#importing libraries
!pip install geopandas
!pip install cenpy
!pip install jenkspy
import numpy as np
import pandas as pd
import geopandas as gpd
import altair as alt
import cenpy as cen
import cenpy.tiger as tiger
import jenkspy
from getpass import getpass

#setting the column width
pd.set_option('display.max_colwidth', 1000)

CENSUS_API_KEY = getpass('Enter your Census API Key: ')

def contrast_theme():
    font = "Helvetica"
    primary_color = "#F63366"
    white = "#ffffff"
    font_color = "#36454f"
    grey_color = "#f0f2f6"
    base_size = 16
    lg_font = base_size * 1.25
    sm_font = base_size * 0.8  # st.table size
    xl_font = base_size * 1.75

    config = {
        "config": {
            "arc": {"fill": primary_color},
            "area": {"fill": primary_color},
            "circle": {"fill": primary_color, "stroke": font_color, "strokeWidth": 0.5},
            "line": {"stroke": primary_color},
            "path": {"stroke": primary_color},
            "point": {"stroke": primary_color},
            "rect": {"fill": primary_color},
            "shape": {"stroke": primary_color},
            "geoshape": {"stroke": white, "strokeWidth": .1},
            "symbol": {"fill": primary_color},
            "title": {
                "font": font,
                "color": font_color,
                "fontSize": lg_font,
                "anchor": "start",
            },
            "axis": {
                "titleFont": font,
                "titleColor": font_color,
                "titleFontSize": sm_font,
                "labelFont": font,
                "labelColor": font_color,
                "labelFontSize": sm_font,
                "gridColor": grey_color,
                "domainColor": font_color,
                "tickColor": "#fff",
            },
            "header": {
                "labelFont": font,
                "titleFont": font,
                "labelFontSize": base_size,
                "titleFontSize": base_size,
            },
            "legend": {
                "titleFont": font,
                "titleColor": font_color,
                "titleFontSize": sm_font,
                "labelFont": font,
                "labelColor": font_color,
                "labelFontSize": sm_font,
                "symbolType": 'square',
            },
            "range": {
                "category": ["#f63366", "#fffd80", "#0068c9", "#ff2b2b", "#09ab3b"],
                "diverging": [
                    "#850018",
                    "#cd1549",
                    "#f6618d",
                    "#fbafc4",
                    "#f5f5f5",
                    "#93c5fe",
                    "#5091e6",
                    "#1d5ebd",
                    "#002f84",
                ],
                "heatmap": [
                    "#ffb5d4",
                    "#ff97b8",
                    "#ff7499",
                    "#fc4c78",
                    "#ec245f",
                    "#d2004b",
                    "#b10034",
                    "#91001f",
                    "#720008",
                ],
                "ramp": [
                    "#ffb5d4",
                    "#ff97b8",
                    "#ff7499",
                    "#fc4c78",
                    "#ec245f",
                    "#d2004b",
                    "#b10034",
                    "#91001f",
                    "#720008",
                ],
                "ordinal": [
                    "#ffb5d4",
                    "#ff97b8",
                    "#ff7499",
                    "#fc4c78",
                    "#ec245f",
                    "#d2004b",
                    "#b10034",
                    "#91001f",
                    "#720008",
                ],
            },
        }
    }
    return config

alt.themes.register("contrast", contrast_theme)
alt.themes.enable("contrast")

availableTables = cen.explorer.available()
acs = cen.products.ACS()

#TOTAL POVERTY STATUS IN THE PAST 12 MONTHS OF INDIVIDUALS BY SEX BY EDUCATIONAL ATTAINMENT	
acs.filter_variables('B17003_001E')

#Interested in change over 2012 & 2019 period
pd.reset_option('display.max_colwidth')
availableTables.loc[availableTables['vintage'] == 2019.0]
availableTables.loc[availableTables['vintage'] == 2012.0]

