geoeconomics

[Proof of concept]

Validating a hypothesis about distribution of streets.

This tool is providing data for my research. It will download osm data, extract street data, and clean up the data. Some other useful data parsing functions are also included.

How to Use

This tool is dockerized and published on docker registry.

docker pull emptymalei/geoeconomics-with-data

Endpoints

distance_calculator: given the geocoordinates, output the distance to all streets in the specified city.

The following options are supported:

--point/-p: geocoordinate, such as -p '(10.2323,52.9384)' ((longitude,latitude) without any white space)
--city/-c: city name; the city name should be specified as a model in config file whose path can be specifid using --config/-cfg parameter if desired
--output/-o: output data path
--verbose/-v (optional): change logging levels
--config/-cfg (optional): path to config file; default config file is located at app/config/geo.yml
--schema/-s (optional): path to schema file being used for data transformations; default schema is located at app/geo/schema/city_streets.json

Example:

[...whatever_docker_path...] distance_calculator -p '(10.2323,52.9384)' -c berlin -o ~/Downloads/berlin.json

Development

Code

The code is localed in the folder app.

app
├── config
│   └── geo.yml
├── distance_calculator.py
└── geo
    ├── config.py
    ├── osm.py
    ├── schema
    │   └── city_streets.json
    ├── sourcing.py
    ├── transformer.py
    └── util.py

Workflow

Download data from some service, and save locally.
Transform data and save locally
Do the calculations and save data to the path specified by --output option

This package uses the open street map data as our data source. The download and transformation are defined in the config file geo.yml under resources variable. For example, we have this for berlin.

- city: berlin
  source: "https://download.geofabrik.de/europe/germany/berlin-latest.osm.pbf"
  pbf_file: "/tmp/germany/berlin-latest.osm.pbf"
  pbf_file_highway: "/tmp/germany/berlin-latest-highway.osm.pbf"
  geojson_file: "/tmp/germany/berlin-latest.geojson"
  transformed_json_file: "/tmp/germany/berlin-latest-transformed.json"

Adding New OSM City Model

Not all cities are included in the config file app/config/geo.yml.

To include a new city, either change add new city model to this default config file or make a copy and specify the path to new config using the --config option.

Adding More Data Fields

The transformed data field saved locally in a file. Here is the example for berlin.

- city: berlin
  source: "https://download.geofabrik.de/europe/germany/berlin-latest.osm.pbf"
  pbf_file: "/tmp/germany/berlin-latest.osm.pbf"
  pbf_file_highway: "/tmp/germany/berlin-latest-highway.osm.pbf"
  geojson_file: "/tmp/germany/berlin-latest.geojson"
  transformed_json_file: "/tmp/germany/berlin-latest-transformed.json"

The config transformed_json_file specifies the path to the transformed data file. This fields in this data file is specified using a schema file. The default schema file is located at app/geo/schema/city_streets.json. A customized schema can be specified using the --schema option. Meanwhile, the transformers should also be included in app/geo/transformer.py.

Adding More Functions

More functions can be added easily by adding new python files and setting up the endpoints in setup.py.

Deployment

Makefile will deal with everything:

make build will build the docker locally
publish will publish the docker image to docker registry

Changes should be made before push using this file.