In going over unfinished libcdio work, is see there still is the problem mentioned in http://lists.gnu.org/archive/html/libcdio-devel/2005-11/msg00001.html
To rephrase the the problem a little, one looks up a name using iso9660_fs_readdir() and the name reported back may be a Joliet name. However when doing a lookup (such as using iso9660_fs_stat()?), a plain ISO 9660 name, not Joliet, is expected so that fails. The above rephrasing (which may or not be correct), is slanted a little becuase perhaps the problem is with iso9660_fs_stat() or whatever is getting used later, rather than providing an Joliet-to-iso9660 name translation or "untranslation" routine. It would help me greatly if someone would create a small test program that exhibits the current problem, so I'm sure I'm fixing the right bug. Side note: When I look at iso9660_fs_readdir() I see that b_mode2 parameter isn't used any more. (That's good because b_mode2 was a hack; a shame though that the parameter has to stick around for compatibility.) _______________________________________________ Libcdio-devel mailing list [email protected] http://lists.gnu.org/mailman/listinfo/libcdio-devel
