NASA PDS DOI Service

The Planetary Data System (PDS) Digital Object Identifier (DOI) Service provides tools for PDS operators to mint DOIs.

Prerequisites

• Python 3.9 or above
• A login to the DOI Service Provider endpoint server (currently DataCite)

User Documentation

Please visit the documentation at: https://nasa-pds.github.io/doi-service/

Developers

JPL Internal Wiki

Get the code and work on a branch:

git clone ...
git checkout -b "#<issue number>"

Install a Python virtual environment, say in a venv directory:

python3 -m venv venv
source venv/bin/activate

Install the package and its dependencies for development into the virtual environment:

pip install --editable '.[dev]'

If you get an error like

src/types.h:36:2: error: You need a compatible libgit2 version (1.1.x)

then you're probably using brew.sh's Python 3.10. Use their Python 3.9 instead.

Update your local configuration to access the DOI service provider's test server.

Create a file in the base directory of the project named pds_doi_service.ini; the following may be used as a template

[SERVICE]
# Should be set to DataCite (case-insensitive)
provider = datacite

[DATACITE]
# Select the appropriate URL endpoint for either a test or production deployment
url = https://api.test.datacite.org/dois
#url = https://api.datacite.org/dois
user = <contact [PDS Help Desk](https://pds.nasa.gov/?feedback=true)>
password = <contact [PDS Help Desk](https://pds.nasa.gov/?feedback=true)>
doi_prefix = 10.17189
validate_against_schema = True

[OSTI]
# This section is kept for posterity, but should be ignored as OSTI is no longer a supported endpoint
url = https://www.osti.gov/iad2test/api/records
#url = https://www.osti.gov/iad2/api/records
user = <contact [PDS Help Desk](https://pds.nasa.gov/?feedback=true)>
password = <contact [PDS Help Desk](https://pds.nasa.gov/?feedback=true)>
doi_prefix = 10.17189
validate_against_schema = True

[PDS4_DICTIONARY]
url = https://pds.nasa.gov/pds4/pds/v1/PDS4_PDS_JSON_1D00.JSON
pds_node_identifier = 0001_NASA_PDS_1.pds.Node.pds.name

[API_AUTHENTICATION]
# Add the issuer of the oauth tokens, for cognito https://cognito-idp.<aws-region>.amazonaws.com/<userpoolID>
jwt_issuer =
# Add the entire content of the JSON file at https://cognito-idp.<aws-region>.amazonaws.com/<userpoolID>/.well-known/jwks.json
json_web_key_set =
jwt_lifetime_seconds = 3600
jwt_algorithm = RS256

[OTHER]
logging_level = INFO
doi_publisher = NASA Planetary Data System
global_keyword_values = PDS,PDS4
pds_uri = https://pds.nasa.gov/pds4/pds/v1/
transaction_dir = ./transaction_history
db_file = doi.db
db_table = doi
api_host = 0.0.0.0
api_port = 8080
api_valid_referrers =
emailer_local_host = localhost
emailer_port       = 25
emailer_sender     = [email protected]
emailer_receivers  = [email protected]

Launch API server

To run the DOI API server, try:

$ pip install pds-doi-service
$ pds-doi-api

The started service documentation is available on http://localhost:8080/PDS_APIs/pds_doi_api/0.2/ui/

👉 Note: When the api_valid_referrers option is set in pds_doi_service.ini, this service documentation UI will be unavailable.

Running with Docker

To run the server on a Docker container, please execute the following from the package directory:

$ # building the image
$ docker image build --tag pds-doi-service --file docker/Dockerfile .
$ # starting up a container
$ docker container run --publish 8080:8080 pds-doi-service

However, note that when launching the container via docker container run, all configuration values are derived from the default INI file bundled with the repository. To override the configuration, it is recommended to launch the service via a Docker Composition:

$ cd docker
$ # Make a copy of the docker composition environment template:
$ cp doi_service.env.in doi_service.env
$ # Edit the environment file, setting the credentials within:
$ vi doi_service.env
$ # Start the composition; on some systems, `docker compose` is `docker-compose`:
$ docker compose up

This will launch the DOI Service container using the docker-compose.yaml file in the docker subdirectory, which specifies that environment variables be imported from doi_service.env. Modify doi_service.env (after copying it from doi_service.env.in) to define any configuration values to override when the service is launched.

Test

Testing details are detailed in this section.

Tox (for developers)

N.B. Updates to pip dependencies are not automatically applied to existing tox virtual environments, to keep unit testing fast. The simplest way to propagate dependency updates is to delete ./.tox and run tox again.

tox is installed automatically during pip install --editable .[dev], and provides virtual environments and run configurations for

• unit/functional testing
• linting
• building the rich documentation.

To launch the full set of tests, simply set the following environment variables:

• CI should be set to true
• DATACITE_USER should be set to the PDS username of the testing Datacite instance
• DATACITE_PASSWORD should be set to that username's password

Then run:

tox

You can also run individual components:

$ tox -e tests  # Run unit, functional, and integration tests
$ tox -e lint  # Run flake8, mypy, and black code reformatting
$ tox -e docs  # Build the documentation to see if that works

It is strongly recommended to add tox -e lint to your pre-commit git hook, and tox -e tests in a pre-push hook, as only linted and test-passing PRs will be merged.

The following linting example is provided for ease of use:

STAGED_FILES=$(git diff --cached --name-only --diff-filter=ACM | grep ".py$")

echo "Linting files"
tox -e lint
git add $STAGED_FILES  # add any lint-related changes to the current commit

if [ $? -ne 0 ]
then
    echo "Initial lint detected errors, re-linting to determine whether errors remain"
    tox -e lint
    if [ $? -ne 0 ]
    then
      exit 1
    fi
fi

exit 0

You can also run pytest, sphinx-build, mypy, etc., if that's more your speed.

Behavioral testing (for Integration & Testing)

Behavioral tests are also pre-installed in the Python virtual environment when you run pip install --editable .[dev]. Launch those by running:

behave

Note this will download reference test data. If they need to be updated you have to first remove your local copy of the reference data (test/aaDOI_production_submitted_labels)

You can also run them for a nicer reporting:

behave -f allure_behave.formatter:AllureFormatter -o ./allure ./features
allure service allure

👉 Note: This assumes you have Allure Test Reporting framework installed.

Testrail Reporting

Test reports can be pushed to Testrail

Project: Planetary Data System (PDS) Test suite: pds-doi-service

Set your environment:

export TESTRAIL_USER=<your email in testrail>
export TESTRAIL_KEY=<your API key in tesrail>

Run the tests:

behave

See the results in https://cae-testrail.jpl.nasa.gov/testrail/index.php?/projects/overview/168

👉 Note: This assumes you have access to the Jet Propulsion Laboratory's Testrail installation.

Documentation Management

Documentation about the documentation is described in this section.

Design

See in this repository:

https://github.com/NASA-PDS/pds-doi-service/tree/main/docs

or the docs directory in the source package.

User Documentation

User documentation is managed with Sphinx, which is also installed in your Python virtual environment when you run pip install --editable .[dev]. You can use tox as described above to make the docs, or by hand at any time by running:

sphinx-build -ab html docs/source docs/build

Build & Release

The build and release process is managed by GitHub Actions and the Roundup.