Subversion Repositories Applications.papyrus

Rev

Details | Last modification | View Log | RSS feed

Rev Author Line No. Line
448 ddelon 1
<?php
1336 neiluj 2
/*vim: set expandtab tabstop=4 shiftwidth=4: */
3
// +------------------------------------------------------------------------------------------------------+
4
// | PHP version 4.1                                                                                      |
5
// +------------------------------------------------------------------------------------------------------+
6
// | Copyright (C) 2004 Tela Botanica (accueil@tela-botanica.org)                                         |
7
// +------------------------------------------------------------------------------------------------------+
8
// | This library is free software; you can redistribute it and/or                                        |
9
// | modify it under the terms of the GNU General Public                                                  |
10
// | License as published by the Free Software Foundation; either                                         |
11
// | version 2.1 of the License, or (at your option) any later version.                                   |
12
// |                                                                                                      |
13
// | This library 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 GNU                                    |
16
// | General Public License for more details.                                                             |
17
// |                                                                                                      |
18
// | You should have received a copy of the GNU General Public                                            |
19
// | License along with this library; if not, write to the Free Software                                  |
20
// | Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA                            |
21
// +------------------------------------------------------------------------------------------------------+
22
// CVS : $Id: ezmlm-parser.php,v 1.3 2007-04-19 15:34:35 neiluj Exp $
23
/**
24
* Application projet
25
*
26
* classe ezmlm_parser pour lire les fichiers d index de ezmlm-idx
27
*
28
*@package projet
29
//Auteur original : ?? recupere dans ezmlm-php
30
*@author        Alexandre Granier <alexandre@tela-botanica.org>
31
*@copyright     Tela-Botanica 2000-2004
32
*@version       $Revision: 1.3 $
33
// +------------------------------------------------------------------------------------------------------+
34
*/
35
 
36
 
37
// +------------------------------------------------------------------------------------------------------+
38
// |                                            ENTETE du PROGRAMME                                       |
39
// +------------------------------------------------------------------------------------------------------+
40
 
41
// $Id: ezmlm-parser.php,v 1.3 2007-04-19 15:34:35 neiluj Exp $
448 ddelon 42
//
43
 
44
require_once("ezmlm.php");
45
require_once("Mail/mimeDecode.php") ;
46
// CLASS: ezmlm-parser
47
class ezmlm_parser extends ezmlm_php {
1336 neiluj 48
    var $headers;           // the full untouched headers of the message
49
    var $body;              // the full untouched (but decoded) body (this is not $this->parts[0]->body)
448 ddelon 50
	var $parts;		// all the parts, if it is a multipart message. each part is an ezmlm_parser object...
51
 
1336 neiluj 52
    // Here's the most accessed headers, everything else can be
53
    // accessed from the $this->headers array.
54
    var $to;                // To:
55
    var $from;              // From:
56
    var $date;              // Date:
57
    var $subject;           // Subject:
58
    var $replyto;           // Reply-To:
59
    var $contenttype;       // Content-Type:
448 ddelon 60
 
61
	var $multipart;		// TRUE if the message is a multipart message
62
 
63
	var $msgfile;		// if parsed from a file, this is the filename...
64
 
65
	// functions
66
 
1336 neiluj 67
	/**
68
     * recent_msgs renvoie les derniers messages de la liste de discussion
69
     * ezmlm
70
     *
71
     * (
72
     * [0] => Array
73
     *   (
74
     *       [1] => sujet
75
     *       [2] => date en anglais
76
     *       [3] => le hash de l auteur
77
     *       [4] => l auteur
78
     *   )
79
     * [1] => ...
80
     * )
81
     * @param	int	le nombre de message a renvoye
82
     * @return array	un tableau contenant les messages
83
     * @access public
84
     */
448 ddelon 85
	function recent_msgs($show = 20, $month = "") {
1336 neiluj 86
 
