2 /* ***** BEGIN LICENSE BLOCK *****
3 * Version: MPL 1.1/GPL 2.0/LGPL 2.1
5 * The contents of this file are subject to the Mozilla Public License Version
6 * 1.1 (the "License"); you may not use this file except in compliance with
7 * the License. You may obtain a copy of the License at
8 * http://www.mozilla.org/MPL/
10 * Software distributed under the License is distributed on an "AS IS" basis,
11 * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
12 * for the specific language governing rights and limitations under the
15 * The Original Code is KaiRo's userAgent detection.
17 * The Initial Developer of the Original Code is
18 * KaiRo - Robert Kaiser.
19 * Portions created by the Initial Developer are Copyright (C) 2003
20 * the Initial Developer. All Rights Reserved.
22 * Contributor(s): Robert Kaiser <kairo@kairo.at>
24 * Alternatively, the contents of this file may be used under the terms of
25 * either the GNU General Public License Version 2 or later (the "GPL"), or
26 * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
27 * in which case the provisions of the GPL or the LGPL are applicable instead
28 * of those above. If you wish to allow use of your version of this file only
29 * under the terms of either the GPL or the LGPL, and not to allow others to
30 * use your version of this file under the terms of the MPL, indicate your
31 * decision by deleting the provisions above and replace them with the notice
32 * and other provisions required by the GPL or the LGPL. If you do not delete
33 * the provisions above, a recipient may use your version of this file under
34 * the terms of any one of the MPL, the GPL or the LGPL.
36 * ***** END LICENSE BLOCK ***** */
39 // userAgent PHP class
40 // get user agent and tell us what Browser is accessing
42 // function userAgent([$ua_string])
43 // CONSTRUCTOR; reads UA string (or takes the optional given UA string) and gets info from that into our variables.
46 // the plain User Agent string
48 // returns the User Agent brand name
50 // the User Agent version
52 // bool: true if this agent is a bot
54 // array of static user agent data (static vars in functions are set for all objects of this class!)
56 // function getBrand()
57 // returns the User Agent Brand Name
58 // function getVersion()
59 // returns the User Agent version
61 // function getAcceptLanguages()
62 // returns an associated array with the accepted languages of this UA
63 // keys are language codes, values are q factors (weights)
65 // function getUAString()
66 // returns the full User Agent string
69 // returns true if User Agent seems to be a bot
71 // returns true if User Agent seems to be Netscape brand, false if not
73 // returns true if User Agent seems to be Netscape Communicator 4.x, false if not
75 // returns true if User Agent seems to be a version of Internet Exploder, false if not
76 // function geckobased()
77 // returns true if User Agent seems to be a Gecko-based browser, false if not
78 // function geckodate()
79 // returns the Gecko date when it's a Gecko-based browser, 0 if not
80 // function khtmlbased()
81 // returns true if User Agent seems to be a KHTML-based browser, false if not
83 // collection of some known User Agent Strings:
84 // Mozilla/5.0 (compatible; Konqueror/3; Linux 2.4.18; X11; i686)
85 // Mozilla/5.0 (X11; U; Linux i686; de-AT; rv:1.3b) Gecko/20030114
86 // Mozilla/4.0 (compatible; MSIE 5.0; Windows 95; DigExt)
87 // Mozilla/4.0 (compatible; MSIE 6.0; Windows 98; Win 9x 4.90)
88 // Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)
89 // Mozilla/4.75 [de] (Win98; U)
90 // Opera/5.12 (Windows 2000; U) [de]
91 // Mozilla/5.0 (Windows; U; Win 9x 4.90; de-DE; m18) Gecko/20010131 Netscape6/6.01
92 // Mozilla/5.0 (Windows; U; Windows NT 5.1; de-DE; rv:1.0.1) Gecko/20020823 Netscape/7.0
93 // Mozilla/5.0 (Macintosh; U; PPC Mac OS X; en) AppleWebKit/51 (like Gecko) Safari/51
