apache/superset · error · Error

Unknown input format

Error message

Unknown input format

What it means

Thrown by PivotData.forEachRecord in the pivot-table plugin's vendored react-pivottable utilities. Upstream pivottable accepts arrays, functions, or jQuery selections of tables as input, but this trimmed port only handles arrays of record objects; anything else is rejected before any record is processed. It is a fail-fast guard against an input contract violation, not a data problem.

Source

Thrown at superset-frontend/plugins/plugin-chart-pivot-table/src/react-pivottable/utilities.ts:1368

        },
        format() {
          return '';
        },
      }
    );
  }
}

// can handle arrays or jQuery selections of tables
PivotData.forEachRecord = function (
  input: unknown,
  processRecord: (record: PivotRecord) => void,
) {
  if (Array.isArray(input)) {
    // array of objects
    return input.map(record => processRecord(record));
  }
  throw new Error(t('Unknown input format'));
};

PivotData.defaultProps = {
  cols: [],
  rows: [],
  vals: [],
  sorters: {},
  rowOrder: 'key_a_to_z',
  colOrder: 'key_a_to_z',
};

PivotData.propTypes = {
  data: PropTypes.oneOfType([PropTypes.array, PropTypes.object, PropTypes.func])
    .isRequired,
  cols: PropTypes.arrayOf(PropTypes.string),
  rows: PropTypes.arrayOf(PropTypes.string),
  vals: PropTypes.arrayOf(PropTypes.string),
  valueFilter: PropTypes.objectOf(PropTypes.objectOf(PropTypes.bool)),

View on GitHub (pinned to f4587218dd)

Solutions

1. Convert your input to a plain array of record objects before passing it: Array.from(input) or [...input] for iterables; call the function yourself if input is a records-producing callback.
2. If you wrapped PivotData, verify the data argument reaches it unmodified (not a DOM/jQuery selection or undefined after an async fetch race).
3. Extend forEachRecord in your fork with the additional branches (function input, jQuery table) copied from upstream react-pivottable if you need them.

Example fix

// before
PivotData.forEachRecord(recordsIterable, processRecord); // throws

// after
PivotData.forEachRecord([...recordsIterable], processRecord);

if (!Array.isArray(input)) {
  input = Array.isArray(input?.records) ? input.records : Array.from(input);
}
PivotData.forEachRecord(input, processRecord);

const isRecordArray = (v: unknown): v is PivotRecord[] =>
  Array.isArray(v) && v.every(r => r != null && typeof r === 'object');

try { PivotData.forEachRecord(data, cb); } catch (e) { if (e instanceof Error && e.message.includes('Unknown input format')) throw new TypeError('PivotData input must be an array of records'); throw e; }

When it happens

Trigger: Calling PivotData.forEachRecord (directly or indirectly via PivotData construction / chart transformProps) with input that is not an Array — e.g. a generator, an iterable, a function returning records, a jQuery/DOM selection, or undefined/null data passed as the records argument.

Common situations: Custom forks or plugins that feed the pivot table with a non-array data source; upstream pivottable code copied over that passes a function; a data connector that returns a Map/stream instead of a plain array; tests that pass malformed query results.

AI-assisted analysis of apache/superset@f4587218dd (2026-08-14). Data as JSON: /api/errors/5140a22eff621afb. Report an issue: GitHub.