On 5/24/11 2:31 PM, Mark Evenson wrote:
On 5/24/11 11:39 , Mark Evenson wrote: […]
Filed as ticket [#150]. Fixing the immediate cause in MAKE-PATHNAME makes other parts of ASDF which I'm looking into.
Looks to be finally fixed with [r13294][]. Fixing this resulted in sharpening our ANSI compliance considerably with Pathname version.
Further analysis of the ANSI test results has revealed a definite bug in that the internal PATHNAME copy constructor just ignored the VERSION component. This was probably why ABCL was sort of working despite some weirdness with version. Ticket #150 has been re-opened and a more comprehensive fix is being prepared.