A recent R-devel commit introduces a change in the way non-printable Unicode characters are shown as an escape code. Whereas large code points were previously printed using an escape code of 8 hexadecimal digits, with initial zeros, the present code (tested with R-devel r79623 on Ubuntu Linux) only prints 6 hex digits. I think this may be problematic: it is now possible that R prints a character string which is not valid when reused as an input. See the following example.
"\U{16FE4}1"
# [1] "\U016fe41"
"\U016fe41"
# Error: invalid \Uxxxxxxxx value 16fe41 (line 1)
Best regards,
- Mikko Korpela
______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Locked
