Numeric Types in Spot Responses

Spot endpoints return price, quantity, average_price, orig_qty, executed_qty, and similar fields as JSON numbers (not strings):

{
  "data": {
    "price": 26000,
    "quantity": 0.0009,
    "average_price": 0,
    "orig_qty": 0.0009,
    "executed_qty": 0
  }
}

Why this matters

Most languages parse JSON numbers as IEEE 754 double-precision floats by default. That's fine for display and arithmetic up to about 15 significant decimal digits, but don't use floats for financial calculations if you care about exact paise/satoshi precision. Common pitfalls:

• 0.1 + 0.2 != 0.3 in IEEE 754
• Quantities like 0.00005 may serialize as 5e-05 (still a valid JSON number, but parsers may surprise you)

Recommendations

Language	Parse strategy
Python	Use decimal.Decimal for arithmetic, parse with json.loads(s, parse_float=Decimal).
Java	Configure Jackson to use BigDecimal: mapper.enable(DeserializationFeature.USE_BIG_DECIMAL_FOR_FLOATS).
JavaScript / Node	Use a string-preserving parser (e.g. lossless-json) or convert to BigNumber/Decimal libraries.
Go	Decode into decimal.Decimal (e.g. shopspring/decimal).

Comparison with Futures

Futures responses use JSON strings for numerics ("price": "26000") — fixed-precision and lossless. Spot responses use JSON numbers.

If you write a single client that handles both surfaces, normalise everything to your preferred decimal type at parse time so downstream code doesn't have to think about the difference.