PyLagoon

PyLagoon

Python client library for Lagoon.

Disclaimer: this library is still in an experimental state. Take a look at the R, Ruby, and cmdline clients if you are looking for a more stable option.

Installation

PyLagoon depends on the SqlAlchemy Python package and uses psycopg2 to generate Postgres queries. This requires that the postgres libraries to be installed on your host, e.g. libpq-dev.

With system dependencies installed, use pip to install via the included setup.py:

pip install .

Usage

Configure a new client connection using a LagoonConfig, for example

from PyLagoon import LagoonConfig, Lagoon

config = 
lagoon = Lagoon(
    config=LagoonConfig.load(yaml_file="../../docker/examples/lagoon-client.yaml")
)

Sources can then be queried using the lagoon.source() method with any of the optional filters described below.

Entire datasets can be downloaded using the Source.download_source() method. Query results can be downloaded using the lagoon.download_query() method.

To generate a query, construct the database schema metadata using PGMeta then use standard SQLAlchemy query methods to generate the query.

from PyLagoon import PGMeta
source = lagoon.sources(name="my-source")[0]
meta = PGMeta(source)
query = meta.query(meta[source]).filter(...)
df = lagoon.download_query(query=query, source=[source])

Source Filters

Lagoon.sources accepts the following filters via keyword arguments:

Key Name	Type	cmdline client equivalent
offset	str	--offset
limit	int	--limit
search	str	--search
ix	int	--ix
tags	list of strs	--tag <foo> --tag <bar>
description	str	--description
name	str	--name
version	int	--version
user	str	--user
columns	list of strs	--column <foo> --column <bar>
createdAfter	str	--created-after
createdBefore	str	--created-before
includeDeprecated	bool	--include-deprecated