As Guy already wrote,
Stabilizing the API is something that will/may happen in the future.
In the near/medium future the internal structures and api's are not
suitable to freeze or stabilize.
It will/may happen in the far future but not now.
Freezing the api's and structures now will only mean we lock ourself into a
set of api's and structures we already know right now are
significantly sub-optimal and sub-standard, and what benefit would
that have?
As of now, the only reason whatsoever there is even a libethereal
shared library is to reduce the size and thus the download bandwidth
requirements for the windows installer. no other reason.
Anyway, libetehreal is GPL and not LGPL which means any and every
program linking with it would have to be GPL as well so those
applications can just as easily be incorporated into the mainline
distribution of ethereal just as well.
I do not see that there is any demand whatsoever from other
open-source GPL projects to be able to link to a stable libethereal
anyway.
And for non opensource non GPL projects it is a moot point since they
would not be able to link with libetehreal anyway,
not now and not in the future.
If you have an application that needs to link to libethereal, send us
the application for review and it might be added to ethereal mainline.
Then we will maintain it and make sure that all the weekly api changes
are propagated to your extension/application.
On 8/11/05, Jeetendra Singh
<jeetendra_singh@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
>
>
>
>
> Are there any plans to formalize the release of an ethereal library?
>
> At some point we'd like to do that.
>
> I don't :-)
> It ties our hand much too tightly and for whos benefits?
> I beg to differ here. Exposing well defined interfaces for (lets say) at
> least registering dissectors, invoking dissectors, creating filters,
> Ethereal control plane commands (like start/stop capture etc) and allowing
> receipt of captured data through an API will only allow users to put more
> value add. As an start, monitoring can be automated and then are are
> limitless possibilities.
> Am not sure, whether community is stopping the users to exploit the
> possibilities which this great software currently provides by not exposing
> the usage interfaces.
> Am also not sure how this ties our hands? There are still well defined
> interfaces for doing the above stuff as far as I know. So why not stabilize
> the interfaces and untie our hands :)
>
>
> IMO we might
> add more applications to the ethereal source tree before we fix the
> libethereal api.
> Adding more application currently I presume will be more easier once
> proper APIs are defined and published.
> And think of the other such possible applications which can be created
> once lib ethereal is there.
> What I would rather propose here to have a clear roadmap of getting this
> thing done? :-)
> regards,
> Jeetendra
>