This repository implements the Google Earth Engine Adapter services that are available in the Resource Watch API.
If you are looking for the API Doc (Info and Usage) please go to the next link: [View the documentation for this API] ()(NOT YET)
To create a new GEE dataset it's necessary to execute the following request.
Important:
- the connectorType value has to be "rest"
- the provider value has to be "gee"
POST: /dataset -> payload:
{
"dataset": {
"application": [<application-name>],
"name": <dataset-name>,
"connectorType": "rest",
"provider": "gee",
"tableName": <table-name>
}
}
POST: https://staging-api.globalforestwatch.org/dataset -> payload:
{
"dataset": {
"application": ["rw"],
"name": "Data about whatever",
"connectorType": "rest",
"provider": "gee",
"tableName": "ft:1qpKIcYQMBsXLA9RLWCaV9D0Hus2cMQHhI-ViKHo"
}
}
Once the dataset has been saved (a few seconds after the creation) you can start doing queries to GEE.
This endpoint returns the available fields in the dataset
GET: /fields/:dataset
Example (copy&paste)
GET: https://staging-api.globalforestwatch.org/fields/68353d61-0f47-4836-9699-72e008cd9f5f
This endpoint returns the execution of the SQL query (sql queryParam is required)
GET: /query/:dataset?sql=<slq_query>
Example (copy&paste)
GET: https://staging-api.globalforestwatch.org/query/68353d61-0f47-4836-9699-72e008cd9f5f?sql=select * from ft:1qpKIcYQMBsXLA9RLWCaV9D0Hus2cMQHhI-ViKHo where width > 100
Download the data in json format (csv coming soon)
GET: /download/:dataset?sql=<slq_query>
Example (copy&paste)
GET: https://staging-api.globalforestwatch.org/download/68353d61-0f47-4836-9699-72e008cd9f5f?sql=select * from ft:1qpKIcYQMBsXLA9RLWCaV9D0Hus2cMQHhI-ViKHo where width > 100