580 |
jpm |
1 |
<?php
|
|
|
2 |
/* vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4: */
|
|
|
3 |
|
|
|
4 |
/**
|
|
|
5 |
* RSS1.1 class for XML_Feed_Parser
|
|
|
6 |
*
|
|
|
7 |
* PHP versions 5
|
|
|
8 |
*
|
|
|
9 |
* LICENSE: This source file is subject to version 3.0 of the PHP license
|
|
|
10 |
* that is available through the world-wide-web at the following URI:
|
|
|
11 |
* http://www.php.net/license/3_0.txt. If you did not receive a copy of
|
|
|
12 |
* the PHP License and are unable to obtain it through the web, please
|
|
|
13 |
* send a note to license@php.net so we can mail you a copy immediately.
|
|
|
14 |
*
|
|
|
15 |
* @category XML
|
|
|
16 |
* @package XML_Feed_Parser
|
|
|
17 |
* @author James Stewart <james@jystewart.net>
|
|
|
18 |
* @copyright 2005 James Stewart <james@jystewart.net>
|
|
|
19 |
* @license http://www.gnu.org/copyleft/lesser.html GNU LGPL 2.1
|
|
|
20 |
* @version CVS: $Id: RSS11.php 304308 2010-10-11 12:05:50Z clockwerx $
|
|
|
21 |
* @link http://pear.php.net/package/XML_Feed_Parser/
|
|
|
22 |
*/
|
|
|
23 |
|
|
|
24 |
/**
|
|
|
25 |
* This class handles RSS1.1 feeds. RSS1.1 is documented at:
|
|
|
26 |
* http://inamidst.com/rss1.1/
|
|
|
27 |
*
|
|
|
28 |
* @author James Stewart <james@jystewart.net>
|
|
|
29 |
* @version Release: @package_version@
|
|
|
30 |
* @package XML_Feed_Parser
|
|
|
31 |
* @todo Support for RDF:List
|
|
|
32 |
* @todo Ensure xml:lang is accessible to users
|
|
|
33 |
*/
|
612 |
jpm |
34 |
class XmlFeedParserRss11 extends XmlFeedParserType {
|
580 |
jpm |
35 |
/**
|
|
|
36 |
* The URI of the RelaxNG schema used to (optionally) validate the feed
|
|
|
37 |
* @var string
|
|
|
38 |
*/
|
|
|
39 |
protected $relax = 'rss11.rng';
|
|
|
40 |
|
|
|
41 |
/**
|
|
|
42 |
* We're likely to use XPath, so let's keep it global
|
|
|
43 |
* @var DOMXPath
|
|
|
44 |
*/
|
|
|
45 |
protected $xpath;
|
|
|
46 |
|
|
|
47 |
/**
|
|
|
48 |
* The feed type we are parsing
|
|
|
49 |
* @var string
|
|
|
50 |
*/
|
|
|
51 |
public $version = 'RSS 1.0';
|
|
|
52 |
|
|
|
53 |
/**
|
|
|
54 |
* The class used to represent individual items
|
|
|
55 |
* @var string
|
|
|
56 |
*/
|
612 |
jpm |
57 |
protected $itemClass = 'XmlFeedParserRss11Element';
|
580 |
jpm |
58 |
|
|
|
59 |
/**
|
|
|
60 |
* The element containing entries
|
|
|
61 |
* @var string
|
|
|
62 |
*/
|
|
|
63 |
protected $itemElement = 'item';
|
|
|
64 |
|
|
|
65 |
/**
|
|
|
66 |
* Here we map those elements we're not going to handle individually
|
|
|
67 |
* to the constructs they are. The optional second parameter in the array
|
|
|
68 |
* tells the parser whether to 'fall back' (not apt. at the feed level) or
|
|
|
69 |
* fail if the element is missing. If the parameter is not set, the function
|
|
|
70 |
* will simply return false and leave it to the client to decide what to do.
|
|
|
71 |
* @var array
|
|
|
72 |
*/
|
|
|
73 |
protected $map = array(
|
|
|
74 |
'title' => array('Text'),
|
|
|
75 |
'link' => array('Text'),
|
|
|
76 |
'description' => array('Text'),
|
|
|
77 |
'image' => array('Image'),
|
|
|
78 |
'updatePeriod' => array('Text'),
|
|
|
79 |
'updateFrequency' => array('Text'),
|
|
|
80 |
'updateBase' => array('Date'),
|
|
|
81 |
'rights' => array('Text'), # dc:rights
|
|
|
82 |
'description' => array('Text'), # dc:description
|
|
|
83 |
'creator' => array('Text'), # dc:creator
|
|
|
84 |
'publisher' => array('Text'), # dc:publisher
|
|
|
85 |
'contributor' => array('Text'), # dc:contributor
|
|
|
86 |
'date' => array('Date') # dc:contributor
|
|
|
87 |
);
|
|
|
88 |
|
|
|
89 |
/**
|
|
|
90 |
* Here we map some elements to their atom equivalents. This is going to be
|
|
|
91 |
* quite tricky to pull off effectively (and some users' methods may vary)
|
|
|
92 |
* but is worth trying. The key is the atom version, the value is RSS2.
