handle ascii range individually in each iconv case
authorRich Felker <dalias@aerifal.cx>
Fri, 10 Nov 2017 18:34:21 +0000 (13:34 -0500)
committerRich Felker <dalias@aerifal.cx>
Fri, 10 Nov 2017 18:34:21 +0000 (13:34 -0500)
short-circuiting low bytes before the switch precluded support for
character encodings that don't coincide with ascii in this range. this
limitation affected iso-2022 encodings, which use the esc byte to
introduce a shift sequence, and things like ebcdic.


No differences found