nautechsystems/nautilus_trader · error

precision exceeded i8 range

Error message

precision exceeded i8 range

What it means

`decode_price_increment` converts a Databento price-increment raw value into a `Price`. The `precision: u8` parameter must fit in `i8` because it is negated into the price exponent; the `.expect` panics when precision exceeds 127. This can only happen with absurd precision values not produced by any real instrument definition.

Solutions

1. Validate the instrument definition's precision before decoding (clamp or reject precision > 127).
2. Check that the tick size / denominator conversion feeding `precision` is correct for the dataset.
3. If decoding third-party definitions, sanitize: use `precision.min(127)` or return a decode error instead of panicking.

Example fix

// before
let price = decode_price_increment(raw_tick, precision_u8);
// after
assert!(precision_u8 <= i8::MAX as u8, "unsupported precision {precision_u8}");
let price = decode_price_increment(raw_tick, precision_u8);

if precision_u8 > i8::MAX as u8 {
    return Err(format!("unsupported precision {precision_u8}"));
}
let price = decode_price_increment(raw_tick, precision_u8);

fn precision_ok(p: u8) -> bool { p <= i8::MAX as u8 }

When it happens

Trigger: Calling `decode_price_increment` (directly or via `decode_currency_pair`/`decode_equity`/`decode_futures_contract`/etc.) with a precision greater than 127, typically from a raw instrument definition whose `raw_tick`/denominator math overflowed or was computed incorrectly.

Common situations: Custom/incorrect Symbology or instrument-definition data from Databento, a hand-rolled decoder passing raw exponent bytes as precision, or a regression in `precision_from_raw` producing garbage values.

Understand the failure class

Background: "value must be between 0 and 1" / "out of range" / "must not be negative" errors: fixing range-validation failures across open-source libraries — this error's family across 42 libraries.

Related errors

• effective raw scale should fit in PriceRaw
• Failed to create avg_px price
• Failed to scale continuous-future adjustment to fixed…
• is not exactly representable with price precision
• invalid betting balance impact

AI-assisted analysis of nautechsystems/nautilus_trader@18893faf8b (2026-09-08). Data as JSON: /api/errors/20c5e467d45840c7. Report an issue: GitHub.

Appendix: source

Thrown at crates/adapters/databento/src/decode/primitives.rs:301

    }
    9 - trailing
}

/// Decodes a minimum price increment from the given value, expressed in units of 1e-9.
///
/// The precision is derived from the actual tick value to avoid truncation of
/// fractional tick sizes (e.g., treasury futures with 1/256 or 1/32 ticks).
/// The derived precision is floored at `precision` (typically the currency precision).
///
/// # Panics
///
/// Panics if `precision` exceeds the supported `Price` precision.
#[inline(always)]
#[must_use]
pub fn decode_price_increment(value: i64, precision: u8) -> Price {
    match value {
        0 | i64::MAX => {
            let exponent = i8::try_from(precision).expect("precision exceeded i8 range");
            Price::from_mantissa_exponent(1, -exponent, precision)
        }
        _ => {
            let derived = precision_from_raw(value).max(precision);
            Price::from_raw(decode_raw_price_i64(value), derived)
        }
    }
}

/// Decodes a quantity from the given value, expressed in standard whole-number units.
#[inline(always)]
#[must_use]
pub fn decode_quantity(value: u64) -> Quantity {
    quantity_from_whole(value)
}

/// Decodes a quantity from the given optional value, where `i64::MAX` indicates missing data.
///

View on GitHub (pinned to 18893faf8b)