87
		$repertoire_archive = opendir($this->listdir . "/archive/");
448 ddelon 88
 
1336 neiluj 89
		$repertoire_message = array() ;
90
		$dernier_repertoire = 0 ;
91
		while (false !== ($item = readdir($repertoire_archive))) {
92
			// $item contient les noms des repertoires
93
			// on ne garde que ceux qui sont des chiffres
94
 
95
			if (preg_match('/[0-9]+/', $item)) {
96
				// on ouvre le fichier d index de chaque repertoire
97
				if ((int) $item > $dernier_repertoire) $dernier_repertoire = (int) $item;
98
 
99
			}
448 ddelon 100
		}
1336 neiluj 101
		$tableau_message = array() ;
102
		$compteur_message = 0 ;
103
		$fichier_index = fopen ($this->listdir.'/archive/'.$dernier_repertoire.'/index', 'r') ;
104
		while (!feof($fichier_index)) {
105
				// Recuperation du numero de message, du hash du sujet et du sujet
106
				$temp = fgets($fichier_index, 4096);
107
				preg_match('/([0-9]+): ([a-z]+) (.*)/', $temp, $match) ;
108
 
109
				// dans la seconde on recupere la date, hash auteur et auteur
110
				$temp = fgets($fichier_index, 4096);
111
				preg_match('/\t([0-9]+) ([a-zA-Z][a-zA-Z][a-zA-Z]) ([0-9][0-9][0-9][0-9]) ([^;]+);([^ ]*) (.*)/', $temp, $match_deuxieme_ligne) ;
112
				if ($match[1] != '') {
113
				$tableau_message[$match[1]] = array ($match[2], $match[3],
114
									$match_deuxieme_ligne[1].' '.$match_deuxieme_ligne[2].' '.$match_deuxieme_ligne[3],
115
									$match_deuxieme_ligne[5],
116
									$match_deuxieme_ligne[6]);
117
				}
118
			}
119
			fclose ($fichier_index);
120
		// on renverse le tableau pour afficher les derniers messages en premier
121
		$tableau_message = array_reverse($tableau_message, true);
448 ddelon 122
 
1336 neiluj 123
		// On compte le nombre de message, s il est inferieur $show et que l on est
124
		// pas dans le premier index, on ouvre le fichier precedent et recupere
125
		// le n dernier message
126
 
127
		if (count ($tableau_message) < $show && $dernier_repertoire != '0') {
128
			$avant_dernier_repertoire = $dernier_repertoire - 1 ;
129
			// On utilise file_get_contents pour renverser le fichier
130
			$fichier_index = array_reverse(
131
									explode ("\n",
132
										preg_replace ('/\n$/', '',
133
											file_get_contents ($this->listdir.'/archive/'.$avant_dernier_repertoire.'/index')) ), true) ;
134
			reset ($fichier_index);
135
			//var_dump ($fichier_index);
136
 
137
			for ($i = count ($tableau_message); $i <= $show; $i++) {
138
				// Recuperation du numero de message, du hash du sujet et du sujet
139
				// dans la seconde on recupere la date, hash auteur et auteur
140
 
141
				preg_match('/\t([0-9]+) ([a-zA-Z][a-zA-Z][a-zA-Z]) ([0-9][0-9][0-9][0-9]) ([^;]+);([^ ]*) (.*)/',
142
									current ($fichier_index), $match_deuxieme_ligne) ;
143
				preg_match('/([0-9]+): ([a-z]+) (.*)/', next($fichier_index), $match) ;
144
				next ($fichier_index);
145
 
146
				if ($match[1] != '') {
147
				$tableau_message[$match[1]] = array ($match[2], $match[3],
148
									$match_deuxieme_ligne[1].' '.$match_deuxieme_ligne[2].' '.$match_deuxieme_ligne[3],
149
									$match_deuxieme_ligne[5],
150
									$match_deuxieme_ligne[6]);
151
				}
448 ddelon 152
			}
1336 neiluj 153
		} else {
154
			// Si le nombre de message est > $show on limite le tableau de retour
155
			$tableau_message = array_slice($tableau_message, 0, $show, true);
448 ddelon 156
		}
1336 neiluj 157
 
