Skip to content

E_LEN_001 — Array length mismatch

Two or more input arrays have different lengths but were expected to be parallel (i.e. each index represents the same time / row across all arrays). Common offenders: OHLCV bars where high / low / close / volume don't match open, or trade arrays where prices / quantities / is_buy don't match.

How to fix

Pad / trim the arrays so they all share the same length, then re-pass to the API:

import numpy as np

# Make sure every array has the same length before passing.
n = min(len(opens), len(highs), len(lows), len(closes))
eng.load_ohlcv({
    "ts": ts[:n],
    "open": opens[:n],
    "high": highs[:n],
    "low": lows[:n],
    "close": closes[:n],
    "volume": volumes[:n],
}, symbol="BTCUSDT")

const n = Math.min(opens.length, highs.length, lows.length, closes.length);
runner.runOhlcv(ts.slice(0, n), closes.slice(0, n), "BTC");

Common causes

  • Forgetting that pandas dropna() operates per-column — re-run on the full DataFrame, not on each Series individually.
  • Indicator computations producing shorter outputs than inputs (rolling windows) — align with the input length before passing.
  • Concatenating data from sources that disagree on the calendar.