94 // Lynx/2.8.4rel.1 libwww-FM/2.14 SSL-MM/1.4.1 OpenSSL/0.9.6g
95 // Mozilla/5.0 (Macintosh; U; PPC Mac OS X; en-US; rv:1.0.1) Gecko/20021109 Chimera/0.6+
96 // Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.7b) Gecko/20040302 Camino/0.7+
97 // Mozilla/5.0 (Windows; U; Win 9x 4.90; en-US; rv:1.3a) Gecko/20021207 Phoenix/0.5
98 // Mozilla/5.0 Galeon/1.2.7 (X11; Linux i686; U;) Gecko/20021204
99 // Mozilla/4.0 (compatible; MSIE 5.0; Windows XP) Opera 6.05 [ja]
100 // Mozilla/4.0 (compatible; MSIE 5.12; Mac_PowerPC) OmniWeb/4.1.1-v424.6
101 // Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4b) Gecko/20030516 Mozilla Firebird/0.6
102 // Mozilla/5.0 (Windows; U; WinNT4.0; en-US; rv:1.5a) Gecko/20030728 Mozilla Firebird/0.6.1
103 // Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.7a) Gecko/20040216 Firefox/0.8.0+
104 // Python-urllib/1.15
105 // *** search bots: ***
106 // W3C_Validator/1.305.2.12 libwww-perl/5.64
109 // Vagabondo/2.0 MT (webagent at wise-guys dot nl)
110 // TurnitinBot/1.5 ( ">http://www.turnitin.com/robot/crawlerinfo.html)
111 // FAST-WebCrawler/3.x Multimedia (mm dash crawler at fast dot no)
112 // Firefly/1.0 (compatible; Mozilla 4.0; MSIE 5.5)
113 // Googlebot/2.1 (+ ">http://www.googlebot.com/bot.html)
114 // Scrubby/2.2 ( ">http://www.scrubtheweb.com/)
115 // psbot/0.1 (+ ">http://www.picsearch.com/bot.html)
117 // URL_Spider_Pro/3.0 ( ">http://www.innerprise.net/usp-spider.asp)"
118 // Pompos/1.3 ">http://dir.com/pompos.html
119 // Szukacz/1.5 (robot; www.szukacz.pl/jakdzialarobot.html; info@szukacz.pl)
122 // NetResearchServer/2.7(loopimprovements.com/robot.html)
123 // dloader(NaverRobot)/1.0
124 // Mozilla/4.0 compatible ZyBorg/1.0 Daily Refresh Beta-d03 (wn.zyborg@looksmart.net;
125 // Mozilla/2.0 (compatible; Ask Jeeves/Teoma)
126 // Mozilla/5.0 (Slurp/si; slurp@inktomi.com; ">http://www.inktomi.com/slurp.html)
127 // Mozilla/5.0 [en] (compatible; Gulper Web Bot 0.2.4 www.ecsl.cs.sunysb.edu/~maxim/cgi-bin/Link/GulperBot)
128 // Mozilla/4.0 (compatible; MSIE 5.0; Windows NT; Girafabot; girafabot at girafa dot com;
129 // Mozilla/4.0 (efp@gmx.net)
130 // Mozilla/4.5 (compatible; HTTrack 3.0x; Windows 98)
131 // PingALink Monitoring Services 1.0 (http://www.pingalink.com)
132 // IlTrovatore-Setaccio (+ ">http://www.iltrovatore.it)
134 // appie 1.1 (www.walhello.com)
135 // larbin_2.6.2 (larbin2.6.2@unspecified.mail)
137 // search.ch V1.4.2 (spiderman@search.ch;
138 // WebFilter Robot 1.0
139 // Openfind data gatherer, Openbot/3.0+(robot-response@openfind.com.tw;+
141 // Internet Explorer 5.5
143 // Mozilla/4.0 (compatible; B-l-i-t-z-B-O-T)
144 // B l i t z B O T @ t r i c u s . n e t (Mozilla compatible)
145 // sitecheck.internetseer.com (For more info see: ">http://sitecheck.internetseer.com)
146 // http://www.almaden.ibm.com/cs/crawler [c01]
150 // HeinrichderMiragoRobot
157 var $uadata = array();
159 function userAgent($ua_string = '') {
160 // *** constructor ***
161 if (strlen($ua_string)) {
162 $this->uastring = $ua_string;
165 // read raw UA string
166 $this->uastring = $_SERVER['HTTP_USER_AGENT'];
169 // get UA brand and version
170 $this->brand = "Unknown"; $this->version = 0;
171 if (ereg("([0-9a-zA-Z\.()-]+)/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
172 $this->brand = $regs[1]; // this is a reasonable default :)
173 $this->version = $regs[2]; // this is a reasonable default :)
175 if (ereg("Netscape6/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
176 $this->brand = "Netscape";
177 $this->version = $regs[1];
179 elseif (ereg("Netscape/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
180 $this->brand = "Netscape";
181 $this->version = $regs[1];
183 elseif (ereg("Chimera/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
184 $this->brand = "Chimera";
185 $this->version = $regs[1];
