Wrapper for easier data management with Earth Engine python sdk

Requires account with access to Google Cloud Storage and Earth Engine.

import eeUtil as eu

# initialize from environment variables
eu.init(bucket='mybucket')

# create image collection
eu.createFolder('mycollection', imageCollection=True)

# upload image to collection
eu.upload('image.tif', 'mycollection/myasset')
eu.setAcl('mycollection', 'public')
eu.ls('mycollection')

Install

pip install -e git+https://github.com/fgassert/eeUtil.git#egg=eeUtil

Develop

git clone https://github.com/fgassert/eeUtil.git
cd eeUtil
pip install -e .

• More consistent python bindings
• GEE paths not starting with / or users/ are relative to your user root folder (users/<username>)
• Upload atomatically stages files via Google Cloud Storage

eeUtil defaults to reading from credentials saved by gcloud auth for Google Cloud Storage and earthengine authenticate for Earth Engine. In your script, initialize these credentials with eeUtil.init(). These credentials are read from environment variables as follows.

# environment variables
export GEE_SERVICE_ACCOUNT=<[email protected]>
export GOOGLE_APPLICATION_CREDENTIALS=<path/to/credentials.json>
export CLOUDSDK_CORE_PROJECT=<my-project>
export GEE_STAGING_BUCKET=<my-bucket>

Alternatively credentials can be provided directly to eeUtil.init() via a json credential file or via a json string in the GEE_JSON environment variable.

eeUtil.init([service_account=], [credential_path=], [project=], [bucket=])

• service_account Service account name. If not specficed, reads defaulds from earthengine authenticate. For more information on GEE service accounts, see: https://developers.google.com/earth-engine/service_account [default: GEE_SERVICE_ACCOUNT]
• credential_path Path to json file containing private key. Required for service accounts. [default: GOOGLE_APPLICATION_CREDENTIALS]
• project GCS project containing bucket. Required if account has access to multiple projects. [default: CLOUDSDK_CORE_PROJECT]
• bucket Storage bucket for staging assets for ingestion. Will create new bucket if none provided. [default: GEE_STAGING_BUCKET]