From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============6590186832363238676==" MIME-Version: 1.0 From: Marcel Holtmann Subject: Re: [PATCH 0/4 v2] Network Time plugin Date: Tue, 21 Dec 2010 06:17:44 -0800 Message-ID: <1292941064.2658.29.camel@aeonflux> In-Reply-To: <1292916842.10058.5.camel@apaila-ThinkPad-T410> List-Id: To: ofono@ofono.org --===============6590186832363238676== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Hi Antti, please do not top-post on this mailing list. > Could someone review this patch. It has been waiting for review a while > now. Thanks. > > This series of patches introduces the network time part of the NITZ fea= ture = > > as outlined in 3GPP spec 22.042. The Network Time plugin has two DBUS = > > interfaces for client applications: notification signal and polling me= thod = > > call. The time information consists of three dictionary entries: 1) ti= me and = > > date in seconds from epoch (renormalized to time indicatation arrival = time); > > 2) daylight saving time; 3) timezone. I am still saying that the plugin should monitor the presence of timed and then just send a D-Bus message to timed. That saves us the problem with the timestamping. And in case timed is not running it should just set the current time directly. I really prefer that we don't hand out timestamps over the D-Bus API. That seems wrong to me. Regards Marcel --===============6590186832363238676==--