Cumulus Library v1.3.1 release

[dogversioning]

The first release of the library since we set up this forum is out over on PyPI - pip install -U cumulus-library to get the latest version, or check out the release notes.

This is mostly bugfixes, but there is one optional feature that we've added with this release that may be of interest if you're thinking about using the library to author your own study, which is around a new (and optional) way to generate count tables.

You can see an instance of this called in the manifest of the core study - we've got a new field, counts_builder_config, which will allow you to use a lot of convenience methods for building count study tables.

Why would you want to do this? Well, these are the same kind of table, over and over, and it's a lot of boilerplate SQL that you need to structure a very specific way. By using one of these counts builders, you can focus more on just the data you're interested in, and less on how to write cube queries.

You can see an example of how these are implemented in the core study counts builder, and the library of available functions available in the base counts builder. There's also some new documentation around using these classes.