From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mga14.intel.com ([143.182.124.37]) by linuxtogo.org with esmtp (Exim 4.72) (envelope-from ) id 1TKxGo-0002Je-Pw for bitbake-devel@lists.openembedded.org; Sun, 07 Oct 2012 22:16:15 +0200 Received: from azsmga001.ch.intel.com ([10.2.17.19]) by azsmga102.ch.intel.com with ESMTP; 07 Oct 2012 13:03:07 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.80,547,1344236400"; d="scan'208";a="201594187" Received: from unknown (HELO helios.localnet) ([10.252.122.60]) by azsmga001.ch.intel.com with ESMTP; 07 Oct 2012 13:03:06 -0700 From: Paul Eggleton To: Chris Larson Date: Sun, 07 Oct 2012 21:03:05 +0100 Message-ID: <3573045.RzU7PMqlKt@helios> Organization: Intel Corporation User-Agent: KMail/4.9.2 (Linux/3.2.0-31-generic-pae; KDE/4.9.2; i686; ; ) In-Reply-To: References: <1349267277-1380-1-git-send-email-paul.eggleton@linux.intel.com> <2059965.DPhKPFNMcC@helios> MIME-Version: 1.0 Cc: bitbake-devel@lists.openembedded.org Subject: Re: [PATCH] fetch2: raise an exception if user specifies protocol=git with http:// X-BeenThere: bitbake-devel@lists.openembedded.org X-Mailman-Version: 2.1.11 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 07 Oct 2012 20:16:15 -0000 Content-Transfer-Encoding: 7Bit Content-Type: text/plain; charset="us-ascii" On Saturday 06 October 2012 15:38:45 Chris Larson wrote: > On Sat, Oct 6, 2012 at 3:32 PM, Paul Eggleton > wrote: > > On Saturday 06 October 2012 15:27:42 Chris Larson wrote: > >> On Wed, Oct 3, 2012 at 5:27 AM, Paul Eggleton > >> > >> wrote: > >> > It is a common mistake to use http:// and protocol=git when attempting > >> > to fetch from a git repository using the http protocol; if this is > >> > detected then throw an error explaining that you need to use git:// > >> > with > >> > protocol=http instead. > >> > > >> > Signed-off-by: Paul Eggleton > >> > >> Perhaps this should check the protocol= value against the protocols > >> supported by wget (ftp, http, https, etc), if the parameter is passed, > >> rather than hardcoding the exceptions. > > > > I don't think protocol= is currently used with the wget fetcher is it? > > It isn't, but it also doesn't seem ideal to hardcode knowledge of the > git fetcher in the wget one. Either way, I just wanted to make sure > both options were considered. I did consider checking for any use of protocol= and failing if any value is specified; perhaps we should still do that; for the purposes of helping people avoid a common pitfall however, it seems reasonable to me to check explicitly for protocol=git and I would still advocate leaving that check and message in even if we did add a general check on protocol. Cheers, Paul -- Paul Eggleton Intel Open Source Technology Centre