3 define('INSTALLDIR', dirname(__FILE__));
4 define('LACONICA', true);
6 require_once(INSTALLDIR . '/lib/common.php');
7 require_once(INSTALLDIR . '/lib/util.php');
9 $output_paths = parse_args();
17 # ------------------------------------------------------------------------------
18 # Main functions: get data out and turn them into sitemaps
19 # ------------------------------------------------------------------------------
21 # Generate index sitemap of all other sitemaps.
22 function index_map() {
24 $output_dir = $output_paths['output_dir'];
25 $output_url = $output_paths['output_url'];
27 foreach (glob("$output_dir*.xml") as $file_name) {
29 # Just the file name please.
30 $file_name = preg_replace("|$output_dir|", '', $file_name);
34 'url' => $output_url . $file_name,
35 'changefreq' => 'daily',
41 write_file($output_paths['index_file'], urlset($index_urls));
45 # Generate sitemap of standard site elements.
46 function standard_map() {
49 $standard_map_urls .= url(
51 'url' => common_local_url('public'),
52 'changefreq' => 'daily',
57 $standard_map_urls .= url(
59 'url' => common_local_url('publicrss'),
60 'changefreq' => 'daily',
65 $docs = array('about', 'faq', 'contact', 'im', 'openid', 'openmublog', 'privacy', 'source');
67 foreach($docs as $title) {
68 $standard_map_urls .= url(
70 'url' => common_local_url('doc', array('title' => $title)),
71 'changefreq' => 'monthly',
77 $urlset_path = $output_paths['output_dir'] . 'standard.xml';
79 write_file($urlset_path, urlset($standard_map_urls));
82 # Generate sitemaps of all notices.
83 function notices_map() {
86 $notices = DB_DataObject::factory('notice');
88 $notices->query('SELECT uri, modified FROM notice');
93 while ($notices->fetch()) {
95 # Maximum 50,000 URLs per sitemap file.
96 if ($notice_count == 50000) {
102 'url' => $notices->uri,
103 'lastmod' => w3cdate($notices->modified),
104 'changefreq' => 'daily',
108 $notice_list[$map_count] .= url($notice);
113 # Make full sitemaps from the lists and save them.
114 array_to_map($notice_list, 'notice');
117 # Generate sitemaps of all users.
118 function user_map() {
119 global $output_paths;
121 $users = DB_DataObject::factory('user');
123 $users->query('SELECT id, nickname FROM user');
128 while ($users->fetch()) {
130 # Maximum 50,000 URLs per sitemap file.
131 if ($user_count == 50000) {
136 $user_args = array('nickname' => $users->nickname);
138 # Define parameters for generating <url></url> elements.
140 'url' => common_local_url('showstream', $user_args),
141 'changefreq' => 'daily',
146 'url' => common_local_url('userrss', $user_args),
147 'changefreq' => 'daily',
152 'url' => common_local_url('all', $user_args),
153 'changefreq' => 'daily',
158 'url' => common_local_url('allrss', $user_args),
159 'changefreq' => 'daily',
164 'url' => common_local_url('replies', $user_args),
165 'changefreq' => 'daily',
169 $replies_rss = array(
170 'url' => common_local_url('repliesrss', $user_args),
171 'changefreq' => 'daily',
176 'url' => common_local_url('foaf', $user_args),
177 'changefreq' => 'weekly',
181 # Construct a <url></url> element for each user facet and add it
182 # to our existing list of those.
183 $user_list[$map_count] .= url($user);
184 $user_rss_list[$map_count] .= url($user_rss);
185 $all_list[$map_count] .= url($all);
186 $all_rss_list[$map_count] .= url($all_rss);
187 $replies_list[$map_count] .= url($replies);
188 $replies_rss_list[$map_count] .= url($replies_rss);
189 $foaf_list[$map_count] .= url($foaf);
194 # Make full sitemaps from the lists and save them.
195 # Possible factoring: put all the lists into a master array, thus allowing
196 # calling with single argument (i.e., array_to_map('user')).
197 array_to_map($user_list, 'user');
198 array_to_map($user_rss_list, 'user_rss');
199 array_to_map($all_list, 'all');
200 array_to_map($all_rss_list, 'all_rss');
201 array_to_map($replies_list, 'replies');
202 array_to_map($replies_rss_list, 'replies_rss');
203 array_to_map($foaf_list, 'foaf');
206 # Generate sitemaps of all avatars.
