| 1 |
# |
| 2 |
# -*- Perl -*- |
| 3 |
# $Id: rfc.pl,v 1.1.1.1 2001/10/20 14:09:01 ｪｲ Exp $ |
| 4 |
# Copyright (C) 1997-2000 Satoru Takabayashi , |
| 5 |
# 1999 NOKUBI Takatsugu All rights reserved. |
| 6 |
# This is free software with ABSOLUTELY NO WARRANTY. |
| 7 |
# |
| 8 |
# This program is free software; you can redistribute it and/or modify |
| 9 |
# it under the terms of the GNU General Public License as published by |
| 10 |
# the Free Software Foundation; either versions 2, or (at your option) |
| 11 |
# any later version. |
| 12 |
# |
| 13 |
# This program is distributed in the hope that it will be useful |
| 14 |
# but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 15 |
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 16 |
# GNU General Public License for more details. |
| 17 |
# |
| 18 |
# You should have received a copy of the GNU General Public License |
| 19 |
# along with this program; if not, write to the Free Software |
| 20 |
# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA |
| 21 |
# 02111-1307, USA |
| 22 |
# |
| 23 |
# This file must be encoded in EUC-JP encoding |
| 24 |
# |
| 25 |
|
| 26 |
package rfc; |
| 27 |
use strict; |
| 28 |
require 'util.pl'; |
| 29 |
require 'gfilter.pl'; |
| 30 |
|
| 31 |
sub mediatype() { |
| 32 |
return ('text/plain; x-type=rfc'); |
| 33 |
} |
| 34 |
|
| 35 |
sub status() { |
| 36 |
return 'yes'; |
| 37 |
} |
| 38 |
|
| 39 |
sub recursive() { |
| 40 |
return 0; |
| 41 |
} |
| 42 |
|
| 43 |
sub pre_codeconv() { |
| 44 |
return 1; # For translated documents. |
| 45 |
} |
| 46 |
|
| 47 |
sub post_codeconv () { |
| 48 |
return 0; |
| 49 |
} |
| 50 |
|
| 51 |
sub add_magic ($) { |
| 52 |
my ($magic) = @_; |
| 53 |
|
| 54 |
$magic->addSpecials("text/plain; x-type=rfc", |
| 55 |
"^Network Working Group", |
| 56 |
"^Request [fF]or Comments", |
| 57 |
"^Obsoletes:", |
| 58 |
"^Category:", |
| 59 |
"^Updates:", |
| 60 |
"^INTERNET[- ]DRAFT", "^Internet[- ]Draft", |
| 61 |
"^draft-[A-Za-z0-9-]+\\.txt", |
| 62 |
"^<draft-[A-Za-z0-9-]+(?:\\.txt)?>", |
| 63 |
"^Expires", |
| 64 |
"^Status of this Memo"); |
| 65 |
$magic->addFileExts('^rfc\d+\.txt$' => 'text/plain; x-type=rfc'); |
| 66 |
$magic->addFileExts('^draft-[A-Za-z0-9-]+\.txt$' => 'text/plain; x-type=rfc'); |
| 67 |
$magic->addFileExts('^fyi\d+\.txt$' => 'text/plain; x-type=rfc'); |
| 68 |
$magic->addFileExts('^bcp\d+\.txt$' => 'text/plain; x-type=rfc'); |
| 69 |
$magic->addFileExts('^std\d+\.txt$' => 'text/plain; x-type=rfc'); |
| 70 |
$magic->addFileExts('^rfc\d+-\w+\.txt$', 'text/plain; x-type=rfc'); |
| 71 |
|
| 72 |
return; |
| 73 |
} |
| 74 |
|
| 75 |
sub filter ($$$$$) { |
| 76 |
my ($orig_cfile, $cont, $weighted_str, $headings, $fields) |
| 77 |
= @_; |
| 78 |
my $cfile = defined $orig_cfile ? $$orig_cfile : ''; |
| 79 |
|
| 80 |
util::vprint("Processing rfc file ...\n"); |
| 81 |
|
| 82 |
rfc_filter($cont, $weighted_str, $fields); |
| 83 |
|
| 84 |
gfilter::line_adjust_filter($cont); |
| 85 |
gfilter::line_adjust_filter($weighted_str); |
| 86 |
gfilter::white_space_adjust_filter($cont); |
| 87 |
gfilter::show_filter_debug_info($cont, $weighted_str, |
| 88 |
$fields, $headings); |
| 89 |
return undef; |
| 90 |
} |
| 91 |
|
| 92 |
# It's not perfect because of fuzziness in RFC documents. |
| 93 |
sub rfc_filter ($$$) { |
| 94 |
my ($contref, $weighted_str, $fields) = @_; |
| 95 |
|
| 96 |
$$contref =~ s/^\s+//s; |
| 97 |
|
| 98 |
# Get an author name first. |
| 99 |
if ($$contref =~ /^Network Working Group\s+(\S.*\S)\s*$/m) { |
| 100 |
$fields->{'author'} = $1; |
| 101 |
} elsif ($$contref =~ /^INTERNET[- ]DRAFT\s+(\S.*\S)\s*$/mi) { |
| 102 |
$fields->{'author'} = $1; |
| 103 |
} |
| 104 |
|
| 105 |
# Handle the header. |
| 106 |
if ($$contref =~ s/((.+\n)+)\s+(.*)//) { |
| 107 |
$$weighted_str .= "\x7f1\x7f$1\x7f/1\x7f\n" if defined $1; |
| 108 |
|
| 109 |
# Get the title. |
| 110 |
my $title = $3 if defined $3; |
| 111 |
$fields->{'title'} = $title; |
| 112 |
|
| 113 |
my $weight = $conf::Weight{'html'}->{'title'}; |
| 114 |
$$weighted_str .= "\x7f$weight\x7f$title\x7f/$weight\x7f\n"; |
| 115 |
} |
| 116 |
|
| 117 |
# Get Summary or Introduction and put it at the beginning |
| 118 |
# for summarization. |
| 119 |
# $$contref =~ s/\A(.+?^(\d+\.\s*)?(Abstract|Introduction)\n\n)//ims; |
| 120 |
# $$contref =~ s/(^[\w\.\s]*(?:Abstract|Introduction|Scope|概要|要約|(?:はじ|始|初)めに|導入|序論|摘要|梗概|イントロダクション|適用範囲))//im; |
| 121 |
# $$contref =~ s/([\s\S]+^(\d+\.\s*)?(Abstract|Introduction)\n\n)//im; |
| 122 |
$$contref =~ s/([\s\S]+^[\w\.\s]*?(?:Abstract|Introduction|Scope|概要|要約|摘要|梗概|(?:はじ|始|初)めに|導入|序論|イントロダクション|適用範囲).*\n\n)//im; |
| 123 |
$$weighted_str .= "\x7f1\x7f$1\x7f/1\x7f\n" if defined $1; |
| 124 |
} |
| 125 |
|
| 126 |
1; |
Parent Directory
| 
Revision Log