# Setting a variable (con**) for making a call to the specific tables.
con19 = cen.remote.APIConnection('ACSDT5Y2019',apikey=CENSUS_API_KEY)
con12 = cen.remote.APIConnection('ACSDT5Y2012',apikey=CENSUS_API_KEY)
#The variables available in these tables.
variables19 = con19.variables
variables12 = con12.variables

# Poverty status and total population
columns = [
    'B01003_001', # Total Population (This is default)
    'B17003_001', # POVERTY STATUS IN THE PAST 12 MONTHS OF INDIVIDUALS BY SEX BY EDUCATIONAL ATTAINMENT
]

# The E was absent from the columns identified. The E identifies the column as an estimate, so we include margins of error with our estimate with the code below.
columns_E = [i+'E' for i in columns]
columns_M = [i+'E' for i in columns]

# Units of geography
# State codes can be found at https://www.census.gov/library/reference/code-lists/ansi/ansi-codes-for-states.html

g_unit = 'county'
g_filter = {'state':'06'} #My focus is on California


# Making an API call for the tables/columns at the county level for the state of California and save to DF

Poverty19 = con19.query(columns_E + columns_M, geo_unit=g_unit, geo_filter=g_filter)

Poverty12 = con12.query(columns_E + columns_M, geo_unit=g_unit, geo_filter=g_filter)

#creating geoids for merge
Poverty19['GEOID'] = Poverty19['state'] + Poverty19['county']
Poverty12['GEOID'] = Poverty12['state'] + Poverty12['county']

#The columns have the same identities/name renaming them to avoid confusion/for an no-error code in the merge, 
Poverty19.columns = ['2019_pop', '2019_PoorPop', '2019_marginOfErrorPop', '2019_marginOfErrorHealth', 'state', 'county', 'GEOID']
Poverty12.columns = ['2012_pop', '2012_PoorPop', '2012_marginOfErrorPop', '2012_marginOfErrorHealth', 'state', 'county', 'GEOID']

#merging
dfCalipoverty = Poverty19.merge(Poverty12, left_on="GEOID", right_on="GEOID")

#tidy up - removing margins of error because they seem fine + very small
dfCalipoverty = dfCalipoverty[['GEOID', '2012_pop', '2019_pop', '2012_PoorPop', '2019_PoorPop']]

#figure out poverty rate across California from 2012 to 2019
dfCalipoverty['2012_poor'] = (dfCalipoverty['2012_PoorPop'].astype(int) / dfCalipoverty['2012_pop'].astype(int)) * 100
dfCalipoverty['2019_poor'] = (dfCalipoverty['2019_PoorPop'].astype(int) / dfCalipoverty['2019_pop'].astype(int)) * 100

#figure out difference between the two (percentage)
dfCalipoverty['change'] = dfCalipoverty['2019_poor'] - dfCalipoverty['2012_poor']
dfCalipoverty['change'] = dfCalipoverty['change']

dfCalipoverty.head()

cxn = tiger.TigerConnection('tigerWMS_ACS2019')

#grab counties
cxn.layers[86].variables

#filter to Californian counties
dfGeo = cxn.query(layer=86, where='STATE = 06')

#set crs
dfGeo = dfGeo.to_crs({'init': 'epsg:4326'})

#chart it to make sure everything is working
alt.Chart(dfGeo).mark_geoshape()

#merge geography with poverty rate data
dfpoorAndGeo = dfGeo.merge(dfCalipoverty, on='GEOID')

#map results
dfMap = alt.Chart(dfpoorAndGeo).mark_geoshape().encode(
    color=alt.Color('change:Q', legend=alt.Legend(title='% change in poverty status')),
    tooltip=[alt.Tooltip('NAME',title='County'),
      alt.Tooltip('2019_pop',title='Population'),
      alt.Tooltip('change',title='% change in poverty status')]
).properties(
    width=700,
    height=700,
    title='% change in poverty status'
).project(
    type='conicConformal',
    parallels=[45 + 37 / 60, 47 + 3 / 60], 
    rotate=[122 + 15 / 60, 0]
)

dfMap