1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace CfdiUtils\Cleaner; |
4
|
|
|
|
5
|
|
|
use CfdiUtils\Cfdi; |
6
|
|
|
use CfdiUtils\Cleaner\BeforeLoad\BeforeLoadCleanerInterface; |
7
|
|
|
use CfdiUtils\Cleaner\Cleaners\SchemaLocationsXsdUrlsFixer; |
8
|
|
|
use CfdiUtils\Utils\SchemaLocations; |
9
|
|
|
use CfdiUtils\Utils\Xml; |
10
|
|
|
use DOMAttr; |
11
|
|
|
use DOMDocument; |
12
|
|
|
use DOMNode; |
13
|
|
|
use DOMNodeList; |
14
|
|
|
use DOMXPath; |
15
|
|
|
|
16
|
|
|
/** |
17
|
|
|
* Class to clean CFDI and avoid bad common practices. |
18
|
|
|
* |
19
|
|
|
* Strictly speaking, CFDI must accomplish all XML rules, including that any other |
20
|
|
|
* XML element must be isolated in its own namespace and follow their own XSD rules. |
21
|
|
|
* |
22
|
|
|
* The common practice (allowed by SAT) is that the CFDI is created, signed and |
23
|
|
|
* some nodes are attached after sign, some of them does not follow the XML standard. |
24
|
|
|
* |
25
|
|
|
* This is why it's better to clear Comprobante/Addenda and remove unused namespaces |
26
|
|
|
*/ |
27
|
|
|
class Cleaner |
28
|
|
|
{ |
29
|
|
|
/** @var DOMDocument|null */ |
30
|
|
|
protected $dom; |
31
|
|
|
|
32
|
|
|
/** @var BeforeLoadCleanerInterface */ |
33
|
|
|
private $beforeLoadCleaner; |
34
|
|
|
|
35
|
14 |
|
public function __construct(string $content, BeforeLoadCleanerInterface $beforeLoadCleaner = null) |
36
|
|
|
{ |
37
|
14 |
|
$this->beforeLoadCleaner = $beforeLoadCleaner ?? new BeforeLoad\BeforeLoadCleaner(); |
38
|
14 |
|
if ('' !== $content) { |
39
|
12 |
|
$this->load($content); |
40
|
|
|
} |
41
|
11 |
|
} |
42
|
|
|
|
43
|
|
|
/** |
44
|
|
|
* Method to clean content and return the result |
45
|
|
|
* If an error occurs, an exception is thrown |
46
|
|
|
* |
47
|
|
|
* @param string $content |
48
|
|
|
* @return string |
49
|
|
|
*/ |
50
|
1 |
|
public static function staticClean($content): string |
51
|
|
|
{ |
52
|
1 |
|
$cleaner = new self($content); |
53
|
1 |
|
$cleaner->clean(); |
54
|
1 |
|
return $cleaner->retrieveXml(); |
55
|
|
|
} |
56
|
|
|
|
57
|
|
|
/** |
58
|
|
|
* Check if the CFDI version is complatible to this class |
59
|
|
|
* |
60
|
|
|
* @param string $version |
61
|
|
|
* @return bool |
62
|
|
|
*/ |
63
|
11 |
|
public static function isVersionAllowed(string $version): bool |
64
|
|
|
{ |
65
|
11 |
|
return in_array($version, ['3.2', '3.3']); |
66
|
|
|
} |
67
|
|
|
|
68
|
|
|
/** |
69
|
|
|
* Check if a given namespace is allowed (must not be removed from CFDI) |
70
|
|
|
* |
71
|
|
|
* @param string $namespace |
72
|
|
|
* @return bool |
73
|
|
|
*/ |
74
|
2 |
|
public static function isNameSpaceAllowed(string $namespace): bool |
75
|
|
|
{ |
76
|
|
|
return ( |
77
|
2 |
|
'http://www.w3.org/' === (substr($namespace, 0, 18) ?: '') |
78
|
2 |
|
|| 'http://www.sat.gob.mx/' === (substr($namespace, 0, 22) ?: '') |
