custom-schema.mjs

/**
 * Custom fields on top of a preset, fields injected into every line-item row, and saving
 * the combination so later calls are one argument.
 *
 *   node examples/custom-schema.mjs invoice.pdf
 */

import { VisionAPI, ConflictError, rows, unwrap, value } from '@devrobotlabs/visionapi';

const [path = 'invoice.pdf'] = process.argv.slice(2);
const vision = new VisionAPI();

const schema = {
  // Plain form: the string is the description. Descriptions are the prompt — the more
  // precisely you say what you want, the better the extraction.
  machine_serial: 'Serial number of the machine being invoiced, without the "SN:" prefix',

  // Typed form. Numbers come back as JSON numbers, dates as ISO YYYY-MM-DD.
  total_net: { type: 'number', description: 'Total before tax' },
  is_paid: { type: 'boolean', description: 'Whether the invoice is stamped or marked PAID' },
  reference_numbers: { type: 'array', description: 'All reference numbers', items: 'string' },

  // `line_item` is reserved: these become extra columns on every row of the preset's
  // line-item array. Only meaningful with a preset that has line items; with a bare
  // custom schema there is nothing to inject into.
  line_item: { lot_number: 'The lot number printed on this line, if present' },
};

const res = await vision.analyze({ file: path, preset: 'invoice', schema });

console.log('serial:', value(res.result, 'machine_serial'));
console.log('net total:', value(res.result, 'total_net'));
console.log('paid:', value(res.result, 'is_paid'));

for (const row of rows(res.result, 'line_item')) {
  const line = unwrap(row);
  console.log(`  ${line.quantity ?? '?'} × ${line.description ?? '(no description)'} ` +
    `= ${line.amount ?? '?'}  [lot ${line.lot_number ?? '—'}]`);
}

// Save it and the next call is `{ schemaName: 'our-invoices' }`. The definition is
// compiled before it is stored, so an invalid schema fails here rather than on the first
// extraction that uses it.
try {
  await vision.createSchema({ name: 'our-invoices', preset: 'invoice', schema });
  console.log('\nsaved as "our-invoices"');
} catch (err) {
  if (!(err instanceof ConflictError)) throw err;
  await vision.updateSchema('our-invoices', { preset: 'invoice', schema });
  console.log('\nupdated "our-invoices"');
}

const again = await vision.analyze({ file: path, schemaName: 'our-invoices' });
console.log('same fields via the saved schema:', Object.keys(again.result).length);