7 * Note: Edits to this file must pass through github. To submit a patch to this
8 * file please contact mang at archive dot org or http://github.com/mangtronix
9 * Direct changes to this file may get clobbered when the code is synchronized
16 // Operators recognized in BookReader download URLs
17 public static $downloadOperators = array('page');
19 // Returns true if can display the book in item with a given prefix (typically the item identifier)
20 public static function canDisplay($item, $prefix, $checkOldScandata = false)
23 // A "book" is an image stack and scandata.
24 // 1. Old items may have scandata.xml or scandata.zip and itemid_{imageformat}.{zip,tar}
25 // 2. Newer items may have multiple {arbitraryname}_scandata.xml and {arbitraryname}_{imageformat}.{zip,tar}
27 $foundScandata = false;
28 $foundImageStack = false;
30 $targetScandata = $prefix . "_scandata.xml";
32 // $$$ TODO add support for jpg and tar stacks
33 // https://bugs.edge.launchpad.net/gnubook/+bug/323003
34 // https://bugs.edge.launchpad.net/gnubook/+bug/385397
35 $imageFormatRegex = '@' . preg_quote($prefix, '@') . '_(jp2|tif|jpg)\.(zip|tar)$@';
37 $baseLength = strlen($item->metadataGrabber->mainDir . '/');
38 foreach ($item->getFiles() as $location => $fileInfo) {
39 $filename = substr($location, $baseLength);
41 if ($checkOldScandata) {
42 if ($filename == 'scandata.xml' || $filename == 'scandata.zip') {
43 $foundScandata = $filename;
47 if ($filename == $targetScandata) {
48 $foundScandata = $filename;
51 if (preg_match($imageFormatRegex, $filename)) {
52 $foundImageStack = $filename;
56 if ($foundScandata && $foundImageStack) {
63 // Finds the prefix to use for the book given the part of the URL trailing after /stream/
64 public static function findPrefix($urlPortion)
66 if (!preg_match('#[^/&?]+#', $urlPortion, $matches)) {
67 // URL portion was empty or started with /, &, or ? -- no item identifier
71 $prefix = $matches[0]; // item identifier
73 // $$$ Currently swallows the rest of the URL.
74 // If we want to support e.g. /stream/itemid/subdir/prefix/page/23 will need to adjust.
75 if (preg_match('#[^/&?]+/([^&?]+)#', $urlPortion, $matches)) {
76 // Match is everything after item identifier and slash, up to end or ? or &
77 // e.g. itemid/{match/these/parts}?foo=bar
78 $prefix = $matches[1]; // sub prefix --
84 // $$$ would be cleaner to use different templates instead of the uiMode param
86 // @param subprefix Optional prefix to display a book inside an item (e.g. if does not match identifier)
87 public static function draw($server, $mainDir, $identifier, $subPrefix, $title,
88 $coverLeaf=null, $titleStart='Internet Archive', $uiMode='full')
90 // Set title to default if not set
92 $title = 'BookReader';
97 // manually update with Launchpad version number at each checkin so that browsers
98 // do not use old cached version
99 // see https://bugs.launchpad.net/gnubook/+bug/330748
102 if (BookReader::getDevHost($server)) {
103 // on dev host - add time to force reload
104 $version .= '_' . time();
108 echo "No identifier specified!";
112 $metaURL = BookReader::jsMetadataURL($server, $identifier, $mainDir, $subPrefix);
115 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
118 <meta name="viewport" content="width=device-width, maximum-scale=1.0" />
119 <meta name="apple-mobile-web-app-capable" content="yes" />
120 <title><? echo $title; ?></title>
122 <meta http-equiv="refresh" content="2; URL=/bookreader/browserunsupported.php?id=<? echo($id); ?>">
124 <link rel="stylesheet" type="text/css" href="/bookreader/BookReader.css?v=<? echo($version); ?>">
125 <? if ($uiMode == "embed") { ?>
126 <link rel="stylesheet" type="text/css" href="/bookreader/BookReaderEmbed.css?v=<? echo($version); ?>">
127 <? } elseif ($uiMode == "touch") { ?>
