README.md

Fateseal

This is a simple utility website that takes an MTG decklist and exports a Tabletop Simulator compatible JSON file. That's it!

Running locally

This website will eventually be hosted, but for now it can be run locally. All that is necessary is cloning the repository and running the following commands:

$ npm install
$ npm run dev

Then you just need to navigate to localhost:3000 and the app is available to use.

Data processing with bun

I previously used jq for quick data processing, but switched over to a script using Bun for better readability. It takes a little longer, but is much easier to maintain. Currently, the 2GB bulk data dump gets converted down into a 20MB JSON.

jq for data processing (old)

I downloaded Scryfall's bulk dataset of "Unique Cards" as the initial JSON dump to work from. I parsed it down to to size from 187 to about 5mb with jq:

$ cat "Unique Cards 20240318.json" | jq -c '[.[] | { name: .name, id: .id, imageUrl: .image_uris.large }]' > cards.json

Here's current pipeline:

$ jq '[.[] | select(.legalities.commander == "legal")]' "Unique Artwork 2024.json" > filtered_output.json
$ jq '[.[] | {name, imageUris: .image_uris, allParts: .all_parts, set, collector_number}]' filtered_output.json > slimmed_output.json
$ jq 'reduce .[] as $item ({};
    .[$item.name] += {
        name: $item.name,
        cards: (.[$item.name].cards + { ($item.set + $item.collector_number): { allParts: (if $item.allParts then $item.allParts | map(select(.component != "combo_piece")) else null end), imageUrl: $item.imageUris.large } })
    }
) | map(.)' slimmed_output.json > collected_output.json

Acknowledgements

MTG-related SVGs courtesy of Investigamer/mtg-vectors