1 | <?php |
---|
2 | /* vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4: */ |
---|
3 | |
---|
4 | /** |
---|
5 | * Class representing feed-level data for an RSS2 feed |
---|
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: RSS2.php,v 1.11 2006/07/27 13:52:05 jystewart Exp $ |
---|
21 | * @link http://pear.php.net/package/XML_Feed_Parser/ |
---|
22 | */ |
---|
23 | |
---|
24 | /** |
---|
25 | * This class handles RSS2 feeds. |
---|
26 | * |
---|
27 | * @author James Stewart <james@jystewart.net> |
---|
28 | * @version Release: 1.0.2 |
---|
29 | * @package XML_Feed_Parser |
---|
30 | */ |
---|
31 | class XML_Feed_Parser_RSS2 extends XML_Feed_Parser_Type |
---|
32 | { |
---|
33 | /** |
---|
34 | * The URI of the RelaxNG schema used to (optionally) validate the feed |
---|
35 | * @var string |
---|
36 | */ |
---|
37 | private $relax = 'rss20.rnc'; |
---|
38 | |
---|
39 | /** |
---|
40 | * We're likely to use XPath, so let's keep it global |
---|
41 | * @var DOMXPath |
---|
42 | */ |
---|
43 | protected $xpath; |
---|
44 | |
---|
45 | /** |
---|
46 | * The feed type we are parsing |
---|
47 | * @var string |
---|
48 | */ |
---|
49 | public $version = 'RSS 2.0'; |
---|
50 | |
---|
51 | /** |
---|
52 | * The class used to represent individual items |
---|
53 | * @var string |
---|
54 | */ |
---|
55 | protected $itemClass = 'XML_Feed_Parser_RSS2Element'; |
---|
56 | |
---|
57 | /** |
---|
58 | * The element containing entries |
---|
59 | * @var string |
---|
60 | */ |
---|
61 | protected $itemElement = 'item'; |
---|
62 | |
---|
63 | /** |
---|
64 | * Here we map those elements we're not going to handle individually |
---|
65 | * to the constructs they are. The optional second parameter in the array |
---|
66 | * tells the parser whether to 'fall back' (not apt. at the feed level) or |
---|
67 | * fail if the element is missing. If the parameter is not set, the function |
---|
68 | * will simply return false and leave it to the client to decide what to do. |
---|
69 | * @var array |
---|
70 | */ |
---|
71 | protected $map = array( |
---|
72 | 'ttl' => array('Text'), |
---|
73 | 'pubDate' => array('Date'), |
---|
74 | 'lastBuildDate' => array('Date'), |
---|
75 | 'title' => array('Text'), |
---|
76 | 'link' => array('Link'), |
---|
77 | 'description' => array('Text'), |
---|
78 | 'language' => array('Text'), |
---|
79 | 'copyright' => array('Text'), |
---|
80 | 'managingEditor' => array('Text'), |
---|
81 | 'webMaster' => array('Text'), |
---|
82 | 'category' => array('Text'), |
---|
83 | 'generator' => array('Text'), |
---|
84 | 'docs' => array('Text'), |
---|
85 | 'ttl' => array('Text'), |
---|
86 | 'image' => array('Image'), |
---|
87 | 'skipDays' => array('skipDays'), |
---|
88 | 'skipHours' => array('skipHours')); |
---|
89 | |
---|
90 | /** |
---|
91 | * Here we map some elements to their atom equivalents. This is going to be |
---|
92 | * quite tricky to pull off effectively (and some users' methods may vary) |
---|
93 | * but is worth trying. The key is the atom version, the value is RSS2. |
---|
94 | * @var array |
---|
95 | */ |
---|
96 | protected $compatMap = array( |
---|
97 | 'title' => array('title'), |
---|
98 | 'rights' => array('copyright'), |
---|
99 | 'updated' => array('lastBuildDate'), |
---|
100 | 'subtitle' => array('description'), |
---|
101 | 'date' => array('pubDate'), |
---|
102 | 'author' => array('managingEditor')); |
---|
103 | |
---|
104 | protected $namespaces = array( |
---|
105 | 'dc' => 'http://purl.org/rss/1.0/modules/dc/', |
---|
106 | 'content' => 'http://purl.org/rss/1.0/modules/content/'); |
---|
107 | |
---|
108 | /** |
---|
109 | * Our constructor does nothing more than its parent. |
---|
110 | * |
---|
111 | * @param DOMDocument $xml A DOM object representing the feed |
---|
112 | * @param bool (optional) $string Whether or not to validate this feed |
---|
113 | */ |
---|
114 | function __construct(DOMDocument $model, $strict = false) |
---|
115 | { |
---|
116 | $this->model = $model; |
---|
117 | |
---|
118 | if ($strict) { |
---|
119 | if (! $this->model->relaxNGValidate($this->relax)) { |
---|
120 | throw new XML_Feed_Parser_Exception('Failed required validation'); |
---|
121 | } |
---|
122 | } |
---|
123 | |
---|
124 | $this->xpath = new DOMXPath($this->model); |
