pygeometa

pygeometa is a Python package to manage metadata for geospatial datasets.

Installation

pygeometa is best installed and used within a Python virtualenv.

Requirements

• Python 3 and above
• Python virtualenv package

Dependencies

Dependencies are listed in requirements.txt. Dependencies are automatically installed during pygeometa's installation.

Installing the Package

python3 -m venv my-env
cd my-env
. bin/activate
git clone https://github.com/geopython/pygeometa.git
cd pygeometa
pip3 install .

Running

From the command line

# show all subcommands
pygeometa

# show all supported schemas
pygeometa metadata schemas

# provide a basic sanity check/report on an MCF
pygeometa metadata info path/to/file.yml

# generate an ISO 19139 document to stdout
pygeometa metadata generate path/to/file.yml --schema=iso19139

# generate an ISO 19139 document to disk
pygeometa metadata generate path/to/file.yml --schema=iso19139 --output=some_file.xml

# generate an ISO 19139 document to disk with debugging (ERROR, WARNING, INFO, DEBUG)
pygeometa metadata generate path/to/file.yml --schema=iso19139 --output=some_file.xml --verbosity=DEBUG # add verbose (ERROR, WARNING, INFO, DEBUG)

# use your own defined schema
pygeometa metadata generate path/to/file.yml --schema_local=/path/to/my-schema --output=some_file.xml  # to file

# validate your MCF
pygeometa metadata validate path/to/file.yml

# import a metadata document to MCF
pygeometa metadata import path/to/file.xml --schema=iso19139

# import a metadata document to MCF, autodetecting the metadata file format
pygeometa metadata import path/to/file.xml --schema=autodetect  # --schema=autodetect is default

# transform from one metadata representation to another
pygeometa metadata transform path/to/file.xml --input-schema=iso19139 --output-schema=oarec-record

# transform from one metadata representation to another, autodetecting the metadata file format
pygeometa metadata transform path/to/file.xml --input-schema=autodetect --output-schema=oarec-record  # --input-schema=autodetect is default

Docker

pygeometa can also be run as a Docker application:

# run pygeometa via Docker, and map local path /path/to/my-data-dir to /data in Docker
docker run -ti -v /path/to/my-data-dir/:/data/ pygeometa:latest metadata generate --schema oarec-record /data/sample.mcf.yml

Supported schemas

Schemas supported by pygeometa:

• CSV on the web, reference
• dcat, reference
• iso19139, reference
• iso19139-hnap, reference
• OGC API - Records - Part 1: Core, record model, reference
• OpenAire metadata schema, reference
• Schema.org, reference
• SpatioTemporal Asset Catalog (STAC)
• iso19139-2, reference
• wmo-cmp, reference
• wmo-wcmp2, reference
• wmo-wigos, reference
• Local schema, specified with --schema_local=/path/to/my-schema

Using the API from Python

from pygeometa.core import read_mcf, render_j2_template

# read from disk
mcf_dict = read_mcf('/path/to/file.yml')
# read from string
mcf_dict = read_mcf(mcf_string)

# choose ISO 19139 output schema
from pygeometa.schemas.iso19139 import ISO19139OutputSchema
iso_os = ISO19139OutputSchema()

# default schema
xml_string = iso_os.write(mcf_dict)

# user-defined schema
xml_string = render_j2_template(mcf_dict, template_dir='/path/to/new-schema')

# write to disk
with open('output.xml', 'wb') as ff:
    ff.write(xml_string)

Development

Setting up a Development Environment

Same as installing a package. Use a virtualenv. Also install developer requirements:

pip3 install ".[dev]"

Adding a Metadata Schema to the Core

Adding an output metadata schemas to pygeometa involves extending pygeometa.schemas.base.BaseOutputSchema and supporting the write function to return a string of exported metadata content. If you are using Jinja2 templates, see the next section. If you are using another means of generating metadata (lxml, xml.etree, json, etc.), override the ABC write class to emit a string using your tooling/workflow accordingly. See the below sections for examples.

Once you have added your metadata schema, you need to register it with pygeometa's schema registry:

vi pygeometa/schemas/__init__.py
# edit the SCHEMAS dict with the metadata schema name and dotted path of class

Jinja2 templates

To add support for a new metadata schema using Jinja2 templates:

cp -r pygeometa/schemas/iso19139 pygeometa/schemas/new-schema

Then modify *.j2 files in the new pygeometa/schemas/new-schema directory to comply to new metadata schema.

Custom tooling

To add support for a new metadata schemas using other tooling/workflow:

mkdir pygeometa/schemas/foo
cp pygeometa/schemas/iso19139/__init__.py pygeometa/schemas/foo
vi pygeometa/schemas/foo/__init__.py
# update class name and super().__init__() function accordingly 

Running Tests

cd tests
python3 run_tests.py

Releasing

# update version
vi pyproject.toml  # update [project]/version
vi debian/changelog  # add changelog entry and summary of updates
git commit -m 'update release version' pyproject.toml debian/changelog
# push changes
git push origin master
git tag -a x.y.z -m 'tagging release x.y.z'
# push tag
git push --tags
rm -fr build dist *.egg-info
python3 -m build
twine upload dist/*

# publish release on GitHub (https://github.com/geopython/pygeometa/releases/new)

# bump version back to dev
vi pyproject.toml  # update [project]/version
git push origin master

Code Conventions

• PEP8

Bugs and Issues

All bugs, enhancements and issues are managed on GitHub.

Contact

• Tom Kralidis
• Alexandre Leroux