# csv-ipld-schema-gen **Repository Path**: mirrors_mikeal/csv-ipld-schema-gen ## Basic Information - **Project Name**: csv-ipld-schema-gen - **Description**: Generate an IPLD Schema template from CSV data - **Primary Language**: Unknown - **License**: Not specified - **Default Branch**: master - **Homepage**: None - **GVP Project**: No ## Statistics - **Stars**: 0 - **Forks**: 0 - **Created**: 2020-08-09 - **Last Updated**: 2025-09-20 ## Categories & Tags **Categories**: Uncategorized **Tags**: None ## README # csv-ipld-schema-gen Generate an IPLD Schema template from CSV data ## CLI ``` csv-ipld-schema-gen [files..] Generate IPLD Struct from source csv files Options: --help Show help [boolean] --version Show version number [boolean] ``` The resulting IPLD Schema will give you a Struct for every csv file. The schema should be considered a base template. You will still want to read and edit it as the script leans more towards reserving the original names rather than producing 100% valid IPLD Schema. * The struct name will be the filename with dots replaced with `_`. These should be edited to conform with the standard camelCase style. * Values are parsed and simple types are inferred: string, int, float. * If any values have a string as well as other types they will be typed as "string" in order to avoid aggressive parsing errors. * If a field has instances of `null` the field will be `nullable`. * If a filed has instances where it is empty the field will be `optional`. * If a field name contains `(*string*)` it will be parsed as a list value. * If all the values found in the field are the same type the schema will contain a typed list definition which the field will be typed as. * If the values in the field vary in type it will generate a kinded union for all the observed types. * The field, union, and list names will all be the original string (w/ `(*string*)` intact) so this won't actually be a valid parsable schema.