1 | <?php |
||
2 | |||
3 | namespace Backend\Modules\Blog\Actions; |
||
4 | |||
5 | use SimpleXMLElement; |
||
6 | use Symfony\Component\Filesystem\Filesystem; |
||
7 | use Backend\Core\Engine\Base\ActionEdit as BackendBaseActionEdit; |
||
8 | use Backend\Core\Engine\Exception; |
||
9 | use Backend\Core\Engine\Model as BackendModel; |
||
10 | use Backend\Core\Language\Language as BL; |
||
11 | use Backend\Core\Engine\Form; |
||
12 | use Backend\Modules\Blog\Engine\Model; |
||
13 | |||
14 | /** |
||
15 | * This import-action will let you import a wordpress blog |
||
16 | */ |
||
17 | class ImportWordpress extends BackendBaseActionEdit |
||
18 | { |
||
19 | /** |
||
20 | * @var array |
||
21 | */ |
||
22 | private $authors = []; |
||
23 | |||
24 | /** |
||
25 | * @var array |
||
26 | */ |
||
27 | private $attachments = []; |
||
28 | |||
29 | /** |
||
30 | * @var Filesystem |
||
31 | */ |
||
32 | private $filesystem; |
||
33 | |||
34 | public function execute(): void |
||
35 | { |
||
36 | parent::execute(); |
||
37 | set_time_limit(0); |
||
38 | $this->filesystem = new Filesystem(); |
||
39 | $this->loadForm(); |
||
40 | $this->validateForm(); |
||
41 | $this->parse(); |
||
42 | $this->display(); |
||
43 | } |
||
44 | |||
45 | private function loadForm(): void |
||
46 | { |
||
47 | $this->form = new Form('import'); |
||
48 | $this->form->addFile('wordpress'); |
||
49 | $this->form->addText('filter', SITE_URL); |
||
50 | } |
||
51 | |||
52 | private function validateForm(): void |
||
53 | { |
||
54 | // Is the form submitted? |
||
55 | if (!$this->form->isSubmitted()) { |
||
56 | return; |
||
57 | } |
||
58 | |||
59 | // Cleanup the submitted fields, ignore fields that were added by hackers |
||
60 | $this->form->cleanupFields(); |
||
61 | |||
62 | // XML provided? |
||
63 | if ($this->form->getField('wordpress')->isFilled()) { |
||
64 | $this->form->getField('wordpress')->isAllowedExtension(['xml'], BL::err('XMLFilesOnly')); |
||
65 | } else { |
||
66 | // No file |
||
67 | $this->form->getField('wordpress')->addError(BL::err('FieldIsRequired')); |
||
68 | } |
||
69 | |||
70 | if (!$this->form->isCorrect()) { |
||
71 | return; |
||
72 | } |
||
73 | |||
74 | // Move the file |
||
75 | $this->form->getField('wordpress')->moveFile(FRONTEND_FILES_PATH . '/wordpress.xml'); |
||
76 | |||
77 | // Process the XML |
||
78 | $this->processXML(); |
||
79 | |||
80 | // Remove the file |
||
81 | $this->filesystem->remove(FRONTEND_FILES_PATH . '/wordpress.xml'); |
||
82 | |||
83 | // Everything is saved, so redirect to the overview |
||
84 | $this->redirect(BackendModel::createUrlForAction('index') . '&report=imported'); |
||
85 | } |
||
86 | |||
87 | private function processXML(): void |
||
88 | { |
||
89 | $reader = new \XMLReader(); |
||
90 | $reader->open(FRONTEND_FILES_PATH . '/wordpress.xml'); |
||
91 | |||
92 | // Loop through the document |
||
93 | while ($reader->read()) { |
||
94 | // Start tag for item? |
||
95 | if ($reader->name !== 'item' && $reader->name !== 'wp:author') { |
||
96 | continue; |
||
97 | } |
||
98 | |||
99 | // End tag? |
||
100 | if ($reader->nodeType === \XMLReader::END_ELEMENT) { |
||
101 | continue; |
||
102 | } |
||
103 | |||
104 | // Get the raw XML |
||
105 | $xmlString = $reader->readOuterXml(); |
||
106 | |||
107 | // Read the XML as an SimpleXML-object |
||
108 | /* @var SimpleXMLElement $xml */ |
||
109 | $xml = @simplexml_load_string($xmlString); |
||
110 | |||
111 | // Skip element if it isn't a valid SimpleXML-object |
||
112 | if ($xml === false) { |
||
113 | continue; |
||
114 | } |
||
115 | |||
116 | // Is it really an item? |
||
117 | if (mb_substr($xmlString, 0, 5) === '<item') { |
||
118 | // What type of content are we dealing with? |
