[Jim Young <jyoung@xxxxxxx>]

Hey Michael,

Are there are any mnemonic option letters available?

Would use of long options be the appropriate solution in this case?

A few years ago I had a need for some additional options for a hacked up version of tshark.  Because there were not enough sensible option letters available I ended up using long options.  This worked out great.   Most of these local long options were very specific to the problem at hand but I suspect the --disable-protocol <proto_name> option might be usable to others.  I've just pushed patch to gerrit (9631) to add --disable-protocol option to tshark.  Perhaps that patch might give you some ideas.

Jim Y.

From: "mmann78@xxxxxxxxxxxx" <mmann78@xxxxxxxxxxxx>
Reply-To: "wireshark-dev@xxxxxxxxxxxxx" <wireshark-dev@xxxxxxxxxxxxx>
Date: Monday, July 13, 2015 8:27 PM
To: "wireshark-dev@xxxxxxxxxxxxx" <wireshark-dev@xxxxxxxxxxxxx>
Subject: Re: [Wireshark-dev] Enabling/disabling ANY heuristic dissector

Command-line option sounds good, but it will probably take longer to figure out the option letter (how many do we have left?) than the functionality that does the enable/disable.  Suggestions for option "letter" to use?  Have we gone beyond just letters yet?  A letter for each enable and disable sounds a bit greedy, so many comma separate "short name" with 0 or 1 for enable/disable?  I also agree that enable/disable protocols for the command line option should be ephemeral, however IF they are launched from Wireshark and then the Enabled Protocol dialog is launched and then saved, the command line option behavior will then be saved to the heuristic dissector file. TShark never has the opportunity to make the change permanent.

 

This should obviously be a separate patch from either the dialog or the preference removal.   I think the heuristic dialog is now ready for submittal (added the short name), but I have been hesitant about "dropping the hammer down" with the preference removal patch.  Having the command-line support before the preferences are removed should at least ease the transition.

 

 

 

-----Original Message-----
From: Pascal Quantin <pascal.quantin@xxxxxxxxx>
To: Developer support list for Wireshark <wireshark-dev@xxxxxxxxxxxxx>
Sent: Mon, Jul 13, 2015 10:03 am
Subject: Re: [Wireshark-dev] Enabling/disabling ANY heuristic dissector

Le 13 juil. 2015 3:32 PM, <mmann78@xxxxxxxxxxxx> a écrit :
>
> I thought somebody might complain about something like this, but I was more focused on the Wireshark (packet) context menu, where I was less inclined to make changes.  This however seems like a more valid use case to consider.  My question back would be - what "string" should be used by tshark?  The "display name" can have some undesirable characters in it from a command line perspective (ie probably require quotes), and the "internal" short name string isn't otherwise exposed for users to learn what is.
> Should the "short name" be exposed on the tabbed dialog so users can learn it to apply it to a (new) tshark option?
>  

I think we should expose the short name to users.
Preferences have their internal name displayed in a pop-up. We could either do the same, or have the internal name explicitly displayed in a column.
Should the enabled / disabled heuristic protocol given in the command line be ephemeral or persistent? I believe it should be the former, like the DL mapping value you can indicate manually in the command line and that does not get stored.

Pascal.

>  
>  
> -----Original Message-----
> From: Pascal Quantin <pascal.quantin@xxxxxxxxx>
> To: Developer support list for Wireshark <wireshark-dev@xxxxxxxxxxxxx>
> Sent: Mon, Jul 13, 2015 9:21 am
> Subject: Re: [Wireshark-dev] Enabling/disabling ANY heuristic dissector
>
>
> Le 13 juil. 2015 3:03 AM, <mmann78@xxxxxxxxxxxx> a écrit :
> >
> > With:
> >  
> > https://code.wireshark.org/review/9508/
> > https://code.wireshark.org/review/9610/
> > (and already submitted https://code.wireshark.org/review/9602/)
> >  
> > I consider this "feature complete enough for now".  If Qt wants to provide a better "user interface" for "heuristics in general", it certainly has some flexibility to do so.  Unless there are major issues/comments, I'll submit in a few days (presuming all pass Petri-Dish)
> Hi Michael,
> Sorry I come late in the discussion. I do not have access to a computer right now so I cannot easily look at the patch (the latest Gerrit diff page is rather smartphone unfriendly) but is there a way to activate heuristic dissectors from tshark / wireshark command line? I use an external tool launching both programs with the right command line and it would be a real functionality loss if it could not be done anymore.
> Note that I consider your overall goal as a good achievement (it was frustrating not to be able to deactivate easily some weak heuristics) but I would dislike losing the ability to activate on demand a given heuristic that is deactivated by default for performance reasons.
> Pascal.
>  
>
> ___________________________________________________________________________
> Sent via:    Wireshark-dev mailing list <wireshark-dev@xxxxxxxxxxxxx>
> Archives:    https://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:   
https://www.wireshark.org/lists/wireshark-dev
Unsubscribe:
https://wireshark.org/mailman/options/wireshark-dev
            
mailto:wireshark-dev-request@xxxxxxxxxxxxx?subject=unsubscribe