Card data

The data model says what each column holds. These pages say what the values mean, where they came from, and how they relate to each other.

Everything here describes card data, which is imported from sources you can check yourself. For the decks and collection that belong to you, see your data.

Pages

• Taxonomy and tagging. The concept vocabulary, the four ways a card gets tagged, and how to argue with a tag.
• Rules and rulings. The Comprehensive Rules, official rulings, and the worked examples built on them. Coming soon.
• Sets and precons. Set metadata, preconstructed deck lists, and the set guides. Coming soon.
• Where a price comes from. Three independent price paths, what each covers, and how to tell how stale a number is.

How the pieces fit

Card text and printings are imported first. Everything else is built on top of them:

• Types and subtypes are parsed out of the type line into their own tables, so t:wolf can match a parsed subtype instead of a substring.
• Concepts and tags are matched against oracle text and against card embeddings, which is what tag: searches.
• Rules and rulings are stored verbatim so they can be quoted with a citation.
• Prices arrive on their own schedule, from different vendors than the card data, which is why a price carries its own timestamp.

Each of those is a separate import with its own cadence, and the data model records which job owns every column.