Closed manics closed 4 years ago
Incidentally, came back to this as https://github.com/ome/omero-metadata/pull/40 now enforces the encoding of CSV files to support Unicode characters. Preliminary testing suggests that using utf-8-sig
should be sufficient to also handle this issue and gracefully handle CSV files with BOM.
So far, this feels like the most convenient solution to this issue as it should equally handle CSV files with or without BOM. I will push extra commits to https://github.com/ome/omero-metadata/pull/40 before putting it up for review.
If a CSV starts with a byte-order-mark
omero metadata populate
fails with an unhelpful message along the lines ofIt should either print out a helpful error message (BOM found) or just handle it.