1 | <?php |
---|
2 | |
---|
3 | // Batch stages. |
---|
4 | define('FEEDS_FETCHING', 'fetching'); |
---|
5 | define('FEEDS_PARSING', 'parsing'); |
---|
6 | define('FEEDS_PROCESSING', 'processing'); |
---|
7 | define('FEEDS_CLEARING', 'clearing'); |
---|
8 | |
---|
9 | /** |
---|
10 | * A FeedsBatch object holds the state of an import or clear batch. |
---|
11 | * |
---|
12 | * Used in FeedsSource class. Counter variables are public for easier access. |
---|
13 | */ |
---|
14 | class FeedsBatch { |
---|
15 | // Total of each stage of this batch. |
---|
16 | protected $total; |
---|
17 | // Progress of each stage of this batch. |
---|
18 | protected $progress; |
---|
19 | public function __construct() { |
---|
20 | $this->total = array(); |
---|
21 | $this->progress = array(); |
---|
22 | } |
---|
23 | |
---|
24 | /** |
---|
25 | * Set the total for a stage. |
---|
26 | */ |
---|
27 | public function setTotal($stage, $total) { |
---|
28 | $this->total[$stage] = $total; |
---|
29 | } |
---|
30 | |
---|
31 | /** |
---|
32 | * Get the total for a stage. |
---|
33 | */ |
---|
34 | public function getTotal($stage) { |
---|
35 | return $this->total[$stage]; |
---|
36 | } |
---|
37 | |
---|
38 | /** |
---|
39 | * Set progress for a stage. |
---|
40 | * |
---|
41 | * @param $stage |
---|
42 | * The stage to set the progress for. One of FEEDS_FETCHING, FEEDS_PARSING, |
---|
43 | * FEEDS_PROCESING or FEEDS_CLEARING. |
---|
44 | * @param $progress |
---|
45 | * The number of items worked off for the given stage. This should be the |
---|
46 | * number of items worked off across all page loads, not just the present |
---|
47 | * page load. |
---|
48 | */ |
---|
49 | public function setProgress($stage, $progress) { |
---|
50 | $this->progress[$stage] = $progress; |
---|
51 | } |
---|
52 | |
---|
53 | /** |
---|
54 | * Report progress. |
---|
55 | * |
---|
56 | * @param $stage |
---|
57 | * The stage to set the progress for. One of FEEDS_FETCHING, FEEDS_PARSING, |
---|
58 | * FEEDS_PROCESING or FEEDS_CLEARING. |
---|
59 | */ |
---|
60 | public function getProgress($stage = NULL) { |
---|
61 | if ($stage) { |
---|
62 | $progress = $this->progress[$stage]; |
---|
63 | if ($progress == FEEDS_BATCH_COMPLETE) { |
---|
64 | return FEEDS_BATCH_COMPLETE; |
---|
65 | } |
---|
66 | $total = $this->total[$stage]; |
---|
67 | } |
---|
68 | else { |
---|
69 | $complete = TRUE; |
---|
70 | $progress = 0; |
---|
71 | foreach ($this->progress as $p) { |
---|
72 | $progress += $p; |
---|
73 | $complete &= $p == FEEDS_BATCH_COMPLETE; |
---|
74 | } |
---|
75 | if ($complete) { |
---|
76 | return FEEDS_BATCH_COMPLETE; |
---|
77 | } |
---|
78 | $total = array_sum($this->total); |
---|
79 | } |
---|
80 | $progress = (1.0 / $total) * $progress; |
---|
81 | return $progress >= FEEDS_BATCH_COMPLETE ? 0.999 : $progress; |
---|
82 | } |
---|
83 | } |
---|
84 | |
---|
85 | /** |
---|
86 | * A FeedsImportBatch wraps the actual content retrieved from a FeedsSource. On |
---|
87 | * import, it is created on the fetching stage and passed through the parsing |
---|
88 | * and processing stage where it is normalized and consumed. |
---|
89 | * |
---|
90 | * A Fetcher must return a FeedsImportBatch object on fetch(). To that end it |
---|
91 | * can use one of the existing FeedsImportBatch classes (FeedsImportBatch, |
