> From: Gavin Smith <gavinsmith0...@gmail.com> > Date: Thu, 17 Apr 2025 22:43:09 +0100 > Cc: "bug-texinfo@gnu.org" <bug-texinfo@gnu.org> > > It's not clear which parts of your patch fix which problem. I would guess > that the part for the MSYS2 console is the following: > > if test -n "$COMSPEC$ComSpec" \ > && uname | $EGREP -iv 'cygwin|msys' >/dev/null \ > - && test "$OSTYPE" != msys ; then > + && test "$OSTYPE" != msys \ > + && test "$OSTYPE" != cygwin ; then > path_sep=";" > else > path_sep=":" > fi > > That seems like a harmless patch to apply, as it mean that cygwin and > msys were treated the same.
But the question is: what does OSTYPE hold in Cygwin? If this also affects Cygwin, we need some Cygwin expert to tell us whether this could break the Cygwin port of texi2dvi. AFAIR, Cygwin emulates Posix, so the correct value of path_sep for it is ":", not ";".