187 elseif (ereg("Camino/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
188 $this->brand = "Camino";
189 $this->version = $regs[1];
191 elseif (ereg("Phoenix/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
192 $this->brand = "Phoenix";
193 $this->version = $regs[1];
195 elseif (ereg("Mozilla Firebird/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
196 $this->brand = "Mozilla Firebird";
197 $this->version = $regs[1];
199 elseif (ereg("Firefox/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
200 $this->brand = "Firefox";
201 $this->version = $regs[1];
203 elseif (ereg("Galeon/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
204 $this->brand = "Galeon";
205 $this->version = $regs[1];
207 elseif (ereg("rv:([0-9a-zA-Z\.+]+)", $this->uastring, $regs) && strstr($this->uastring, "Mozilla/") && strstr($this->uastring, "Gecko/")) {
208 $this->brand = "Mozilla";
209 $this->version = $regs[1];
211 elseif (ereg("Opera[ /]([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
212 $this->brand = "Opera";
213 $this->version = $regs[1];
215 elseif (ereg("OmniWeb/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
216 $this->brand = "OmniWeb";
217 $this->version = $regs[1];
219 elseif (ereg("Konqueror/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
220 $this->brand = "Konqueror";
221 $this->version = $regs[1];
223 elseif (ereg("Safari/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
224 $this->brand = "Safari";
225 $this->version = $regs[1];
227 elseif (ereg("AppleWebKit/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
228 $this->brand = "AppleWebKit";
229 $this->version = $regs[1];
231 elseif (ereg("W3C_Validator/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
232 $this->brand = "W3C_Validator";
233 $this->version = $regs[1];
236 elseif (ereg("ZyBorg/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
237 $this->brand = "ZyBorg";
238 $this->version = $regs[1];
241 elseif (ereg("Ask Jeeves/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
242 $this->brand = "Ask Jeeves";
243 $this->version = $regs[1];
246 elseif (ereg("Slurp/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
247 $this->brand = "Slurp";
248 $this->version = $regs[1];
251 elseif (ereg("Openbot/([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
252 $this->brand = "Openbot";
253 $this->version = $regs[1];
256 elseif (ereg("Mercator-([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
257 $this->brand = "Mercator";
258 $this->version = $regs[1];
261 elseif (ereg("appie ([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
262 $this->brand = "appie";
263 $this->version = $regs[1];
266 elseif (ereg("larbin_([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
267 $this->brand = "larbin";
268 $this->version = $regs[1];
271 elseif (ereg("Gulper Web Bot ([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
272 $this->brand = "Gulper Web Bot";
273 $this->version = $regs[1];
276 elseif (ereg("HTTrack ([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
277 $this->brand = "HTTrack";
278 $this->version = $regs[1];
281 elseif (ereg("OWR_Crawler ([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
282 $this->brand = "OWR_Crawler";
283 $this->version = $regs[1];
286 elseif (ereg("search.ch ([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
287 $this->brand = "search.ch";
288 $this->version = $regs[1];
291 elseif (ereg("WebFilter Robot ([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
292 $this->brand = "WebFilter Robot";
293 $this->version = $regs[1];
296 elseif (ereg("^LinkWalker", $this->uastring)) {
297 $this->brand = "LinkWalker";
301 elseif (ereg("^Internet Explorer 5.5", $this->uastring)) {
302 $this->brand = "Unknown bot (IE5.5)";
306 elseif (ereg("^Mozilla[\s ]*$", $this->uastring)) {
307 $this->brand = "Unknown bot (Mozilla)";
311 elseif (ereg("^BaiDuSpider", $this->uastring)) {
