How to write a list of Polygons to a .shp file in Python

Question

I create a list of Polygons and MultiPolygons outmulti as follows:

for pol in polygonToMulti:
    for pol2 in tre:
        if pol.intersects(pol2)==True:
            counter = counter+1
            # If they intersect, create a new polygon that is
            # essentially pol minus the intersection
            nonoverlap = (pol.symmetric_difference(pol2)).difference(pol2)
            outmulti.append(nonoverlap)


finalpol = MultiPolygon(outmulti)

I now want to save these Polygons and MultiPolygons of the outmulti list in a shp file & then be able to plot and visualize all the objects. Firstly, I thought of keeping the values of wkt into a different list like this:

list = []
for i in range(len(outmulti)):
    list.append(outmulti[i].wkt)

and then I thought of creating a GeoDataFrame having as the geometry attribute the above list list.

gdf3=gpd.GeoDataFrame(geometry=list)

however this gives a TypeError saying it's not a valid geometry object.

Secondly, I tried creating a GeoDataFrame with a column name "geometry" and then writing all the wkt values of outmulti into that column and then saving the final GeoDataFrame into a file like:

#Make list of Polygons -> Geo Data Frame
gdf3=gpd.GeoDataFrame(columns={"geometry"})

for i in range(len(outmulti)):
    temp = gpd.GeoSeries(outmulti[i].wkt)
    gdf3= gdf3.append({'geometry':temp}, ignore_index=True)

    gpd.geometry


gdf3.to_file(filename='dPolygons.shp', driver='ESRI Shapefile')

But in that case I get the error:

AttributeError: 'Series' object has no attribute 'has_z'

I guess this is because the shp file must have some specific attributes in order to be created, but I cannot understand which and where to get those from.

Answer 1

You already have the geometry as a shapely multipolygon/list of polygons. Add it directly to the geodataframe.

gdf3 = gpd.GeoDataFrame(geometry=[finalpol])  # Note GeoDataFrame geometry requires a list
gdf3.to_file(filename='dPolygons.shp', driver='ESRI Shapefile')

# Or

gdf3 = gpd.GeoDataFrame(geometry=outmulti)  # outmulti is already a list
gdf3.to_file(filename='dPolygons.shp', driver='ESRI Shapefile')