|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* @license LGPLv3, http://opensource.org/licenses/LGPL-3.0 |
|
5
|
|
|
* @copyright Aimeos (aimeos.org), 2019 |
|
6
|
|
|
* @package Controller |
|
7
|
|
|
* @subpackage Jobs |
|
8
|
|
|
*/ |
|
9
|
|
|
|
|
10
|
|
|
|
|
11
|
|
|
namespace Aimeos\Controller\Jobs\Product\Import\Xml; |
|
12
|
|
|
|
|
13
|
|
|
|
|
14
|
|
|
/** |
|
15
|
|
|
* Job controller for XML product imports |
|
16
|
|
|
* |
|
17
|
|
|
* @package Controller |
|
18
|
|
|
* @subpackage Jobs |
|
19
|
|
|
*/ |
|
20
|
|
|
class Standard |
|
21
|
|
|
extends \Aimeos\Controller\Jobs\Base |
|
22
|
|
|
implements \Aimeos\Controller\Jobs\Iface |
|
23
|
|
|
{ |
|
24
|
|
|
use \Aimeos\Controller\Common\Common\Import\Xml\Traits; |
|
25
|
|
|
|
|
26
|
|
|
|
|
27
|
|
|
/** |
|
28
|
|
|
* Returns the localized name of the job. |
|
29
|
|
|
* |
|
30
|
|
|
* @return string Name of the job |
|
31
|
|
|
*/ |
|
32
|
|
|
public function getName() |
|
33
|
|
|
{ |
|
34
|
|
|
return $this->getContext()->getI18n()->dt( 'controller/jobs', 'Product import XML' ); |
|
35
|
|
|
} |
|
36
|
|
|
|
|
37
|
|
|
|
|
38
|
|
|
/** |
|
39
|
|
|
* Returns the localized description of the job. |
|
40
|
|
|
* |
|
41
|
|
|
* @return string Description of the job |
|
42
|
|
|
*/ |
|
43
|
|
|
public function getDescription() |
|
44
|
|
|
{ |
|
45
|
|
|
return $this->getContext()->getI18n()->dt( 'controller/jobs', 'Imports new and updates existing products from XML files' ); |
|
46
|
|
|
} |
|
47
|
|
|
|
|
48
|
|
|
|
|
49
|
|
|
/** |
|
50
|
|
|
* Executes the job. |
|
51
|
|
|
* |
|
52
|
|
|
* @throws \Aimeos\Controller\Jobs\Exception If an error occurs |
|
53
|
|
|
*/ |
|
54
|
|
|
public function run() |
|
55
|
|
|
{ |
|
56
|
|
|
$context = $this->getContext(); |
|
57
|
|
|
$config = $context->getConfig(); |
|
58
|
|
|
$logger = $context->getLogger(); |
|
59
|
|
|
|
|
60
|
|
|
/** controller/jobs/product/import/xml/location |
|
61
|
|
|
* File or directory where the content is stored which should be imported |
|
62
|
|
|
* |
|
63
|
|
|
* You need to configure the XML file or directory with the XML files that |
|
64
|
|
|
* should be imported. It should be an absolute path to be sure but can be |
|
65
|
|
|
* relative path if you absolutely know from where the job will be executed |
|
66
|
|
|
* from. |
|
67
|
|
|
* |
|
68
|
|
|
* @param string Absolute file or directory path |
|
69
|
|
|
* @since 2019.04 |
|
70
|
|
|
* @category Developer |
|
71
|
|
|
* @category User |
|
72
|
|
|
* @see controller/jobs/product/import/xml/container/type |
|
73
|
|
|
* @see controller/jobs/product/import/xml/container/content |
|
74
|
|
|
* @see controller/jobs/product/import/xml/container/options |
|
75
|
|
|
*/ |
|
76
|
|
|
$location = $config->get( 'controller/jobs/product/import/xml/location' ); |
|
77
|
|
|
|
|
78
|
|
|
try |
|
79
|
|
|
{ |
|
80
|
|
|
$msg = sprintf( 'Started product import from "%1$s" (%2$s)', $location, __CLASS__ ); |
|
81
|
|
|
$logger->log( $msg, \Aimeos\MW\Logger\Base::INFO ); |
|
82
|
|
|
|
|
83
|
|
|
if( !file_exists( $location ) ) |
|
84
|
|
|
{ |
|
85
|
|
|
$msg = sprintf( 'File or directory "%1$s" doesn\'t exist', $location ); |
|
86
|
|
|
throw new \Aimeos\Controller\Jobs\Exception( $msg ); |