79
|
|
|
); |
80
|
|
|
} |
81
|
|
|
|
82
|
|
|
/** |
83
|
|
|
* Apply all removals (Addenda, Non SAT Nodes and Non SAT namespaces) |
84
|
|
|
* |
85
|
|
|
* @return void |
86
|
|
|
*/ |
87
|
1 |
|
public function clean() |
88
|
|
|
{ |
89
|
1 |
|
$this->removeAddenda(); |
90
|
1 |
|
$this->removeIncompleteSchemaLocations(); |
91
|
1 |
|
$this->removeNonSatNSNodes(); |
92
|
1 |
|
$this->removeNonSatNSschemaLocations(); |
93
|
1 |
|
$this->removeUnusedNamespaces(); |
94
|
1 |
|
$this->collapseComprobanteComplemento(); |
95
|
1 |
|
$this->fixKnownSchemaLocationsXsdUrls(); |
96
|
1 |
|
} |
97
|
|
|
|
98
|
|
|
/** |
99
|
|
|
* Load the string content as a CFDI |
100
|
|
|
* This is exposed to reuse the current object instead of create a new instance |
101
|
|
|
* |
102
|
|
|
* @param string $content |
103
|
|
|
* |
104
|
|
|
* @throws CleanerException when the content is not valid xml |
105
|
|
|
* @throws CleanerException when the document does not use the namespace http://www.sat.gob.mx/cfd/3 |
106
|
|
|
* @throws CleanerException when cannot find a Comprobante version (or Version) attribute |
107
|
|
|
* @throws CleanerException when the version is not compatible |
108
|
|
|
* |
109
|
|
|
* @return void |
110
|
|
|
*/ |
111
|
14 |
|
public function load(string $content) |
112
|
|
|
{ |
113
|
|
|
try { |
114
|
14 |
|
$content = $this->beforeLoadCleaner->clean($content); |
115
|
14 |
|
$cfdi = Cfdi::newFromString($content); |
116
|
3 |
|
} catch (\Throwable $exception) { |
117
|
3 |
|
throw new CleanerException($exception->getMessage(), $exception->getCode(), $exception->getPrevious()); |
118
|
|
|
} |
119
|
11 |
|
$version = $cfdi->getVersion(); |
120
|
11 |
|
if (! $this->isVersionAllowed($version)) { |
121
|
2 |
|
throw new CleanerException("The CFDI version '$version' is not allowed"); |
122
|
|
|
} |
123
|
9 |
|
$this->dom = $cfdi->getDocument(); |
124
|
9 |
|
} |
125
|
|
|
|
126
|
|
|
/** |
127
|
|
|
* Get the XML content of the CFDI |
128
|
|
|
* |
129
|
|
|
* @return string |
130
|
|
|
*/ |
131
|
5 |
|
public function retrieveXml(): string |
132
|
|
|
{ |
133
|
5 |
|
return $this->dom()->saveXML(); |
134
|
|
|
} |
135
|
|
|
|
136
|
|
|
/** |
137
|
|
|
* Get a clone of the XML DOM Document of the CFDI |
138
|
|
|
* |
139
|
|
|
* @return DOMDocument |
140
|
|
|
*/ |
141
|
1 |
|
public function retrieveDocument(): DOMDocument |
142
|
|
|
{ |
143
|
1 |
|
return clone $this->dom(); |
144
|
|
|
} |
145
|
|
|
|
146
|
|
|
/** |
147
|
|
|
* Procedure to remove the Comprobante/Addenda node |
148
|
|
|
* |
149
|
|
|
* @return void |
150
|
|
|
*/ |
151
|
1 |
|
public function removeAddenda() |
152
|
|
|
{ |
153
|
1 |
|
$query = '/cfdi:Comprobante/cfdi:Addenda'; |
154
|
1 |
|
$addendas = $this->xpathQuery($query); |
155
|
1 |
|
foreach ($addendas as $addenda) { |
156
|
1 |
|
$addenda->parentNode->removeChild($addenda); |
