example
# 
# This notebook walks you through some of the high level aspects of the Planetoid API

# In[2]:


import pandas as pd
import numpy as np
from matplotlib import colors
from sklearn.datasets import make_blobs
import planetoids as pt
np.random.seed(42)


# Using scikit-learn, we will generate some random data that we will use to seed a Planetoid

# In[3]:


X, y = make_blobs(n_samples=200,
                  n_features=2,
                  centers=3,
                  cluster_std=1.1,
                  center_box=(0,10),
                  random_state=42)
data = pd.DataFrame(X)
data['Cluster'] = y

#plot
data.plot(kind='scatter', x=0, y=1, c='Cluster', cmap='tab10')


# ## Seed the planet

# In[4]:


blob_planet = pt.Planetoid(data, x=0, y=1, cluster_field='Cluster')


# ## Fit & Terraform in a single step
# 
# This will generate all of the data needed to terraform and render our `blob_planet` but calling `fit` and then `terraform`.

# In[5]:


blob_planet.fit_terraform(planet_name='fit_terraform demo')


# # Several high level controls are available to you to change how your world will be terraformed

# ## Ecology
# 
# Ecology is one of the simplest changes you can make that produces dramatic difference in the look and feel of a Planetoid. Ecology is considered a key seed element for your planet, as such, you will have to explicitly set it using its own method if you want to change it.
# 
# ## Topography Levels
# 
# This controls the level of detail in the generated land mass contours. Lower values will produce fewer contours and consequently shallower land masses, higher values will help land masses rise out of the 'ocean'. This can result in fundamental shifts in the generated ecology.
# 
# ## Lighting Levels
# 
# This controls the level of detail in the lighting effect. This is meant to add an additional level of depth to your Planetoid. Terraforming allows you to selectively choose whether or not to show the hillshade and highlight independantly of one another.
# 
# ## Relief Density
# 
# This controls the level of detail in the relief. These lines represent the gradient of the generated land mass physical features.
# 
# ## Projection
# 
# This is the map projection that is used to render the planetoid. The default 'orthographic' will produce a 3D sphere effect. Use an 'equirectangular' projection to show the data in a more traditional euclidean space. Interesting distortion effects using some of the more exotic projections are also possible.

# We're going to generate 3 planetoids using a different mix of these elements to illustrate some of the variation possible with the same seed data.

# In[6]:


for ix, projection in enumerate(["orthographic", "natural earth", "azimuthal equidistant"]):
    ecology = ['gist_earth', 'gnuplot', 'rainbow_r']
    topography_levels = int(np.random.choice([x for x in range(10,35,5)],1)[0])
    lighting_levels = int(np.random.choice([x for x in range(10,35,5)],1)[0])
    relief_density = np.random.randint(1,5)
    plot_highlight = np.random.choice([True, False])
    plot_hillshade = np.random.choice([True, False])
    
    print("Ecology: {}".format(ecology[ix]))
    print("Topography Levels: {}".format(topography_levels))
    print("Lighting Levels: {}".format(lighting_levels))
    print("Relief Density: {}".format(relief_density))
    print("Projection: {}".format(projection))
    print("Highlight: {}".format(plot_highlight))
    print("Hillshade: {}".format(plot_hillshade))

    blob_planet.change_ecology(ecology[ix])
    blob_planet.fit_terraform(topography_levels=topography_levels,
                              lighting_levels=lighting_levels,
                              relief_density=relief_density,
                              plot_highlight=plot_highlight,
                              plot_hillshade=plot_hillshade,
                              projection=projection,
                              planet_name=projection)