I have been having a lot of difficulty with data exported from my nation's People database. The problem is with accented characters and sometimes things like em and en dashes, apostrophes and inverted commas.

I generally have to read the CSV file into an intelligent text editor (like BBEdit in Mac OS), convert it to a known-good character set and then re-import it to the spreadsheet program.

But recently I've encountered a CSV export where I cannot find a character set that allows me to render accented characters correctly. I've tried UTF-8 (BOM and no BOM) ISO-Latin-1, ISO-Latin-9, Windows-Latin-1 and Mac OS Roman. All garble the accented characters.

And yet when I look at the user's profile in NationBuilder - specifically the custom fields I've had to set up - the data is fine. Accented characters render correctly.

This adds a lot of manual labour to the process of exporting (in my case) hundreds of records, because if I cannot get export working for accented characters I have to copy and paste data for each field for each record into the exported document.