Another thing worth mentioning is that not all invalid Unicode is Latin1, of course. It could just as well be CP-850 or KOI-8R or ISO2022-JP or KSC-5601 or GB2312 or any of at least a hundred other legacy encodings.
Another thing worth mentioning is that not all invalid Unicode is Latin1, of course. It could just as well be CP-850 or KOI-8R or ISO2022-JP or KSC-5601 or GB2312 or any of at least a hundred other legacy encodings.