[pcap-ng-format] The "scope" of the Name Resolution Block

[Hadriel Kaplan]

On Tue, Sep 1, 2015 at 11:27 PM, Guy Harris <guy at alum.mit.edu> wrote:
>
> I'd prefer not to have to start a new section just because I clamshelled my laptop, with Wireshark running, in San Francisco, and reopened it in New York, so it might be nice to at least try to support changes to if_tzone - and to any future option with an IANA tz database tzid - in the process.

I'm fairly certain you're one of the few people on the planet who
would actually experience that scenario. ;)

But in all seriousness, that's a fairly rare thing and starting a new
SHB section in that rare event doesn't seem unreasonable. (or even
just adding a new set of IDBs in the same SHB, for the previous
interfaces, without creating a new SHB... or even for wireshark to
just error out and require you to start a new capture altogether)


> The application would have to manage its own map of packet ranges to time zone information, but it'll see the Info Changed messages, so it's at least *capable* of doing so.

Yeah, but that's exactly what I'm trying to avoid: keeping packet
ranges. So far nothing has required us to do that, and it adds
complexity for this rare event.

-hadriel