From skafetzo@netmode.ntua.gr Sun Jun 2 12:04:24 2013 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9478D21F8657 for ; Sun, 2 Jun 2013 12:04:24 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.11 X-Spam-Level: X-Spam-Status: No, score=-1.11 tagged_above=-999 required=5 tests=[BAYES_05=-1.11] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KkkbVlEnBxwq for ; Sun, 2 Jun 2013 12:04:18 -0700 (PDT) Received: from ulysses.noc.ntua.gr (ulysses.noc.ntua.gr [IPv6:2001:648:2000:de::230]) by ietfa.amsl.com (Postfix) with ESMTP id 63AED21F8607 for ; Sun, 2 Jun 2013 12:04:15 -0700 (PDT) Received: from netmode.ece.ntua.gr (dolly.netmode.ece.ntua.gr [147.102.13.10]) by ulysses.noc.ntua.gr (8.14.5/8.14.5) with ESMTP id r52J4DPN085170 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Sun, 2 Jun 2013 22:04:13 +0300 (EEST) (envelope-from skafetzo@netmode.ntua.gr) Received: from [192.168.2.6] (ppp141237156212.access.hol.gr [141.237.156.212]) (authenticated bits=0) by netmode.ece.ntua.gr (8.14.4/8.14.3) with ESMTP id r52J3ls4003282 (version=TLSv1/SSLv3 cipher=DHE-RSA-CAMELLIA256-SHA bits=256 verify=NOT) for ; Sun, 2 Jun 2013 22:03:48 +0300 (EEST) (envelope-from skafetzo@netmode.ntua.gr) Message-ID: <51AB9713.3040104@netmode.ntua.gr> Date: Sun, 02 Jun 2013 22:03:47 +0300 From: Stella Kafetzoglou User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6 MIME-Version: 1.0 To: dtn-users@irtf.org Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.2.7 (ulysses.noc.ntua.gr [147.102.222.230]); Sun, 02 Jun 2013 22:04:13 +0300 (EEST) X-Virus-Scanned: clamav-milter 0.97.5 at ulysses.noc.ntua.gr X-Virus-Status: Clean Subject: [dtn-users] New convergence layer X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 02 Jun 2013 19:23:01 -0000 Hi all, I am new to DTN so this may seem a naive question. Nevertheless, I wish to implement a new Convergence layer for DTN2. Is there is a manual that I can get advice from? Moreover, I saw that the external convergence layer is deprecated. Is this the case or should I use this? Thank you in advance Stella From sickmind@lavabit.com Sun Jun 2 12:45:06 2013 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D7E2421F918C for ; Sun, 2 Jun 2013 12:45:05 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tidiiOSVw0hA for ; Sun, 2 Jun 2013 12:45:00 -0700 (PDT) Received: from karen.lavabit.com (karen.lavabit.com [72.249.41.33]) by ietfa.amsl.com (Postfix) with ESMTP id 4EBF321F9052 for ; Sun, 2 Jun 2013 12:45:00 -0700 (PDT) Received: from e.earth.lavabit.com (e.earth.lavabit.com [192.168.111.14]) by karen.lavabit.com (Postfix) with ESMTP id 7903E11BBE8; Sun, 2 Jun 2013 14:44:59 -0500 (CDT) Received: from localhost (pppoe.178-66-161-34.dynamic.avangarddsl.ru [178.66.161.34]) by lavabit.com with ESMTP id 6KIG360BCSID; Sun, 02 Jun 2013 14:44:59 -0500 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=lavabit; d=lavabit.com; b=U5NXkbJAE9MbiJxBIi09BQ1X39VE6eAT/DjEX4+whRrJr3fBwWhJCn6lgR/7c3s1sJWqA/rudmO+PkRkaqbqhkoq0XYGDgEkibDA4Z4wi+2/bNkpq0HbUtO4L2gPtUpXSjJH1Smr3j2EJPtO3ZzxyrcRYR03IQQB1Va6lJNEL1o=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version:Content-Type:Content-Disposition:In-Reply-To:User-Agent; Date: Sun, 2 Jun 2013 19:39:38 +0000 From: Lana Black To: Stella Kafetzoglou Message-ID: <20130602193938.GB16333@glow> References: <51AB9713.3040104@netmode.ntua.gr> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <51AB9713.3040104@netmode.ntua.gr> User-Agent: Mutt/1.5.21 (2010-09-15) Cc: dtn-users@irtf.org Subject: Re: [dtn-users] New convergence layer X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 02 Jun 2013 19:45:06 -0000 On 22:03 Sun 02 Jun , Stella Kafetzoglou wrote: > Hi all, > I am new to DTN so this may seem a naive question. Nevertheless, I wish > to implement a new Convergence layer for DTN2. Is there is a manual that > I can get advice from? > Moreover, I saw that the external convergence layer is deprecated. Is > this the case or should I use this? > > Thank you in advance > Stella Hi Stella, I suppose the current implementation of external convergence layers interface is deprecated, yes (I wonder why never caught attention). The best manual at this point is the code itself which is pretty straightforward. There is also a functional specification by Elwyn Davies and Avri Doria (www.n4c.eu/Download/n4c-wp2-023-dtn-infrastructure-fs-12.pdf) which you may find quite useful. With