---|
92 | * FeedsFileBatch or FeedsHTTPBatch) or provide its own as a direct or indirect |
---|
93 | * extension of FeedsImportBatch. |
---|
94 | * |
---|
95 | * A Parser must populate a FeedsImportBatch object upon parse(). For instance: |
---|
96 | * |
---|
97 | * @code |
---|
98 | * $batch->items = $parsed_rows; |
---|
99 | * $batch->title = 'My imported document'; |
---|
100 | * @endcode |
---|
101 | * |
---|
102 | * Finally, a processor can work off the information produced on the parsing |
---|
103 | * stage by consuming items with $batch->shiftItem(). |
---|
104 | * |
---|
105 | * @code |
---|
106 | * while ($item = $batch->shiftItem()) { |
---|
107 | * $object = $this->map($item); |
---|
108 | * $object->save(); |
---|
109 | * } |
---|
110 | * @endcode |
---|
111 | * |
---|
112 | * If a processing task is very slow, it can be batched over multiple page |
---|
113 | * loads. For batching the consumer loop can be left while the current progress |
---|
114 | * is set on the batch object. If the current progress is not |
---|
115 | * FEEDS_BATCH_COMPLETE the processor will be called again on a subsequent page |
---|
116 | * load to continue where it has left off. For an example, see |
---|
117 | * FeedsNodeProcessor::process(). |
---|
118 | * |
---|
119 | * @code |
---|
120 | * $created = 0; |
---|
121 | * while ($item = $batch->shiftItem()) { |
---|
122 | * $object = $this->map($item); |
---|
123 | * $object->save(); |
---|
124 | * $created++; // Created in this page load. |
---|
125 | * $batch->created++; // Created total. |
---|
126 | * if ($created > MAX_CREATED) { |
---|
127 | * $batch->setProgress(FEEDS_PROCESSING, $batch->created); |
---|
128 | * return; |
---|
129 | * } |
---|
130 | * } |
---|
131 | * $batch->setProgress(FEEDS_PROCESSING, FEEDS_BATCH_COMPLETE); |
---|
132 | * @endcode |
---|
133 | * |
---|
134 | * Note: Knowledge of the internal structure of a single item in the $items |
---|
135 | * array is managed by the mapping API specified in FeedsParser class and |
---|
136 | * FeedsProcessor class. |
---|
137 | * |
---|
138 | * @see FeedsBatch |
---|
139 | * @see FeedsFileBatch |
---|
140 | * @see FeedsHTTPBatch |
---|
141 | */ |
---|
142 | class FeedsImportBatch extends FeedsBatch { |
---|
143 | protected $current_item; |
---|
144 | protected $raw; |
---|
145 | public $title; |
---|
146 | public $description; |
---|
147 | public $link; |
---|
148 | public $items; |
---|
149 | public $feed_nid; |
---|
150 | public $created; |
---|
151 | public $updated; |
---|
152 | |
---|
153 | public function __construct($raw = '', $feed_nid = 0) { |
---|
154 | parent::__construct(); |
---|
155 | $this->progress = array( |
---|
156 | FEEDS_FETCHING => FEEDS_BATCH_COMPLETE, |
---|
157 | FEEDS_PARSING => FEEDS_BATCH_COMPLETE, |
---|
158 | FEEDS_PROCESSING => FEEDS_BATCH_COMPLETE, |
---|
159 | ); |
---|
160 | $this->total = array( |
---|
161 | FEEDS_FETCHING => 0, |
---|
162 | FEEDS_PARSING => 0, |
---|
163 | FEEDS_PROCESSING => 0, |
---|
164 | ); |
---|
165 | $this->title = ''; |
---|
166 | $this->description = ''; |
---|
167 | $this->link = ''; |
---|
168 | $this->items = array(); |
---|
169 | $this->raw = $raw; |
---|
170 | $this->feed_nid = $feed_nid; |
---|
171 | $this->created = 0; |
---|
172 | $this->updated = 0; |
