Ethereal-dev: Re: AW: AW: [Ethereal-dev] Plugin naming conventions?

Note: This archive is from the project's previous web site, ethereal.com. This list is no longer active.

From: Jaap Keuter <jaap.keuter@xxxxxxxxx>
Date: Fri, 17 Mar 2006 16:22:18 +0100 (CET)
Hi,

Refer to the comments in epan/prefs.c in the neighborhood of the
g_assert's for the reason why.

BTW: shouldn't these be DISSECTOR_ASSERT(), or does that only hold for the
dissection code itself.

Thanx,
Jaap

On Fri, 17 Mar 2006, Lars Roland wrote:

> Am 17.03.2006, 14:03 Uhr, schrieb Grossert, Alexander
> <Alexander.Grossert@xxxxxxxxxxx>:
>
> > Sure. It was just there because the plugin_register is of course only
> > called, when the code is compiled as a dll. In case of a built-in
> > dissector the test proto_register_xxx is called directly.
> >
> > By the way....PROBLEM SOLVED!
> >
> > I rebuilt the dev sources of ethereal completely and added the plugin
> > there, instead of the standard distribution. When I started it, a so
> > called "debug console" popped up to tell me that the protocol filter
> > name (it is described by parameter abbreviation in funtion
> > proto_register_protocol) contained "illegal" characters. In my case it
> > was just a capital character.
> >
> > Well, who would expect a capital letter in the protocol abbreviation to
> > crash the registration of the register process?
> >
> > Does someone have an idea why it has been implemented that way and why
> > registering chrashes due to this character?
> >
>
> I really don't know why capital letters aren't allowed in protocol filter
> names. There are display filter names with capital letters, e.g.
> "sip.Status-Code".
>
> The crash without debug message on Windows was a bug that was fixed after
> 0.10.14 release.
>
> Anyhow you should not use plugins with any not corresponding ethereal
> binary.
> If you wan't to use a plugin with official 0.10.14 binaries, you should
> also use the official ethereal sources for this version to build the
> plugin. The API and the ABI of libethereal is not stable.
>
> Regards,
> Lars
> _______________________________________________
> Ethereal-dev mailing list
> Ethereal-dev@xxxxxxxxxxxx
> http://www.ethereal.com/mailman/listinfo/ethereal-dev
>