Skip to content

Latest commit

 

History

History
44 lines (24 loc) · 2.58 KB

contributing.md

File metadata and controls

44 lines (24 loc) · 2.58 KB
Raw

Contributing Guidelines

We welcome contributions to both the contents of the database and as well as our data model. If you are unsure if your data would make a good fit for inclusion into ReadAct feel free to open an issue or otherwise get in touch.

We cannot accept PRs that contain personal information, as the result of field work. All data must satisfy the GDPR.

Research Assistants (HiWis)

Student assistants who join our project should always consult the step-by-step guidelines for how to use git, work with spreadsheets, best practices, etc. in the wiki.

Data Directory

The main data files are in csv format and located inside the csv/data/ directory. The source encoding is utf-8 and uses commas , as field separators. The csv files are the authoritative files used for populating the SQL database as well as for generating the tei-xml files driving the ReadAct web application.

Validation and consistency

To maintain data consistency we use two test suites, which are automatically carried out by our CI service on each commit or PR. Only PRs that satisfy our tests can be reviewed and merged.

Structural changes, e.g. new files or new columns, as opposed to new entries or corrections, require edits to the schema files, the data-dictionary located in data/, and the bats tests.

CI and schema files

The schemas for validating csv can be found in csv/schema. The validation runs on frictionlessdata. Please follow their installation instructions to validate your data locally.

Each Schema is named after the corresponding csv table, so e.g. Act.json validates Act.csv. These are referenced in the datapackage.json to validate:

goodtables validate csv/datapackage.json

Alternatively you can take a look at this webpage

Auxiliary Files

We automatically generate both tei-xml authority files, and gexf networks.

  • TEI (xml/*.xml)
  • networks (xml/*.gexf)

New diagrams or new network layouts should be added as separate files in the respective parent directories, and be accompanied with a short description in your PR.

Views

View data for visualization from the database is similarly automatically generated on CI. To add new views, place the underlying data table in the csv/views folder, and add a corresponding entry in the datapackage.json. To display the view on the readchina wepage open a corresponding pull request in its repo.