||
119 | switch ($xml->children('wp', true)->post_type) { |
||
120 | case 'post': |
||
121 | // Process as post |
||
122 | $this->importPost($xml); |
||
123 | break; |
||
124 | |||
125 | case 'attachment': |
||
126 | // Process as attachment |
||
127 | $this->importAttachment($xml); |
||
128 | break; |
||
129 | |||
130 | default: |
||
131 | // Don't do anything |
||
132 | break; |
||
133 | } |
||
134 | } elseif (mb_substr($xmlString, 0, 10) === '<wp:author') { |
||
135 | // Process the authors |
||
136 | $this->authors[(string) $xml->children('wp', true)->author_login] = [ |
||
137 | 'id' => (string) $xml->children('wp', true)->author_id, |
||
138 | 'login' => (string) $xml->children('wp', true)->author_login, |
||
139 | 'email' => (string) $xml->children('wp', true)->author_email, |
||
140 | 'display_name' => (string) $xml->children('wp', true)->author_display_name, |
||
141 | 'first_name' => (string) $xml->children('wp', true)->author_first_name, |
||
142 | 'last_name' => (string) $xml->children('wp', true)->author_last_name, |
||
143 | ]; |
||
144 | } |
||
145 | |||
146 | // End |
||
147 | if (!$reader->read()) { |
||
148 | break; |
||
149 | } |
||
150 | } |
||
151 | |||
152 | // close |
||
153 | $reader->close(); |
||
154 | } |
||
155 | |||
156 | private function importPost(SimpleXMLElement $xml): bool |
||
157 | { |
||
158 | // Are we really working with a post? |
||
159 | if ($xml->children('wp', true)->post_type != 'post') { |
||
0 ignored issues
–
show
introduced
by
Loading history...
|
|||
160 | return false; |
||
161 | } |
||
162 | |||
163 | // This is a deleted post, don't import |
||
164 | if ($xml->children('wp', true)->status === 'trash') { |
||
165 | return false; |
||
166 | } |
||
167 | |||
168 | // Mapping for wordpress status => fork status |
||
169 | $statusses = [ |
||
170 | 'draft' => 'draft', |
||
171 | 'pending' => 'draft', |
||
172 | 'private' => 'private', |
||
173 | 'publish' => 'active', |
||
174 | 'future' => 'publish', |
||
175 | ]; |
||
176 | $commentStatusses = [ |
||
177 | 'open' => true, |
||
178 | 'closed' => false, |
||
179 | ]; |
||
180 | |||
181 | // Prepare item |
||
182 | $item = []; |
||
183 | $item['user_id'] = $this->handleUser((string) $xml->children('dc', true)->creator); |
||
184 | $item['title'] = (string) $xml->title; |
||
185 | $item['text'] = $this->handleUrls( |
||
186 | (string) $xml->children('content', true)->encoded, |
||
187 | $this->form->getField('filter')->getValue() |
||
188 | ); |
||
189 | $item['created_on'] = (string) $xml->children('wp', true)->post_date; |
||
190 | $item['publish_on'] = (string) $xml->children('wp', true)->post_date; |
||
191 | $item['edited_on'] = (string) $xml->children('wp', true)->post_date; |
||
192 | $item['status'] = $statusses[(string) $xml->children('wp', true)->status]; |
||
193 | $item['allow_comments'] = $commentStatusses[(string) $xml->children('wp', true)->comment_status]; |
||
194 | |||
195 | // Some status corrections |
||
196 | if ($item['status'] === 'draft') { |
||
197 | $item['hidden'] = true; |
||
198 | } elseif ($item['status'] === 'private') { |
||
199 | $item['status'] = 'publish'; |
||
200 | $item['hidden'] = true; |
||
201 | } |
||
202 | |||
203 | // Prepare meta |
||
204 | $meta = []; |
||
205 | $meta['url'] = (string) $xml->children('wp', true)->post_name; |
||
206 | |||
207 | // Prepare tags |
||
208 | $tags = []; |
||
209 | |||
210 | // Walk through wp categories |
||
211 | foreach ($xml->category as $category) { |
||
212 | /* @var SimpleXMLElement $category */ |
||
213 | switch ($category->attributes()->domain) { |
||
214 | case 'category': |
||
215 | $item['category_id'] = $this->handleCategory((string) $category); |
||