158
 
159
		return $tableau_message ;
448 ddelon 160
	}
161
 
162
 
163
	// parse_file - opens a file and feeds the data to parse, file can be relative to the listdir
164
	function parse_file($file,$simple = FALSE) {
165
		if (!is_file($file)) {
166
			if (is_file($this->listdir . "/" . $file)) { $file = $this->listdir . "/" . $file; }
167
			else if (is_file($this->listdir . "/archive/" . $file)) { $file = $this->listdir . "/archive/" . $file; }
168
			else { return FALSE; }
169
		}
170
 
171
		$this->msgfile = $file;
172
        $data = '' ;
173
		$fd = fopen($file, "r");
174
		while (!feof($fd)) { $data .= fgets($fd,4096); }
175
		fclose($fd);
176
		return $this->parse($data,$simple);
177
	}
178
 
1336 neiluj 179
    // parse_file_headers - ouvre un fichier et analyse les entête
448 ddelon 180
	function parse_file_headers($file,$simple = FALSE) {
181
		if (!is_file($file)) {
182
			if (is_file($this->listdir . "/" . $file)) { $file = $this->listdir . "/" . $file; }
183
			else if (is_file($this->listdir . "/archive/" . $file)) { $file = $this->listdir . "/archive/" . $file; }
184
			else { return FALSE; }
185
		}
186
 
187
		$this->msgfile = $file;
188
        $data = file_get_contents ($file) ;
189
        $message = file_get_contents($file) ;
190
        $mimeDecode = new Mail_mimeDecode($message) ;
191
        $mailDecode = $mimeDecode->decode() ;
192
        return $mailDecode ;
193
	}
194
 
195
	// this does all of the work (well it calls two functions that do all the work :)
196
	// all the decoding a part breaking follows RFC2045 (http://www.faqs.org/rfcs/rfc2045.html)
197
	function parse($data,$simple = FALSE) {
198
 
199
		if (($this->_get_headers($data,$simple)) && $this->_get_body($data,$simple)) { return TRUE; }
200
		return FALSE;
201
	}
202
 
203
	// all of these are internal functions, you shouldn't call them directly...
204
 
205
	// _ct_parse: parse Content-Type headers -> $ct[0] = Full header, $ct[1] = Content-Type, $ct[2] ... $ct[n] = AP's
206
	function _ct_parse() {
207
		$instr = $this->headers['content-type'];
208
		preg_replace('/\(.*\)/','',$instr); // strip rfc822 comments
209
		if (preg_match('/: /', $instr)) {
210
			$ct = preg_split('/:/',trim($instr),2);
211
			$ct = preg_split('/;/',trim($ct[1]));
212
		} else {
213
			$ct = preg_split('/;/',trim($instr));
214
		}
215
		if (isset($ct[1])) $attrs = preg_split('/[\s\n]/',$ct[1]);
216
		$i = 2;
217
		$ct[1] = $ct[0];
218
		$ct[0] = $this->headers['content-type'];
219
        if (isset($attrs) && is_array($attrs)) {
220
            while (list($key, $val) = each($attrs)) {
221
                if ($val == '') continue;
222
                $ap = preg_split('/=/',$val,2);
223
                if (preg_match('/^"/',$ap[1])) { $ap[1] = substr($ap[1],1,strlen($ap[1])-2); }
224
                $ct[$i] = $ap;
225
                $i++;
226
            }
227
        }
228
		// are we a multipart message?
229
		if (preg_match('/^multipart/i', $ct[1])) { $this->multipart = TRUE; }
230
 
231
		return $ct;
232
	}
233
 
234
	// _get_headers: pulls the headers out of the data and builds the $this->headers array
235
	function _get_headers($data,$simple = FALSE) {
236
		$lines = preg_split('/\n/', $data);
237
		while (list($key, $val) = each($lines)) {
238
			$val = trim($val);
239
			if ($val == "") break;
240
			if (preg_match('/^From[^:].*$/', $val)) continue;	/* strips out any From lines added by the MTA */
241
 