best regards. From elwynd@folly.org.uk Sun Jun 2 16:34:34 2013 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 78F8621F85E0 for ; Sun, 2 Jun 2013 16:34:34 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HRCgiaYFd+2P for ; Sun, 2 Jun 2013 16:34:33 -0700 (PDT) Received: from auth.a.painless.aa.net.uk (a.painless.aa.net.uk [IPv6:2001:8b0:0:30::51bb:1e33]) by ietfa.amsl.com (Postfix) with ESMTP id 387BC21F8556 for ; Sun, 2 Jun 2013 16:34:33 -0700 (PDT) Received: from 247.254.187.81.in-addr.arpa ([81.187.254.247]) by a.painless.aa.net.uk with esmtpsa (TLSv1:RC4-MD5:128) (Exim 4.77) (envelope-from ) id 1UjHnA-0004S3-7p; Mon, 03 Jun 2013 00:34:30 +0100 Date: Mon, 03 Jun 2013 00:34:28 +0100 Message-ID: <01fb8c6ti2plgawksx0ody9j.1370211648119@email.android.com> From: Elwyn Davies To: Lana Black MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: base64 X-Spam-Score-a.painless.aa.net.uk: -4.0 Cc: dtn-users@irtf.org Subject: Re: [dtn-users] New convergence layer X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 02 Jun 2013 23:34:34 -0000 VGhlIEV4dGVybmFsIENvbnZlcmdlbmNlIExheWVyIGlzIG5vdCBhY3R1YWxseSBkZXByZWNhdGVk IC0gaXQgaXMganVzdCB0aGF0IGl0IGhhcyBzZWVuIHZlcnkgbGl0dGxlIHVzYWdlLiAgVG8gdGhl IGJlc3Qgb2YgbXkga25vd2xlZGdlIHRoaXMgcGFwZXIKaHR0cDovL3d3dy5jbC5jYW0uYWMudWsv fmZiMzc1L2V4dHJlbWVjb20vMjAxMC9QZXR6RXh0cmVtZUNvbTEwLnBkZiBpcyB0aGUgbW9zdCBy ZWNlbnQgYXBwbGljYXRpb24gIHRvIHRyeSBpdC4gIEhvd2V2ZXIgYWx0aG91Z2ggdGhlcmUgaXMg YSBwYXBlciwgdGhlIGNvZGUgZG9lc24ndCBzZWVtIHRvIGhhdmUgYmVlbiBtYWRlIHB1YmxpYy4g IFlvdSBjb3VsZCB0cnkgZW1haWxpbmcgQWdvc3RvbiBQZXR6IHdobyBpcyBzdGlsbCBkb2luZyBo aXMgUGhEIGF0IFUgb2YgVGV4YXMvQXVzdGluLgoKQW5vdGhlciBjYXNlIHRoYXQgaGFzIHVzZWQg dGhlIEV4dGVybmFsIENMIGlzIHRoZSBjb25uZWN0aW9uIHRvIHRoZSBPTkUgc2ltbGF0b3IgLSBz ZWUgaHR0cDovL3d3dy5uZXRsYWIudGtrLmZpL3R1dGtpbXVzL2R0bi90aGVvbmUvcHViL3RoZV9v bmVfc2ltdXRvb2xzLnBkZiAoVGhlIE9ORSBTaW11bGF0b3IgZm9yIERUTiBQcm90b2NvbCBFdmFs dWF0aW9uIC0gVEtLKSAgU2VlIFNlY3Rpb24gNC4yIGFuZCB0aGVuIHJlZmVyZW5jZSAxMi4gIEFn YWluIHRoZSBzb3VyY2UgY29kZSBkb2Vzbid0IHNlZW0gdG8gYmUgb25saW5lIGJ1dCBJJ2xsICBi dWcgdGhlIE9ORSB0ZWFtLgoKSWYgeW91IHdhbnQgdG8gYnVpbGQgYW4gaW50ZWdyYXRlZCBDTCBp biBEVE4yLCBtYWtlIHN1cmUgeW91IGhhdmUgYSBsb29rIGF0IHRoZSB2YXJpb3VzIGJhc2UgY2xh c3NlcyB0aGF0IGV4aXN0IC0gZm9yIGV4YW1wbGUsIHRoZSBUQ1AgY29udmVyZ2VuY2UgbGF5ZXIg YW5kIEJsdWV0b290aCBjb252ZXJnZW5jZSBsYXllcnMgcmVseSBvbiBhIGNvbW1vbiBiYXNlIGNs YXNzIHRoYXQgaXMgY29udmVuaWVudCBmb3IgYW55IHNvcnQgb2Ygc3RyZWFtIHByb3RvY29sLiAg Q0xzIGFyZSBxdWl0ZSBjb21wbGljYXRlZCBkdWUgdG8gdGhlIHVzdWFsIG5lZWQgZm9yIHJlY2Vp dmUgYW5kIHNlbmQgdGhyZWFkcyBpbiBhZGRpdGlvbiB0byB0aGUgbWFpbiBsaW5lIGNvZGUuCgpS ZWdhcmRzLApFbHd5biBEYXZpZXMKU2VudCBmcm9tIG15IEFTVVMgUGFkCgpMYW5hIEJsYWNrIDxz aWNrbWluZEBsYXZhYml0LmNvbT4gd3JvdGU6Cgo+T24gMjI6MDMgU3VuIDAyIEp1biAgICAgLCBT dGVsbGEgS2FmZXR6b2dsb3Ugd3JvdGU6Cj4+IEhpIGFsbCwKPj4gSSBhbSBuZXcgdG8gRFROIHNv IHRoaXMgbWF5IHNlZW0gYSBuYWl2ZSBxdWVzdGlvbi4gTmV2ZXJ0aGVsZXNzLCBJIHdpc2ggCj4+ IHRvIGltcGxlbWVudCBhIG5ldyBDb252ZXJnZW5jZSBsYXllciBmb3IgRFROMi4gSXMgdGhlcmUg aXMgYSBtYW51YWwgdGhhdCAKPj4gSSBjYW4gZ2V0IGFkdmljZSBmcm9tPwo+PiBNb3Jlb3Zlciwg SSBzYXcgdGhhdCB0aGUgZXh0ZXJuYWwgY29udmVyZ2VuY2UgbGF5ZXIgaXMgZGVwcmVjYXRlZC4g SXMgCj4+IHRoaXMgdGhlIGNhc2Ugb3Igc2hvdWxkIEkgdXNlIHRoaXM/Cj4+IAo+PiBUaGFuayB5 b3UgaW4gYWR2YW5jZQo+PiBTdGVsbGEKPgo+SGkgU3RlbGxhLAo+Cj5JIHN1cHBvc2UgdGhlIGN1 cnJlbnQgaW1wbGVtZW50YXRpb24gb2YgZXh0ZXJuYWwgY29udmVyZ2VuY2UgbGF5ZXJzCj5pbnRl cmZhY2UgaXMgZGVwcmVjYXRlZCwgeWVzIChJIHdvbmRlciB3aHkgbmV2ZXIgY2F1Z2h0IGF0dGVu dGlvbikuIFRoZQo+YmVzdCBtYW51YWwgYXQgdGhpcyBwb2ludCBpcyB0aGUgY29kZSBpdHNlbGYg d2hpY2ggaXMgcHJldHR5IAo+c3RyYWlnaHRmb3J3YXJkLgo+Cj5UaGVyZSBpcyBhbHNvIGEgZnVu Y3Rpb25hbCBzcGVjaWZpY2F0aW9uIGJ5IEVsd3luIERhdmllcyBhbmQgQXZyaSBEb3JpYQo+KHd3 dy5uNGMuZXUvRG93bmxvYWQvbjRjLXdwMi0wMjMtZHRuLWluZnJhc3RydWN0dXJlLWZzLTEyLnBk Zikgd2hpY2ggeW91Cj5tYXkgZmluZCBxdWl0ZSB1c2VmdWwuCj4KPldpdGggYmVzdCByZWdhcmRz Lgo+Cj5fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXwo+ZHRu LXVzZXJzIG1haWxpbmcgbGlzdAo+ZHRuLXVzZXJzQGlydGYub3JnCj5odHRwczovL3d3dy5pcnRm Lm9yZy9tYWlsbWFuL2xpc3RpbmZvL2R0bi11c2Vycwo= From sickmind@lavabit.com Sun Jun 2 19:10:03 2013 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6A2A221F8E89 for ; Sun, 2 Jun 2013 19:10:03 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pU50nt0uXi63 for ; Sun, 2 Jun 2013 19:09:59 -0700 (PDT) Received: from karen.lavabit.com (karen.lavabit.com [72.249.41.33]) by ietfa.amsl.com (Postfix) with ESMTP id 