[csv-loader] Support for generating unique id's

[marceloverdijk]

Hi,

First of all thank you for this project.

I have some csv files without a defined id column, nor a column which values are unique itself.
E.g. a csv with a year and month column.

Would it be an idea to provide a (new) CSVLoaderOptions option to automatically generate an id?
Or to have a Function option to generate an id for each row?
E.g. (row) => `${year}-${month}`

[marceloverdijk]

I was looking at the papaparse options and thought the step might be useful for this:

  loader: csvLoader({
    fileName: 'data/my-data.csv',
    idField: 'id',
    parserOptions: {
      step: (result) => {
        const row = result.data;
        row.id = `${row.year}-${row.month};
      },
    },
  }),

However that option is omitted unfortunately:

https://github.com/ascorbic/astro-loaders/blob/main/packages/csv/src/csv-loader.ts#L25-L28

  parserOptions?: Omit<
    Papa.ParseConfig,
    "header" | "dynamicTyping" | "transformHeader" | "step" | "complete"
  >;

for the other options I understand why they are omitted, but not for the step:

    const csvStream = Papa.parse(Papa.NODE_STREAM_INPUT, {
      dynamicTyping: true,
      ...parserOptions,
      header: true,
      transformHeader: transformHeader === false ? undefined : transformHeader,
    });