242
			$hdr = preg_split('/: /', $val, 2);
243
			if (count($hdr) == 1) {
244
				// this is a continuation of the last header (like a recieved from line)
245
				$this->headers[$last] .= $val;
246
			} else {
247
				$this->headers[strtolower($hdr[0])] = $hdr[1];
248
                //echo htmlspecialchars($this->headers['from'])."<br />" ;
249
				$last = strtolower($hdr[0]);
250
			}
251
		}
252
        // ajout alex
253
        // pour supprimer le problème des ISO...
254
        // a déplacer ailleur, et appelé avant affichage
255
 
256
        if (preg_match ('/windows-[0-9][0-9][0-9][0-9]/', $this->headers['subject'], $nombre)) {
257
            $reg_exp = $nombre[0] ;
258
        } else {
259
            $reg_exp = 'ISO-8859-15?' ;
260
        }
261
        if (preg_match ('/UTF/i', $this->headers['subject'])) $reg_exp = 'UTF-8' ;
262
        preg_match_all ("/=\?$reg_exp\?(Q|B)\?(.*?)\?=/i", $this->headers['subject'], $match, PREG_PATTERN_ORDER)  ;
263
        for ($i = 0; $i < count ($match[0]); $i++ ) {
264
 
265
                if ($match[1][$i] == 'Q') {
266
                    $decode = quoted_printable_decode ($match[2][$i]) ;
267
                } elseif ($match[1][$i] == 'B') {
268
                    $decode = base64_decode ($match[2][$i]) ;
269
                }
270
                $decode = preg_replace ("/_/", " ", $decode) ;
271
            if ($reg_exp == 'UTF-8') {
272
                $decode = utf8_decode ($decode) ;
273
            }
274
            $this->headers['subject'] = str_replace ($match[0][$i], $decode, $this->headers['subject']) ;
275
        }
276
		// sanity anyone?
277
		if (!$this->headers['content-type']) { $this->headers['content-type'] = "text/plain; charset=us-ascii"; }
278
		if (!$simple) { $this->headers['content-type'] = $this->_ct_parse(); }
279
 
280
 
281
		return TRUE;
282
	}
283
 
284
	// _get_body: pulls the body out of the data and fills $this->body, decoding the data if nessesary.
285
	function _get_body($data,$simple = FALSE) {
286
		$lines = preg_split('/\n/', $data);
287
		$doneheaders = FALSE;
288
 
289
		$data = "";
290
		while (list($key,$val) = each($lines)) {
291
            //echo htmlspecialchars($val)."<br>";
292
			if (($val == '') and (!$doneheaders)) {
293
				$doneheaders = TRUE;
294
				continue;
295
			} else if ($doneheaders) {
296
				$data .= $val . "\n";
297
			}
298
		}
299
 
300
		// now here comes the fun part... decoding.
301
		switch($this->headers['content-transfer-encoding']) {
302
			case 'binary':
303
				$this->body = $this->_cte_8bit($this->_cte_qp($this->_cte_binary($data)),$simple);
304
				break;
305
 
306
			case 'base64':
307
				$this->body = $this->_cte_8bit($this->_cte_qp($this->_cte_base64($data)),$simple);
308
				break;
309
 
310
			case 'quoted-printable':
311
				$this->body = $this->_cte_8bit($this->_cte_qp($data),$simple);
312
				break;
313
 
314
			case '8bit':
315
				$this->body = $this->_cte_8bit($data,$simple);
316
				break;
317
 
318
			case '7bit':		// 7bit doesn't need to be decoded
319
			default:		// And the fall through as well...
320
				$this->body = $data;
321
				break;
322
		}
323
        //echo  $this->headers['content-type'][2][1];
324
        if (isset($this->headers['content-type'][2][1]) && $this->headers['content-type'][2][1] == 'UTF-8') {
325
                //$this->body = utf8_decode ($this->body) ;
326
                //echo quoted_printable_decode(utf8_decode ($this->body)) ;
327
        }
328
		if ($simple) { return TRUE; }
329
 
