Library, Sitemap generator

app/common/library/Sitemap.php

@@ -0,0 +1,196 @@
+<?php
+
+namespace Baseapp\Library;
+
+/**
+ * Sitemap Library
+ *
+ * @package     base-app
+ * @category    Library
+ * @version     2.0
+ */
+class Sitemap
+{
+
+    /**
+     * @var DOMDocument
+     */
+    protected $_xml;
+
+    /**
+     * @var DOMElement
+     */
+    protected $_root;
+
+    /**
+     * @var boolean Enable gzip compression
+     */
+    public $gzip = false;
+
+    /**
+     * @var integer Compression level
+     */
+    public $compression = 7;
+
+    /**
+     * Setup the XML document
+     */
+    public function __construct()
+    {
+        // Load sitemap config from config.ini
+        if (isset(\Phalcon\DI::getDefault()->getShared('config')->sitemap) && $config = \Phalcon\DI::getDefault()->getShared('config')->sitemap) {
+            foreach ($config as $key => $value) {
+                $this->$key = $value;
+            }
+        }
+
+        // XML document
+        $this->_xml = new \DOMDocument('1.0', 'UTF-8');
+
+        // Attributes
+        $this->_xml->formatOutput = true;
+
+        // Root element
+        $this->_root = $this->_xml->createElement('urlset');
+
+        // Append to XML document
+        $this->_xml->appendChild($this->_root);
+    }
+
+    /**
+     * @param Sitemap_URL $object
+     */
+    public function add(Sitemap\URL $object)
+    {
+        $url = $object->create();
+
+        // Decorate the urlset
+        $object->root($this->_root);
+
+        // Append URL to root element
+        $this->_root->appendChild($this->_xml->importNode($url, true));
+    }
+
+    /**
+     * Ping web services
+     *
+     * @param string $sitemap Full website path to sitemap
+     * @return array Service key with the HTTP response code as the value.
+     */
+    public static function ping($sitemap)
+    {
+        if (!isset(\Phalcon\DI::getDefault()->getShared('config')->sitemap->ping)) {
+            return null;
+        }
+
+        // URLs to ping
+        $ping = \Phalcon\DI::getDefault()->getShared('config')->sitemap->ping;
+
+        // Main handle
+        $master = curl_multi_init();
+
+        $handles = array();
+
+        // Create handles for each URL and add them to the main handle.
+        foreach ($ping as $key => $val) {
+            $handles[$key] = curl_init(sprintf($val, $sitemap));
+
+            curl_setopt($handles[$key], CURLOPT_FOLLOWLOCATION, true);
+            curl_setopt($handles[$key], CURLOPT_RETURNTRANSFER, true);
+            curl_setopt($handles[$key], CURLOPT_USERAGENT, 'Mozilla/5.0 (X11; U; Linux x86_64; en-GB; rv:1.9.2.3) Gecko/20100423 Ubuntu/10.04 (lucid) Firefox/3.6.3');
+
+            curl_multi_add_handle($master, $handles[$key]);
+        }
+
+        do {
+            curl_multi_exec($master, $still_running);
+        } while ($still_running > 0);
+
+        $info = array();
+
+        // Build an array of the execution information.
+        foreach (array_keys($ping) as $key) {
+            $info[$key] = curl_getinfo($handles[$key], CURLINFO_HTTP_CODE);
+
+            // Close the handles while we're here.
+            curl_multi_remove_handle($master, $handles[$key]);
+        }
+
+        // and finally close the master handle.
+        curl_multi_close($master);
+
+        return $info;
+    }
+
+    /**
+     * UTF8 encode a string
+     *
+     * @access public
+     * @param string $string
+     * @return string
+     */
+    public static function encode($string)
+    {
+        $string = htmlspecialchars($string, ENT_QUOTES, 'UTF-8');
+
+        // This is a rather ugly hack. Basically urlencode and rawurlencode use RFC 1738
+        // encoding. This brings it up to date (RFC 3986); The newer RFC has a different
+        // set of reserved characters. Credit goes to davis dot peixoto at gmail dot com
+        // God bless PHP comments.
+        $entities = array('%21', '%2A', '%27', '%28', '%29', '%3B', '%3A', '%40',
+            '%26', '%3D', '%2B', '%24', '%2C', '%2F', '%3F', '%23', '%5B', '%5D');
+
+        $replacements = array('!', '*', "'", "(", ")", ";", ":", "@", "&", "=", "+",
+            "$", ",", "/", "?", "#", "[", "]");
+
+        $string = str_replace($entities, $replacements, rawurlencode($string));
+
+        return str_replace('&#039;', '&apos;', $string);
+    }
+
+    /**
+     * Format a unix timestamp into W3C Datetime
+     *
+     * @access public
+     * @see http://www.w3.org/TR/NOTE-datetime
+     * @param string $unix Unixtimestamp
+     * @return string W3C Datetime
+     */
+    public static function date_format($unix)
+    {
+        if (is_numeric($unix) AND $unix <= PHP_INT_MAX) {
+            return date('Y-m-d\TH:i:sP', $unix);
+        }
+
+        throw new \InvalidArgumentException('Must be a unix timestamp');
+    }
+
+    /**
+     * @return string Either an XML document or a gzipped file
+     */
+    public function render()
+    {
+        // Default uncompressed
+        $response = $this->_xml->saveXML();
+
+        if ($this->gzip) {
+            // Try and gzip the file before we send it off.
+            try {
+                $response = gzencode($response, $this->compression);
+            } catch (ErrorException $e) {
+                \Baseapp\Bootstrap::exception($e);
+            }
+        }
+
+        return $response;
+    }
+
+    /**
+     * @return string XML output.
+     */
+    public function __toString()
+    {
+        return $this->render();
+    }
+
+}

