NFT Metadata JSON Generator 0.1

NFT Metadata JSON Generator takes a CSV file with the metadata of an NFT Collection and creates a JSON file for each item with the specifications required by NFT Marketplaces such as OpenSea.

Considerations for version 0.1

The first version of NFT Metadata JSON Generator was tailored taking into consideration the specifications of the Sublimart NFT Collection. Consider it a PoC that may evolve into a more flexible software, suitable for the needs of most NFT projects.

Usage

To run the program, clone the repository, navigate to /metadata-generator, and execute the following command:

node generate-metadata --input <PATH_TO_CSV> --ipfs <IPFS_BASE_URL>

The command takes two arguments:

• input: path to the csv file with your metadata.
• ipfs: base URL of the images directory.

CSV File

In this case the csv file should include the following columns:

• name
• description
• image (file name)
• external URL
• attributes[artist]
• attributes[burn_date]
• attributes[burn_location]
• attributes[technique]
• attributes[curves]
• width
• heigh
• id
• cnt
• ligh_x
• bgcolor__001
• bgcolor__002
• bgcolor__003

IPFS URL

Before generating the set of JSON files, all item images should have been uploaded to IPFS or any other decentralized storage platform, as the program requires the URL of the directory with all the images. E.g: ipfs://bafybeifgk4maoi7b436httwca643uq2u6yhoenc3o7yj4jsfbuatbontga/

Output

All generated JSON files are stored inside the generated-metadata directory.

Generated JSON example:

{
    "name":"Lo Abierto #1",
    "description":"Florencia Giovanni Pacini is an Argentinian artist whose work has been exhibited in several national galleries. Merging both neo-romanticism and magical realism, she describes her work as figurative, poetic, narrative and dramatic, as well as being driven by feminism and ecology.",
    "image":"ipfs://bafybeifjvaqfaw62pvtmsctognkawvxjsdaiemelqf4quvdgoinsfwbw7q/5-82.png",
    "external_url":"https://www.sublim.art/artwork/lo-abierto@flo_giovanni_pacini",
    "traits":  [
        {"trait_type":"Burn date","value":"2022-09-16"},
        {"trait_type":"Burn location","value":"Buenos Aires"},
        {"trait_type":"Curves","value":"14"},
        {"trait_type":"Technique","value":"Oil on canvas"},
        {"trait_type":"Artist","value":"Flo Giovanni Pacini"},
        {"trait_type":"Shadow","value":"Left"},
        {"trait_type":"Background color","value":"dimgray"},
        {"trait_type":"Width","value":"8.5"}
    ]
}

Clarifications on JSON Data Construction

1. name, description and external_url fields in the CSV file directly map to name, description and external URL in the JSON output.

2. image is constructed by appending the file_name string to the ipfs constant passed as an argument.

3. The array of traits represents the atttributes of the item.

4. Burn date, Burn location, Curves, Technique, and Artist values are directly taken from their respective fields in the CSV file.

5. Shadow is calculated taking the light_x field, representing the direction of the projected shadow.

6. Background color is derived from bgcolor__001, bgcolor__002, bgcolor__003 values, which together form an RGB color code. This code is then translated into a more friendly color name using the color-namer library.