pub fn rate_limit_is_explicit_null(json: &Value) -> boolExpand description
True when the usage payload is a valid response that explicitly reports no
active rate-limit window ("rate_limit": null).
The ChatGPT backend returns this when there is no usage recorded in the
current window (and it transiently returned it for every account during an
upstream incident). It is a benign “no data yet” state, not a malformed
payload, so callers should degrade gracefully (serve cache / show n/a)
rather than reporting an error. This mirrors the official codex client,
which maps a null rate_limit to empty windows instead of failing.