Hi,
I'd say using a generated field is more elegant :-)
/Anders
-----Original Message-----
From: wireshark-dev-bounces@xxxxxxxxxxxxx [mailto:wireshark-dev-bounces@xxxxxxxxxxxxx] On Behalf Of Roland Knall
Sent: den 31 oktober 2011 10:51
To: Developer support list for Wireshark
Subject: Re: [Wireshark-dev] Is it still ok to create hidden items ?
Hi
As I just came across something regarding this issue, there is a counter argument to the whole "if it is not there, the user may not find it" idea. Looking at the way the IP dissector is used, hidden fields have their merits. ip.addr is a more generic way of avoiding ( ip.src == x || ip.dest == x ). I plan to use it in the same way in the openSAFETY dissector, where I have the fields opensafety.msg.sender and opensafety.msg.receiver, and I am currently implementing a opensafety.msg.node matching either one.
The most elegant solution for such a case is still using hidden fields.
regards,
Roland
On Thu, Oct 27, 2011 at 4:04 PM, Teto <mattator@xxxxxxxxx> wrote:
> Thanks for both of your ideas. What bothers me with Michaels'idea is
> that I wonder how many wireshark users know of or use "contains" and
> "matches" compared to eq or == keywords. From that point of view,
> Jeff's idea looks as a good idea.
>
> On Thu, Oct 27, 2011 at 3:34 PM, Jeff Morriss <jeff.morriss.ws@xxxxxxxxx> wrote:
>>
>> Teto wrote:
>>>
>>> Hi,
>>>
>>> Just had a question about what's the best practice. I have a packet
>>> with a field contianing several keywords. I intend to split those
>>> keywords so that one can filter display based upon a keyword.
>>> My problem is am compelled to display each keyword separately (one
>>> itemp per kewyord and group them in a subtree) or could I display
>>> all of them in one item in the main tree (my preference) and then
>>> create several hidden fields (one per keyword). I wonder if that
>>> last
>>
>> Why not combine the two? Put one item (or maybe even just a text entry--from proto_tree_add_text()) with all the keywords (possibly added with proto_tree_append_text()) and then create a subtree below that with each keyword individually?
>>
>> This is how we get, for example, nice summary lines for the TCP protocol (including port numbers, etc.) while keeping the port numbers themselves as separate filterable items in the TCP subtree.
> ______________________________________________________________________
> _____ Sent via: Wireshark-dev mailing list
> <wireshark-dev@xxxxxxxxxxxxx>
> Archives: http://www.wireshark.org/lists/wireshark-dev
> Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
>
> mailto:wireshark-dev-request@xxxxxxxxxxxxx?subject=unsubscribe
___________________________________________________________________________
Sent via: Wireshark-dev mailing list <wireshark-dev@xxxxxxxxxxxxx>
Archives: http://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
mailto:wireshark-dev-request@xxxxxxxxxxxxx?subject=unsubscribe