216 | break; |
||
217 | |||
218 | case 'post_tag': |
||
219 | $tags[] = (string) $category; |
||
220 | break; |
||
221 | |||
222 | default: |
||
223 | // Do nothing |
||
224 | break; |
||
225 | } |
||
226 | } |
||
227 | |||
228 | // Prepare comments |
||
229 | $comments = []; |
||
230 | |||
231 | // Walk through wp comments |
||
232 | foreach ($xml->children('wp', true)->comment as $comment) { |
||
233 | /* @var SimpleXMLElement $comment */ |
||
234 | $comments[] = [ |
||
235 | 'author' => (string) $comment->children('wp', true)->comment_author, |
||
236 | 'email' => (string) $comment->children('wp', true)->comment_author_email, |
||
237 | 'text' => filter_var((string) $comment->children('wp', true)->comment_content, FILTER_SANITIZE_STRING), |
||
238 | 'created_on' => (string) $comment->children('wp', true)->comment_date, |
||
239 | 'status' => ((string) $comment->children('wp', true)->comment_approved == '1') ? 'published' : 'moderation', |
||
240 | ]; |
||
241 | } |
||
242 | |||
243 | // Make the call |
||
244 | Model::insertCompletePost($item, $meta, $tags, $comments); |
||
245 | |||
246 | return true; |
||
247 | } |
||
248 | |||
249 | private function importAttachment(SimpleXMLElement $xml): bool |
||
250 | { |
||
251 | // Are we really working with a post? |
||
252 | if ($xml->children('wp', true)->post_type !== 'attachment') { |
||
253 | return false; |
||
254 | } |
||
255 | |||
256 | // Set paths |
||
257 | $imagesPath = FRONTEND_FILES_PATH . '/Core/CKFinder/images/blog'; |
||
258 | $imagesUrl = FRONTEND_FILES_URL . '/Core/CKFinder/images/blog'; |
||
259 | |||
260 | // Create directory if needed |
||
261 | if (!file_exists($imagesPath) || !is_dir($imagesPath)) { |
||
262 | $this->filesystem->mkdir($imagesPath); |
||
263 | } |
||
264 | |||
265 | $file = (string) $xml->children('wp', true)->attachment_url; |
||
266 | $guid = (string) $xml->guid; |
||
267 | $fileId = (string) $xml->children('wp', true)->post_id; |
||
268 | |||
269 | // Set filename |
||
270 | $destinationFile = $fileId . '_' . basename($file); |
||
271 | |||
272 | // Download the file |
||
273 | try { |
||
274 | $this->filesystem->dumpFile( |
||
275 | $imagesPath . '/' . $destinationFile, |
||
276 | file_get_contents($file) |
||
277 | ); |
||
278 | } catch (Exception $e) { |
||
279 | // Ignore |
||
280 | } |
||
281 | |||
282 | // Keep a log of downloaded files |
||
283 | $this->attachments[mb_strtolower($file)] = $imagesUrl . '/' . $destinationFile; |
||
284 | $this->attachments[mb_strtolower($guid)] = $imagesUrl . '/' . $destinationFile; |
||
285 | |||
286 | return true; |
||
287 | } |
||
288 | |||
289 | /** |
||
290 | * Handle the user of a post |
||
291 | * |
||
292 | * We'll try and match the original user with a fork user. |
||
293 | * If we find no matches, we'll assign to the main fork user. |
||
294 | * |
||
295 | * @param string $username The original user name |
||
296 | * |
||
297 | * @return int |
||
298 | */ |
||
299 | private function handleUser(string $username = ''): int |
||
300 | { |
||
301 | // Does someone with this username exist? |
||
302 | /* @var \SpoonDatabase $database */ |
||
303 | $database = BackendModel::getContainer()->get('database'); |
||
304 | $id = (int) $database->getVar( |
||
305 | 'SELECT id FROM users WHERE email=? AND active=? AND deleted=?', |
||
306 | [mb_strtolower($this->authors[(string) $username]['email']), true, false] |
||
307 | ); |
||
308 | |||
309 | // We found an id! |
||
310 | if ($id > 0) { |
||
311 | return $id; |
||
312 | } |
||
313 | |||
314 | // Assign to main user |
||
315 | return 1; |
||
316 | } |
||
317 | |||
318 | /** |
||
319 | * Handle the urls inside a post |
||
320 | * |
||
321 | * We'll try and download images, and replace their urls |
