On Sun, 2015-07-26 at 21:51 +0200, Michael Šimáček wrote: > It would probably be nicer to do the full cycle, but I'd like to > avoid > changes in behavior when porting from one library to another. And the > code above doesn't actually hold any connection, so it would require > more refactoring to make that happen. For now I would follow what the > original code was doing. As for the exceptions, I think it would > actually be justifiable to use the raw api's init_sec_context, > because > the high level api would just do the same call + the exception > handling > magic, which we want to avoid for now. Please let me know what do you > think. > Attaching updated patch that uses 'unicode' instead of > raw.display_name > and reverts back to using init_sec_context.
Sorry, but we should really not use the raw API here. If it means more changes to the code, so be it, please us the high level API as recommended by Robbie, we wrote a better API so that people would use it, and we want to apply best practices when changing code in IPA. Simo. -- Manage your subscription for the Freeipa-devel mailing list: https://www.redhat.com/mailman/listinfo/freeipa-devel Contribute to FreeIPA: http://www.freeipa.org/page/Contribute/Code