Am Samstag 16 Januar 2010 14:06:58 schrieb Jean-Michel Pouré: > Le samedi 16 janvier 2010 à 08:20 +0200, Martin Paljak a écrit : > > I tried to answer twice to the ticket but my network connection > > dropped... > > Looking at the implementation of --pristine it is a "hidden" option > > and as that, I don't know how useful it is. > > I think we should be able to run pkcs15-init -E --create-pkcs15 in any > situation.
no, if there is nothing to erase, that will fail. "rm foo" also fails if there is no file "foo". maybe we should disable that combination at all, as it has problems: is the pin given with "--pin" for the erase, for the create or both? it is much easier if you run erase first (if you want that) and then create. > > In the long run, the current regression test suite should be replaced > > by a card specific testsuite and a "generic" testsuite (some basic > > features that should work on 80% of cards) > > I can help writing a test suite (to improve my knowledge) as I really > need to test and validate cards. But I need a reliable pkcs15-init -E > --create-pkcs15. Otherwise, the test suite will always fail. IMHO the best is to "fix" pkcs15-init and disallow giving both "erase" and "create" at the same time. Regards, Andreas _______________________________________________ opensc-devel mailing list opensc-devel@lists.opensc-project.org http://www.opensc-project.org/mailman/listinfo/opensc-devel
