On Thu, Jun 30, 2011 at 11:05 PM, Jean-Claude Beaudoin < jean.claude.beaudoin@gmail.com> wrote:
This makes, in a multi-threaded SBCL, impossible to predict the value of:
(equal (truename #P"") (truename #P""))
It may return T most of the time but not always since the current working directory may have been changed by some other thread between the two "truename" calls.
After verification I have to add that the statement here above is partially inexact for SBCL. In fact, SBCL seems to ignore the process current working directory after sampling it on startup and relies entirely on *default-pathname-defaults* ever after. But SBCL keeps *default-pathname-defaults* a true global variable and if one is not careful to rebind it locally in its own thread then you get the same unpredictable behavior stated above but for a somewhat different reason
mkcl-devel mailing list mkcl-devel@common-lisp.net http://lists.common-lisp.net/cgi-bin/mailman/listinfo/mkcl-devel