pub fn decode_fatal(bytes: &[u8]) -> Option<&str>Expand description
What:
Decode record bytes the way a fatal TextDecoder does:
valid UTF-8 only,
with one
leading byte-order mark removed.
Why:
The incumbent decodes owner and journal records with new TextDecoder('utf-8', { fatal: true }),
which rejects invalid bytes and drops a leading mark.
In TS you’d write (pseudocode):
new TextDecoder('utf-8', { fatal: true }).decode(bytes)