157
|
|
|
} |
158
|
1 |
|
} |
159
|
|
|
|
160
|
|
|
/** |
161
|
|
|
* Procedure to drop schemaLocations where second part does not ends with '.xsd' |
162
|
|
|
* |
163
|
|
|
* @return void |
164
|
|
|
*/ |
165
|
1 |
|
public function removeIncompleteSchemaLocations() |
166
|
|
|
{ |
167
|
1 |
|
foreach ($this->obtainXsiSchemaLocations() as $element) { |
168
|
1 |
|
$element->nodeValue = $this->removeIncompleteSchemaLocationPrivate($element->nodeValue); |
169
|
|
|
} |
170
|
1 |
|
} |
171
|
|
|
|
172
|
|
|
/** |
173
|
|
|
* @param string $source |
174
|
|
|
* @return string |
175
|
|
|
* @deprecated 2.12.0: This function is internal and visibility should be private, use SchemaLocations |
176
|
|
|
* @internal |
177
|
|
|
*/ |
178
|
|
|
public function removeIncompleteSchemaLocation(string $source): string |
179
|
|
|
{ |
180
|
|
|
trigger_error('This method is deprecated, should not be used from outside this class', E_USER_DEPRECATED); |
181
|
|
|
return $this->removeIncompleteSchemaLocationPrivate($source); |
182
|
|
|
} |
183
|
|
|
|
184
|
1 |
|
private function removeIncompleteSchemaLocationPrivate(string $source): string |
185
|
|
|
{ |
186
|
1 |
|
$schemaLocations = SchemaLocations::fromStingStrictXsd($source); |
187
|
1 |
|
foreach ($schemaLocations->getNamespacesWithoutLocation() as $namespace) { |
188
|
1 |
|
$schemaLocations->remove($namespace); |
189
|
|
|
} |
190
|
1 |
|
return $schemaLocations->asString(); |
191
|
|
|
} |
192
|
|
|
|
193
|
|
|
/** |
194
|
|
|
* Procedure to drop schemaLocations that are not allowed |
195
|
|
|
* If the schemaLocation is empty then remove the attribute |
196
|
|
|
* |
197
|
|
|
* @return void |
198
|
|
|
*/ |
199
|
3 |
|
public function removeNonSatNSschemaLocations() |
200
|
|
|
{ |
201
|
3 |
|
$schemaLocations = $this->obtainXsiSchemaLocations(); |
202
|
3 |
|
foreach ($schemaLocations as $element) { |
203
|
3 |
|
$this->removeNonSatNSschemaLocation($element); |
204
|
|
|
} |
205
|
2 |
|
} |
206
|
|
|
|
207
|
3 |
|
private function removeNonSatNSschemaLocation(DOMAttr $schemaLocation) |
208
|
|
|
{ |
209
|
3 |
|
$source = $schemaLocation->nodeValue; |
210
|
|
|
// load locations |
211
|
3 |
|
$schemaLocations = SchemaLocations::fromString($source, true); |
212
|
3 |
|
if ($schemaLocations->hasAnyNamespaceWithoutLocation()) { |
213
|
1 |
|
throw new CleanerException( |
214
|
1 |
|
sprintf("The schemaLocation value '%s' must have even number of URIs", $source) |
215
|
|
|
); |
216
|
|
|
} |
217
|
|
|
// filter |
218
|
2 |
|
foreach ($schemaLocations as $namespace => $location) { |
219
|
2 |
|
if (! $this->isNameSpaceAllowed($namespace)) { |
220
|
2 |
|
$schemaLocations->remove($namespace); |
221
|
|
|
} |
222
|
|
|
} |
223
|
|
|
// apply |
224
|
2 |
|
$modified = $schemaLocations->asString(); |
225
|
2 |
|
if ($schemaLocations->isEmpty()) { // remove node |