|
87
|
|
|
} |
|
88
|
|
|
|
|
89
|
|
|
$files = []; |
|
90
|
|
|
|
|
91
|
|
|
if( is_dir( $location ) ) |
|
92
|
|
|
{ |
|
93
|
|
|
foreach( new \DirectoryIterator( $location ) as $entry ) |
|
94
|
|
|
{ |
|
95
|
|
|
if( strncmp( $entry->getFilename(), 'product', 7 ) === 0 && $entry->getExtension() === 'xml' ) { |
|
96
|
|
|
$files[] = $entry->getPathname(); |
|
97
|
|
|
} |
|
98
|
|
|
} |
|
99
|
|
|
} |
|
100
|
|
|
else |
|
101
|
|
|
{ |
|
102
|
|
|
$files[] = $location; |
|
103
|
|
|
} |
|
104
|
|
|
|
|
105
|
|
|
sort( $files ); |
|
106
|
|
|
$total = 0; |
|
107
|
|
|
|
|
108
|
|
|
foreach( $files as $filepath ) { |
|
109
|
|
|
$total += $this->import( $filepath ); |
|
110
|
|
|
} |
|
111
|
|
|
|
|
112
|
|
|
$msg = 'Finished product import from "%1$s": %2$s total (%3$s MB)'; |
|
113
|
|
|
$mem = number_format( memory_get_peak_usage() / 1024 / 1024, 2 ); |
|
114
|
|
|
|
|
115
|
|
|
$logger->log( sprintf( $msg, $location, $total, $mem ), \Aimeos\MW\Logger\Base::INFO ); |
|
116
|
|
|
} |
|
117
|
|
|
catch( \Exception $e ) |
|
118
|
|
|
{ |
|
119
|
|
|
$logger->log( 'Product import error: ' . $e->getMessage() . "\n" . $e->getTraceAsString() ); |
|
120
|
|
|
throw $e; |
|
121
|
|
|
} |
|
122
|
|
|
} |
|
123
|
|
|
|
|
124
|
|
|
|
|
125
|
|
|
/** |
|
126
|
|
|
* Imports the XML file given by its path |
|
127
|
|
|
* |
|
128
|
|
|
* @param string $filename Absolute or relative path to the XML file |
|
129
|
|
|
* @return integer Total number of imported products |
|
130
|
|
|
*/ |
|
131
|
|
|
protected function import( $filename ) |
|
132
|
|
|
{ |
|
133
|
|
|
$context = $this->getContext(); |
|
134
|
|
|
$config = $context->getConfig(); |
|
135
|
|
|
$domains = ['attribute', 'media', 'price', 'product', 'product/property', 'text']; |
|
136
|
|
|
|
|
137
|
|
|
/** controller/jobs/product/import/xml/domains |
|
138
|
|
|
* List of item domain names that should be retrieved along with the product items |
|
139
|
|
|
* |
|
140
|
|
|
* This configuration setting overwrites the shared option |
|
141
|
|
|
* "controller/common/product/import/xml/domains" if you need a |
|
142
|
|
|
* specific setting for the job controller. Otherwise, you should |
|
143
|
|
|
* use the shared option for consistency. |
|
144
|
|
|
* |
|
145
|
|
|
* @param array Associative list of MShop item domain names |
|
146
|
|
|
* @since 2019.04 |
|
147
|
|
|
* @category Developer |
|
148
|
|
|
* @see controller/jobs/product/import/xml/backup |
|
149
|
|
|
* @see controller/jobs/product/import/xml/max-query |
|
150
|
|
|
*/ |
|
151
|
|
|
$domains = $config->get( 'controller/jobs/product/import/xml/domains', $domains ); |
|
152
|
|
|
|
|
153
|
|
|
/** controller/jobs/product/import/xml/backup |
|
154
|
|
|
* Name of the backup for sucessfully imported files |
|
155
|
|
|
* |
|
156
|
|
|
* After a XML file was imported successfully, you can move it to another |
|
157
|
|
|
* location, so it won't be imported again and isn't overwritten by the |
|
158
|
|
|
* next file that is stored at the same location in the file system. |
|
159
|
|
|
* |
|
160
|
|
|
* You should use an absolute path to be sure but can be relative path |
|
161
|
|
|
* if you absolutely know from where the job will be executed from. The |