app/common/library/Sitemap/Code.php

@@ -0,0 +1,139 @@
+<?php
+
+namespace Baseapp\Library\Sitemap;
+
+class Code implements \Baseapp\Library\Sitemap\SitemapInterface
+{
+
+    private $_attributes = array(
+        'filetype' => null,
+        'license' => null,
+        'filename' => null,
+        'packageurl' => null,
+        'packagemap' => null
+    );
+    protected $_licenses = array(
+        'aladdin', 'artistic', 'apache', 'apple', 'bsd', 'cpl', 'gpl', 'lgpl', 'disclaimer',
+        'ibm', 'lucent', 'mit', 'mozilla', 'nasa', 'python', 'qpl', 'sleepycat', 'zope'
+    );
+    protected $_archives = array(
+        '.tar', '.tar.z', '.tar.gz', '.tgz', '.tar.bz2', '.tbz', '.tbz2', '.zip'
+    );
+
+    /**
+     * @param string $type Case-insensitive. The value "archive" indicates that
+     * the file is an archive file. For source code files, the value defines the
+     * the source code language. Examples include "C", "Python", "C#", "Java", "Vim".
+     * For source code language, the Short Name, as specified in the list of supported
+     * languages, must be used. The value must be printable ASCII characters, and
+     * no white space is allowed.
+     *
+     * @see http://www.google.com/support/webmasters/bin/answer.py?answer=75252
+     */
+    public function set_file_type($type)
+    {
+        $type = (string) $type;
+
+        if (!preg_match('/^[a-z][a-z0-9+#]*$/i', $type)) {
+            throw new \InvalidArgumentException('Type must only contain a-z, 0-9, + and #');
+        }
+
+        $this->_attributes['filetype'] = $type;
+
+        return $this;
+    }
+
+    /**
+     * @param string $license Case-insensitive. The name of the software license.
+     * For archive files, this indicates the default license for files in the archive.
+     * Examples include "GPL", "BSD", "Python", "disclaimer". You must use the Short
+     * Name, as specified in the list of supported licenses.
+     *
+     * @see http://www.google.com/support/webmasters/bin/answer.py?answer=75256
+     */
+    public function set_license($license)
+    {
+        $license = (string) $license;
+
+        if (!in_array($license, $this->_licenses)) {
+            throw new \InvalidArgumentException('Invalid license type. See http://www.google.com/support/webmasters/bin/answer.py?answer=75256 for details');
+        }
+
+        $this->_attributes['license'] = $license;
+
+        return $this;
+    }
+
+    /**
+     * @param string $file_name The name of the actual file. This is useful if the
+     * URL ends in something like download.php?id=1234 instead of the actual filename.
+     * The name can contain any character except "/". If the file is an archive file,
+     * it will be indexed only if it has one of the supported archive suffixes.
+     *
+     * @see http://www.google.com/support/webmasters/bin/answer.py?answer=75259
+     */
+    public function set_file_name($file_name)
+    {
+        $file_name = (string) $file_name;
+
+        if ($this->_attributes['filetype'] === 'archive') {
+            if (!in_array(pathinfo($file_name, PATHINFO_EXTENSION), $this->_archives)) {
+                throw new \InvalidArgumentException('Not a valid archive type');
+            }
+        }
+
+        $this->_attributes['filename'] = basename($file_name);
+
+        return $this;