226
|
1 |
|
$schemaLocation->ownerElement->removeAttributeNode($schemaLocation); |
227
|
1 |
|
} elseif ($source !== $modified) { // replace node content and is different |
228
|
1 |
|
$schemaLocation->nodeValue = $modified; |
229
|
|
|
} |
230
|
2 |
|
} |
231
|
|
|
|
232
|
|
|
/** |
233
|
|
|
* Procedure to remove all nodes that are not from an allowed namespace |
234
|
|
|
* |
235
|
|
|
* @return void |
236
|
|
|
*/ |
237
|
1 |
|
public function removeNonSatNSNodes() |
238
|
|
|
{ |
239
|
1 |
|
$nss = $this->obtainNamespaces(); |
240
|
1 |
|
foreach ($nss as $namespace) { |
241
|
1 |
|
if (! $this->isNameSpaceAllowed($namespace)) { |
242
|
1 |
|
$this->removeNonSatNSNode($namespace); |
243
|
|
|
} |
244
|
|
|
} |
245
|
1 |
|
} |
246
|
|
|
|
247
|
|
|
/** |
248
|
|
|
* Procedure to remove all nodes from an specific namespace |
249
|
|
|
* |
250
|
|
|
* @param string $namespace |
251
|
|
|
* @return void |
252
|
|
|
*/ |
253
|
1 |
|
private function removeNonSatNSNode(string $namespace) |
254
|
|
|
{ |
255
|
1 |
|
foreach ($this->dom()->getElementsByTagNameNS($namespace, '*') as $children) { |
256
|
1 |
|
$children->parentNode->removeChild($children); |
257
|
|
|
} |
258
|
1 |
|
} |
259
|
|
|
|
260
|
|
|
/** |
261
|
|
|
* Procedure to remove not allowed xmlns definitions |
262
|
|
|
* |
263
|
|
|
* @return void |
264
|
|
|
*/ |
265
|
1 |
|
public function removeUnusedNamespaces() |
266
|
|
|
{ |
267
|
1 |
|
$nss = []; |
268
|
1 |
|
$dom = $this->dom(); |
269
|
1 |
|
$namespaces = $this->obtainNamespaces(); |
270
|
1 |
|
foreach ($namespaces as $namespace) { |
271
|
1 |
|
if (! $namespace || $this->isNameSpaceAllowed($namespace)) { |
272
|
1 |
|
continue; |
273
|
|
|
} |
274
|
1 |
|
$prefix = $dom->lookupPrefix($namespace); |
275
|
1 |
|
$nss[$prefix] = $namespace; |
276
|
|
|
} |
277
|
1 |
|
$documentElement = Xml::documentElement($dom); |
278
|
1 |
|
foreach ($nss as $prefix => $namespace) { |
279
|
1 |
|
$documentElement->removeAttributeNS($namespace, $prefix); |
280
|
|
|
} |
281
|
1 |
|
} |
282
|
|
|
|
283
|
|
|
/** |
284
|
|
|
* Procedure to collapse Complemento elements from Comprobante |
285
|
|
|
* Collapse will take its children and put then on the first Complemento found |
286
|
|
|
* |
287
|
|
|
* @return void |
288
|
|
|
*/ |
289
|
3 |
|
public function collapseComprobanteComplemento() |
290
|
|
|
{ |
291
|
3 |
|
$comprobante = Xml::documentElement($this->dom()); |
292
|
3 |
|
$complementos = $this->xpathQuery('./cfdi:Complemento', $comprobante); |
293
|
3 |
|
if ($complementos->length < 2) { |
294
|
1 |
|
return; // nothing to do, there are less than 2 complemento |
295
|
|
|
} |
296
|
2 |
|
$first = null; |
297
|
|
|
/** @var DOMNode $extra */ |
298
|
2 |
|
foreach ($complementos as $extra) { // iterate over all extra children |
299
|
2 |
|
if (null === $first) { |
300
|
2 |
|
$first = $extra; |