561B821F8EAE for ; Sun, 2 Jun 2013 19:09:57 -0700 (PDT) Received: from f.earth.lavabit.com (f.earth.lavabit.com [192.168.111.15]) by karen.lavabit.com (Postfix) with ESMTP id DDCAB11BB2D; Sun, 2 Jun 2013 21:09:54 -0500 (CDT) Received: from localhost (pppoe.178-66-134-134.dynamic.avangarddsl.ru [178.66.134.134]) by lavabit.com with ESMTP id RGB0DNR4LXKC; Sun, 02 Jun 2013 21:09:54 -0500 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=lavabit; d=lavabit.com; b=RFGgb39AXWFPOycrG4dMsOZhzu2pDIM7LEGRtCojk1uZSRPKa9r8CtFOv71MthtGgUfD0ocXyw1BzCQ0ntaPcPZWaHHCYmPUGI6+FkqBeuvUnXLhEgrP5z0/bf9DGMque60g3j9XbG0iPqwFYUjDO2D5AGPldR5SgpSItB+xOks=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version:Content-Type:Content-Disposition:In-Reply-To:User-Agent; Date: Mon, 3 Jun 2013 02:04:34 +0000 From: Lana Black To: Elwyn Davies Message-ID: <20130603020434.GC16333@glow> References: <01fb8c6ti2plgawksx0ody9j.1370211648119@email.android.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <01fb8c6ti2plgawksx0ody9j.1370211648119@email.android.com> User-Agent: Mutt/1.5.21 (2010-09-15) Cc: dtn-users@irtf.org Subject: Re: [dtn-users] New convergence layer X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 03 Jun 2013 02:10:03 -0000 On 00:34 Mon 03 Jun , Elwyn Davies wrote: > The External Convergence Layer is not actually deprecated - it is just that it has seen very little usage. To the best of my knowledge this paper > http://www.cl.cam.ac.uk/~fb375/extremecom/2010/PetzExtremeCom10.pdf is the most recent application to try it. However although there is a paper, the code doesn't seem to have been made public. You could try emailing Agoston Petz who is still doing his PhD at U of Texas/Austin. > > Another case that has used the External CL is the connection to the ONE simlator - see http://www.netlab.tkk.fi/tutkimus/dtn/theone/pub/the_one_simutools.pdf (The ONE Simulator for DTN Protocol Evaluation - TKK) See Section 4.2 and then reference 12. Again the source code doesn't seem to be online but I'll bug the ONE team. > > If you want to build an integrated CL in DTN2, make sure you have a look at the various base classes that exist - for example, the TCP convergence layer and Bluetooth convergence layers rely on a common base class that is convenient for any sort of stream protocol. CLs are quite complicated due to the usual need for receive and send threads in addition to the main line code. > > Regards, > Elwyn Davies > Sent from my ASUS Pad Looks like there has been a misunderstanding. Its is not the external convergence layer itself deprecated, but several helper classes. See Christopher Small's commit from Nov 25 2008. I think this deserves some attention. From elwynd@folly.org.uk Mon Jun 3 02:10:13 2013 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9767821F95E9 for ; Mon, 3 Jun 2013 02:10:12 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id MioEQlhEgk0Z for ; Mon, 3 Jun 2013 02:10:07 -0700 (PDT) Received: from a.painless.aa.net.uk (a.painless.aa.net.uk [IPv6:2001:8b0:0:30::51bb:1e33]) by ietfa.amsl.com (Postfix) with ESMTP id 1292E21F90EF for ; Mon, 3 Jun 2013 02:09:25 -0700 (PDT) Received: from mightyatom.folly.org.uk ([81.187.254.250]) by a.painless.aa.net.uk with esmtp (Exim 4.77) (envelope-from ) id 1UjQlJ-0004IG-ON; Mon, 03 Jun 2013 10:09:09 +0100 From: Elwyn Davies To: Lana Black In-Reply-To: <20130603020434.GC16333@glow> References: <01fb8c6ti2plgawksx0ody9j.1370211648119@email.android.com> <20130603020434.GC16333@glow> Content-Type: text/plain Organization: Folly Consulting Date: Mon, 03 Jun 2013 09:59:30 +0100 Message-Id: <1370249971.4596.7785.camel@mightyatom> Mime-Version: 1.0 X-Mailer: Evolution 2.26.3 Content-Transfer-Encoding: 7bit Cc: dtn-users@irtf.org Subject: Re: [dtn-users] New convergence layer X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 03 Jun 2013 09:10:13 -0000 On Mon, 2013-06-03 at 02:04 +0000, Lana Black wrote: > On 00:34 Mon 03 Jun , Elwyn Davies wrote: > > The External Convergence Layer is not actually deprecated - it is just that it has seen very little usage. To the best of my knowledge this paper > > http://www.cl.cam.ac.uk/~fb375/extremecom/2010/PetzExtremeCom10.pdf is the most recent application to try it. However although there is a paper, the code doesn't seem to have been made public. You could try emailing Agoston Petz who is still doing his PhD at U of Texas/Austin. > > > > Another