[Bug 63982] CombinedRealm makes assumptions about principal implementation

2020-08-25 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=63982 --- Comment #9 from Michael Osipov --- (In reply to Ole Ostergaard from comment #8) > I'm hit by the upcoming removal of Realm#getRoles(). I was using the > Realm#getRoles() to get the Roles of a Principal. Is there an alternate way > to perfor

[Bug 63982] CombinedRealm makes assumptions about principal implementation

2020-08-25 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=63982 --- Comment #8 from Ole Ostergaard --- I'm hit by the upcoming removal of Realm#getRoles(). I was using the Realm#getRoles() to get the Roles of a Principal. Is there an alternate way to perform this action, in regard of the upcoming removal of

[Bug 63982] CombinedRealm makes assumptions about principal implementation

2019-12-06 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=63982 Michael Osipov changed: What|Removed |Added Status|NEW |RESOLVED Resolution|---

[Bug 63982] CombinedRealm makes assumptions about principal implementation

2019-12-03 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=63982 --- Comment #6 from Mark Thomas --- (In reply to Michael Osipov from comment #5) > Even if the behavior is not documented and an implementation detail? How > would you properly call #getRoles() from the CombinedRealm then? I wouldn't. I'd lea

[Bug 63982] CombinedRealm makes assumptions about principal implementation

2019-12-03 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=63982 --- Comment #5 from Michael Osipov --- (In reply to Mark Thomas from comment #4) > (In reply to Michael Osipov from comment #3) > > Even if, it has to be supported until Tomcat 10. Do you consider returning > > null is better here? That would m

[Bug 63982] CombinedRealm makes assumptions about principal implementation

2019-12-03 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=63982 --- Comment #4 from Mark Thomas --- (In reply to Michael Osipov from comment #3) > Even if, it has to be supported until Tomcat 10. Do you consider returning > null is better here? That would make like in CombinedRealm easier. It has to be pre

[Bug 63982] CombinedRealm makes assumptions about principal implementation

2019-12-03 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=63982 --- Comment #3 from Michael Osipov --- (In reply to Mark Thomas from comment #2) > I think getRoles() can be deprecated. It isn't used anywhere now. It was > added to support the failed GSoC JASPIC work. Even if, it has to be supported until T

[Bug 63982] CombinedRealm makes assumptions about principal implementation

2019-12-02 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=63982 --- Comment #2 from Mark Thomas --- I think getRoles() can be deprecated. It isn't used anywhere now. It was added to support the failed GSoC JASPIC work. The proposed solution for hasRole() looks reasonable to me. -- You are receiving this

[Bug 63982] CombinedRealm makes assumptions about principal implementation

2019-12-02 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=63982 --- Comment #1 from Remy Maucherat --- It is obvious reading the code in the realm package that it is assumed GenericPrincipal will have to be used, so that applies to this hypothetical CustomRealm as well. Of course, there are plenty of people