Den 28. april 2017 17:37, skrev Richard Heck:
I agree it is a problem, but I've leanred enough about encodings to
know that dealing with them is difficult. If one has a file (or
string), it can be difficult to tell what the encoding is. But we
should perhaps do something when we cut and paste into LyX to try to
make sure we have UTF-8.
That would be very nice. If the original encoding is known, use a proper
conversion. If not, make an assumption about the encoding and convert
based on that. (A default encoding, or an attempt at autodetection.) If
the assumption is wrong, the string might look weird but the user should
notice that. The document will still compile & print, although it might
look ugly containing the wrong string.
The point is that ideally, LyX itself should never create an
uncompileable .lyx file. (Unless the user writes errors in ERT, includes
a bad .tex or mess up the .lyx file using some other editor.)
Helge Hafting
