[Guy Harris <gharris@xxxxxxxxx>]

On Tue, Dec 18, 2001 at 10:10:21AM -0600, Frank Singleton wrote:
> On the other point about name collisions, we would need a
> nice "decode as" feature as I have discussed before. This one
> would have to handle selections to decode similar operation names
> both within the same IDL (subdissector), and in different IDL's.
> 
> Could we not extend the current "decode as" function to
> allow users to select which one they want. ie: "Decode as" can
> resolve not only down to subdissector level, but also down to registered
> functions within a subdissector.

If the names are "similar" but not the same, that shouldn't be an issue,
right?  The names should be sufficient to determine which operation it
is.

Is there ever a case where the *exact same* operation name is used twice
within the same IDL?  If so, how does that work?

If the same operation name is used in two *different* IDLs, that doesn't
require resolution to functions within a subdissector.  That just
requires a way of specifying - for a particular conversation, presumably
- which IDL's subdissector to use.