Putting it all together

1.py

from descartes import PolygonPatch
import fiona
from functools import partial, reduce
from math import sqrt
from itertools import imap
import pylab
from pyproj import Proj, transform

fiona.open = fiona.collection

BLUE = "#6699cc"
RED = "#cc6666"

# The one polygon rendering function we'll reuse throughout this module.
def add_record(fig, rec, fc=BLUE, ec=BLUE, alpha=1.0, zorder=2):
    """Makes a patch of the record's geometry, adds it to the default axes of
    the given figure and returns the figure. The color and alpha parameters
    can be constants or unary functions of the given record.

    Note: not strictly functional because we're mutating the figure."""
    facecolor = callable(fc) and fc(rec) or fc
    edgecolor = callable(ec) and ec(rec) or ec
    alphaval = callable(alpha) and alpha(rec) or alpha
    zorderval = callable(zorder) and zorder(rec) or zorder
    fig.gca().add_patch(
        PolygonPatch(
            rec['geometry'], 
            fc=facecolor, 
            ec=edgecolor, 
            alpha=alphaval, 
            zorder=zorderval))
    return fig

add_feature_alpha = partial(add_record, alpha=0.5)

with fiona.open("docs/data/test_uk.shp") as features:
    fig = reduce(add_feature_alpha, features, figure(figsize=(8, 8)))
    fig.gca().autoscale(tight=False)
    fig.savefig("test_uk.png")

2.py

# Add in map projection capabilities

def transform_coords(func, rec):
    # Transform record geometry coordinates using the provided function.
    # Returns a record or None.
    try:
        assert rec['geometry']['type'] == "Polygon"
        record = rec.copy()
        new_coords = list(
            zip(*func(*zip(*ring))
            ) for ring in rec['geometry']['coordinates'])
        record['geometry']['coordinates'] = new_coords
        return record
    except Exception, e:
        # Writing untransformed features to a different shapefile
        # is another option.
        logging.exception(
            "Error transforming record %s:", rec)
        return None

# The transform_coords() function calls another function for the actual
# transformation of coordinate values. We'll make one by "freezing" the
# input and output projections for pyproj.transform().
epsg4326 = Proj(init="epsg:4326", no_defs=True)
robin = Proj(
    proj="robin",
    lon_0=0,
    x_0=0,
    y_0=0,
    ellps="WGS84",
    datum="WGS84",
    units="m",
    no_defs=True)
transform_robin = partial(transform, epsg4326, robin)

# Here's partial function that makes a Robinson projection of feature records.
# Usage: projected_features = map(robinson, features)
robinson = partial(transform_coords, transform_robin)

with fiona.open(
        "/Users/seang/work/UVA-SL/world_borders/world_borders.shp"
        ) as features:
    fig = reduce(
        add_feature_alpha, imap(robinson, features), figure(figsize=(24, 12)))
    fig.gca().autoscale(tight=False)
    fig.savefig("world-robinson.png")

3.py

# The rendering function allows image alpha to vary from feature to feature.
# We'll create a weighting function to control the alpha value, normalizing
# population density to 80 per square km.
def weight(rec):
    props = rec['properties']
    dens = props['POP_CNTRY']/props['AREA']
    return min(sqrt(dens/80.0), 1.0)

# Here's our new rendering partial function.
add_feature_choropleth = partial(add_record, fc=RED, ec=RED, alpha=weight)

# And just for fun, we'll make a partial from reduce().
choropleth = partial(reduce, add_feature_choropleth)

with fiona.open(
        "/Users/seang/work/UVA-SL/world_borders/world_borders.shp"
        ) as features:
    fig = choropleth(imap(robinson, features), figure(figsize=(24, 12)))
    fig.gca().autoscale(tight=False)
    fig.savefig("world-dens-robinson.png")