128 <link rel="stylesheet" type="text/css" href="/bookreader/touch/BookReaderTouch.css?v=<? echo($version); ?>">
130 <script src="/includes/jquery-1.4.2.min.js" type="text/javascript"></script>
131 <script type="text/javascript" src="http://www.archive.org/includes/analytics.js?v=2"></script>
132 <script type="text/javascript" src="/bookreader/jquery-ui-1.8.1.custom.min.js?v=<? echo($version); ?>"></script>
133 <script type="text/javascript" src="/bookreader/dragscrollable.js?v=<? echo($version); ?>"></script>
134 <script type="text/javascript" src="/bookreader/BookReader.js?v=<? echo($version); ?>"></script>
136 <body style="background-color: #FFFFFF;">
138 <? if ($uiMode == 'full') { ?>
139 <div id="BookReader" style="left:10px; right:200px; top:10px; bottom:2em;">Internet Archive BookReader <noscript>requires JavaScript to be enabled.</noscript></div>
141 <div id="BookReader" style="left:0; right:0; top:0; bottom:0; border:0">Internet Archive Bookreader <noscript>requires JavaScript to be enabled.</noscript></div>
144 <script type="text/javascript">
145 // Set some config variables -- $$$ NB: Config object format has not been finalized
147 <? if ($uiMode == 'embed') { ?>
148 brConfig["mode"] = 1;
149 brConfig["reduce"] = 8;
150 brConfig["ui"] = "embed";
152 brConfig["mode"] = 2;
155 <!-- The script included below is dynamically generated JavaScript that includes the book metadata and page image access functions -->
156 <script type="text/javascript" src="<? echo($metaURL); ?>"></script>
158 <? if ($uiMode == 'full') { ?>
159 <div id="BookReaderSearch" style="width:190px; right:0px; top:10px; bottom:2em;">
160 <form action='javascript:' onsubmit="br.search($('#BookReaderSearchBox').val());">
161 <p style="display: inline">
162 <input id="BookReaderSearchBox" type="text" size="20" value="search..." onfocus="if('search...'==this.value)this.value='';" /><input type="submit" value="go" />
165 <div id="BookReaderSearchResults">
172 <div class="BRlogotype">
173 <a href="http://archive.org/" class="BRblack">Internet Archive</a>
175 <div class="BRnavlinks">
176 <!-- <a class="BRblack" href="http://openlibrary.org/dev/docs/bookreader">About the Bookreader</a> | -->
177 <a class="BRblack" href="http://www.archive.org/about/faqs.php#Report_Item">Content Problems</a> |
178 <a class="BRblack" href="https://bugs.launchpad.net/bookreader/+filebug">Report Bugs</a> |
179 <a class="BRblack" href="http://www.archive.org/details/texts">Texts Collection</a> |
180 <a class="BRblack" href="http://www.archive.org/about/contact.php">Contact Us</a>
185 <script type="text/javascript">
186 // $$$ hack to workaround sizing bug when starting in two-up mode
187 $(document).ready(function() {
188 $(window).trigger('resize');
192 if(window.archive_analytics) { window.archive_analytics.values['bookreader'] = 'open'}; </script>
198 // Returns the user part of dev host from URL, or null
199 public static function getDevHost($server)
201 if (preg_match("/^www-(\w+)/", $_SERVER["SERVER_NAME"], $match)) {
209 public static function serverBaseURL($server)
211 // Check if we're on a dev vhost and point to JSIA in the user's public_html
213 // $$$ the remapping isn't totally automatic yet and requires user to
214 // ln -s ~/petabox/www/datanode/BookReader ~/public_html/BookReader
215 // so we enable it only for known hosts
216 $devhost = BookReader::getDevHost($server);
217 $devhosts = array('mang', 'testflip', 'rkumar');
218 if (in_array($devhost, $devhosts)) {
219 $server = $server . "/~" . $devhost;
225 public static function jsMetadataURL($server, $identifier, $mainDir, $subPrefix = '')
227 $serverBaseURL = BookReader::serverBaseURL($server);
229 $params = array( 'id' => $identifier, 'itemPath' => $mainDir, 'server' => $server );
231 $params['subPrefix'] = $subPrefix;
