Return-Path: From: Alok To: BlueZ development In-Reply-To: <20080227134910.GA26414@localhost> References: <1204114452.25999.8.camel@greatbear> <20080227134910.GA26414@localhost> Date: Wed, 27 Feb 2008 19:38:30 +0530 Message-Id: <1204121310.25999.17.camel@greatbear> Mime-Version: 1.0 Subject: Re: [Bluez-devel] [HFP][Patch] Plugin. Reply-To: BlueZ development List-Id: BlueZ development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Sender: bluez-devel-bounces@lists.sourceforge.net Errors-To: bluez-devel-bounces@lists.sourceforge.net Hi Johan, Find my responses inline. On Wed, 2008-02-27 at 15:49 +0200, Johan Hedberg wrote: > Hi Alok, > > Some comments below. > > On Wed, Feb 27, 2008, Alok wrote: > > + AC_ARG_ENABLE(plugin, AC_HELP_STRING([--enable-plugin], [enable telephony plugins]), [ > > + plugin_enable=${enableval} > > + ]) > > + > > Maybe --enable-hfp-plugin would be better since it's not clear what > plugin is in question. > sure. "--enable-hfp-plugin" sounds fine. > > +if PLUGIN > > +plugindir = $(libdir)/bluetooth > > + > > +plugin_LTLIBRARIES = hfp_plugin_test.la > > + > > +hfp_plugin_test_la_SOURCES = hfp_plugin_test.c hfp_plugin.h > > +hfp_plugin_test_la_LDFLAGS = -module > > +hfp_plugin_test_la_LIBADD = @GLIB_LIBS@ > > +hfp_plugin_test_la_CFLAGS = @GLIB_CFLAGS@ > > +endif > > Isn't there rist that these files get left out of "make dist" if you > didn't give --enable-plugin to configure? > I am not sure I understand. Is there some rule missing or something? > > +int load_hfp_plugin(GKeyFile *config) > > +{ > > + gchar *str = NULL; > > + GError *err = NULL; > > + char filename[PATH_MAX]; > > + > > + str = g_key_file_get_string(config, "Headset", "Plugin", &err); > > Incorrect indentation and missing check for config == NULL Will fix that. > > > + if (load_hfp_plugin(config) < 0) > > + return -1; > > > I think we should keep the current policy that the audio service gets > run successfully with some sane set of defaults if there is no audio.conf > present in the filesystem. This change looks like it will make the audio > service fail either if audio.conf doesn't exist or if it doesn't contain > a Plugin entry. > As far as I remember, Marcel had commented that there should not be a default plugin. (in case of no audio.conf). But I am open to the idea. We can have the test plugin as the default plugin. > > --- audio/unix.c 8 Feb 2008 17:43:48 -0000 1.64 > > +++ audio/unix.c 26 Feb 2008 15:19:44 -0000 > > @@ -36,10 +36,11 @@ > > > > #include > > #include > > #include > > #include > > +#include > Why do you need to add this include since there are no other changes to > this file in your patch? > You are right. This is unnecessary. will remove it. > Johan > Thanks, Alok. ------------------------------------------------------------------------- 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/ _______________________________________________ Bluez-devel mailing list Bluez-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bluez-devel