we-promise/sure · warning · Import::MaxRowCountExceededError

max_row_count_exceeded

max_row_count_exceeded

Error message

Import was uploaded but has too many rows to publish automatically.

What it means

Import#publish_later refuses to enqueue when rows_count exceeds max_row_count (10,000 in the base Import class at app/models/import.rb:529; AccountImport overrides; ImportSession checks the sum against SureImport.max_row_count). MaxRowCountExceededError means the file parsed and uploaded fine but is too large to auto-publish — the limit bounds job memory.

Source

Thrown at app/models/import.rb:236

          parsed_count += 1
          reasonable_count += 1 if reasonable_range.cover?(date)
        end

        { format: fmt, parsed: parsed_count, reasonable: reasonable_count }
      end

      # Filter to candidates that parsed at least one sample
      viable = scored.select { |s| s[:parsed] > 0 }
      return fallback if viable.empty?

      best = viable.max_by { |s| [ s[:parsed], s[:reasonable] ] }
      best[:format]
    end
  end

  def publish_later
    raise MaxRowCountExceededError if row_count_exceeded?
    raise "Import is not publishable" unless publishable?

    update! status: :importing

    ImportJob.perform_later(self)
  end

  # Whether import! already committed rows for this import. Distinguishes a
  # job that died mid-import (single transaction → rolled back, nothing
  # attached) from one that died after the data landed but before the status
  # write. Covers entry-producing imports and account-producing ones
  # (AccountImport creates accounts, not entries).
  def data_committed?
    entries.exists? || accounts.exists?
  end

  # Reaper guard: still wedged in a job-owned status and untouched since the
  # sweep's cutoff. Called under the record's row lock (fresh read).

View on GitHub (pinned to e69894adb9)

Solutions

1. Split the file into chunks under the limit and publish each as its own import
2. Pre-check import.rows_count > import.max_row_count before publish_later and tell the user to split, rather than letting the exception fire
3. If the limit is genuinely wrong for your install, deliberately raise Import#max_row_count (or SureImport.max_row_count) — do not silently bypass it

Example fix

// before
import.publish_later # raises Import::MaxRowCountExceededError

// after
if import.rows_count > import.max_row_count
  # surface a split-the-file message to the user
else
  import.publish_later
end

import.rows_count <= import.max_row_count # false => publish_later raises MaxRowCountExceededError; Preflight also warns at lint time

rescue Import::MaxRowCountExceededError in publish endpoints and return an unprocessable-entity response instructing the user to split the file

When it happens

Trigger: Uploading a CSV/QIF whose parsed rows_count exceeds max_row_count and then calling publish_later. Import::Preflight already emits the warning 'Row count exceeds this import type's publish limit.' at lint time (app/models/import/preflight.rb:149).

Common situations: Multi-year bank statement exports; users concatenating statements into one file; import types whose subclass limit differs from the base 10,000.

Related errors

• Import::MaxRowCountExceededError
• duplicate_headers
• file_too_large
• content_too_large
• invalid_file_type

AI-assisted analysis of we-promise/sure@e69894adb9 (2026-08-21). Data as JSON: /api/errors/d5c9e82ab8b24d9a. Report an issue: GitHub.