Re: [GNUnet-developers] Binding scheduler with another event loop, gnunet-search and gnunet-publish metadata

[Amirouche Boubekki]

On 2019-03-22 18:31, Christian Grothoff wrote:
> On 3/22/19 5:37 PM, Amirouche Boubekki wrote:
> > Hello all,
> >
> >
> > A) I would like to know how to bind gnunet event loop
> >    aka. scheduler to another event loop that will be
> >    the main event loop without relying on threads.
> >    The reason is in scheme I can use call/cc to implement
> >    something similar to python async / await the end result
> >    is that there no more or at least less callbacks in my code.
> >
> >    Does anyone successfully integrated gnunet with another
> >    event loop?
> >
> >    Almost exactly last year, there was some discussion about
> >    this topic on IRC and gnunet-web.
>
> AFAIK only the existing 'select' implementation was finished. The API 
> to
> do this exists though: GNUNET_SCHEDULER_driver_init().
>
> > B) Does it make sense to group search queries using a single call
> >    to gnunet-search? Does it save bandwidth or something?
>
> It would not save bandwidth.
>
> > C) In gnunet-publish there is an option -m aka. meta:
> >
> >
> >    -m, --meta=TYPE:VALUE      set the meta-data for the given
> >                               TYPE to the given VALUE
> >
> >    How this metadata is used?
>
> It is included in the search result for the file.

Another batch of questions:

D) If I download a file, do I automatically share it and take part
   in distributing it or not?

E) Depending on the answer of the previous question, do I need to 
unindex
   a file I downloaded because I am not interested in sharing it for 
some
   reason?

F) Does it help to publish a file I downloaded?