Please sign in to comment.
Do not try to convert from ascii to utf8
ASCII is compatible with UTF8 so we need no conversion. If ASCII detection was bogus then we hope the file is really UTF8 already, in which case we loose NO characters. If it's something else we'll just loose the character (as it is already the case). Basically this change fixes the "misdetected UTF8" in a better way. See #904 for further details
- Loading branch information...
Showing with 13 additions and 16 deletions.