---|
125 | foreach ($this->namespaces as $key => $value) { |
---|
126 | $this->xpath->registerNamespace($key, $value); |
---|
127 | } |
---|
128 | $this->numberEntries = $this->count('item'); |
---|
129 | } |
---|
130 | |
---|
131 | /** |
---|
132 | * Retrieves an entry by ID, if the ID is specified with the guid element |
---|
133 | * |
---|
134 | * This is not really something that will work with RSS2 as it does not have |
---|
135 | * clear restrictions on the global uniqueness of IDs. But we can emulate |
---|
136 | * it by allowing access based on the 'guid' element. If DOMXPath::evaluate |
---|
137 | * is available, we also use that to store a reference to the entry in the array |
---|
138 | * used by getEntryByOffset so that method does not have to seek out the entry |
---|
139 | * if it's requested that way. |
---|
140 | * |
---|
141 | * @param string $id any valid ID. |
---|
142 | * @return XML_Feed_Parser_RSS2Element |
---|
143 | */ |
---|
144 | function getEntryById($id) |
---|
145 | { |
---|
146 | if (isset($this->idMappings[$id])) { |
---|
147 | return $this->entries[$this->idMappings[$id]]; |
---|
148 | } |
---|
149 | |
---|
150 | $entries = $this->xpath->query("//item[guid='$id']"); |
---|
151 | if ($entries->length > 0) { |
---|
152 | $entry = new $this->itemElement($entries->item(0), $this); |
---|
153 | if (in_array('evaluate', get_class_methods($this->xpath))) { |
---|
154 | $offset = $this->xpath->evaluate("count(preceding-sibling::item)", $entries->item(0)); |
---|
155 | $this->entries[$offset] = $entry; |
---|
156 | } |
---|
157 | $this->idMappings[$id] = $entry; |
---|
158 | return $entry; |
---|
159 | } |
---|
160 | } |
---|
161 | |
---|
162 | /** |
---|
163 | * Get a category from the element |
---|
164 | * |
---|
165 | * The category element is a simple text construct which can occur any number |
---|
166 | * of times. We allow access by offset or access to an array of results. |
---|
167 | * |
---|
168 | * @param string $call for compatibility with our overloading |
---|
169 | * @param array $arguments - arg 0 is the offset, arg 1 is whether to return as array |
---|
170 | * @return string|array|false |
---|
171 | */ |
---|
172 | function getCategory($call, $arguments = array()) |
---|
173 | { |
---|
174 | $categories = $this->model->getElementsByTagName('category'); |
---|
175 | $offset = empty($arguments[0]) ? 0 : $arguments[0]; |
---|
176 | $array = empty($arguments[1]) ? false : true; |
---|
177 | if ($categories->length <= $offset) { |
---|
178 | return false; |
---|
179 | } |
---|
180 | if ($array) { |
---|
181 | $list = array(); |
---|
182 | foreach ($categories as $category) { |
---|
183 | array_push($list, $category->nodeValue); |
---|
184 | } |
---|
185 | return $list; |
---|
186 | } |
---|
187 | return $categories->item($offset)->nodeValue; |
---|
188 | } |
---|
189 | |
---|
190 | /** |
---|
191 | * Get details of the image associated with the feed. |
---|
192 | * |
---|
193 | * @return array|false an array simply containing the child elements |
---|
194 | */ |
---|
195 | protected function getImage() |
---|
196 | { |
---|
197 | $images = $this->model->getElementsByTagName('image'); |
---|
198 | if ($images->length > 0) { |
---|
199 | $image = $images->item(0); |
---|
200 | $desc = $image->getElementsByTagName('description'); |
---|
201 | $description = $desc->length ? $desc->item(0)->nodeValue : false; |
---|
202 | $heigh = $image->getElementsByTagName('height'); |
---|
203 | $height = $heigh->length ? $heigh->item(0)->nodeValue : false; |
---|
204 | $widt = $image->getElementsByTagName('width'); |
---|
205 | $width = $widt->length ? $widt->item(0)->nodeValue : false; |
---|
206 | return array( |
---|
207 | 'title' => $image->getElementsByTagName('title')->item(0)->nodeValue, |
---|
208 | 'link' => $image->getElementsByTagName('link')->item(0)->nodeValue, |
---|
209 | 'url' => $image->getElementsByTagName('url')->item(0)->nodeValue, |
---|
210 | 'description' => $description, |
---|
211 | 'height' => $height, |
---|
212 | 'width' => $width); |
---|
213 | } |
---|
214 | return false; |
---|
215 | } |
---|
216 | |
---|
217 | /** |
---|
218 | * The textinput element is little used, but in the interests of |
---|
219 | * completeness... |
---|
220 | * |
---|
221 | * @return array|false |