---|
173 | } |
---|
174 | |
---|
175 | /** |
---|
176 | * @return |
---|
177 | * The raw content from the source as a string. |
---|
178 | * |
---|
179 | * @throws Exception |
---|
180 | * Extending classes MAY throw an exception if a problem occurred. |
---|
181 | */ |
---|
182 | public function getRaw() { |
---|
183 | return $this->raw; |
---|
184 | } |
---|
185 | |
---|
186 | /** |
---|
187 | * @return |
---|
188 | * A path to a file containing the raw content as a source. |
---|
189 | * |
---|
190 | * @throws Exception |
---|
191 | * If an unexpected problem occurred. |
---|
192 | */ |
---|
193 | public function getFilePath() { |
---|
194 | if (!isset($this->file_path)) { |
---|
195 | $dir = file_directory_path() .'/feeds'; |
---|
196 | if (!file_check_directory($dir, TRUE)) { |
---|
197 | throw new Exception(t('Feeds directory either cannot be created or is not writable.')); |
---|
198 | } |
---|
199 | $dest = file_destination($dir . '/' . get_class($this) .'_'. drupal_get_token($this->url) .'_'. time(), FILE_EXISTS_RENAME); |
---|
200 | $this->file_path = file_save_data($this->getRaw(), $dest); |
---|
201 | if ($this->file_path === 0) { |
---|
202 | throw new Exception(t('Cannot write content to %dest', array('%dest' => $dest))); |
---|
203 | } |
---|
204 | } |
---|
205 | return $this->file_path; |
---|
206 | } |
---|
207 | |
---|
208 | /** |
---|
209 | * @deprecated |
---|
210 | * This method is deprecated. Access directly instead. For example: |
---|
211 | * $node = node_load($batch->feed_nid); |
---|
212 | * |
---|
213 | * Return the feed node related to this batch object. |
---|
214 | */ |
---|
215 | public function feedNode() { |
---|
216 | if ($this->feed_nid) { |
---|
217 | return node_load($this->feed_nid); |
---|
218 | } |
---|
219 | } |
---|
220 | |
---|
221 | /** |
---|
222 | * @deprecated |
---|
223 | * This method is deprecated. Access directly instead. For example: |
---|
224 | * $title = $batch->title; |
---|
225 | * |
---|
226 | * @return |
---|
227 | * A string that is the feed's title. |
---|
228 | */ |
---|
229 | public function getTitle() { |
---|
230 | return $this->title; |
---|
231 | } |
---|
232 | |
---|
233 | /** |
---|
234 | * @deprecated |
---|
235 | * This method is deprecated. Access directly instead. For example: |
---|
236 | * $description = $batch->description; |
---|
237 | * |
---|
238 | * @return |
---|
239 | * A string that is the feed's description. |
---|
240 | */ |
---|
241 | public function getDescription() { |
---|
242 | return $this->description; |
---|
243 | } |
---|
244 | |
---|
245 | /** |
---|
246 | * @deprecated |
---|
247 | * This method is deprecated. Access directly instead. For example: |
---|
248 | * $link = $batch->link; |
---|
249 | * |
---|
250 | * @return |
---|
251 | * A string that is the link to the feed's site (not the actual URL of the |
---|
252 | * feed). Falls back to URL if not available. |
---|
253 | */ |
---|
254 | public function getLink() { |
---|
255 | return $this->link; |
---|
256 | } |
---|
257 | |
---|
258 | /** |
---|
259 | * @todo Move to a nextItem() based approach, not consuming the item array. |
---|
260 | * Can only be done once we don't cache the entire batch object between page |
---|
261 | * loads for batching anymore. |
---|
262 | * |
---|