234 $keys = array_keys($params);
235 $lastParam = end($keys);
236 $url = "http://{$serverBaseURL}/BookReader/BookReaderJSIA.php?";
237 foreach($params as $param=>$value) {
238 $url .= $param . '=' . $value;
239 if ($param != $lastParam) {
247 // Return the URL for the requested /download/$path, or null
248 public static function getURL($path, $item) {
249 // $path should look like {itemId}/{operator}/{filename}
250 // Other operators may be added
252 $urlParts = BookReader::parsePath($path);
254 // Check for non-handled cases
255 $required = array('identifier', 'operator', 'operand');
256 foreach ($required as $key) {
257 if (!array_key_exists($key, $urlParts)) {
262 $identifier = $urlParts['identifier'];
263 $operator = $urlParts['operator'];
264 $filename = $urlParts['operand'];
265 $subPrefix = $urlParts['subPrefix'];
267 $serverBaseURL = BookReader::serverBaseURL($item->getServer());
269 // Baseline query params
272 'itemPath' => $item->getMainDir(),
273 'server' => $serverBaseURL
276 $query['subPrefix'] = $subPrefix;
282 // Look for old-style preview request - e.g. {identifier}_cover.jpg
283 if (preg_match('/^(.*)_((cover|title|preview).*)/', $filename, $matches) === 1) {
284 // Serve preview image
286 $query['page'] = $page;
287 return 'http://' . $serverBaseURL . '/BookReader/BookReaderPreview.php?' . http_build_query($query, '', '&');
290 // New-style preview request - e.g. cover_thumb.jpg
291 if (preg_match('/^(cover|title|preview)/', $filename, $matches) === 1) {
292 $query['page'] = $filename;
293 return 'http://' . $serverBaseURL . '/BookReader/BookReaderPreview.php?' . http_build_query($query, '', '&');
296 // Asking for a non-preview page
297 $query['page'] = $filename;
298 return 'http://' . $serverBaseURL . '/BookReader/BookReaderImages.php?' . http_build_query($query, '', '&');
305 return null; // was not handled
308 public static function browserFromUserAgent($userAgent) {
309 $browserPatterns = array(
311 'iphone' => '/iPhone/', // Also cover iPod Touch
312 'android' => '/Android/',
315 foreach ($browserPatterns as $browser => $pattern) {
316 if (preg_match($pattern, $userAgent)) {
324 // $$$ Ideally we will not rely on user agent, but for the moment we do
325 public static function paramsFromUserAgent($userAgent) {
326 // $$$ using 'embed' here for devices with assumed small screens -- really should just use CSS3 media queries
327 $browserParams = array(
328 'ipad' => array( 'ui' => 'touch' ),
329 'iphone' => array( 'ui' => 'embed', 'mode' => '1up' ),
330 'android' => array( 'ui' => 'embed', 'mode' => '1up' ),
333 $browser = BookReader::browserFromUserAgent($userAgent);
335 return $browserParams[$browser];
340 public static function parsePath($path) {
341 // Parse the BookReader path and return the parts
342 // e.g. itemid/some/sub/dir/page/cover.jpg -> array( 'identifier' => 'itemid', 'subPrefix' => 'some/sub/dir',
343 // 'operator' => 'page', 'filename' => 'cover.jpg')
347 // Pull off query, e.g. ?foo=bar
348 if (preg_match('#(.*?)(\?.*)#', $path, $matches) === 1) {
349 $parts['query'] = $matches[2];
353 // Pull off identifier
354 if (preg_match('#[^/&?]+#', $path, $matches) === 0) {
358 $parts['identifier'] = $matches[0];
359 $path = substr($path, strlen($matches[0]));
361 // Look for operators
362 // The sub-prefix can be arbitrary, so we match up until the first operator
363 $operators = '(' . join('|', self::$downloadOperators) . ')';
364 $pattern = '#(?P<subPrefix>.*?)/(?P<operator>' . $operators . ')/(?P<operand>.*)#';
365 if (preg_match($pattern, $path, $matches) === 1) {
366 $parts['subPrefix'] = substr($matches['subPrefix'], 1); // remove leading '/'
367 $parts['operator'] = $matches['operator'];
368 $parts['operand'] = $matches['operand'];
370 $parts['subPrefix'] = $path;