On Mon, May 09, 2005 at 09:19:14PM +0200, Michael Schaap wrote: >On 9-May-2005 19:22, David Rothenberger wrote: > >> On 5/8/2005 7:26 PM, Christopher Faylor wrote: >> >>> On Sun, May 08, 2005 at 08:21:26PM -0400, Christopher Faylor wrote: >>> >>>> Ultimately, I just have to make strace and cygcheck understand the >>>> cygwin arguments and environment variables. Then we won't need this. >>> >>> >>> I would appreciate it if people would check out the latest snapshot to >>> verify if I actually got this working in all scenarios (directories >>> mounted with -X, -x, not mounted at all, or mounted without -X and -x). >> >> >> Does cygstart also need to be fixed? I've found that it doesn't >> propagate the full Cygwin environment when /bin is mounted in cygexec >> mode. >> >> % cygstart -- /bin/rxvt -e bash -c "'env; read x'" >> >> prints out a small set of environment variables when /bin is mounted >> in cygexec. When /bin is mounted normally, it gets the full environment. >> >Well, cygstart is a proper Cygwin executable. However, it does use a >Windows API call (ShellExecute, see "cygstart --reference") to execute >whatever needs to be started, so I can see how it might depend on a >properly synchronized Windows environment. > >If anyone can tell me how to do this, I'll be happy to make the change >to cygstart.
This shows what I did to cygcheck: http://sources.redhat.com/cgi-bin/cvsweb.cgi/winsup/utils/cygcheck.cc.diff?cvsroot=uberbaum&r1=1.65&r2=1.66 It's possible that the environment code may just work. Otherwise, you'll have to grab the environment from cygwin and build a new environment block, I assume. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
