Again I think there are two different issues: 1. the API that classes must follow 2. serialization
The issue I point here is not related to how serialization works. It's a matter of "if a program faces a CLR classes and the program's logic depend on the classes' declaration (attributes, interfaces...), then mono classes should implement the same declaration as .NET ones". Actually, I came to this situation because the application I'm creating draw a map of classes' relationships. And I discover that my graphical chart of these relationships were different between Mono and .NET for DateTime... Then, relatively to the bug you mentioned, what I suggest should have no impact. Indeed it's quite easy to implement ISerializable without affecting Mono behavior (GetObject and the deserialization constructor would simply explicitly mimic what's implicitly done today) Do you see what I want to achieve? Robert Jordan wrote: > > Zacman wrote: >> I saw that one, but that's two different subjects. I'm not speaking about >> binary serialization or internal implementation, but about the public API >> that classes should expose. Today the public API of the DateTime class >> does >> not match .NET one (and bug 325067 is rather about how it works >> internally). >> >> So DateTime could implement ISerializable the way it wants (either to >> solve >> the bug mentioned or even by doing nothing special to match current >> behavior), but it should implement it, no? > > The bug mentions why ISerializable is of no use. You just have > to go through the whole thread. > > Robert > > _______________________________________________ > Mono-devel-list mailing list > [email protected] > http://lists.ximian.com/mailman/listinfo/mono-devel-list > > -- View this message in context: http://www.nabble.com/Mono%27s-DateTime-not-implement-same-interfaces-as-.NET-tp23175058p23196787.html Sent from the Mono - Dev mailing list archive at Nabble.com. _______________________________________________ Mono-devel-list mailing list [email protected] http://lists.ximian.com/mailman/listinfo/mono-devel-list