|
|
|
93 |
* @var array
|
|
|
94 |
*/
|
|
|
95 |
protected $compatMap = array(
|
|
|
96 |
'title' => array('title'),
|
|
|
97 |
'link' => array('link'),
|
|
|
98 |
'subtitle' => array('description'),
|
|
|
99 |
'author' => array('creator'),
|
|
|
100 |
'updated' => array('date'));
|
|
|
101 |
|
|
|
102 |
/**
|
|
|
103 |
* We will be working with multiple namespaces and it is useful to
|
|
|
104 |
* keep them together. We will retain support for some common RSS1.0 modules
|
|
|
105 |
* @var array
|
|
|
106 |
*/
|
|
|
107 |
protected $namespaces = array(
|
|
|
108 |
'rdf' => 'http://www.w3.org/1999/02/22-rdf-syntax-ns#',
|
|
|
109 |
'rss' => 'http://purl.org/net/rss1.1#',
|
|
|
110 |
'dc' => 'http://purl.org/rss/1.0/modules/dc/',
|
|
|
111 |
'content' => 'http://purl.org/rss/1.0/modules/content/',
|
|
|
112 |
'sy' => 'http://web.resource.org/rss/1.0/modules/syndication/');
|
|
|
113 |
|
|
|
114 |
/**
|
|
|
115 |
* Our constructor does nothing more than its parent.
|
|
|
116 |
*
|
|
|
117 |
* @param DOMDocument $xml A DOM object representing the feed
|
|
|
118 |
* @param bool (optional) $string Whether or not to validate this feed
|
|
|
119 |
*/
|
612 |
jpm |
120 |
function __construct(DOMDocument $model, $strict = false) {
|
580 |
jpm |
121 |
$this->model = $model;
|
|
|
122 |
|
|
|
123 |
if ($strict) {
|
|
|
124 |
if (! $this->relaxNGValidate()) {
|
|
|
125 |
throw new XML_Feed_Parser_Exception('Failed required validation');
|
|
|
126 |
}
|
|
|
127 |
}
|
|
|
128 |
|
|
|
129 |
$this->xpath = new DOMXPath($model);
|
|
|
130 |
foreach ($this->namespaces as $key => $value) {
|
|
|
131 |
$this->xpath->registerNamespace($key, $value);
|
|
|
132 |
}
|
|
|
133 |
$this->numberEntries = $this->count('item');
|
|
|
134 |
}
|
|
|
135 |
|
|
|
136 |
/**
|
|
|
137 |
* Attempts to identify an element by ID given by the rdf:about attribute
|
|
|
138 |
*
|
|
|
139 |
* This is not really something that will work with RSS1.1 as it does not have
|
|
|
140 |
* clear restrictions on the global uniqueness of IDs. We will employ the
|
|
|
141 |
* _very_ hit and miss method of selecting entries based on the rdf:about
|
|
|
142 |
* attribute. Please note that this is even more hit and miss with RSS1.1 than
|
|
|
143 |
* with RSS1.0 since RSS1.1 does not require the rdf:about attribute for items.
|
|
|
144 |
*
|
|
|
145 |
* @param string $id any valid ID.
|
|
|
146 |
* @return XML_Feed_Parser_RSS1Element
|
|
|
147 |
*/
|
612 |
jpm |
148 |
function getEntryById($id) {
|
580 |
jpm |
149 |
if (isset($this->idMappings[$id])) {
|
|
|
150 |
return $this->entries[$this->idMappings[$id]];
|
|
|
151 |
}
|
|
|
152 |
|
|
|
153 |
$entries = $this->xpath->query("//rss:item[@rdf:about='$id']");
|
|
|
154 |
if ($entries->length > 0) {
|
|
|
155 |
$classname = $this->itemClass;
|
|
|
156 |
$entry = new $classname($entries->item(0), $this);
|
|
|
157 |
return $entry;
|
|
|
158 |
}
|
|
|
159 |
return false;
|
|
|
160 |
}
|
|
|
161 |
|
|
|
162 |
/**
|
|
|
163 |
* Get details of the image associated with the feed.
