xsd format with imported BITS Textual Conventions error

First, sorry for the last post and it's attempted recall.
Next,
I have a file (test_tc.mi2 attaced) with a "BITS" TextualConvention declaration "LocationCategory". I also have an object "myGoodLocation" declared as being of type "LocationCategory". The resulting xsd file is correct. It contains the TC definition, the object definition and a new TC declaration "myGoodLocationBits".
I also import the TC in a file (test.mi2) and declare an object "myBadLocation" declared as being of type "LocationCategory". The resulting xsd file is NOT correct. The file contains "myBadLocation" declared as being of type "myBadLocationCategoryBits". However, the declaration of the "myBadLocationCategoryBits" TC is not present.
Steve

Hi Steve!
On Mon, Jan 27, 2003 at 11:25:17AM -0500, Steve Bucknam wrote:
I have a file (test_tc.mi2 attaced) with a "BITS" TextualConvention declaration "LocationCategory". I also have an object "myGoodLocation" declared as being of type "LocationCategory". The resulting xsd file is correct. It contains the TC definition, the object definition and a new TC declaration "myGoodLocationBits".
I also import the TC in a file (test.mi2) and declare an object "myBadLocation" declared as being of type "LocationCategory". The resulting xsd file is NOT correct. The file contains "myBadLocation" declared as being of type "myBadLocationCategoryBits". However, the declaration of the "myBadLocationCategoryBits" TC is not present.
Thanks for your bug report. It was the right motivation to change the behavior of smidump regarding to BITS related types :-). I think the best solution is to treat BITS based types exactly the same as the other types. If the type is an implicit type in the mib it will be also implicit in the schema. For explicit types and TCs, there will be only one type in the schema (and not three as there used to be). This generated type will have the same name as in the MIB.
Please note that I have not been able to test the new code yet. I will do that tomorrow (if I manage to repair my test program).
Regards, Torsten
participants (2)
-
Steve Bucknam
-
Torsten Klie