[Mono-bugs] [Bug 75246][Nor] Changed - KeyPair of AssemblyName used to create assembly has no effect on AssemblyName of AssemblyBuilder

[bugzilla-daemon at bugzilla.ximian.com]

Please do not reply to this email- if you want to comment on the bug, go to the
URL shown below and enter your comments there.

Changed by gert.driesen at pandora.be.

http://bugzilla.ximian.com/show_bug.cgi?id=75246

--- shadow/75246	2005-06-13 15:07:22.000000000 -0400
+++ shadow/75246.tmp.25044	2005-06-13 16:02:03.000000000 -0400
@@ -81,6 +81,37 @@
 split the tests.
 
 ------- Additional Comments From sebastien at ximian.com  2005-06-13 15:07 -------
 Created an attachment (id=15320)
 bug75246.diff
 
+
+------- Additional Comments From gert.driesen at pandora.be  2005-06-13 16:02 -------
+Sebastien,
+
+Not sure if I'll have time to look into it today, but here's already 
+a little feedback on your first remark:
+
+For a "manually" instantiated AssemblyName, MS.NET returns NULL for 
+GetPublicKey and GetPublicKeyToken (unless these have been explictly 
+set using SetPublicKey/SetPublicKeyToken, I assume).
+
+For an AssemblyName constructed from AssemblyBuilder or for a 
+previously emitted assembly, MS.NET 1.x returns NULL for 
+GetPublicKey and GetPublicKeyToken if the assembly is not signed.
+MS.NET 2.0 returns a zero-length byte array in this case.
+
+I think, we really need to make a difference between 
+- an AssemblyName that is constructed "manually"
+- an AssemblyName for an AssemblyBuilder
+- an AssemblyName for a previously emitted assembly
+
+and in some cases we need to have a different behaviour depending on 
+the profile (1.x/2.0).
+
+I'd be happy to assist in creating unit tests and modifications in 
+managed code, but I'm not enough familiar with the runtime (and I'm 
+afraid most of the work must be done there).
+
+Note: I have not verified this yet !! But I'm pretty sure lots (?) 
+of our AssemblyName tests (if they're not disabled)fail on MS.NET 
+1.x (and definitely on 2.0)