312 $this->brand = "BaiDuSpider";
316 elseif (ereg("^ia_archiver", $this->uastring)) {
317 $this->brand = "ia_archiver";
321 elseif (ereg("^PingALink", $this->uastring)) {
322 $this->brand = "PingALink";
326 elseif (ereg("^IlTrovatore-Setaccio", $this->uastring)) {
327 $this->brand = "IlTrovatore-Setaccio";
331 elseif (ereg("^Nutch", $this->uastring)) {
332 $this->brand = "Nutch";
336 elseif (ereg("^dumbBot", $this->uastring)) {
337 $this->brand = "dumbBot";
341 elseif (ereg("http://www.almaden.ibm.com/cs/crawler", $this->uastring)) {
342 $this->brand = "almaden crawler";
346 elseif (ereg("B-l-i-t-z-B-O-T", $this->uastring) || ereg("B l i t z B O T @ t r i c u s . n e t", $this->uastring)) {
347 $this->brand = "BlitzBOT";
351 elseif (ereg("sitecheck.internetseer.com", $this->uastring)) {
352 $this->brand = "internetseer";
356 elseif (ereg("Girafabot", $this->uastring)) {
357 $this->brand = "Girafabot";
361 elseif (ereg("efp@gmx.net", $this->uastring)) {
362 $this->brand = "efp";
366 elseif (ereg("HeinrichderMiragoRobot", $this->uastring)) {
367 $this->brand = "HeinrichderMiragoRobot";
371 elseif (ereg("MSIE ([0-9a-zA-Z\.+]+)", $this->uastring, $regs)) {
372 $this->brand = "Microsoft Internet Explorer";
373 $this->version = $regs[1];
375 elseif (ereg("Mozilla/([0-9a-zA-Z\.+]+)", $this->uastring, $regs) && !strstr($this->uastring, "compatible;") && !strstr($this->uastring, "Gecko/")) {
376 $this->brand = "Netscape";
377 $this->version = $regs[1];
378 if (intval($this->version) == 4) { $this->brand .= " Communicator"; }
381 $botArray = array('Scooter','Spinne','Vagabondo','TurnitinBot','FAST-WebCrawler','Firefly','Googlebot',
382 'Scrubby','psbot','NG','URL_Spider_Pro','Pompos','Szukacz','ASPseek','NPBot-1',
383 'dloader(NaverRobot)','NetResearchServer','','','','','','','');
385 if (in_array($this->brand, $botArray)) {
390 function getBrand() { return $this->brand; }
391 function getVersion() { return $this->version; }
393 function getAcceptLanguages() {
394 if (!isset($this->uadata['accept-languages'])) {
395 $headers = getAllHeaders();
396 $accLcomp = explode(',', $headers['Accept-Language']);
398 foreach ($accLcomp as $lcomp) {
399 if (strlen($lcomp)) {
400 $ldef = explode(';', $lcomp);
401 $accLang[$ldef[0]] = (float)((strpos($ldef[1],'q=')===0)?substr($ldef[1],2):1);
404 $this->uadata['accept-languages'] = $accLang;
406 return $this->uadata['accept-languages'];
409 function getUAString() { return $this->uastring; }
410 function isbot() { return $this->bot; }
413 if (!isset($this->uadata['is_ns'])) {
414 $this->uadata['is_ns'] = false;
415 if (strstr($this->brand, 'Netscape')) {
416 $this->uadata['is_ns'] = true;
419 return $this->uadata['is_ns'];
423 if (!isset($this->uadata['is_ns4'])) {
424 $this->uadata['is_ns4'] = false;
425 if (strstr($this->brand, 'Netscape') && (intval($this->version) == 4)) {
426 $this->uadata['is_ns4'] = true;
429 return $this->uadata['is_ns4'];
433 if (!isset($this->uadata['is_ie'])) {
435 if (strstr($this->brand, 'Internet Explorer')) {
436 $this->uadata['is_ie'] = true;
439 return $this->uadata['is_ie'];
442 function geckobased() {
443 if (!isset($this->uadata['is_gecko'])) {
444 $this->uadata['is_gecko'] = false;
445 if (strstr($this->uastring, 'Gecko/')) {
446 $this->uadata['is_gecko'] = true;
449 return $this->uadata['is_gecko'];
452 function geckodate() {
453 if (!isset($this->uadata['gdate'])) {
454 $this->uadata['gdate'] = 0;
455 if (ereg('Gecko/([0-9]+)', $this->uastring, $regs)) {
456 $this->uadata['gdate'] = $regs[1];
459 return $this->uadata['gdate'];
462 function khtmlbased() {
463 if (!isset($this->uadata['is_khtml'])) {
464 $this->uadata['is_khtml'] = false;
465 if (strstr($this->brand, 'Konqueror') || strstr($this->brand, 'Safari') || strstr($this->brand, 'AppleWebKit')) {
466 $this->uadata['is_khtml'] = true;
469 return $this->uadata['is_khtml'];