FlowiseAI/Flowise · error · Error

Column ${column} not found in CSV file.

Error message

Column ${column} not found in CSV file.

What it means

CsvLoader resolved header fields successfully but the requested column was not among them. column may be a string name or a numeric index; when numeric it is first mapped to fields[column]. If that mapped name (or the literal string) is not in fields, this throws.

Source

Thrown at packages/components/nodes/documentloaders/Csv/CsvLoader.ts:65

            meta: { fields = [] }
        } = Papa.parse<{ [K: string]: string }>(raw.trim(), {
            delimiter: separator,
            header: true
        })

        if (column !== undefined) {
            if (!fields.length) {
                throw new Error(`Unable to resolve fields from header.`)
            }

            let searchIdx = column

            if (typeof column == 'number') {
                searchIdx = fields[column]
            }

            if (!fields.includes(searchIdx as string)) {
                throw new Error(`Column ${column} not found in CSV file.`)
            }

            // Note TextLoader will raise an exception if the value is null.
            return parsed.map((row) => row[searchIdx])
        }

        return parsed.map((row) => fields.map((key) => `${key.trim() || '_0'}: ${row[key]?.trim()}`).join('\n'))
    }
}

View on GitHub (pinned to abe4a8601a)

Solutions

1. Print the resolved fields (Papa meta.fields) and match the column name exactly, including case and whitespace.
2. If passing a numeric index, remember it is 0-based and must be within fields.length.
3. Normalize header whitespace upstream: trim each field name before comparison.

Example fix

// before
if (!fields.includes(searchIdx as string)) {
  throw new Error(`Column ${column} not found in CSV file.`)
}
// after - list available fields to aid debugging
if (!fields.includes(searchIdx as string)) {
  throw new Error(`Column ${JSON.stringify(column)} not found. Available: ${JSON.stringify(fields)}`)
}

import Papa from 'papaparse'

function resolveColumn(raw: string, column: string | number, separator?: string): string {
  const { meta: { fields = [] } } = Papa.parse(raw.trim(), { delimiter: separator, header: true, preview: 1 })
  const trimmed = fields.map((f) => f.trim())
  const resolved = typeof column === 'number' ? trimmed[column] : column
  if (!trimmed.includes(typeof column === 'number' ? resolved : column.trim())) {
    throw new Error(`Column ${JSON.stringify(column)} not found. Available: ${JSON.stringify(trimmed)}`)
  }
  return resolved
}

function columnExists(raw: string, column: string | number, separator?: string): boolean {
  const { meta: { fields = [] } } = Papa.parse(raw.trim(), { delimiter: separator, header: true, preview: 1 })
  const trimmed = fields.map((f) => f.trim())
  if (typeof column === 'number') return column >= 0 && column < trimmed.length
  return trimmed.includes(column.trim())
}

When it happens

Trigger: User typed 'email_address' but the header is 'email'; numeric index out of range (e.g., 5 in a 3-column file maps to undefined); trailing whitespace in the header name (e.g., 'email ' vs 'email'); case mismatch.

Common situations: Schema drift between the expected and actual CSV header; column name copy-pasted from a spec but the file uses different naming; 1-based vs 0-based indexing confusion when passing a number.

Related errors

• Unable to resolve fields from header.
• Invalid JSON in the Chat NVIDIA NIM's baseOptions: ${excepti
• Invalid JSON in the OpenAIEmbedding's BaseOptions:
• Invalid JSON in the ChatOpenAI's BaseOptions:
• Invalid JSON in the OpenAI's BaseOptions:

AI-assisted analysis of FlowiseAI/Flowise@abe4a8601a (2026-08-12). Data as JSON: /api/errors/1f37e4d8ec71e908. Report an issue: GitHub.