case that has used the External CL is the connection to the ONE simlator - see http://www.netlab.tkk.fi/tutkimus/dtn/theone/pub/the_one_simutools.pdf (The ONE Simulator for DTN Protocol Evaluation - TKK) See Section 4.2 and then reference 12. Again the source code doesn't seem to be online but I'll bug the ONE team. > > > > If you want to build an integrated CL in DTN2, make sure you have a look at the various base classes that exist - for example, the TCP convergence layer and Bluetooth convergence layers rely on a common base class that is convenient for any sort of stream protocol. CLs are quite complicated due to the usual need for receive and send threads in addition to the main line code. > > > > Regards, > > Elwyn Davies > > Sent from my ASUS Pad > > Looks like there has been a misunderstanding. Its is not the external > convergence layer itself deprecated, but several helper classes. See > Christopher Small's commit from Nov 25 2008. I think this deserves some > attention. > True - but this doesn't stop the External CL compiling and linking when using GNU/g++ - I have it available in a more or less up to date version. I think we are supposed to use unordered_map instead of hash_map but the lack of usage has not made updating the code a priority for anyone. And there are several other usages of hash_map in more mainstream areas that probably ought to be fixed some time (notably Oasys StringUtils). OTOH this doesn't seem to have been a problem for anyone recently (e.g., someone not using GNU). /Elwyn From sickmind@lavabit.com Mon Jun 3 02:42:05 2013 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C233C21F93D4 for ; Mon, 3 Jun 2013 02:42:04 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fWYW0gUOPJIX for ; Mon, 3 Jun 2013 02:41:59 -0700 (PDT) Received: from karen.lavabit.com (karen.lavabit.com [72.249.41.33]) by ietfa.amsl.com (Postfix) with ESMTP id 636FE21F8609 for ; Mon, 3 Jun 2013 02:41:59 -0700 (PDT) Received: from d.earth.lavabit.com (d.earth.lavabit.com [192.168.111.13]) by karen.lavabit.com (Postfix) with ESMTP id A1EE511BB66; Mon, 3 Jun 2013 04:41:58 -0500 (CDT) Received: from localhost (193.160.158.7) by lavabit.com with ESMTP id 9XSY47ZAQLGS; Mon, 03 Jun 2013 04:41:58 -0500 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=lavabit; d=lavabit.com; b=J3I7XPMFh1Qrxl7ces9OkumXlhp0t+3ow9nTsZQzQUmfQTBsYR6ePrvlJ8bvM014jcQ25E9p/65rGr23PUssY0EEH6ffyewiCYZDpN+5c0IbK68PgT3Yk7SjWSkBzuIKf9IXL7Gd4/5l7H8Y4S7gpMIHOuxrG1E/YrqqGJ6acYQ=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version:Content-Type:Content-Disposition:In-Reply-To:User-Agent; Date: Mon, 3 Jun 2013 09:40:40 +0000 From: Lana Black To: Elwyn Davies Message-ID: <20130603094040.GA1879@kasumi> References: <01fb8c6ti2plgawksx0ody9j.1370211648119@email.android.com> <20130603020434.GC16333@glow> <1370249971.4596.7785.camel@mightyatom> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1370249971.4596.7785.camel@mightyatom> User-Agent: Mutt/1.5.21 (2010-09-15) Cc: dtn-users@irtf.org Subject: Re: [dtn-users] New convergence layer X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 03 Jun 2013 09:42:05 -0000 On 09:59 Mon 03 Jun , Elwyn Davies wrote: > On Mon, 2013-06-03 at 02:04 +0000, Lana Black wrote: > > On 00:34 Mon 03 Jun , Elwyn Davies wrote: > > > The External Convergence Layer is not actually deprecated - it is just that it has seen very little usage. To the best of my knowledge this paper > > > http://www.cl.cam.ac.uk/~fb375/extremecom/2010/PetzExtremeCom10.pdf is the most recent application to try it. However although there is a paper, the code doesn't seem to have been made public. You could try emailing Agoston Petz who is still doing his PhD at U of Texas/Austin. > > > > > > Another case that has used the External CL is the connection to the ONE simlator - see http://www.netlab.tkk.fi/tutkimus/dtn/theone/pub/the_one_simutools.pdf (The ONE Simulator for DTN Protocol Evaluation - TKK) See Section 4.2 and then reference 12. Again the source code doesn't seem to be online but I'll bug the ONE team. > > > > > > If you want to build an integrated CL in DTN2, make sure you have a look at the various base classes that exist - for example, the TCP convergence layer and Bluetooth convergence layers rely on a common base class that is convenient for any sort of stream protocol. CLs are quite complicated due to the usual need for receive and send threads in addition to the main line code. > > > > > > Regards, > > > Elwyn Davies > > > Sent from my ASUS Pad > > > > Looks like there has been a misunderstanding. Its is not the external > > convergence layer itself deprecated, but several helper classes. See > > Christopher Small's commit from Nov 25 2008. I think this deserves some > > attention. > > > True - but this doesn't stop the External CL compiling and linking when > using GNU/g++ - I have it available in a more or less up to date > version. > > I think we are supposed to use unordered_map instead of hash_map but the > lack of usage has not made updating the code a priority for anyone. > > And there are several other usages of hash_map in more mainstream areas > that probably ought to be fixed some time (notably Oasys StringUtils). > OTOH this doesn't seem to have been a problem for anyone recently (e.g., > someone not using GNU). > > /Elwyn I have found the similiar code from oasys causing compilation errors [1] when compiling with libc++ instead of the GNU C++ library. Many tests performed during ./configure run also fail for unknown reasons when using other than GNU libraries. Looks like we are facing portability issues here. Do we still need to support gcc versions 2 or 3? [1] Error message example: /util/StringUtils.h:70:22: error: no member named '__stl_hash_string' in namespace '__gnu_cxx' return _std::__stl_hash_string(str.c_str()); From elwynd@folly.org.uk Mon Jun 3 04:38:55 2013 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 96CBB21F961F for ; Mon, 3 Jun 2013 04:38:55 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0LXO4062M828 for ; Mon, 3 Jun 2013 04:38:51 -0700 (PDT) Received: from b.painless.aa.net.uk (b.painless.aa.net.uk [IPv6:2001:8b0:0:30::51bb:1e34]) by ietfa.amsl.com (Postfix) with ESMTP id 7A86221F9485 for ; Mon, 3 Jun 2013 04:38:51 -0700 (PDT) Received: from mightyatom.folly.org.uk ([81.187.254.250]) by b.painless.aa.net.uk with esmtp (Exim 4.72) (envelope-from ) id 1UjT68-0002VD-0O; Mon, 03 Jun 2013 12:38:48 +0100 From: Elwyn Davies To: Lana Black In-Reply-To: <20130603094040.GA1879@kasumi> References: <01fb8c6ti2plgawksx0ody9j.1370211648119@email.android.com> <20130603020434.GC16333@glow> <1370249971.4596.7785.camel@mightyatom> <20130603094040.GA1879@kasumi> Content-Type: text/plain Organization: Folly Consulting Date: Mon, 03 Jun 2013 12:29:08 +0100 Message-Id: <1370258948.4596.7787.camel@mightyatom> Mime-Version: 1.0 X-Mailer: Evolution 2.26.3 Content-Transfer-Encoding: 7bit X-ACL-Warn: Blacklisted URL in message. (ac.uk) in. See http://lookup.uribl.com. Cc: dtn-users@irtf.org Subject: Re: [dtn-users] New convergence layer X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 03 Jun 2013 11:38:55 -0000 On Mon, 2013-06-03 at 09:40 +0000, Lana Black wrote: > On 09:59 Mon 03 Jun , Elwyn Davies wrote: > > On Mon, 2013-06-03 at 02:04 +0000, Lana Black wrote: > > > On 00:34 Mon 03 Jun , Elwyn Davies wrote: > > > > The External Convergence Layer is not actually deprecated - it is just that it has seen very little usage. To the best of my knowledge this paper > > > > http://www.cl.cam.ac.uk/~fb375/extremecom/2010/PetzExtremeCom10.pdf is the most recent application to try it. However although there is a paper, the code doesn't seem to have been made public. You could try emailing Agoston Petz who is still doing his PhD at U of Texas/Austin. > > > > > > > > Another case that has used the External CL is the connection to the ONE simlator - see http://www.netlab.tkk.fi/tutkimus/dtn/theone/pub/the_one_simutools.pdf (The ONE Simulator for DTN Protocol Evaluation - TKK) See Section 4.2 and then reference 12. Again the source code doesn't seem to be online but I'll bug the ONE team. > > > > > > > > If you want to build an integrated CL in DTN2, make sure you have a look at the various base classes that exist - for example, the TCP convergence layer and Bluetooth convergence layers rely on a common base class that is convenient for any sort of stream protocol. CLs are quite complicated due to the usual need for receive and send threads in addition to the main line code. > > > > > > > > Regards, > > > > Elwyn Davies > > > > Sent from my ASUS Pad > > > > > > Looks like there has been a misunderstanding. Its is not the external > > > convergence layer itself deprecated, but several helper classes. See > > > Christopher Small's commit from Nov 25 2008. I think this deserves some > > > attention. > > > > > True - but this doesn't stop the External CL compiling and linking when > > using GNU/g++ - I have it available in a more or less up to date > > version. > > > > I think we are supposed to use unordered_map instead of hash_map but the > > lack of usage has not made updating the code a priority for anyone. > > > > And there are several other usages of hash_map in more mainstream areas > > that probably ought to be fixed some time (notably Oasys StringUtils). > > OTOH this doesn't seem to have been a problem for anyone recently (e.g., > > someone not using GNU). > > > > /Elwyn > > I have found the similiar code from oasys causing compilation errors [1] > when compiling with libc++ instead of the GNU C++ library. Many tests > performed during ./configure run also fail for unknown reasons when > using other than GNU libraries. Looks like we are facing portability > issues here. Do we still need to support gcc versions 2 or 3? > > [1] Error message example: > /util/StringUtils.h:70:22: error: no member named '__stl_hash_string' in > namespace '__gnu_cxx' > return _std::__stl_hash_string(str.c_str()); > Hi, Lana. Did you come up with a fix? Or resign yourself to living with GNU? I remember you mentioned something about other issues a while back. /Elwyn From sickmind@lavabit.com Mon Jun 3 05:27:57 2013 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B94DB21F85F7 for ; Mon, 3 Jun 2013 05:27:57 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ebtzSrRfE1qJ for ; Mon, 3 Jun 2013 05:27:52 -0700 (PDT) Received: from karen.lavabit.com (karen.lavabit.com [72.249.41.33]) by ietfa.amsl.com (Postfix) with ESMTP id 8206121F9648 for ; Mon, 3 Jun 2013 05:27:49 -0700 (PDT) Received: from f.earth.lavabit.com (f.earth.lavabit.com [192.168.111.15]) by karen.lavabit.com (Postfix) with ESMTP id ABCF411BB95; Mon, 3 Jun 2013 07:27:48 -0500 (CDT) Received: from localhost (pppoe.178-66-134-134.dynamic.avangarddsl.ru [178.66.134.134]) by lavabit.com with ESMTP id 0HR4M7OQ217S; Mon, 03 Jun 2013 07:27:48 -0500 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=lavabit; d=lavabit.com; b=L2wlfL7oLdD3D9oDLmsV8WQkj69YnXZrJlbpuS8phBd4HLqIQ+gxkstA0OiHdP4feq1qSljbO5ltSL2I+dj25eKKjL15hKM6zK3kCbxvSO8EWNMF6vtkc/GrB641YqbJgHCeJavKurCCKTRe7ojj3uxad2xGMLzQPhlNtB+gUPw=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version:Content-Type:Content-Disposition:In-Reply-To:User-Agent; Date: Mon, 3 Jun 2013 12:22:28 +0000 From: Lana Black To: Elwyn Davies Message-ID: <20130603122228.GA13214@glow> References: <01fb8c6ti2plgawksx0ody9j.1370211648119@email.android.com> <20130603020434.GC16333@glow> <1370249971.4596.7785.camel@mightyatom> <20130603094040.GA1879@kasumi> <1370258948.4596.7787.camel@mightyatom> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1370258948.4596.7787.camel@mightyatom> User-Agent: Mutt/1.5.21 (2010-09-15) Cc: dtn-users@irtf.org Subject: Re: [dtn-users] New convergence layer X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 03 Jun 2013 12:27:57 -0000 On 12:29 Mon 03 Jun , Elwyn Davies wrote: > On Mon, 2013-06-03 at 09:40 +0000, Lana Black wrote: > > I have found the similiar code from oasys causing compilation errors > > [1] > > when compiling with libc++ instead of the GNU C++ library. Many > > tests > > performed during ./configure run also fail for unknown reasons when > > using other than GNU libraries. Looks like we are facing portability > > issues here. Do we still need to support gcc versions 2 or 3? > > > > [1] Error message example: > > /util/StringUtils.h:70:22: error: no member named > > '__stl_hash_string' in > > namespace '__gnu_cxx' > > return _std::__stl_hash_string(str.c_str()); > > > > Hi, Lana. > > Did you come up with a fix? Or resign yourself to living with GNU? I > remember you mentioned something about other issues a while back. > > /Elwyn The most adequate solution would be rewriting the code in question to use unordered_{set,map} containers instead of hash_{set,map}, but by doing so we limit ourselves to C++11-compliant toolchains. We also can borrow hash_{set,map} implementations from libc++ (it has MIT license, so no problem with that) and including those into oasys. Both ways move us closer to the standard compliance, but the latter adds a little bit more code. What do you think, Elwyn? From elwynd@folly.org.uk Mon Jun 3 07:17:36 2013 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5107221F998D for ; Mon, 3 Jun 2013 07:17:36 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[AWL=0.000, BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LVxYN6vegg2P for ; Mon, 3 Jun 2013 07:17:16 -0700 (PDT) Received: from a.painless.aa.net.uk (a.painless.aa.net.uk [IPv6:2001:8b0:0:30::51bb:1e33]) by ietfa.amsl.com (Postfix) with ESMTP id A76BF21F8D16 for ; Mon, 3 Jun 2013 07:17:15 -0700 (PDT) Received: from mightyatom.folly.org.uk ([81.187.254.250]) by a.painless.aa.net.uk with esmtp (Exim 4.77) (envelope-from ) id 1UjVZP-0000cG-NL; Mon, 03 Jun 2013 15:17:11 +0100 From: Elwyn Davies To: Lana Black In-Reply-To: <20130603122228.GA13214@glow> References: <01fb8c6ti2plgawksx0ody9j.1370211648119@email.android.com> <20130603020434.GC16333@glow> <1370249971.4596.7785.camel@mightyatom> <20130603094040.GA1879@kasumi> <1370258948.4596.7787.camel@mightyatom> <20130603122228.GA13214@glow> Content-Type: text/plain Organization: Folly Consulting Date: Mon, 03 Jun 2013 15:07:29 +0100 Message-Id: <1370268449.4596.7920.camel@mightyatom> Mime-Version: 1.0 X-Mailer: Evolution 2.26.3 Content-Transfer-Encoding: 7bit Cc: dtn-users@irtf.org Subject: Re: [dtn-users] New convergence layer X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 03 Jun 2013 14:17:36 -0000 On Mon, 2013-06-03 at 12:22 +0000, Lana Black wrote: > On 12:29 Mon 03 Jun , Elwyn Davies wrote: > > On Mon, 2013-06-03 at 09:40 +0000, Lana Black wrote: > > > I have found the similiar code from oasys causing compilation errors > > > [1] > > > when compiling with libc++ instead of the GNU C++ library. Many > > > tests > > > performed during ./configure run also fail for unknown reasons when > > > using other than GNU libraries. Looks like we are facing portability > > > issues here. Do we still need to support gcc versions 2 or 3? > > > > > > [1] Error message example: > > > /util/StringUtils.h:70:22: error: no member named > > > '__stl_hash_string' in > > > namespace '__gnu_cxx' > > > return _std::__stl_hash_string(str.c_str()); > > > > > > > Hi, Lana. > > > > Did you come up with a fix? Or resign yourself to living with GNU? I > > remember you mentioned something about other issues a while back. > > > > /Elwyn > > The most adequate solution would be rewriting the code in question to > use unordered_{set,map} containers instead of hash_{set,map}, but by > doing so we limit ourselves to C++11-compliant toolchains. We also can > borrow hash_{set,map} implementations from libc++ (it has MIT license, > so no problem with that) and including those into oasys. Both ways move > us closer to the standard compliance, but the latter adds a little bit > more code. > > What do you think, Elwyn? > I feel a bit of fiddling with autoconf coming on. There are 3 cases seemingly: 1. Systems that have hash_map and hash_set as first class citizens so std::hash_map etc work. 2. Intermediate versions of GCC from v2.95 to v4.2.x that have hash_map and hash_set as extensions where we need to use __gnu_cxx::hash_map etc. These may also have experimental versions of unordered_* as std:tr1:unordered_map etc. 3. Recent versions of GCC (>v4.3.0) and other compilers that have unordered_map and unordered_hash as first class citizens so std::unordered_map etc work. Autoconf can check which of these applies and provide some macros. Case 3 conforms to C++11 standard as you mention. For GNU I suspect that you don't actually need to tell the compiler to conform to C++11 just to get the unordered_* library - haven't yet tried 4.3.x but 4.1.2 with std::tr1:: works without extra compile flags. Of course this may not be the case for other compilers that might be more pernickety. Do you have any evidence for other compilers? I'll see whether I can sort this out later this week. Regards, Elwyn From sickmind@lavabit.com Mon Jun 3 13:31:30 2013 Return-Path: X-Original-To: dtn-users@ietfa.amsl.com Delivered-To: dtn-users@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5BB9B11E80B8 for ; Mon, 3 Jun 2013 13:31:30 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[AWL=0.000, BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xwcsA-jTh7S0 for ; Mon, 3 Jun 2013 13:31:16 -0700 (PDT) Received: from karen.lavabit.com (karen.lavabit.com [72.249.41.33]) by ietfa.amsl.com (Postfix) with ESMTP id 5D8CA11E80A5 for ; Mon, 3 Jun 2013 13:21:37 -0700 (PDT) Received: