Received: with ECARTIS (v1.0.0; list gopher); Sun, 23 Dec 2007 17:51:00 -0600 (CST) Received: from floodgap.com ([66.159.214.137] ident=elvis) by glockenspiel.complete.org with esmtp (Exim 4.63) id 1J6aaw-00009C-KT for gopher@complete.org; Sun, 23 Dec 2007 17:51:00 -0600 Received: (from spectre@localhost) by floodgap.com (6.6.6.666.1/2007.10.21) id lBNNouSL012062 for gopher@complete.org; Sun, 23 Dec 2007 15:50:56 -0800 From: Cameron Kaiser Message-Id: <200712232350.lBNNouSL012062@floodgap.com> Subject: [gopher] Re: Final dot required on server request? In-Reply-To: <20071223234049.GA10710@pongonova.net> from "brian@pongonova.net" at "Dec 23, 7 05:40:49 pm" To: gopher@complete.org Date: Sun, 23 Dec 2007 15:50:56 -0800 (PST) X-Mailer: ELM [version 2.4ME+ PL39 (25)] MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8bit X-Spam-Status: No (score 0.0): AWL=0.004 X-Virus-Scanned: by Exiscan on glockenspiel.complete.org at Sun, 23 Dec 2007 17:51:00 -0600 X-archive-position: 1764 X-ecartis-version: Ecartis v1.0.0 Sender: gopher-bounce@complete.org Errors-to: gopher-bounce@complete.org X-original-sender: spectre@floodgap.com Precedence: bulk Reply-to: gopher@complete.org List-help: List-unsubscribe: List-software: Ecartis version 1.0.0 List-Id: Gopher X-List-ID: Gopher List-subscribe: List-owner: List-post: List-archive: X-list: gopher > I noticed that PyGopherd does not appear to send a final "." character > prior to terminating the connection, but Bucktooth does. A quick read > of the RFC leads me to believe the "." is to be expected. Has this > been deprecated? It's not a big deal, just need to know how to write > a bit of code to either expect or not expect the dot. My interpretation of the standard is that it is recommended, but that clients should still handle an end-of-data condition gracefully. To this end, I've deliberately taken a compromise view in Bucktooth where "text" data has the final "." but binary data (or what Bucktooth perceives is binary) does not, for those clients that simply suck all data and assume that's the file. However, PyGopherd would also be spec-compliant insofar as the clients should handle the condition. -- ------------------------------------ personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser@floodgap.com -- This signature is free of dihydrogen monoxide! Ban it now! www.dhmo.org ----