https://github.com/dpriskorn/geojson2osmnotes

Script to upload notes based on a source geojson to OSM. This is an alternative to directly import the data in the database.
https://github.com/dpriskorn/geojson2osmnotes

Last synced: 4 days ago
JSON representation

Script to upload notes based on a source geojson to OSM. This is an alternative to directly import the data in the database.

• Host: GitHub
• URL: https://github.com/dpriskorn/geojson2osmnotes
• Owner: dpriskorn
• License: gpl-3.0
• Created: 2023-09-27T03:38:49.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-08-13T12:10:33.000Z (5 months ago)
• Last Synced: 2024-12-14T22:41:05.753Z (21 days ago)
• Language: Python
• Size: 706 KB
• Stars: 0
• Watchers: 1
• Forks: 1
• Open Issues: 8
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# geojson2osmnotes

**Warning: automatic note creation is not endorsed by the community. MapRoulette or a regular import should be used instead, see https://wiki.openstreetmap.org/wiki/Notes**

The purpose of this script is to import geojson features as osm notes.

The purpose is to help curate OSM based on messy data from some source.

In the example below we have unreliable data for bathing sites from a Swedish government agency.

They don't have names for the site, nor a stable identifier. 

Still the points are useful for OSM editors to know about and check out on the ground.

# Features

Compares a dataset with OSM and creates notes 

if the source has more information than OSM

# Limitations

* None known

# Installation

Clone the repository.

Run

`$ pip install poetry && poetry install`

Copy config.

`$cp sample_config.py config.py`

Edit the config and add your credentials.

# Use

## CLI

`python main.py --source-geojson source.geojson --osm-geojson osm.geojson --notes-file notes.csv  --bounding-box 12.0541,59.8233,16.9100,62.3042`

The bounding box is on the same format as the Openstreetmap API, 

see https://wiki.openstreetmap.org/wiki/Bounding_Box

# Examples

## Bathing sites

First we get the geojson from the source.

In this case we got the data  

from the Swedish Lantmäteriet and extracted the geojson

via https://github.com/dpriskorn/LantmaterietHikingPaths/blob/main/extract_bathing_site_pois.py

The OSM data was fetched via overpass-api using this query:

```

/*

This has been generated by the overpass-turbo wizard.

The original search was:

“amenity=public_bath or leisure=bathing_place in Sweden”

*/

[out:json][timeout:25];

// fetch area “Sweden” to search in

{{geocodeArea:Sweden}}->.searchArea;

// gather results

(

  // query part for: “amenity=public_bath”

  node["amenity"="public_bath"](area.searchArea);

  way["amenity"="public_bath"](area.searchArea);

  relation["amenity"="public_bath"](area.searchArea);

  // query part for: “leisure=bathing_place”

  node["leisure"="bathing_place"](area.searchArea);

  way["leisure"="bathing_place"](area.searchArea);

  relation["leisure"="bathing_place"](area.searchArea);

);

// print results

out body;

>;

out skel qt;

```

Both files were fed to the script like this

`$ python main.py --source-geojson badplatser_sverige.geojson --osm-geojson export.geojson --notes-file notes.csv  --bounding-box 12.0541,59.8233,16.9100,62.3042`

# License

GPLv3+

# Thanks

To the Magnus Sälgö and the Swedish OSM community for feedback.

# What I learned

* Chatgpt is not good at handling shapely.geometry.point.Point 

without explicitly telling it to.

* Geopandas and geopy are super nice and fast.

* Lambda + apply is a pain in the ass. Fortunately chatgpt provided a working workaround.

* Oauth2 for a command line script is easy once you have a good example to follow :)