[I think @tmtmtmtm and @chrismytton would be in a better position to write this ticket, so please assume that this pretty much a placeholder, or an opportunity to refine our collective understanding of what's required.]
The role of these tools is to take data from Wikidata about politicians in a particular country and generate easily usable data: CSV files and Popolo JSON or Popolo JSON-LD.
These tools should in some way allow us to do sanity checks on the changes in data coming from Wikidata; this might be via a similar mechanism as for everypolitician-data (where a new version of the data is generated, and a pull request is made with that which someone has to review) or there could be some other interface to let us or partners check the incoming changes and fix any problems.
(This should be easier than with everypolitician-data pull requests since there is only a single source of data (Wikidata) not multiple sources that are being merged.)
To set up a new country repository, I'd anticipate the process being something like:
Add a config.json to the repository with the Wikidata item ID for the country
Add a Gemfile / Gemfile.lock and Rakefile
Run something like bundle install && bundle exec rake clean default
... which would then build the CSV and JSON.
Note that:
These repositories will include historic data, not just current representatives.
These repositories will also include boundary data.
Some of these countries will have many legislatures, so we need to think about what directory structure to use for the data within them.
[I think @tmtmtmtm and @chrismytton would be in a better position to write this ticket, so please assume that this pretty much a placeholder, or an opportunity to refine our collective understanding of what's required.]
The role of these tools is to take data from Wikidata about politicians in a particular country and generate easily usable data: CSV files and Popolo JSON or Popolo JSON-LD.
These tools should in some way allow us to do sanity checks on the changes in data coming from Wikidata; this might be via a similar mechanism as for everypolitician-data (where a new version of the data is generated, and a pull request is made with that which someone has to review) or there could be some other interface to let us or partners check the incoming changes and fix any problems.
(This should be easier than with everypolitician-data pull requests since there is only a single source of data (Wikidata) not multiple sources that are being merged.)
To set up a new country repository, I'd anticipate the process being something like:
bundle install && bundle exec rake clean default
... which would then build the CSV and JSON.
Note that: