> It compiles but have never been tested on any packets.
I've tested it on a number of captures, and cleaned up some stuff.
> I tried to keep the functionality as close as possible to the original
> dissector, but there is one exception:
> I assume that every instance of a string for a ServerName is always exactly
> 16 characters (which might not
> be 0 terminated).
I changed it back to the original, as per my other mail.
I also changed some "proto_tree_add_uint()" calls to use
"proto_tree_add_item()", changed the fields' names to begin with
"browser" rather than "browse" (as "browser" is the display filter name
of the protocol), split the "election version" field into major and
minor OS version numbers (as per the spec I sent out; in some traces,
the numbers looked like NT major and minor version numbers, although a
domain/workgroup announcement puts something weird in there - I don't
believe NetMon's claim that it's a "comment pointer"), and made some
other changes.