330
		// if we are a multipart message then break up the parts and decode, set the appropriate variables.
331
		// here comes the best part about making ezmlm-php OOP. since each part is just really a little message
332
		// in itself each part becomes a new parser object and all the wheels turn again... :)
333
		if ($this->multipart) {
334
 
335
			$boundary = '';
336
			for ($i = 2; $i <= count($this->headers['content-type']); $i++) {
337
				if (preg_match('/boundary/i', $this->headers['content-type'][$i][0])) {
338
					$boundary = $this->headers['content-type'][$i][1];
339
 
340
				}
341
			}
342
			if ($boundary != '') {
343
				$this->_get_parts($this->body,$boundary);
344
			} else {
345
				// whoopps... something's not right here. we were told that the message is supposed
346
				// to be a multipart message, yet the boundary wasn't set in the content type.
347
				// mark the message as non multipart and add a message to the top of the body.
348
				$this->multipart = FALSE;
349
				$this->body = "PARSER ERROR:\nWHILE PARSING THIS MESSAGE AS A MULTIPART MESSAGE AS DEFINED IN RFC2045 THE BOUNDARY IDENTIFIER WAS NOT FOUND!\nTHIS MESSAGE WILL NOT DISPLAY CORRECTLY!\n\n" . $this->body;
350
			}
351
		}
352
 
353
		return TRUE;
354
	}
355
 
356
	// _get_parts: breaks up $data into parts based on $boundary following the rfc specs
357
	// detailed in section 5 of RFC2046 (http://www.faqs.org/rfcs/rfc2046.html)
358
	// After the parts are broken up they are then turned into parser objects and the
359
	// resulting array of parts is set to $this->parts;
360
	function _get_parts($data,$boundary) {
361
		$inpart = -1;
362
		$lines = preg_split('/\n/', $data);
363
        // La première partie contient l'avertissement pour les client mail ne supportant pas
364
        // multipart, elle est stocké dans parts[-1]
365
		while(list($key,$val) = each($lines)) {
366
			if ($val == "--" . $boundary) { $inpart++; continue; } // start of a part
367
			else if ($val == "--" . $boundary . "--") { break; } // the end of the last part
368
			else { $parts[$inpart] .= $val . "\n"; }
369
		}
370
 
371
		for ($i = 0; $i < count($parts) - 1; $i++) {    // On saute la première partie
372
			$part[$i] = new ezmlm_parser();
373
			$part[$i]->parse($parts[$i]);
374
			$this->parts[$i] = $part[$i];
375
            //echo $this->parts[$i]."<br>" ;
376
		}
377
 
378
	}
379
 
380
	// _cte_8bit: decode a content transfer encoding of 8bit
381
	// NOTE: this function is a little bit special. Since the end result will be displayed in
382
	// a web browser _cte_8bit decodes ASCII characters > 127 (the US-ASCII table) into the
383
	// html ordinal equivilant, it also ensures that the messages content-type is changed
384
	// to include text/html if it changes anything...
385
	function _cte_8bit($data,$simple = FALSE) {
386
		if ($simple) { return $data; }
387
		$changed = FALSE;
388
		$chars = preg_split('//',$data);
389
		while (list($key,$val) = each($chars)) {
390
			if (ord($val) > 127) { $out .= '&#' . ord($val) . ';'; $changed = TRUE; }
391
			else { $out .= $val; }
392
		}
393
		if ($changed) { $this->headers['content-type'][1] = 'text/html'; }
394
		return $out;
395
	}
396
 
397
	// _cte_binary: decode a content transfer encoding of binary
398
	function _cte_binary($data) { return $data; }
399
 
400
	// _cte_base64: decode a content transfer encoding of base64
401
	function _cte_base64($data) { return base64_decode($data); }
402
 
403
	// _cte_qp: decode a content transfer encoding of quoted_printable
404
	function _cte_qp($data) {
405
		// For the time being we'll use PHP's function, it seems to work well enough.
406
		return quoted_printable_decode($data);
407
	}
408
 
409
}