A basic module for interacting with the GenoLogics LIMS server via its REST API. The goal is to provide simple access to the most common entities and their attributes in a reasonably Pythonic fashion.
All instances of Project, Sample, Artifact, etc should be obtained using the get_* methods of the Lims class, which keeps an internal cache of current instances. The idea is to create one and only one instance in a running script for representing an item in the database. If one has more than one instance representing the same item, there is a danger that one of them gets updated and not the others.
An instance of Project, Sample, Artifact, etc, retrieves lazily (i.e. only when required) its XML representation from the database. This is parsed and kept as an ElementTree within the instance. All access to predefined attributes goes via descriptors which read from or modify the ElementTree. This simplifies writing back an updated instance to the database.
The 'genologics' directory should be made accessible in your Python path, by whatever method suits your installation.
The client script imports the class Lims from the genologics.lims module, and instantiates it with the required arguments:
- Base URI of the server, including the port number, but excluding the '/api/v1' segment of the path.
- User name of the account on the server.
- Password of the account on the server.
Usage example scripts are provided in the subdirectory 'examples'.
NOTE: The example files rely on specific entities and configurations on the server, and use base URI, user name and password, so to work for your server, all these must be reviewed and modified.
The interface has not been used much yet, so it is not properly debugged.
Known issues:
- Artifact state is part of its URL (as a query parameter). It is not entirely clear how to deal with this in the Lims.cache: Currently, an artifact that has the current state may be represented by a URL that includes the state, and another that does not contain it.