https://bugzilla.novell.com/show_bug.cgi?id=672206
https://bugzilla.novell.com/show_bug.cgi?id=672206#c1 Sebastien Pouliot <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED CC| |[email protected] Resolution| |INVALID --- Comment #1 from Sebastien Pouliot <[email protected]> 2011-02-16 01:56:26 UTC --- That rule was a bit difficult to name (IMO fxcop equivalent is even more confusing) and maybe the documentation could be better (suggestions welcomed) although its example is a lot like your test case. Anyway the rule's goal is to promote API that do not require caller to specify generic types. By contrast your test case requires the caller to specify the type, which is a defect as far as the rule is concerned, because that makes the API harder to consume. Now if the generic type was used with one (or more) parameter(s) then the compiler could infer it and the API could be used without specifying the type (leading to a simpler to consume API, i.e. just like non-generic calls). Personally I believe there are valid cases for such API (and Gendarme test harness use some ;-). OTOH I understand that this is not something that should be abused since it can, often, be avoided, hence the rule's pertinence. Now FxCop only applies the rule to visible methods [1] because its 'FX' cop (even if it slowly tends to get less FX, and more application, oriented). In general Gendarme promotes that same rules, either externally for reusable frameworks or internally inside applications, even if, in some cases, the severity on non-visible code is reduced [2]. [1] you can configure Gendarme to skip non-visible code (for some, or all, rules) to get a more FX-cop like behavior. [2] which is not the case (presently) for this rule -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the QA contact for the bug. You are the assignee for the bug. _______________________________________________ mono-bugs maillist - [email protected] http://lists.ximian.com/mailman/listinfo/mono-bugs
