From mboxrd@z Thu Jan 1 00:00:00 1970 X-Msuck: nntp://news.gmane.io/gmane.emacs.gnus.general/24032 Path: main.gmane.org!not-for-mail From: Toby Speight Newsgroups: gmane.emacs.gnus.general Subject: Re: Unintended base64 encoding Date: 09 Jul 1999 12:55:51 +0100 Organization: Citrix Systems (Cambridge) Sender: owner-ding@hpc.uh.edu Message-ID: References: <874skefuuh.fsf@pc-hrvoje.srce.hr> NNTP-Posting-Host: coloc-standby.netfonds.no Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: main.gmane.org 1035161670 6518 80.91.224.250 (21 Oct 2002 00:54:30 GMT) X-Complaints-To: usenet@main.gmane.org NNTP-Posting-Date: Mon, 21 Oct 2002 00:54:30 +0000 (UTC) Return-Path: Original-Received: from farabi.math.uh.edu (farabi.math.uh.edu [129.7.128.57]) by sclp3.sclp.com (8.8.5/8.8.5) with ESMTP id IAA15178 for ; Fri, 9 Jul 1999 08:00:34 -0400 (EDT) Original-Received: from sina.hpc.uh.edu (lists@Sina.HPC.UH.EDU [129.7.3.5]) by farabi.math.uh.edu (8.9.1/8.9.1) with ESMTP id HAB19987; Fri, 9 Jul 1999 07:00:27 -0500 (CDT) Original-Received: by sina.hpc.uh.edu (TLB v0.09a (1.20 tibbs 1996/10/09 22:03:07)); Fri, 09 Jul 1999 07:00:54 -0500 (CDT) Original-Received: from sclp3.sclp.com (root@sclp3.sclp.com [204.252.123.139]) by sina.hpc.uh.edu (8.9.3/8.9.3) with ESMTP id HAA10122 for ; Fri, 9 Jul 1999 07:00:45 -0500 (CDT) Original-Received: from ironside.ansa.co.uk (ironside.ansa.co.uk [192.5.254.44]) by sclp3.sclp.com (8.8.5/8.8.5) with ESMTP id HAA15168 for ; Fri, 9 Jul 1999 07:59:43 -0400 (EDT) Original-Received: from lanber.cam.citrix.com (LANBER.ansa.co.uk [10.70.1.235]) by ironside.ansa.co.uk (Netscape Mail Server v2.02) with ESMTP id AAA82; Fri, 9 Jul 1999 13:00:19 +0100 Original-To: ding@gnus.org Original-Lines: 21 In-Reply-To: Lars Magne Ingebrigtsen's message of "Sat, 03 Jul 1999 09:22:12 GMT" X-Author-Info: Precedence: list X-Majordomo: 1.94.jlt7 Xref: main.gmane.org gmane.emacs.gnus.general:24032 X-Report-Spam: http://spam.gmane.org/gmane.emacs.gnus.general:24032 Lars> Lars Magne Ingebrigtsen 0> In , Lars wrote: Lars> Toby Speight writes: >> Why does it do that, rather than inspecting the content to decide >> the appropriate encoding? Lars> What is the appropriate encoding? base64 is always appropriate. Some time back (a couple of years, I think), I wrote code for TM which inspected the message to determine the most space-efficient encoding (if 8bit/binary weren't possible). So bodyparts containing mostly US-ASCII printable characters got q-p encoding, and others got base64, *irrespective of content-type*. AIUI, pgnus uses content-type or disposition as the primary heuristic. [For large parts, it may be a useful optimisation to only scan perhaps the first 1K and the last 1K and assume they are representative of the whole.]