Question

Some byte sequences are apparently invalid in Unicode encodings, and I know that some languages (Python for one) throw an error when that happens.

My question is: what happens in Javascript when receiving such a sequence during an XMLHttpRequest or XDomainRequest? Does the resulting string:

1. Get truncated when that happens?
2. Skip the bad sequence and start at the next byte(s)?
3. Continue decoding and only show the replacement � character when displayed?

If 3, then does the charCodeAt function return a valid character code?

Answer 1

Number 3 happens. It shows the � when displayed, and charCodeAt returns 0xFFFD, the � unicode character.