This script downloads copernicus data from the Copernicus Data Space Ecosystem
from cdsetool.query import query_features, shape_to_wkt
from cdsetool.credentials import Credentials
from cdsetool.download import download_features
from cdsetool.monitor import StatusMonitor
from datetime import date
features = query_features(
"Sentinel1",
{
"startDate": "2020-12-20",
"completionDate": date(2020, 12, 25),
"processingLevel": "LEVEL1",
"sensorMode": "IW",
"productType": "IW_GRDH_1S",
"geometry": shape_to_wkt("path/to/shapefile.shp"),
},
)
list(
download_features(
features,
"path/to/output/folder/",
{
"concurrency": 4,
"monitor": StatusMonitor(),
"credentials": Credentials("username", "password"),
},
)
)
Or use the CLI:
cdsetool query search Sentinel2 --search-term startDate=2020-01-01 --search-term completionDate=2020-01-10 --search-term processingLevel=S2MSI2A --search-term box="4","51","4.5","52"
cdsetool download Sentinel2 PATH/TO/DIR --concurrency 4 --search-term startDate=2020-01-01 --search-term completionDate=2020-01-10 --search-term processingLevel=S2MSI2A --search-term box="4","51","4.5","52"
- CDSETool
Install cdsetool
using pip:
pip install cdsetool==0.2.13
Querying is always done in batches, returning len(results) <= maxRecords
records each time.
A local buffer is filled and gradually emptied as results are yielded. When the buffer is empty,
more results will be requested and the process repeated until no more results are available, or
the iterator is discarded.
Since downloading features is the most common use-case, query_features
assumes that the query will run till the end.
Because of this, the batch size is set to 2000
, which is the size limit set by CDSE.
from cdsetool.query import query_features
collection = "Sentinel2"
search_terms = {
"maxRecords": "100", # batch size, between 1 and 2000 (default: 2000).
"startDate": "1999-01-01",
"processingLevel": "S2MSI1C"
}
# wait for a single batch to finish, yield results immediately
for feature in query_features(collection, search_terms):
# do something with feature
# wait for all batch requests to complete, returning list
features = list(query_features(collection, search_terms))
# manually iterate
iterator = query_features(collection, search_terms)
featureA = next(iterator)
featureB = next(iterator)
# ...
To query by shapes, you must first convert your shape to Well Known Text (WKT). The included
shape_to_wkt
can solve this.
from cdsetool.query import query_features, shape_to_wkt
geometry = shape_to_wkt("path/to/shape.shp")
features = query_features("Sentinel3", {"geometry": geometry})
Most search terms only accept a single argument. To query by a list of arguments, loop the arguments and pass them one by one to the query function.
from cdsetool.query import query_features
tile_ids = ["32TPT", "32UPU", "32UPU", "31RFL", "37XDA"]
for tile_id in tile_ids:
features = query_features("Sentinel2", {"tileId": tile_id})
for feature in features:
# do things with feature
Its quite common to query for features created before, after or between dates.
from cdsetool.query import query_features
from datetime import date, datetime
date_from = date(2020, 1, 1) # or datetime(2020, 1, 1, 23, 59, 59, 123456) or "2020-01-01" or "2020-01-01T23:59:59.123456Z"
date_to = date(2020, 12, 31)
features = query_features("Sentinel2", {"startDate": date_from, "completionDate": date_to})
To get a list of all search terms for a given collection, you may either use the describe_collection
function or
use the CLI:
from cdsetool.query import describe_collection
search_terms = describe_collection("Sentinel2").keys()
print(search_terms)
And the CLI:
$ cdsetool query search-terms Sentinel2
An account is required to download features from the Copernicus distribution service.
To authenticate using an account, instantiate Credentials
and pass your username and password
from cdsetool.credentials import Credentials
username = "[email protected]"
password = "password123"
credentials = Credentials(username, password)
Alternatively, Credentials
can pull from ~/.netrc
when username and password are left blank.
# ~/.netrc
machine https://identity.dataspace.copernicus.eu/auth/realms/CDSE/protocol/openid-connect/token
login konata@izumi.com
password password123
# main.py
from cdsetool.credentials import Credentials
credentials = Credentials()
The credentials object may then be passed to a download function. If left out, the download
functions will default to using .netrc
.
credentials = Credentials()
download_features(features, "/some/download/path", {"credentials": credentials})
Credentials can be validated using the validate_credentials
function which will return a boolean.
from cdsetool.credentials import validate_credentials
validate_credentials(username='user', password='password')
If None are passed to username and password, validate_credentials
will validate .netrc
CDSETool provides a method for concurrently downloading features. The concurrency level should match your accounts privileges. See CDSE quotas
The downloaded feature ids are yielded, so its required to await the results.
from cdsetool.query import query_features
from cdsetool.download import download_features
features = query_features("Sentinel2")
download_path = "/path/to/download/folder"
downloads = download_features(features, download_path, {"concurrency": 4})
for id in downloads:
print(f"feature {id} downloaded")
# or
list(downloads)
Its possible to download features sequentially in a single thread if desired.
from cdsetool.query import query_features
from cdsetool.download import download_feature
features = query_features("Sentinel2")
download_path = "/path/to/download/folder"
for feature in features:
download_feature(feature, download_path)
- Query schema validation
- High-level API
- Query features
- Download features
- Download single feature
- Download list of features
- Download by ID
- Download by URL
- Command-Line Interface
- Update to match the high-level API
- Better
--help
messages - Quickstart guide in README.md
- Test suite
- Query
- Credentials
- Download
- Monitor
- Strategy for handling HTTP and connection errors
Any contributions you make are greatly appreciated.
If you have a suggestion that would make this better, please fork the repo and create a pull request. You can also simply open an issue with the tag "enhancement".
- Fork the Project
- Create your Feature Branch (
git checkout -b feature/cool-new-feature
) - Commit your Changes (
git commit -m 'Add some feature'
) - Push to the Branch (
git push origin feature/cool-new-feature
) - Open a Pull Request
Distributed under the MIT License. See LICENSE for more information.