Cross-posted from The State Decoded
Version 0.9 Released
Version 0.9 of The State Decoded is now available on GitHub. This release completes the work to store multiple editions of a code started in v0.8. Since we have made fundamental changes to how laws are stored, potentially breaking previous installations, we have added it as an additional pre-1.0 release.
Site admins can now choose to import a new edition of the legal code, or replace the current edition of code. This allows multiple editions of each code to live in parallel with the others – so when your locality updates their code, you can create a historical record of previous versions. Right now you can browse and search previous editions of the code, and all related data is kept isolate to each version. In the future this will enable us to compare and show the differences between versions of the code, as Wikipedia and GitHub do.
We now allow users to choose between the default Solr search engine and a new, optional MySQL search adapter. The MySQL adapter uses the built-in database to search, with no additional packages needed. This means that The State Decoded can now be run on most standard hosting without any extra setup. If your host can run WordPress, you can run State Decoded. Solr is still a far more powerful and robust solution, but users now have an easier setup option.
To support this, we’ve refactored the search system to better abstract the code. The result of this is that data is directly imported into the search system from the database instead of having to jump through XSLT hoops. This abstraction also means that new search engines can be used easily, by writing a new search interface adapter. For example, if anyone wanted to write an adapter to use Elasticsearch, that’s now much easier.
Command Line Interface and Database Versioning
There is now a new command line tool that comes with the package called
statedecoded. This command can be used to run various tasks, including updating the database when a new version of State Decoded comes out, and more. Data imports can be run via the tool as well, so it can now easily be scheduled with
cron to import automatically on a regular basis. You can also create your own tasks to run using the tool.
The database structure itself can now be updated with this tool. This removes the need to delete the entire database and install from scratch every time the structure is changed.
We’ve added an example parser for working with Municipal Code Corporation’s XML format, to go along with our parser for American Legal Publishing Corporation’s data, and our default, generic parser. This improves the ability for localities to import data from their current codifier.
And Much More
We’ve also made a large number of other bug fixes, design tweaks, and accessibility improvements.
As always, this work wouldn’t be possible without the generous support of The John S. and James L. Knight Foundation, who fund The OpenGov Foundation‘s work on The State Decoded. We also couldn’t have done this without the help of Grant Ingersoll ofLucidworks fixing our Solr issues, and support from other amazing members of the Solr community, including Doug Turnbull, Erik Hatcher, and John Berryman. Design suggestions were, as always, made by John Athayde of Meticulous.