Wireshark · Ethereal-dev: Re: [Ethereal-dev] IrDA dissector plugin

[Jan Kiszka <jan.kiszka@xxxxxx>]

Guy Harris schrieb:
> On Wed, Jan 07, 2004 at 11:52:26AM -0800, Jean Tourrilhes wrote:
>
> > 	Well, it might be a case where the API/semantic of the
> > function would become so complex that the benefit of merging the code
> > would not pan out, because you would not actually reduce the
> > maintainance overhead.
>
>
> 85 lines removed from "packet-irda.c", 44 lines removed from "xdlc.c",
> and 51 lines added to "xdlc.h" (the latter two involve moving some
> #defines to "xdlc.h") as a result of the change.  The API isn't much
> more complicated - you have to pass a table of pointers to hf_ values
> for subfields of the control field (which also means that the rest of
> the dissectors using it have to add those fields - but that also means
> you can filter on those fields, e.g. asking for all XID frames).

Ok, I will have a look at it tomorrow - my online connection is too slow 
where I am today. Are there any changes on surface (item names and 
ordering), or is it just an internal restructuring?

> In the process of doing that, I noticed that the code to set the address
> columns was treating source and destination addresses as little-endian
> but the code to put them into the protocol tree treated them as
> big-endian.  The code now treats them as little-endian in both places.

Thanks for fixing. Little endian is the correct order.

Jan

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature