/**
 * Field inference
 * Derives a field list and column types from sample records, so a source can be
 * configured from real data rather than an assumed shape. Used by the CSV
 * suggest endpoint and by bank-feed sampling.
 */

const ISO_DATE_RE = /^\d{4}-\d{2}-\d{2}(T[\d:.Z+-]+)?$/;

/**
 * @param records  array of flat objects
 * @param limit    how many rows to look at
 * @returns { fields: [{name, type}], sampleRows }
 *
 * Keys are unioned across the sample, not read off the first row — API feeds
 * omit optional keys entirely on records that lack them.
 */
function inferFields(records, limit = 50) {
    const sampleRows = records.slice(0, limit);

    const keys = [];
    for (const row of sampleRows) {
        for (const key of Object.keys(row)) {
            if (!keys.includes(key)) keys.push(key);
        }
    }

    const fields = keys.map(key => {
        const vals = sampleRows.map(r => r[key]).filter(v => v !== '' && v != null);
        let type = 'text';
        if (vals.length > 0 && vals.every(v => !isNaN(parseFloat(v)) && isFinite(v) && String(v).charAt(0) !== '0')) {
            type = 'numeric';
        } else if (vals.length > 0 && vals.every(v => ISO_DATE_RE.test(String(v)))) {
            type = 'date';
        }
        return { name: key, type };
    });

    return { fields, sampleRows };
}

module.exports = { inferFields };