Geo-Bento

Bento plugins to transform geographic coordinates from a stream.

Bento is the swiss army of stream processing: Bento solves common data engineering tasks such as transformations, integrations, and multiplexing with declarative and unit testable configuration.

This repo contains multiple Bento plugins as Go modules, that you can build on demand (see cmd/geo-bento).

Note that the h3 plugin is using a CGO free version.

Get the country for a latitude and longitude

Use country with the following parameters: latitude, longitude.

An example position.json:

{"id":42, "lat": 48.86, "lng": 2.34}

A country.yaml pipeline.

input:
  file:
    paths: ["testdata/position.json"]
    codec: all-bytes

pipeline:
  threads: 1
  processors:
  - mapping: |
      #!blobl
      root = this
      root.country = country(this.lat, this.lng)

output:
  label: "out"
  stdout:
    codec: lines

Enrich the input with the timezone:

go build -o geo-0bento ./cmd/geo-bento
./geo-bento -c testdata/country.yaml
{"country":["France"],"id":42,"lat":48.86,"lng":2.34}

country module is using coord2country.

Transform latitude and longitude into an Uber h3 cell

Use h3 with the following parameters: latitude, longitude, resolution.

An example position.json:

{"id":42, "lat": 48.86, "lng": 2.34}

A h3.yaml pipeline.

input:
  file:
    paths: ["testdata/position.json"]
    codec: all-bytes

pipeline:
  threads: 1
  processors:
  - mapping: |
      #!blobl
      root = this
      root.h3 = h3(this.lat, this.lng, 5)

output:
  label: "out"
  stdout:
    codec: lines

Enrich the input with the h3 cell:

go build -o geo-bento ./cmd/geo-bento
./geo-bento -c testdata/h3.yaml
{"h3":"851fb467fffffff","id":42,"lat":48.86,"lng":2.34}

Transform latitude and longitude into a Google s2 cell

Use geos2 with the following parameters: latitude, longitude, resolution.

An example position.json:

{"id":42, "lat": 48.86, "lng": 2.34}

A s2.yaml pipeline.

input:
  file:
    paths: ["testdata/position.json"]
    codec: all-bytes

pipeline:
  threads: 1
  processors:
  - mapping: |
      #!blobl
      root = this
      root.s2 = geos2(this.lat, this.lng, 15)

output:
  label: "out"
  stdout:
    codec: lines

Enrich the input with the s2 cell:

go build -o geo-bento ./cmd/geo-bento
./geo-bento -c testdata/s2.yaml
{"id":42,"lat":48.86,"lng":2.34,"s2":"2/033303031301002"}

Get the timezone for a given latitude and longitude

Use tz with the following parameters: latitude, longitude.

An example position.json:

{"id":42, "lat": 48.86, "lng": 2.34}

A tz.yaml pipeline.

input:
  file:
    paths: ["testdata/position.json"]
    codec: all-bytes

pipeline:
  threads: 1
  processors:
  - mapping: |
      #!blobl
      root = this
      root.tz = tz(this.lat, this.lng)

output:
  label: "out"
  stdout:
    codec: lines

Enrich the input with the timezone:

go build -o geo-bento ./cmd/geo-bento
./geo-bento -c testdata/tz.yaml
{"tz":"Europe/Paris","id":42,"lat":48.86,"lng":2.34}

tz module is using tzf.

Generate random position in a range (mainly for debug)

This is an input plugin that will generate random coordinates in your range.

input:
  randpos:
    min_lat: 46.0
    max_lat: 48.0
    min_lng: 2.0
    max_lng: 2.3

output:
  label: "out"
  stdout:
    codec: lines

Example

Use the random position generator, to use all plugins.

input:
  randpos:
    min_lat: 46.0
    max_lat: 48.0
    min_lng: 2.0
    max_lng: 2.3
pipeline:
  threads: 1
  processors:
  - mapping: |
      #!blobl
      root = this
      root.h3 = h3(this.lat, this.lng, 12)
      root.country = country(this.lat, this.lng)
      root.s2 = geos2(this.lat, this.lng, 15)
      root.tz = tz(this.lat, this.lng)
      
output:
  label: "out"
  stdout:
    codec: lines

Live Testing

Run this command and point your browser to http://localhost:4195/

./geo-bento blobl server --no-open --host 0.0.0.0 --input-file ./testdata/position.json -m testdata/s2_mapping.txt

Docker

A pre built binary is also availale as a docker image:

 docker pull ghcr.io/akhenakh/geo-bento:latest

TODO

• s2 shape index to perform PIP
• spatialite lookup to perform PIP
• random points in a rect
• lat lng to h3
• lat lng to s2
• lat lng to tz
• lat lng to country