illumina.util.load_csv assumes UTF-8, but in case there happens to be, say, an ISO/IEC 8859-1 0xCA (Ê) inserted into the file for some reason it'll crash with:
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xca in position 5534: invalid continuation byte
What's the "right" behavior here? Intentionally throw an exception for this? Allow these to be automatically stripped out with a warning?
ressy
commented
4 years ago
illumina.util.load_csvassumes UTF-8, but in case there happens to be, say, an ISO/IEC 8859-1 0xCA (Ê) inserted into the file for some reason it'll crash with:What's the "right" behavior here? Intentionally throw an exception for this? Allow these to be automatically stripped out with a warning?