+    }
+
+    /**
+     * @param <type> $package_type For use only when the value of codesearch:filetype
+     * is not "archive". The URL truncated at the top-level directory for the package.
+     * For example, the file http://path/Foo/1.23/bar/file.c could have the package URL
+     * http://path/Foo/1.23. All files in a package should have the same packageurl.
+     * This tells us which files belong together.
+     */
+    public function set_package_url($package_type)
+    {
+        $this->_attributes['packageurl'] = $package_type;
+    }
+
+    /**
+     * @param string $package_map Case-sensitive. For use only when codesearch:filetype
+     * is "archive". The name of the packagemap file inside the archive. Just like a
+     * Sitemap is a list of files on a web site, a packagemap is a list of files in
+     * a package.
+     *
+     * @see http://www.google.com/help/codesearch_packagemap.html
+     */
+    public function set_package_map($package_map)
+    {
+        $this->_attributes['packagemap'] = $package_map;
+    }
+
+    public function create()
+    {
+        // Here we need to create a new DOMDocument. This is so we can re-import the
+        // DOMElement at the other end.
+        $document = new \DOMDocument;
+
+        // Mobile element
+        $code = $document->createElement('codesearch:codesearch');
+
+        // Append attributes
+        foreach ($this->_attributes as $name => $value) {
+            if (null !== $value) {
+                $code->appendChild($document->createElement('codesearch:' . $name, $value));
+            }
+        }
+
+        return $code;
+    }
+
+    public function root(\DOMElement & $root)
+    {
+        $root->setAttributeNS('http://www.w3.org/2000/xmlns/', 'xmlns:codesearch', 'http://www.google.com/codesearch/schemas/sitemap/1.0');
+    }
+
+}

app/common/library/Sitemap/Geo.php

@@ -0,0 +1,50 @@
+<?php
+
+namespace Baseapp\Library\Sitemap;
+
+class Geo implements \Baseapp\Library\Sitemap\SitemapInterface
+{
+
+    protected $_format = null;
+    protected $_allowed_formats = array(
+        'kml', 'kmz', 'georss'
+    );
+
+    /**
+     * @param string $format Case-insensitive. Specifies the format of the geo content.
+     * Examples include "kml" and "georss". Only supported formats will be indexed.
+     *
+     * @see http://www.google.com/support/webmasters/bin/answer.py?answer=94556
+     */
+    public function set_format($format)
+    {
+        if (in_array((string) $format, $this->_allowed_formats)) {
+            $this->_format = $format;
+
+            return $this;
+        }
+
+        throw new \InvalidArgumentException('The format must either be kml, kmlz or georss');
+    }
+
+    public function create()
+    {
+        // Here we need to create a new DOMDocument. This is so we can re-import the
+        // DOMElement at the other end.
+        $document = new \DOMDocument;
+
+        // Mobile element
+        $geo = $document->createElement('geo:geo');
+
+        // Add format
+        $geo->appendChild($document->createElement('geo:format', $this->_format));
+
+        return $geo;
+    }
+
+    public function root(\DOMElement & $root)
+    {
+        $root->setAttributeNS('http://www.w3.org/2000/xmlns/', 'xmlns:geo', 'http://www.google.com/geo/schemas/sitemap/1.0');
+    }
+
+}