207 function avatar_map() {
208 global $output_paths;
210 $avatars = DB_DataObject::factory('avatar');
212 $avatars->query('SELECT url, modified FROM avatar');
217 while ($avatars->fetch()) {
219 # We only want the original size and 24px thumbnail version - skip 96px.
220 if (preg_match('/-96-/', $avatars->url)) {
224 # Maximum 50,000 URLs per sitemap file.
225 if ($avatar_count == 50000) {
229 w3cdate($avatars->modified);
231 'url' => $avatars->url,
232 'lastmod' => w3cdate($avatars->modified),
233 'changefreq' => 'monthly',
237 # Construct a <url></url> element for each avatar and add it
238 # to our existing list of those.
239 $avatar_list[$map_count] .= url($image);
242 array_to_map($avatar_list, 'avatars');
245 # ------------------------------------------------------------------------------
246 # XML generation functions
247 # ------------------------------------------------------------------------------
249 # Generate a <url></url> element.
250 function url($url_args) {
251 $url = preg_replace('/&/', '&', $url_args['url']); # escape ampersands for XML
252 $lastmod = $url_args['lastmod'];
253 $changefreq = $url_args['changefreq'];
254 $priority = $url_args['priority'];
257 error("url() arguments require 'url' value.");
260 $url_out = "\t<url>\n";
261 $url_out .= "\t\t<loc>$url</loc>\n";
264 $url_out .= "\t\t<changefreq>$changefreq</changefreq>\n";
268 $url_out .= "\t\t<lastmod>$lastmod</lastmod>\n";
272 $url_out .= "\t\t<priority>$priority</priority>\n";
275 $url_out .= "\t</url>\n";
280 # Generate a <urlset></urlset> element.
281 function urlset($urlset_text) {
282 $urlset = '<?xml version="1.0" encoding="UTF-8"?>' . "\n" .
283 '<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">' . "\n" .
290 # Generate a sitemap from an array containing <url></url> elements and write it to a file.
291 function array_to_map($url_list, $filename_prefix) {
292 global $output_paths;
295 # $map_urls is a long string containing concatenated <url></url> elements.
296 while (list($map_idx, $map_urls) = each($url_list)) {
297 $urlset_path = $output_paths['output_dir'] . "$filename_prefix-$map_idx.xml";
299 write_file($urlset_path, urlset($map_urls));
304 # ------------------------------------------------------------------------------
306 # ------------------------------------------------------------------------------
308 # Parse command line arguments.
309 function parse_args() {
310 $args = getopt('f:d:u:');
312 if (is_null($args[f]) && is_null($args[d]) && is_null($args[u])) {
313 error('Mandatory arguments: -f <index file path> -d <output directory path> -u <URL of sitemaps directory>');
316 if (is_null($args[f])) {
317 error('You must specify an index file name with the -f option.');
320 if (is_null($args[d])) {
321 error('You must specify a directory for the output file with the -d option.');
324 if (is_null($args[u])) {
325 error('You must specify a URL for the directory where the sitemaps will be kept with the -u option.');
328 $index_file = $args[f];
329 $output_dir = $args[d];
330 $output_url = $args[u];
332 if (file_exists($output_dir)) {
333 if (is_writable($output_dir) === FALSE) {
334 error("$output_dir is not writable.");
337 error("output directory $output_dir does not exist.");
341 'index_file' => $index_file,
342 'output_dir' => trailing_slash($output_dir),
343 'output_url' => trailing_slash($output_url),
349 # Format database timestamps as W3C DTF.
350 function w3cdate ($timestamp) {
351 preg_match('/(\d{4})-(\d\d)-(\d\d) (\d\d):(\d\d):(\d\d)/', $timestamp, $date);
353 return date(DATE_W3C, mktime($date[4], $date[5], $date[6], $date[2], $date[3], $date[1]));
356 # Ensure paths end with a "/".
357 function trailing_slash($path) {
358 if (preg_match('/\/$/', $path) == 0) {
365 # Write data to disk.
366 function write_file($path, $data) {
367 if (is_null($path)) {
368 error('No path specified for writing to.');
369 } elseif (is_null($data)) {
370 error('No data specified for writing.');
373 if (($fh_out = fopen($path,'w')) === FALSE) {
374 error("couldn't open $path for writing.");
377 if (fwrite($fh_out, $data) === FALSE) {
378 error("couldn't write to $path.");
382 # Display an error message and exit.
383 function error ($error_msg) {
384 if (is_null($error_msg)) {
385 $error_msg = 'error() was called without any explanation!';
388 echo "Error: $error_msg\n";