|
|
|
164 |
*
|
|
|
165 |
* @return array|false an array simply containing the child elements
|
|
|
166 |
*/
|
612 |
jpm |
167 |
protected function getImage() {
|
580 |
jpm |
168 |
$images = $this->model->getElementsByTagName('image');
|
|
|
169 |
if ($images->length > 0) {
|
|
|
170 |
$image = $images->item(0);
|
|
|
171 |
$details = array();
|
|
|
172 |
if ($image->hasChildNodes()) {
|
|
|
173 |
$details = array(
|
|
|
174 |
'title' => $image->getElementsByTagName('title')->item(0)->value,
|
|
|
175 |
'url' => $image->getElementsByTagName('url')->item(0)->value);
|
|
|
176 |
if ($image->getElementsByTagName('link')->length > 0) {
|
|
|
177 |
$details['link'] =
|
|
|
178 |
$image->getElementsByTagName('link')->item(0)->value;
|
|
|
179 |
}
|
|
|
180 |
} else {
|
|
|
181 |
$details = array('title' => false,
|
|
|
182 |
'link' => false,
|
|
|
183 |
'url' => $image->attributes->getNamedItem('resource')->nodeValue);
|
|
|
184 |
}
|
|
|
185 |
$details = array_merge($details,
|
|
|
186 |
array('description' => false, 'height' => false, 'width' => false));
|
|
|
187 |
if (! empty($details)) {
|
|
|
188 |
return $details;
|
|
|
189 |
}
|
|
|
190 |
}
|
|
|
191 |
return false;
|
|
|
192 |
}
|
|
|
193 |
|
|
|
194 |
/**
|
|
|
195 |
* The textinput element is little used, but in the interests of
|
|
|
196 |
* completeness we will support it.
|
|
|
197 |
*
|
|
|
198 |
* @return array|false
|
|
|
199 |
*/
|
612 |
jpm |
200 |
protected function getTextInput() {
|
580 |
jpm |
201 |
$inputs = $this->model->getElementsByTagName('textinput');
|
|
|
202 |
if ($inputs->length > 0) {
|
|
|
203 |
$input = $inputs->item(0);
|
|
|
204 |
$results = array();
|
|
|
205 |
$results['title'] = isset(
|
|
|
206 |
$input->getElementsByTagName('title')->item(0)->value) ?
|
|
|
207 |
$input->getElementsByTagName('title')->item(0)->value : null;
|
|
|
208 |
$results['description'] = isset(
|
|
|
209 |
$input->getElementsByTagName('description')->item(0)->value) ?
|
|
|
210 |
$input->getElementsByTagName('description')->item(0)->value : null;
|
|
|
211 |
$results['name'] = isset(
|
|
|
212 |
$input->getElementsByTagName('name')->item(0)->value) ?
|
|
|
213 |
$input->getElementsByTagName('name')->item(0)->value : null;
|
|
|
214 |
$results['link'] = isset(
|
|
|
215 |
$input->getElementsByTagName('link')->item(0)->value) ?
|
|
|
216 |
$input->getElementsByTagName('link')->item(0)->value : null;
|
|
|
217 |
if (empty($results['link']) and
|
|
|
218 |
$input->attributes->getNamedItem('resource')) {
|
|
|
219 |
$results['link'] = $input->attributes->getNamedItem('resource')->nodeValue;
|
|
|
220 |
}
|
|
|
221 |
if (! empty($results)) {
|
|
|
222 |
return $results;
|
|
|
223 |
}
|
|
|
224 |
}
|
|
|
225 |
return false;
|
|
|
226 |
}
|
|
|
227 |
|
|
|
228 |
/**
|
|
|
229 |
* Attempts to discern authorship
|
|
|
230 |
*
|
|
|
231 |
* Dublin Core provides the dc:creator, dc:contributor, and dc:publisher
|
|
|
232 |
* elements for defining authorship in RSS1. We will try each of those in
|
|
|
233 |
* turn in order to simulate the atom author element and will return it
|
|
|
234 |
* as text.
|
|
|
235 |
*
|
|
|
236 |
* @return array|false
|
|
|
237 |
*/
|
612 |
jpm |
238 |
function getAuthor() {
|
580 |
jpm |
239 |
$options = array('creator', 'contributor', 'publisher');
|
|
|
240 |
foreach ($options as $element) {
|
|
|
241 |
$test = $this->model->getElementsByTagName($element);
|
|
|
242 |
if ($test->length > 0) {
|
|
|
243 |
return $test->item(0)->value;
|
|
|
244 |
}
|
|
|
245 |
}
|
|
|
246 |
return false;
|
|
|
247 |
}
|
|
|
248 |
|
|
|
249 |
/**
|
|
|
250 |
* Retrieve a link
|
|
|
251 |
*
|
|
|
252 |
* In RSS1 a link is a text element but in order to ensure that we resolve
|
|
|
253 |
* URLs properly we have a special function for them.
|
|
|
254 |
*
|
|
|
255 |
* @return string
|
|
|
256 |
*/
|
612 |
jpm |
257 |
function getLink($offset = 0, $attribute = 'href', $params = false) {
|
580 |
jpm |
258 |
$links = $this->model->getElementsByTagName('link');
|
|
|
259 |
if ($links->length <= $offset) {
|
|
|
260 |
return false;
|
|
|
261 |
}
|
|
|
262 |
$link = $links->item($offset);
|
|
|
263 |
return $this->addBase($link->nodeValue, $link);
|
|
|
264 |
}
|
|
|
265 |
}
|
612 |
jpm |
266 |
?>
|