---|
222 | */ |
---|
223 | function getTextInput() |
---|
224 | { |
---|
225 | $inputs = $this->model->getElementsByTagName('input'); |
---|
226 | if ($inputs->length > 0) { |
---|
227 | $input = $inputs->item(0); |
---|
228 | return array( |
---|
229 | 'title' => $input->getElementsByTagName('title')->item(0)->value, |
---|
230 | 'description' => |
---|
231 | $input->getElementsByTagName('description')->item(0)->value, |
---|
232 | 'name' => $input->getElementsByTagName('name')->item(0)->value, |
---|
233 | 'link' => $input->getElementsByTagName('link')->item(0)->value); |
---|
234 | } |
---|
235 | return false; |
---|
236 | } |
---|
237 | |
---|
238 | /** |
---|
239 | * Utility function for getSkipDays and getSkipHours |
---|
240 | * |
---|
241 | * This is a general function used by both getSkipDays and getSkipHours. It simply |
---|
242 | * returns an array of the values of the children of the appropriate tag. |
---|
243 | * |
---|
244 | * @param string $tagName The tag name (getSkipDays or getSkipHours) |
---|
245 | * @return array|false |
---|
246 | */ |
---|
247 | protected function getSkips($tagName) |
---|
248 | { |
---|
249 | $hours = $this->model->getElementsByTagName($tagName); |
---|
250 | if ($hours->length == 0) { |
---|
251 | return false; |
---|
252 | } |
---|
253 | $skipHours = array(); |
---|
254 | foreach($hours->item(0)->childNodes as $hour) { |
---|
255 | if ($hour instanceof DOMElement) { |
---|
256 | array_push($skipHours, $hour->nodeValue); |
---|
257 | } |
---|
258 | } |
---|
259 | return $skipHours; |
---|
260 | } |
---|
261 | |
---|
262 | /** |
---|
263 | * Retrieve skipHours data |
---|
264 | * |
---|
265 | * The skiphours element provides a list of hours on which this feed should |
---|
266 | * not be checked. We return an array of those hours (integers, 24 hour clock) |
---|
267 | * |
---|
268 | * @return array |
---|
269 | */ |
---|
270 | function getSkipHours() |
---|
271 | { |
---|
272 | return $this->getSkips('skipHours'); |
---|
273 | } |
---|
274 | |
---|
275 | /** |
---|
276 | * Retrieve skipDays data |
---|
277 | * |
---|
278 | * The skipdays element provides a list of days on which this feed should |
---|
279 | * not be checked. We return an array of those days. |
---|
280 | * |
---|
281 | * @return array |
---|
282 | */ |
---|
283 | function getSkipDays() |
---|
284 | { |
---|
285 | return $this->getSkips('skipDays'); |
---|
286 | } |
---|
287 | |
---|
288 | /** |
---|
289 | * Return content of the little-used 'cloud' element |
---|
290 | * |
---|
291 | * The cloud element is rarely used. It is designed to provide some details |
---|
292 | * of a location to update the feed. |
---|
293 | * |
---|
294 | * @return array an array of the attributes of the element |
---|
295 | */ |
---|
296 | function getCloud() |
---|
297 | { |
---|
298 | $cloud = $this->model->getElementsByTagName('cloud'); |
---|
299 | if ($cloud->length == 0) { |
---|
300 | return false; |
---|
301 | } |
---|
302 | $cloudData = array(); |
---|
303 | foreach ($cloud->item(0)->attributes as $attribute) { |
---|
304 | $cloudData[$attribute->name] = $attribute->value; |
---|
305 | } |
---|
306 | return $cloudData; |
---|
307 | } |
---|
308 | |
---|
309 | /** |
---|
310 | * Get link URL |
---|
311 | * |
---|
312 | * In RSS2 a link is a text element but in order to ensure that we resolve |
---|
313 | * URLs properly we have a special function for them. We maintain the |
---|
314 | * parameter used by the atom getLink method, though we only use the offset |
---|
315 | * parameter. |
---|
316 | * |
---|
317 | * @param int $offset The position of the link within the feed. Starts from 0 |
---|
318 | * @param string $attribute The attribute of the link element required |
---|
319 | * @param array $params An array of other parameters. Not used. |
---|
320 | * @return string |
---|
321 | */ |
---|
322 | function getLink($offset, $attribute = 'href', $params = array()) |
---|
323 | { |
---|
324 | $links = $this->model->getElementsByTagName('link'); |
---|
325 | |
---|
326 | if ($links->length <= $offset) { |
---|
327 | return false; |
---|
328 | } |
---|
329 | $link = $links->item($offset); |
---|
330 | return $this->addBase($link->nodeValue, $link); |
---|
331 | } |
---|
332 | } |
---|
333 | |
---|
334 | ?> |
---|