From mboxrd@z Thu Jan  1 00:00:00 1970
X-Msuck: nntp://news.gmane.io/gmane.emacs.gnus.general/68175
Path: news.gmane.org!not-for-mail
From: jidanni@jidanni.org
Newsgroups: gmane.discuss,gmane.emacs.gnus.general
Subject: gmane_link -- extract some URLs to a gmane article
Date: Sun, 18 Jan 2009 05:50:19 +0800
Message-ID: <87ocy5d2lg.fsf@jidanni.org>
References: <87hc424lwh.fsf@jidanni.org>
NNTP-Posting-Host: lo.gmane.org
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Trace: ger.gmane.org 1232229042 10864 80.91.229.12 (17 Jan 2009 21:50:42 GMT)
X-Complaints-To: usenet@ger.gmane.org
NNTP-Posting-Date: Sat, 17 Jan 2009 21:50:42 +0000 (UTC)
Cc: ding@gnus.org
To: gmane-discuss@hawk.netfonds.no
Original-X-From: gmane-discuss-bounces@hawk.netfonds.no Sat Jan 17 22:51:55 2009
Return-path: <gmane-discuss-bounces@hawk.netfonds.no>
Envelope-to: gd-gmane-discuss@m.gmane.org
Original-Received: from hawk.netfonds.no ([80.91.224.246])
	by lo.gmane.org with esmtp (Exim 4.50)
	id 1LOJ4z-0004gJ-8e
	for gd-gmane-discuss@m.gmane.org; Sat, 17 Jan 2009 22:51:45 +0100
Original-Received: from localhost ([127.0.0.1] helo=hawk.netfonds.no)
	by hawk.netfonds.no with esmtp (Exim 3.36 #1 (Debian))
	id 1LOJ3i-0003M0-00; Sat, 17 Jan 2009 22:50:26 +0100
Original-Received: from sd-green-bigip-177.dreamhost.com ([208.97.132.177]
	helo=homiemail-a1.g.dreamhost.com)
	by hawk.netfonds.no with esmtp (Exim 3.36 #1 (Debian))
	id 1LOJ3h-0003Lv-00
	for <gmane-discuss@hawk.netfonds.no>; Sat, 17 Jan 2009 22:50:25 +0100
Original-Received: from jidanni.org (122-127-32-36.dynamic.hinet.net [122.127.32.36])
	(using TLSv1 with cipher AES256-SHA (256/256 bits))
	(No client certificate requested)
	by homiemail-a1.g.dreamhost.com (Postfix) with ESMTP id D26D8119E23;
	Sat, 17 Jan 2009 13:50:21 -0800 (PST)
X-BeenThere: gmane-discuss@hawk.netfonds.no
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: gmane-discuss.hawk.netfonds.no
List-Unsubscribe: <http://hawk.netfonds.no/cgi-bin/mailman/listinfo/gmane-discuss>,
	<mailto:gmane-discuss-request@hawk.netfonds.no?subject=unsubscribe>
List-Archive: <http://hawk.netfonds.no/pipermail/gmane-discuss>
List-Post: <mailto:gmane-discuss@hawk.netfonds.no>
List-Help: <mailto:gmane-discuss-request@hawk.netfonds.no?subject=help>
List-Subscribe: <http://hawk.netfonds.no/cgi-bin/mailman/listinfo/gmane-discuss>,
	<mailto:gmane-discuss-request@hawk.netfonds.no?subject=subscribe>
Original-Sender: gmane-discuss-bounces@hawk.netfonds.no
Errors-To: gmane-discuss-bounces@hawk.netfonds.no
Xref: news.gmane.org gmane.discuss:12104 gmane.emacs.gnus.general:68175
Archived-At: <http://permalink.gmane.org/gmane.emacs.gnus.general/68175>

#!/usr/bin/perl

# Gentlemen, as you know on Gmane some articles, e.g.,
# http://article.gmane.org/gmane.org.wikimedia.mediawiki.api/873/raw
# lack the Archived-At: <http://permalink.gmane.org/gmane.org.wikimedia.mediawiki.api/873>
# header, even though that URL very much exists.

# That's where I step in. My masterfully crafted script below extracts
# the pertinent substitute links. Whilst not total "breast
# reconstruction" for the damaged article, at least the two substitute
# URLs should mean no more groping for the right links to tell your
# friends.

# In fact we print all the Gmane links for the article possible to
# compute offline we can, including those computed from any
# Archived-At header actually present.

# One might ask "why not fix the damage at the source?" Well it's
# attitudes like that lead to long unemployment lines!

# gmane_link -- extract some URLs to a gmane article fed in on stdin
# Copyright       : http://www.fsf.org/copyleft/gpl.html
# Author          : http://jidanni.org/
# Created On      : Sun Jan 18 04:24:07 2009
# Last Modified On: Sun Jan 18 05:48:58 2009
# Update Count    : 41
use strict;
use warnings FATAL => 'all';
while (<>) {
    if (/^Message-ID: <(.*)>/) {
        print <<EOF;
http://mid.gmane.org/$1
http://news.gmane.org/find-root.php?message_id=$1
EOF
    }
    elsif (/^Archived-At: <(.*)>/) {
        $_ = $1;
        print "$_\n";
        s/permalink(?=\.gmane)/article/ or die 'Not Gmane?';
        print "$_\n";
        s@$@/raw@;
        print "$_\n";
    }
    elsif (/^$/) { exit; }
}