||
322 | * We'll also check for links to schrijf.be and try to replace them |
||
323 | * |
||
324 | * @param string $text The post text |
||
325 | * @param string $filter The text that needs to be in a url before we start replacing it. |
||
326 | * |
||
327 | * @return string |
||
328 | */ |
||
329 | private function handleUrls(string $text, string $filter = ''): string |
||
330 | { |
||
331 | // Check for images and download them, replace urls |
||
332 | preg_match_all('/<img.*src="(.*)".*\/>/Ui', $text, $matchesImages); |
||
333 | |||
334 | if (isset($matchesImages[1]) && !empty($matchesImages[1])) { |
||
335 | // Walk through image links |
||
336 | foreach ($matchesImages[1] as $key => $file) { |
||
337 | // Should we bother looking at this file? |
||
338 | if (!empty($filter) && !mb_stristr($file, $filter)) { |
||
339 | continue; |
||
340 | } |
||
341 | |||
342 | $noSize = preg_replace('/\-\d+x\d+/i', '', $file); |
||
343 | |||
344 | if (isset($this->attachments[mb_strtolower($file)])) { |
||
345 | $text = str_replace($file, $this->attachments[mb_strtolower($file)], $text); |
||
346 | } elseif (isset($this->attachments[mb_strtolower($noSize)])) { |
||
347 | $text = str_replace($file, $this->attachments[mb_strtolower($noSize)], $text); |
||
348 | } |
||
349 | } |
||
350 | } |
||
351 | |||
352 | // Check for links to schrijf.be and try to replace them |
||
353 | preg_match_all('/<a.*href="(.*)".*\/>/Ui', $text, $matchesLinks); |
||
354 | |||
355 | if (isset($matchesLinks[1]) && !empty($matchesLinks[1])) { |
||
356 | // Walk through links |
||
357 | foreach ($matchesLinks[1] as $key => $link) { |
||
358 | // Should we bother looking at this file? |
||
359 | if (!empty($filter) && !mb_stristr($link, $filter)) { |
||
360 | continue; |
||
361 | } |
||
362 | |||
363 | $noSize = preg_replace('/\-\d+x\d+/i', '', $link); |
||
364 | |||
365 | if (isset($this->attachments[mb_strtolower($link)])) { |
||
366 | $text = str_replace($link, $this->attachments[mb_strtolower($link)], $text); |
||
367 | } elseif (isset($this->attachments[mb_strtolower($noSize)])) { |
||
368 | $text = str_replace($link, $this->attachments[mb_strtolower($noSize)], $text); |
||
369 | } |
||
370 | } |
||
371 | } |
||
372 | |||
373 | return $text; |
||
374 | } |
||
375 | |||
376 | /** |
||
377 | * Handle the category of a post |
||
378 | * |
||
379 | * We'll check if the category exists in the fork blog module, and create it if it doesn't. |
||
380 | * |
||
381 | * @param string $category The post category |
||
382 | * |
||
383 | * @return int |
||
384 | */ |
||
385 | private function handleCategory(string $category = ''): int |
||
386 | { |
||
387 | // Does a category with this name exist? |
||
388 | /* @var \SpoonDatabase $database */ |
||
389 | $database = BackendModel::getContainer()->get('database'); |
||
390 | $id = (int) $database->getVar( |
||
391 | 'SELECT id FROM blog_categories WHERE title=? AND language=?', |
||
392 | [$category, BL::getWorkingLanguage()] |
||
393 | ); |
||
394 | |||
395 | // We found an id! |
||
396 | if ($id > 0) { |
||
397 | return $id; |
||
398 | } |
||
399 | |||
400 | // Return default if we got an empty string |
||
401 | if (trim($category) === '') { |
||
402 | return 2; |
||
403 | } |
||
404 | |||
405 | // We should create a new category |
||
406 | $cat = []; |
||
407 | $cat['language'] = BL::getWorkingLanguage(); |
||
408 | $cat['title'] = $category; |
||
409 | $meta = []; |
||
410 | $meta['keywords'] = $category; |
||
411 | $meta['description'] = $category; |
||
412 | $meta['title'] = $category; |
||
413 | $meta['url'] = $category; |
||
414 | |||
415 | return Model::insertCategory($cat, $meta); |
||
416 | } |
||
417 | } |
||
418 |