Wireshark-bugs: [Wireshark-bugs] [Bug 3548] Cleanup header_field_info definitions
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3548
--- Comment #83 from Kovarththanan Rajaratnam <kovarththanan.rajaratnam@xxxxxxxxx> 2009-07-08 04:08:28 PDT ---
(In reply to comment #82)
> (In reply to comment #75)
> > Created an attachment (id=3285)
--> (https://bugs.wireshark.org/bugzilla/attachment.cgi?id=3285) [details] [details]
> > Enforce new constraints
>
> I don't know if we should enforce all this with asserts. We do have Lua
> scripts and Python bindings which can add fields, and the assert messages are
> not easy to understand.
We could use REPORT_DISSECTOR_BUG() instead and provide the user with some
guidance on how to correct the error ("Use BASE_NONE, 0, NULL...."). Would that
help?
> Does this extra checks increase the startup time?
We've started to use strlen() to detect trailing spaces so we might see an
increase here. I haven't done any benchmark on this. Could we guard the new
checks and provide an option to use them or not?
>
> Another constraint I'm unsure about is FT_BOOLEAN with 'display' value 8 and
> 'bitmask' value 0, which is the value used when BASE_NONE. I think we should
> allow this, as it describes the length (even if it's default).
I see your point but why allow this special case? To me, it would be like
allowing to specify 'display' value 32 for FT_UINT32. Why not be consistent and
simply disallow it? What do you think?
--
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