from f.earth.lavabit.com (f.earth.lavabit.com [192.168.111.15]) by karen.lavabit.com (Postfix) with ESMTP id 0178111BB42; Mon, 3 Jun 2013 15:21:37 -0500 (CDT) Received: from localhost (pppoe.178-66-134-134.dynamic.avangarddsl.ru [178.66.134.134]) by lavabit.com with ESMTP id 41MUC02AY9AS; Mon, 03 Jun 2013 15:21:36 -0500 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=lavabit; d=lavabit.com; b=GSbznIlN1rOiVF4CqHbxtra10FRI3AviNiOGceIutQWqqhid1ldY2X5WP/qlUxVMeiIOaofs0WWr/hNn3RqOtY7JaTb0KzaxqJHjBHzOWOGDh+cW5DUkrPBV1XrLUFc2KxDCJ9A43ud84gq1AIE44LZPrZF6qpuZSE3YuZbb5UQ=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version:Content-Type:Content-Disposition:In-Reply-To:User-Agent; Date: Mon, 3 Jun 2013 20:16:17 +0000 From: Lana Black To: Elwyn Davies Message-ID: <20130603201617.GB13214@glow> References: <01fb8c6ti2plgawksx0ody9j.1370211648119@email.android.com> <20130603020434.GC16333@glow> <1370249971.4596.7785.camel@mightyatom> <20130603094040.GA1879@kasumi> <1370258948.4596.7787.camel@mightyatom> <20130603122228.GA13214@glow> <1370268449.4596.7920.camel@mightyatom> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1370268449.4596.7920.camel@mightyatom> User-Agent: Mutt/1.5.21 (2010-09-15) Cc: dtn-users@irtf.org Subject: Re: [dtn-users] New convergence layer X-BeenThere: dtn-users@irtf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 03 Jun 2013 20:31:30 -0000 On 15:07 Mon 03 Jun , Elwyn Davies wrote: > On Mon, 2013-06-03 at 12:22 +0000, Lana Black wrote: > > On 12:29 Mon 03 Jun , Elwyn Davies wrote: > > > On Mon, 2013-06-03 at 09:40 +0000, Lana Black wrote: > > > > I have found the similiar code from oasys causing compilation errors > > > > [1] > > > > when compiling with libc++ instead of the GNU C++ library. Many > > > > tests > > > > performed during ./configure run also fail for unknown reasons when > > > > using other than GNU libraries. Looks like we are facing portability > > > > issues here. Do we still need to support gcc versions 2 or 3? > > > > > > > > [1] Error message example: > > > > /util/StringUtils.h:70:22: error: no member named > > > > '__stl_hash_string' in > > > > namespace '__gnu_cxx' > > > > return _std::__stl_hash_string(str.c_str()); > > > > > > > > > > Hi, Lana. > > > > > > Did you come up with a fix? Or resign yourself to living with GNU? I > > > remember you mentioned something about other issues a while back. > > > > > > /Elwyn > > > > The most adequate solution would be rewriting the code in question to > > use unordered_{set,map} containers instead of hash_{set,map}, but by > > doing so we limit ourselves to C++11-compliant toolchains. We also can > > borrow hash_{set,map} implementations from libc++ (it has MIT license, > > so no problem with that) and including those into oasys. Both ways move > > us closer to the standard compliance, but the latter adds a little bit > > more code. > > > > What do you think, Elwyn? > > > > I feel a bit of fiddling with autoconf coming on. > There are 3 cases seemingly: > 1. Systems that have hash_map and hash_set as first class citizens so > std::hash_map etc work. > 2. Intermediate versions of GCC from v2.95 to v4.2.x that have hash_map > and hash_set as extensions where we need to use __gnu_cxx::hash_map etc. > These may also have experimental versions of unordered_* as > std:tr1:unordered_map etc. > 3. Recent versions of GCC (>v4.3.0) and other compilers that have > unordered_map and unordered_hash as first class citizens so > std::unordered_map etc work. > > Autoconf can check which of these applies and provide some macros. > > Case 3 conforms to C++11 standard as you mention. For GNU I suspect > that you don't actually need to tell the compiler to conform to C++11 > just to get the unordered_* library - haven't yet tried 4.3.x but 4.1.2 > with std::tr1:: works without extra compile flags. Of course this may > not be the case for other compilers that might be more pernickety. Do > you have any evidence for other compilers? > > I'll see whether I can sort this out later this week. > > Regards, > Elwyn Clang uses either present GCC headers, or libc++ if invoked with -stdlib=libc++ flag. Libc++ entered the linux realm quite recently, and was confined to osx/bsd systems in the past, but it contains both ext/hash_{set,map} and unordered_{set,map}, so its support should not be a problem unless we try to use the library internals like in StringUtils.h.