Spec

Extraits de code

curl "https://api.oanor.com/ieee754-api/v1/meta" \
  -H "x-oanor-key: oanor_test_..."

await fetch("https://api.oanor.com/ieee754-api/v1/meta", {
  headers: { "x-oanor-key": "oanor_test_..." }
});

$ch = curl_init("https://api.oanor.com/ieee754-api/v1/meta");
curl_setopt($ch, CURLOPT_HTTPHEADER, ["x-oanor-key: oanor_test_..."]);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
$out = curl_exec($ch);

import requests
requests.get(
    "https://api.oanor.com/ieee754-api/v1/meta",
    headers={"x-oanor-key": "oanor_test_..."}
)

Exemple de réponse

Une réponse réelle de ce endpoint, capturée lors du dernier contrôle de santé.

{
    "data": {
        "name": "IEEE 754 API",
        "notes": "Double has an 11-bit exponent (bias 1023) and 52-bit mantissa; single has 8-bit exponent (bias 127) and 23-bit mantissa. Encoding to single rounds the value. Bytes are laid out big-endian (most significant first). This inspects float bits — for integer base conversion use a base-convert API. Nothing is stored.",
        "version": "v1",
        "endpoints": [
            {
                "path": "/v1/encode",
                "params": {
                    "value": "the number (or inf, -inf, nan)",
                    "precision": "double (default, 64-bit) or single (32-bit)"
                },
                "returns": "the sign, exponent, mantissa, hex and binary layout"
            },
            {
                "path": "/v1/decode",
                "params": {
                    "hex": "the hex word (8 or 16 digits)",
                    "bits": "alternatively a 32- or 64-bit binary string",
                    "precision": "single or double (auto-detected from length)"
                },
                "returns": "the number and its IEEE 754 fields"
            },
            {
                "path": "/v1/meta",
                "params": [],
                "returns": "this document"
            }
        ],
        "description": "Inspect and build IEEE 754 floating-point numbers. The encode endpoint takes a number and shows exactly how it is stored in single (32-bit) or double (64-bit) precision: the sign bit, the raw and unbiased exponent, the mantissa, the full binary layout split into sign/exponent/mantissa, the hex word, and the classification (normal, subnormal, zero, infinity or NaN). For single precision it also shows the actual stored value after rounding. The decode endpoint goes the other way — give it a hex word or a binary string and it returns the number it represents and the same field breakdown. Pure local, no key."
    },
    "meta": {
        "timestamp": "2026-06-03T17:42:19.441Z",
        "request_id": "176e4023-5c92-4c4c-9e22-2b2ea9d21db5"
    },
    "status": "ok",
    "message": "Meta",
    "success": true
}