On tor, 2008-05-15 at 16:11 +0200, Jan Safranek wrote:
> Magnus Fromreide wrote:
> > Do the rewrite affect both stream and packet sockets or only stream
> > sockets?
> 
> Currently both, but if there will be flag allowing asynchronous write, 
> like hardaker is suggesting, I'll have to rewrite it either.
> 
> > What action should be taken if the remote party fails to read and to
> > respond? Is this when OP_SEND_FAILED is called?
> 
> That's the question. Currently OP_SEND_FAILED, but discussion is welcome.
> 
> >> Comments? Thoughts?
> > 
> > Could you please show us what you have got?
> Today's snapshot is at http://people.redhat.com/jsafrane/net-snmp. Is 
> there any better way you prefer, like svn branches? It's really work in 
> progress.

patches are fine with me.

Now on to commenting the code:

You are changing the interface of snmp_select_info, that is a big no-no
since the interface is expected to be kept stable, at least in all 5.x
versions. This is also true for snmp_sess_select_info.

Many coments talk about read but then the code handles write. (But then
this is an early draft)

Please try to keep to 80 char line width. The SNMP_MALLOC_STRUCT and
SNMP_MALLOC_TYPEDEF macros can help with this.

Names starting with _ are reserved for the lib implementation - do not
use them here.

Good that you make your helper functions static.

Will look more later.

/MF



-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft 
Defy all challenges. Microsoft(R) Visual Studio 2008. 
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to