zoff-kollektiv / cifar

The website documents persons under EU sanctions and analyzes sanctions as a tool against corruption.

Cifar - sanctions watch

Structure

If you only want to update some content displayed, have a look into the data directory, as if contains most of the displayed texts.

• data/ Contains all data displayed on the website
• data/countries Contains all country relevant data
• data/home Contains all the content displayed in the blocks of the homepage
• data/pages Contains the about and report page
• data/people Contains the import of all persons from the google sheet
• data/translations Contains the translations of the people keys

General site data, such as the twitter/ facebook link or the page title is located in gatsby-config.js.

All pages can contain markdown in the body text. Some frontmatter fields of a person can also contain markdown. Supported fields: gatsby-node.js.

Report

The report (located in data/pages/report.md) is capable of rendering footnotes, which doesn't follow a markdown standard. In order to add a footnote you have to use:

Lorem ipsum dolor. ##Footnote Text##

The footnote text can contain markdown, to e.g. render links.

Lorem ipsum dolor. ##Footnote Text with [link](http://sanctionswatch.cifar.eu).##

Images

Portraits of the people under sanctions are stored in static/media. The filenames always have to start with the ID of the person and have to be at least 600px wide.

avatar.png is the fallback, which is displayed if a person doesn't have an image.

Development

npm run develop

You can manually lint all js and jsx files by running npm run lint. Before each commit, all staged files are linted automatically.

Release process

This project follows git-flow.

1. Create a release branch from develop and name it release/[version].
2. Create PR, wait for builds to be green.
3. Merge release/[version] into master.
4. git tag [version]
5. git push --tags
6. Update release description on github
7. Merge master into develop

Production build

npm run build

Import people

All people are currently fetched from a google spreadsheet. To re-import the data run node scripts/import-persons.js and commit the updated markdown files.

The import will omit the following values: Unknown, Unkown, .. All column headers will be transformed into a proper key. The original translation is stored in data/translations/columns.json.