Hi, Jörg Schaible wrote:
> Hello guys, > > the current implementation of the UserAuthenticationData only allows us to > keep character arrays as values for the user authentication. However, for > FTPS or SFTP (maybe more - HTTPS, WebDAVS) we may also have to use private > keys or client-side certificates. Unfortunately I cannot keep those in the > UserAuthenticationData. > > This results in half-baked solutions like provided in the patch of VFS-283 > where the passphase is delivered with the "standard" mechanism, but the > key data is individually set and on top of it a private UserAuthenticator > implementation that tries to overcome the limits of the > UserAuthenticationData. > > Therefore I'd like to apply my proposed changes in VFS-466, where I keep > binary compatibility, but have to deprecate the old getter/setter in favor > of new ones dealing with arbitrary data (defined by the enhanced inner > Type). > > WDYT? I reverted this for now. There are too many implications and its probably not the best time to start a refactoring near a new release. It might be better to think about the complete UserAuthenticator concept for 3.0. Sorry for the noise, Jörg --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
