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

# In[1]:


import json
import geopandas as gpd
import matplotlib.pyplot as plt
import pandas as pd
import requests
import yaml

from geopandas.tools import sjoin
import plotly.graph_objects as go


# Load parameters to populate iNaturalist API request from YAML configuration file `params.yaml`

# In[2]:


# API docs in https://www.inaturalist.org/pages/api+reference
params = yaml.load(open('params.yml'), yaml.SafeLoader)
api_params = params['api']
poly_endpoint = params['join']['url']
api_baseurl = params['api']['url']


# Load polygons to be intersected with iNaturalist point data (ideally endpoint serves GeoJSON)

# In[3]:


dfpoly = gpd.read_file(poly_endpoint)


# Define the method to fetch the data from the iNaturalist REST API

# In[4]:


def get_data(api_params={}):
    print("⏳ Fetch iNaturalist data from API... page %d" % api_params['page'])
    r = requests.get(api_baseurl, params=api_params, headers={
        'Content-Type': 'application/json'})
    if r.status_code != 200:
        raise Exception(
            'Yikes - iNaturalist API error: {}'.format(
                r.status_code)
        )
    df = pd.json_normalize(r.json(), record_path=['results'])
    # Extract lat-Lon and keep critical columns only to speed things up
    df[['latitude', 'longitude']] = df['location'].str.split(',', 1, expand=True)
    df = df[['latitude', 'longitude', 'taxon.id', 'taxon.name', 'taxon.preferred_common_name']]

    if len(df.index) == api_params['per_page']:  # more pages to fetch
       api_params['page'] += 1
       df = df.append(get_data(api_params))
    
    return df


# In[5]:


df = get_data(api_params)
print("✔ %d records downloaded from the iNaturalist API" % len(df.index))


# Calculate inner join and manipulate data to be used in the plots below

# In[6]:


gdf = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.longitude, df.latitude))
gdf.set_crs(crs='EPSG:4326', inplace=True)
pointInPolys = sjoin(gdf, dfpoly, how='left')

def calc_within_area(row):
    return ('Outside' if pd.isnull(row['OBJECTID']) else 'Inside')

pointInPolys['state'] = pointInPolys.apply(calc_within_area, axis=1)
dftaxon = pointInPolys.groupby(['taxon.id','state']).size().reset_index(name='counts')
subdf = pointInPolys.loc[pointInPolys['Name'] != pd.NA]


# In[7]:


get_ipython().run_line_magic('matplotlib', 'inline')
pd.options.plotting.backend = "plotly"


# In[8]:


fig = pd.value_counts(pointInPolys['state']).plot.bar(title='Observations in and outside MRs', labels=dict(index="", value="# Observations", variable=""))
fig.update_layout(showlegend=False)
fig.show()


# In[9]:


fig = pd.value_counts(dftaxon['state']).plot.bar(title='Taxon richness in and outside MRs', labels=dict(index="", value="# Taxon", variable=""))
fig.update_layout(showlegend=False)
fig.show()


# In[10]:


fig = pd.value_counts(subdf['Name']).plot.bar(title='Observations inside MRs', labels=dict(index="", value="# Observations", variable="MR"))
fig.update_layout(showlegend=False)
fig.show()


# In[11]:


dftaxon = pointInPolys.groupby(['Name','taxon.id']).size().reset_index(name='counts')
fig = pd.value_counts(dftaxon['Name']).plot.bar(title='Taxon richness inside MRs', labels=dict(index="", value="# Taxon", variable=""))
fig.update_layout(showlegend=False)
fig.show()


# In[12]:


subdff = subdf.groupby(['taxon.name','Name']).size().reset_index(name='counts').sort_values(by=['Name', 'counts'])
fig = go.Figure(data=[go.Table(
    header=dict(values=list(subdff.columns),
                fill_color='paleturquoise',
                align='left'),
    cells=dict(values=[subdff.Name, subdff['taxon.name'], subdff.counts],
               fill_color='lavender',
               align='left'))
])

fig.show()