ANNOUNCEMENT: Live Wireshark University & Allegro Packets online APAC Wireshark Training Session
April 17th, 2024 | 14:30-16:00 SGT (UTC+8) | Online

Wireshark-dev: Re: [Wireshark-dev] get_pdu_len signature change (TCP re-assembly) and SVN versi

From: Jaap Keuter <jaap.keuter@xxxxxxxxx>
Date: Tue, 21 Nov 2006 15:21:30 +0100 (CET)
Hi,

Solved in revision 19949. Involved dissector code and header,
README.developer and docbook source.

Thanx,
Jaap

On Tue, 21 Nov 2006, Jacques, Olivier (PD&E IT Test) wrote:

> Hello,
>
> with recent Wireshark SVN versions, my TCP plugins which use re-assembly
> are throwing exceptions.
> I found that there is a change that occured in "get_pdu_len" signature
> that happened in SVN 19751
> (http://www.mail-archive.com/wireshark-commits@xxxxxxxxxxxxx/msg01181.ht
> ml
> <http://www.mail-archive.com/wireshark-commits@xxxxxxxxxxxxx/msg01181.ht
> ml> ).
>
> The corresponding documentation in README.developer should change too to
> reflect that.
> In 2.7.1 Using tcp_dissect_pdus(), the text should read:
> -----------------------------------
>  a routine that takes as arguments a packet_info pointer, a tvbuff
> pointer and an offset
>  value representing the offset into the tvbuff at which a PDU
>  begins and should return - *without* throwing an exception (it
>  is guaranteed that the number of bytes specified by the previous
>  argument to tcp_dissect_pdus is available, but more data might
>  not be available, so don't refer to any data past that) - the
>  total length of the PDU, in bytes;
> -----------------------------------
> I updated my plugins and it seems to be OK.
>
> The ideal would be to protect the plugins from being incompatible from
> one Wireshark version to another (I know binary compatibility is not
> guaranteed, but I would like to make this as transparent as possible).
> The first idea that comes to me is having access to the SVN release at
> run time (compile time is not enough as we provide plugin binaries) and
> test against this.
>
> But maybe there is something that already exist for that?
>
> Thanks,
> Olivier.
>