aboutsummaryrefslogtreecommitdiffhomepage
path: root/readityourself.php
blob: c59bb9d1226a03aa5f74670546400f8d9a3c1a92 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
<?php

define("VERSION", "0.0.3");

header('Content-type:text/html; charset=utf-8');
// Set locale to French
setlocale(LC_ALL, 'fr_FR');

// set timezone to Europe/Paris
date_default_timezone_set('Europe/Paris');

// set charset to utf-8 important since all pages will be transform to utf-8
header('Content-Type: text/html;charset=utf-8');

// get readability library
require_once dirname(__FILE__).'/inc/Readability.php';

// get Encoding library.
require_once dirname(__FILE__).'/inc/Encoding.php';

// appel de la libraire RainTPL.
require_once dirname(__FILE__).'/inc/rain.tpl.class.php';

// FUNCTIONS BEGIN



function url(){
  $protocol = "http";
  if(isset($_SERVER['HTTPS']))
    if($_SERVER['HTTPS'] != "off")
      $protocol = "https";
  
  return $protocol . "://" . $_SERVER['HTTP_HOST'] . $_SERVER['REQUEST_URI'];
}

function generate_page($url,$title,$content) {
	raintpl::$tpl_dir = './tpl/'; // template directory
	raintpl::$cache_dir = "./cache/"; // cache directory
	raintpl::$base_url = url(); // base URL of blog
	raintpl::configure( 'path_replace', false );
	raintpl::configure('debug', false); 

	$tpl = new raintpl(); //include Rain TPL

	$tpl->assign( "url", $url);
	$tpl->assign( "title", $title);
	$tpl->assign( "content", $content);
	
	$tpl->assign( "version", VERSION);
	
	$tpl->draw( "index"); // draw the template
}

// function define to retrieve url content
function get_external_file($url, $timeout) {
	// spoofing FireFox 18.0
	$useragent="Mozilla/5.0 (Windows NT 5.1; rv:18.0) Gecko/20100101 Firefox/18.0";

	if  (in_array  ('curl', get_loaded_extensions())) {	
		// Fetch feed from URL
		$curl = curl_init();
		curl_setopt($curl, CURLOPT_URL, $url);
		curl_setopt($curl, CURLOPT_TIMEOUT, $timeout);
		curl_setopt($curl, CURLOPT_FOLLOWLOCATION, true);
		curl_setopt($curl, CURLOPT_RETURNTRANSFER, true);
		curl_setopt($curl, CURLOPT_HEADER, false);
		 
		// FeedBurner requires a proper USER-AGENT...
		curl_setopt($curl, CURL_HTTP_VERSION_1_1, true);
		curl_setopt($curl, CURLOPT_ENCODING, "gzip, deflate");
		curl_setopt($curl, CURLOPT_USERAGENT, $useragent);

		$data = curl_exec($curl);
		
		$httpcode = curl_getinfo($curl, CURLINFO_HTTP_CODE);
		
		$httpcodeOK = isset($httpcode) and ($httpcode == 200 or $httpcode == 301);
		
		curl_close($curl);
	} else {

		// create http context and add timeout and user-agent
		$context = stream_context_create(array('http'=>array('timeout' => $timeout, // Timeout : time until we stop waiting for the response.
																					'header'=> "User-Agent: ".$useragent, // spoot Mozilla Firefox
																					'follow_location' => true
														)));

		// only download page lesser than 4MB
		$data = @file_get_contents($url, false, $context, -1, 4000000); // We download at most 4 MB from source.
		//	echo "<pre>http_response_header : ".print_r($http_response_header);
		
		if(isset($http_response_header) and isset($http_response_header[0])) {
			$httpcodeOK = isset($http_response_header) and isset($http_response_header[0]) and ((strpos($http_response_header[0], '200 OK') !== FALSE) or (strpos($http_response_header[0], '301 Moved Permanently') !== FALSE));
		}
	}

	// if response is not empty and response is OK
	if (isset($data) and isset($httpcodeOK) and $httpcodeOK ) {

		// take charset of page and get it
		preg_match('#<meta .*charset=.*>#Usi', $data, $meta);

		// if meta tag is found
		if (!empty($meta[0])) {
			// retrieve encoding in $enc
			preg_match('#charset="?(.*)"#si', $meta[0], $enc);

			// if charset is found set it otherwise, set it to utf-8
			$html_charset = (!empty($enc[1])) ? strtolower($enc[1]) : 'utf-8';

		} else { 
			$html_charset = 'utf-8';
			$enc[1] = '';
		}

		// replace charset of url to charset of page
		$data = str_replace('charset='.$enc[1], 'charset='.$html_charset, $data);

		return $data;
	}
	else {
		return FALSE;
	}
}

function rel2abs($rel, $base)
{
    /* return if already absolute URL */
    if (parse_url($rel, PHP_URL_SCHEME) != '') return $rel;

    /* queries and anchors */
    if ($rel[0]=='#' || $rel[0]=='?') return $base.$rel;

    /* parse base URL and convert to local variables:
       $scheme, $host, $path */
    extract(parse_url($base));

    /* remove non-directory element from path */
    $path = preg_replace('#/[^/]*$#', '', $path);

    /* destroy path if relative url points to root */
    if ($rel[0] == '/') $path = '';

    /* dirty absolute URL */
    $abs = "$host$path/$rel";

    /* replace '//' or '/./' or '/foo/../' with '/' */
    $re = array('#(/\.?/)#', '#/(?!\.\.)[^/]+/\.\./#');
    for($n=1; $n>0; $abs=preg_replace($re, '/', $abs, -1, $n)) {}

    /* absolute URL is ready! */
    return $scheme.'://'.$abs;
}

// $str=preg_replace('#(href|src)="([^:"]*)("|(?:(?:%20|\s|\+)[^"]*"))#','$1="http://wintermute.com.au/$2$3',$str);

function absolutes_links($data, $base) {
	// cherche les balises 'a' qui contiennent un href
	$matches = array();
	preg_match_all('#(href|src)="([^:"]*)("|(?:(?:%20|\s|\+)[^"]*"))#Si', $data, $matches, PREG_SET_ORDER);

	// ne conserve que les liens ne commen�ant pas par un protocole � protocole:// � ni par une ancre � # �
	foreach($matches as $i => $link) {
		$link[1] = trim($link[1]);

		if (!preg_match('#^(([a-z]+://)|(\#))#', $link[1]) ) {

			$absolutePath=rel2abs($link[2],$base);

			$data = str_replace($matches[$i][2], $absolutePath, $data);
		}

	}
	return $data;
}


// FUNCTIONS END

// EXUCUTION CODE


if(isset($_GET['url']) && $_GET['url'] != null && trim($_GET['url']) != "") {
	// get url link
	if(strlen(trim($_GET['url'])) > 2048) {
		echo "Error URL is too large !!";
	} else {
		$url = trim($_GET['url']);

		// decode it
		$url = html_entity_decode($url);
		
		// if url use https protocol change it to http
		if (!preg_match('!^https?://!i', $url)) $url = 'http://'.$url;
		
		// convert page to utf-8
		$html = Encoding::toUTF8(get_external_file($url,15));
		
		if(isset($html) and strlen($html) > 0) {
		
			// send result to readability library
			$r = new Readability($html, $url);

			if($r->init()) {
				generate_page($url,$r->articleTitle->innerHTML,$r->articleContent->innerHTML);
			} else {
				// return data into an iframe
				echo "<iframe id='readabilityframe'>".$html."</iframe>";
			}
		} else {
			echo "Error unable to get link : ".$url;
		}
	}
}
?>