|
162
|
|
|
* name of the new backup location can contain placeholders understood |
|
163
|
|
|
* by the PHP strftime() function to create dynamic paths, e.g. "backup/%Y-%m-%d" |
|
164
|
|
|
* which would create "backup/2000-01-01". For more information about the |
|
165
|
|
|
* strftime() placeholders, please have a look into the PHP documentation of |
|
166
|
|
|
* the {@link http://php.net/manual/en/function.strftime.php strftime() function}. |
|
167
|
|
|
* |
|
168
|
|
|
* '''Note:''' If no backup name is configured, the file or directory |
|
169
|
|
|
* won't be moved away. Please make also sure that the parent directory |
|
170
|
|
|
* and the new directory are writable so the file or directory could be |
|
171
|
|
|
* moved. |
|
172
|
|
|
* |
|
173
|
|
|
* @param integer Name of the backup file, optionally with date/time placeholders |
|
174
|
|
|
* @since 2019.04 |
|
175
|
|
|
* @category Developer |
|
176
|
|
|
* @see controller/jobs/product/import/xml/domains |
|
177
|
|
|
* @see controller/jobs/product/import/xml/max-query |
|
178
|
|
|
*/ |
|
179
|
|
|
$backup = $config->get( 'controller/jobs/product/import/xml/backup' ); |
|
180
|
|
|
|
|
181
|
|
|
/** controller/jobs/product/import/xml/max-query |
|
182
|
|
|
* Maximum number of XML nodes processed at once |
|
183
|
|
|
* |
|
184
|
|
|
* Processing and fetching several product items at once speeds up importing |
|
185
|
|
|
* the XML files. The more items can be processed at once, the faster the |
|
186
|
|
|
* import. More items also increases the memory usage of the importer and |
|
187
|
|
|
* thus, this parameter should be low enough to avoid reaching the memory |
|
188
|
|
|
* limit of the PHP process. |
|
189
|
|
|
* |
|
190
|
|
|
* @param integer Number of XML nodes |
|
191
|
|
|
* @since 2019.04 |
|
192
|
|
|
* @category Developer |
|
193
|
|
|
* @category User |
|
194
|
|
|
* @see controller/jobs/product/import/xml/domains |
|
195
|
|
|
* @see controller/jobs/product/import/xml/backup |
|
196
|
|
|
*/ |
|
197
|
|
|
$maxquery = $config->get( 'controller/jobs/product/import/xml/max-query', 1000 ); |
|
198
|
|
|
|
|
199
|
|
|
|
|
200
|
|
|
$nodes = []; |
|
201
|
|
|
$total = $slice = 0; |
|
202
|
|
|
$xml = new \XMLReader(); |
|
203
|
|
|
|
|
204
|
|
|
if( $xml->open( $filename, LIBXML_COMPACT | LIBXML_PARSEHUGE ) === false ) { |
|
205
|
|
|
throw new \Aimeos\Controller\Jobs\Exception( sprintf( 'No XML file "%1$s" found', $filename ) ); |
|
206
|
|
|
} |
|
207
|
|
|
|
|
208
|
|
|
while( $xml->read() === true ) |
|
209
|
|
|
{ |
|
210
|
|
|
if( $xml->depth === 1 && $xml->nodeType === \XMLReader::ELEMENT && $xml->name === 'productitem' ) |
|
211
|
|
|
{ |
|
212
|
|
|
if( ( $dom = $xml->expand() ) === false ) |
|
213
|
|
|
{ |
|
214
|
|
|
$msg = sprintf( 'Expanding "%1$s" node failed', 'productitem' ); |
|
215
|
|
|
throw new \Aimeos\Controller\Jobs\Exception( $msg ); |
|
216
|
|
|
} |
|
217
|
|
|
|
|
218
|
|
|
$nodes[] = $dom; |
|
219
|
|
|
|
|
220
|
|
|
if( $slice++ >= $maxquery ) |
|
221
|
|
|
{ |
|
222
|
|
|
$this->importNodes( $nodes, $domains ); |
|
223
|
|
|
unset( $nodes ); |
|
224
|
|
|
$nodes = []; |
|
225
|
|
|
$slice = 0; |
|
226
|
|
|
} |
|
227
|
|
|
|
|
228
|
|
|
$total++; |
|
229
|
|
|
} |
|
