make_maps.py

import re
import json
import copy
from datetime import timedelta, datetime

# Data is downloaded from https://maps.massgis.digital.mass.gov/MassMapper/MassMapper.html
# Using layer: Massachusetts Municipalities Multipar Polygons
# and exported as lat/lon as shapefile in WGS84
# Downloaded data is passed to https://mapshaper.org/
# and simplified to > 5 % of the original file size, fixing any line crossings.
# The resulting data is saved as MA_towns.json into this repository.

# Date a which the timeline ends. Needs to be later than the last date in the data
# but it's eaiser to update here by hand than to search through the data to find it.
enddate = "01/02/2026"

with open('MA_towns.json') as f:
    data = json.load(f)

for i, f in enumerate(data['features']):
    f['id'] = f['properties']['town']

# Parse input data in three files
fossil_fuel_free_table = re.compile("(?P<name>[a-zA-Z\ ]+)\ (?P<date>[0-9/]+)")
stretch_code_table = re.compile("(?P<name>[a-zA-Z\ ]+)\ [0-9\,]+ (?P<date>[0-9/]+)")
specialized_code_table = re.compile("(?P<name>[a-zA-Z\ ]+)\ [0-9\,]+ (?P<date>[0-9/]+) (?P<specialdate>[0-9/]*)")

# Stretch code
stretch_code = {}
specialized_opt_in = {}
with open('code_dates.dat') as f:
    for line in f:
        out = stretch_code_table.search(line)
        if out is not None:
            stretch_code[out.group('name')] = out.group('date')
        out = specialized_code_table.search(line)
        if out is not None:
            specialized_opt_in[out.group('name')] = out.group('specialdate')

# Approved list of fossil-fuel free pilot towns
fossil_fuel_free = {}
with open('fossil_fuel_free.dat') as f:
    for line in f:
        out = fossil_fuel_free_table.search(line)
        if out is None:
            # Just the name of the town, no date
            fossil_fuel_free[line.strip()] = "draft"
        else:
            # Name and date
            fossil_fuel_free[out.group("name")] = out.group("date")

print(fossil_fuel_free)

print(f'Number of municipalities in geojson file: {len(data["features"])}')
print(f'Number of municipalities with base code: {len(data["features"]) - len(stretch_code)}')
print(f'Number of municipalities with stretch code (incl. those with opt-in stretch code): {len(stretch_code) - len(specialized_opt_in)}')
print(f'Number of municipalities with specialized opt-in stretch code: {len(specialized_opt_in)}')
print(f'Number of municipalities with fossil-fuel free: {len(fossil_fuel_free)}')

def USdate_to_ISO(datestr, subtractdays=0):
    return (datetime.strptime(datestr, '%m/%d/%Y') -
            timedelta(days=subtractdays)).isoformat()

# Now we build up the geojson files with the data we want to display
featurelist = []  # for timeline
featurelist_future = []  # for plain version, capture last known date

for town in data['features']:
    name = town['properties']['town'].title()
    feature = {'type': 'Feature',
               'geometry': town['geometry'],
               'properties': town['properties']}
    feature['properties']['stretchcode'] = stretch_code.get(name, 'not yet')
    feature['properties']['optinstretchcode'] = specialized_opt_in.get(name, 'not yet')
    if name in fossil_fuel_free:
        feature["properties"][
            "fossilfuel"
        ] = f"Fossil Fuel Free prioritized community: {fossil_fuel_free[name]}"
    else:
        feature["properties"]["fossilfuel"] = ""

    # Special case for towns that left the stretch code program
    if name == 'Essex':
        feature['properties']['stretchcode'] = '1/1/2015 - 5/8/2023'
        dates = [
            ("01/01/2009", 0),
            ("01/01/2015", 1),
            ("05/08/2023", -1),
            (enddate, None),
        ]
    elif name == 'Rochester':
        feature['properties']['stretchcode'] = '1/1/2019 - 5/22/2023'
        dates = [
            ("01/01/2009", 0),
            ("01/01/2019", 1),
            ("05/22/2023", -1),
            (enddate, None),
        ]
    else:
        dates = [("01/01/2009", 0), (enddate, None)]
        if name in stretch_code:
            dates.append((stretch_code.get(name), 1))
        if name in specialized_opt_in:
            dates.append((specialized_opt_in.get(name), 1))
        if name in fossil_fuel_free:
            if fossil_fuel_free[name] == "draft":
                dates.append(("07/01/2024", 10))
            else:
                dates.append((fossil_fuel_free[name], 10))

    dates.sort(key=lambda x: datetime.strptime(x[0], '%m/%d/%Y'))

    codecolor = 0
    for i, d in enumerate(dates[:-1]):
        codecolor = codecolor + d[1]
        code = copy.deepcopy(feature)
        code['properties']['code_color'] = codecolor
        code['properties']['start'] = USdate_to_ISO(d[0])
        code['properties']['end'] = USdate_to_ISO(dates[i + 1][0], subtractdays=1)
        featurelist.append(code)

    # Add last known data point - the one most in the future
    # i.e. the last thing the town has adopted so far
    featurelist_future.append(code)


# And write it into docs/MA_energy_codes_town.js such that it's read directly as a javascript file
# That way, we get a file very similar to https://leafletjs.com/examples/choropleth/
outdata = {'type': 'FeatureCollection', 'features': featurelist}
with open('docs/MA_energy_codes_town.js', 'w') as f:
    f.write('var townsData = ' + json.dumps(outdata) + ';')

outdata = {'type': 'FeatureCollection', 'features': featurelist_future}
with open('docs/MA_energy_codes_town_future.js', 'w') as f:
    f.write('var townsData = ' + json.dumps(outdata) + ';')


base_code = []
for town in data['features']:
    name = town['properties']['town'].title()
    if name not in stretch_code:
         base_code.append(name)
base_code.sort()

for name in base_code:
    print(name)

print(len(base_code))