263 | * @return |
---|
264 | * Next available item or NULL if there is none. Every returned item is |
---|
265 | * removed from the internal array. |
---|
266 | */ |
---|
267 | public function shiftItem() { |
---|
268 | $this->current_item = array_shift($this->items); |
---|
269 | return $this->current_item; |
---|
270 | } |
---|
271 | |
---|
272 | /** |
---|
273 | * @return |
---|
274 | * Current feed item. |
---|
275 | */ |
---|
276 | public function currentItem() { |
---|
277 | return empty($this->current_item) ? NULL : $this->current_item; |
---|
278 | } |
---|
279 | |
---|
280 | /** |
---|
281 | * Set title. |
---|
282 | * |
---|
283 | * @deprecated |
---|
284 | * This method is deprecated. Access directly instead. For example: |
---|
285 | * $batch->title = $title; |
---|
286 | */ |
---|
287 | public function setTitle($title) { |
---|
288 | $this->title = $title; |
---|
289 | } |
---|
290 | |
---|
291 | /** |
---|
292 | * Set description. |
---|
293 | * |
---|
294 | * @deprecated |
---|
295 | * This method is deprecated. Access directly instead. For example: |
---|
296 | * $batch->description = $description; |
---|
297 | */ |
---|
298 | public function setDescription($description) { |
---|
299 | $this->description = $description; |
---|
300 | } |
---|
301 | |
---|
302 | /** |
---|
303 | * Set link. |
---|
304 | * |
---|
305 | * @deprecated |
---|
306 | * This method is deprecated. Access directly instead. For example: |
---|
307 | * $batch->link = $link; |
---|
308 | */ |
---|
309 | public function setLink($link) { |
---|
310 | $this->link = $link; |
---|
311 | } |
---|
312 | |
---|
313 | /** |
---|
314 | * Set items. |
---|
315 | * |
---|
316 | * @deprecated |
---|
317 | * This method is deprecated. Access directly instead. For example: |
---|
318 | * $batch->items = $items; |
---|
319 | * |
---|
320 | * @param $items |
---|
321 | * An array of the items in the feed. Cannot be NULL. |
---|
322 | */ |
---|
323 | public function setItems($items) { |
---|
324 | $this->items = $items; |
---|
325 | } |
---|
326 | |
---|
327 | /** |
---|
328 | * Add an item. |
---|
329 | * |
---|
330 | * @deprecated |
---|
331 | * This method is deprecated. Access directly instead. For example: |
---|
332 | * $batch->items[] = $item; |
---|
333 | */ |
---|
334 | public function addItem($item) { |
---|
335 | $this->items[] = $item; |
---|
336 | } |
---|
337 | |
---|
338 | /** |
---|
339 | * Get number of items. |
---|
340 | * |
---|
341 | * @deprecated |
---|
342 | * This method is deprecated. Access directly instead. For example: |
---|
343 | * $count = count($batch->items); |
---|
344 | */ |
---|
345 | public function getItemCount() { |
---|
346 | return count($this->items); |
---|
347 | } |
---|
348 | } |
---|
349 | |
---|
350 | /** |
---|
351 | * Batch class for batched deleting of items. |
---|
352 | */ |
---|
353 | class FeedsClearBatch extends FeedsBatch { |
---|
354 | // Number of items deleted. |
---|
355 | public $deleted; |
---|
356 | public function __construct() { |
---|
357 | parent::__construct(); |
---|
358 | $this->progress = array( |
---|
359 | FEEDS_CLEARING => FEEDS_BATCH_COMPLETE, |
---|
360 | ); |
---|
361 | $this->total = array( |
---|
362 | FEEDS_CLEARING => 0, |
---|
363 | ); |
---|
364 | $this->deleted = 0; |
---|
365 | } |
---|
366 | } |
---|