230
|
|
|
} |
|
231
|
|
|
|
|
232
|
|
|
$this->importNodes( $nodes, $domains ); |
|
233
|
|
|
unset( $nodes ); |
|
234
|
|
|
|
|
235
|
|
|
if( !empty( $backup ) && @rename( $filename, strftime( $backup ) ) === false ) |
|
236
|
|
|
{ |
|
237
|
|
|
$msg = sprintf( 'Unable to move imported file "%1$s" to "%2$s"', $filename, $backup ); |
|
238
|
|
|
throw new \Aimeos\Controller\Jobs\Exception( $msg ); |
|
239
|
|
|
} |
|
240
|
|
|
|
|
241
|
|
|
return $total; |
|
242
|
|
|
} |
|
243
|
|
|
|
|
244
|
|
|
|
|
245
|
|
|
/** |
|
246
|
|
|
* Imports the given DOM nodes |
|
247
|
|
|
* |
|
248
|
|
|
* @param \DomElement[] $nodes List of nodes to import |
|
249
|
|
|
* @param string[] $ref List of domain names whose referenced items will be updated in the product items |
|
250
|
|
|
*/ |
|
251
|
|
|
protected function importNodes( array $nodes, array $ref ) |
|
252
|
|
|
{ |
|
253
|
|
|
$codes = $map = []; |
|
254
|
|
|
|
|
255
|
|
|
foreach( $nodes as $node ) |
|
256
|
|
|
{ |
|
257
|
|
|
if( ( $attr = $node->attributes->getNamedItem( 'ref' ) ) !== null ) { |
|
258
|
|
|
$codes[$attr->nodeValue] = null; |
|
259
|
|
|
} |
|
260
|
|
|
} |
|
261
|
|
|
|
|
262
|
|
|
$manager = \Aimeos\MShop::create( $this->getContext(), 'product' ); |
|
263
|
|
|
$search = $manager->createSearch()->setSlice( 0, count( $codes ) ); |
|
264
|
|
|
$search->setConditions( $search->compare( '==', 'product.code', array_keys( $codes ) ) ); |
|
265
|
|
|
|
|
266
|
|
|
foreach( $manager->searchItems( $search, $ref ) as $item ) { |
|
267
|
|
|
$map[$item->getCode()] = $item; |
|
268
|
|
|
} |
|
269
|
|
|
|
|
270
|
|
|
foreach( $nodes as $node ) |
|
271
|
|
|
{ |
|
272
|
|
|
if( ( $attr = $node->attributes->getNamedItem( 'ref' ) ) !== null && isset( $map[$attr->nodeValue] ) ) { |
|
273
|
|
|
$item = $this->process( $map[$attr->nodeValue], $node ); |
|
274
|
|
|
} else { |
|
275
|
|
|
$item = $this->process( $manager->createItem(), $node ); |
|
|
|
|
|
|
276
|
|
|
} |
|
277
|
|
|
|
|
278
|
|
|
$manager->saveItem( $item ); |
|
279
|
|
|
} |
|
280
|
|
|
} |
|
281
|
|
|
|
|
282
|
|
|
|
|
283
|
|
|
/** |
|
284
|
|
|
* Updates the product item and its referenced items using the given DOM node |
|
285
|
|
|
* |
|
286
|
|
|
* @param \Aimeos\MShop\Product\Item\Iface $item Product item object to update |
|
287
|
|
|
* @param \DomElement $node DOM node used for updateding the product item |
|
288
|
|
|
* @return \Aimeos\MShop\Product\Item\Iface $item Updated product item object |
|
289
|
|
|
*/ |
|
290
|
|
|
protected function process( \Aimeos\MShop\Product\Item\Iface $item, \DomElement $node ) |
|
291
|
|
|
{ |
|
292
|
|
|
$list = []; |
|
293
|
|
|
|
|
294
|
|
|
foreach( $node->attributes as $attr ) { |
|
295
|
|
|
$list[$attr->nodeName] = $attr->nodeValue; |
|
296
|
|
|
} |
|
297
|
|
|
|
|
298
|
|
|
foreach( $node->childNodes as $tag ) |
|
299
|
|
|
{ |
|
300
|
|
|
if( in_array( $tag->nodeName, ['lists', 'property'] ) ) { |
|
301
|
|
|
$item = $this->getProcessor( $tag->nodeName )->process( $item, $tag ); |
|
302
|
|
|
} else { |
|
303
|
|
|
$list[$tag->nodeName] = $tag->nodeValue; |
|
304
|
|
|
} |
|
305
|
|
|
} |
|
306
|
|
|
|
|
307
|
|
|
return $item->fromArray( $list, true ); |
|
308
|
|
|
} |
|
309
|
|
|
} |
|
310
|
|
|
|