#!/usr/bin/perl

## Copyright (C) 2003 Jeremy Andrews
## This is free software; you can redistribute it and/or modify it under the
## terms of the GNU General Public License as published by the Free Software
## Foundation;  either version 2 of the License, or (at your option) any later
## version.
##
## This program is distributed in the hope that it will be useful, but WITHOUT
## ANY WARRANTY; without even the implied warrenty of MERCHANTABILITY or
## FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
## more details.
##
## Version 1.03  (October 08, 2003)
## 
## Modified:
##  08/2003 - Jeremy Andrews 
##    - cleaned up lots of false-positive bugs
##    - avoid encoding common html tags
##  10/2003 - JA 
##    - add Rusty Russel
##  11/2003 - JA
##    - switch to absolute paths (otherwise emails link to localhost)
##
## Simple perl script prepares page of emails for kerneltrap posting by:
##  - inserting <!--break--> separater at top of file for Drupal
##  - inserting horizontal rule at top of file
##  - wrapping block of emails with <pre></pre> tags
##  - strips email address (replacing with [email blocked])
##  - encodes < and > with &lt; and &gt; repectively
##  - urlifying http, telnet and ftp links
##  - wraps 'From: Some Body' in <b></b>'
##  - adds links to KernelTrap interviews, where applicable
##  - strips extraneous email headers
##  - strips all but first 'To:' header
##  - adds standard "related links" block at bottom
##
## Bugs:
##  - Doesn't handle html already part of email bodies well
##     (tends to encode when it shouldn't, and relinks html links, etc...
##  - Doesn't handle From: when name is in quotes
##  - Doesn't handle From: when full name is more than two names
##
## If you're good at perl, and can make this script better...  Please do!
## I can be reached by email at jeremy@kerneltrap.org

## We only allow one 'To:' header
$to=0;

## Begin with <!--break--> for Drupal, add line break and open <pre>
print "<!--break-->\n<hr><pre>\n";

$urls='(http|telnet|ftp)';
$ltrs = '\w';
$gunk = '/#~:.?+=&%@!\-';
$punc = '.:?\-';
$any = "${ltrs}${gunk}${punc}";
$strip_headers='(^To:|^Sender:|^Cc:|^Organization:|^X-Mailer:|^User-Agent:)';

while (<>) {

  ## Catch the unwrapped ones in the To: line...
  s{
    (
      To:
    )
    (
      ((\s*\w+)(-*))+(\w+)
    )
    (
      (@)((\w+)(\.))+(\w+)
    )
  }{To: $2}x;

  ## Try to strip email addresses...
  s{
    (
      (\s+)(<*)((\w+)(-*))+(\w+)(@)((\w+)(\.))+(\w+)(>*)
    )
  }{ [email blocked]}x;


  ## encode < and >
  if (!(m'<br>|<br />|<b>|</b>|<pre>|</pre>|<p>|</p>|<ul>|</ul>|<li>|<hr>')) {
    s/</&lt;/gx;
    s/>/&gt;/gx;
  }

  ## add html links
  s{
    \b	
    (
      $urls  :
      [$any] +?
    ) 
    (?= 
      [$punc]*
      [^$any] 
      | 
      $ 
    )
  }{<a href="$1" target="new">$1</a>}igox;

  ## Wrap 'From:' line in <b> html tags
  s{
    (
      From: 
      (\s+\w+)+
    )
  }{<b>$1</b>}ix;

## add links to anyone that's been interviewed
  interview_links("David Weinehall", 2229);
  interview_links("Marcelo Tosatti", 1880);
  interview_links("Rusty Russell", 892);
  interview_links("Nick Piggin", 657);
  interview_links("Ingo Molnar", 517);
  interview_links("Daniel Hartmeier", 477);
  interview_links("Con Kolivas", 465);
  interview_links("Robert Love", 336);
  interview_links("Rob Love", 336);
  interview_links("Marc-Christian Peterson", 293);
  interview_links("Jordan Hubbard", 278);
  interview_links("Peter Chubb", 242);
  interview_links("Larry McVoy", 222);
  interview_links("William Lee Irwin", 80);
  interview_links("Rik van Riel", 46);
  interview_links("Andrew Morton", 10);
  interview_links("Alan Cox", 9);
  interview_links("Matthew Dillon", 8);
  interview_links("Dave Jones", 7);
  interview_links("Theo de Raadt", 6);
  interview_links("Neal Walfield", 5);
  interview_links("John Levon", 4);
  interview_links("Keith Owens", 3);
  interview_links("Russell King", 2);

## display the prep'd result, stripping unwanted mail headers
  if (m/To: /i) {
    if (!$to) {
      $to=1;
      print;
    }
  } elsif (!(m/$strip_headers /i)) {
    print;
  }
}

## display the footer stuff
  print "</pre>\n";
  print "\n<hr>\n";
  print "<b>Related Links:</b><br />\n";
  print "<ul>\n";
  print "  <li><a href=\"\" target=\"new\">Archive of above thread</a>\n";
  if ($links) { print "$links"; }
  print "</ul>\n";


#############################
## Helper functions:
sub interview_links ($test) {
  my $name = shift;
  my $nid = shift;
  if (s/From:(\s+)$name/From: <a href="http:\/\/kerneltrap.org\/node\/view\/$nid">$name<\/a>/) {
    if ($links !~ m/$name/) {
      $links .= "  <li><a href=\"http:\/\/kerneltrap.org\/node\/view\/$nid\">KernelTrap interview with $name</a>\n";
    }
  }
}