301
|
2 |
|
continue; |
302
|
|
|
} |
303
|
2 |
|
$comprobante->removeChild($extra); // remove extra child from parent |
304
|
2 |
|
while ($extra->childNodes->length > 0) { // append extra child contents into first child |
305
|
|
|
/** @var DOMNode $child */ |
306
|
2 |
|
$child = $extra->childNodes->item(0); |
307
|
2 |
|
$extra->removeChild($child); |
308
|
2 |
|
$first->appendChild($child); |
309
|
|
|
} |
310
|
|
|
} |
311
|
2 |
|
} |
312
|
|
|
|
313
|
|
|
/** |
314
|
|
|
* Procedure to fix XSD known location paths for CFDI and TFD |
315
|
|
|
* |
316
|
|
|
* @return void |
317
|
|
|
*/ |
318
|
1 |
|
public function fixKnownSchemaLocationsXsdUrls() |
319
|
|
|
{ |
320
|
1 |
|
$xsiLocations = $this->obtainXsiSchemaLocations(); |
321
|
1 |
|
$schemasFixer = SchemaLocationsXsdUrlsFixer::createWithKnownSatUrls(); |
322
|
1 |
|
foreach ($xsiLocations as $xsiSchemaLocation) { |
323
|
1 |
|
$schemasFixer->fixSchemaLocationAttribute($xsiSchemaLocation); |
324
|
|
|
} |
325
|
1 |
|
} |
326
|
|
|
|
327
|
|
|
/** @return DOMNodeList|DOMAttr[] */ |
328
|
3 |
|
private function obtainXsiSchemaLocations(): DOMNodeList |
329
|
|
|
{ |
330
|
|
|
// Do not assume that prefix for http://www.w3.org/2001/XMLSchema-instance is "xsi" |
331
|
3 |
|
$xsi = $this->dom()->lookupPrefix('http://www.w3.org/2001/XMLSchema-instance'); |
332
|
3 |
|
if (! $xsi) { |
333
|
|
|
return new DOMNodeList(); |
334
|
|
|
} |
335
|
3 |
|
return $this->xpathQuery("//@$xsi:schemaLocation"); |
336
|
|
|
} |
337
|
|
|
|
338
|
|
|
/** @return string[] */ |
339
|
1 |
|
private function obtainNamespaces(): array |
340
|
|
|
{ |
341
|
1 |
|
return array_unique(array_column(iterator_to_array($this->xpathQuery('//namespace::*')), 'nodeValue')); |
342
|
|
|
} |
343
|
|
|
|
344
|
|
|
/** |
345
|
|
|
* Helper function to perform a XPath query using an element (or root element) |
346
|
|
|
* |
347
|
|
|
* @param string $query |
348
|
|
|
* @param DOMNode|null $element |
349
|
|
|
* @return DOMNodeList |
350
|
|
|
*/ |
351
|
5 |
|
private function xpathQuery(string $query, DOMNode $element = null): DOMNodeList |
352
|
|
|
{ |
353
|
5 |
|
if (null === $element) { |
354
|
3 |
|
$document = $this->dom(); |
355
|
3 |
|
$element = Xml::documentElement($document); |
356
|
|
|
} else { |
357
|
3 |
|
$document = Xml::ownerDocument($element); |
358
|
|
|
} |
359
|
|
|
/** @var DOMNodeList|false $nodelist phpstan does not know that query can return false */ |
360
|
5 |
|
$nodelist = (new DOMXPath($document))->query($query, $element); |
361
|
5 |
|
if (false === $nodelist) { |
362
|
|
|
$nodelist = new DOMNodeList(); |
363
|
|
|
} |
364
|
5 |
|
return $nodelist; |
365
|
|
|
} |
366
|
|
|
|
367
|
7 |
|
private function dom(): DOMDocument |
368
|
|
|
{ |
369
|
7 |
|
if (null === $this->dom) { |
370
|
|
|
throw new \LogicException('No document has been loaded'); |
371
|
|
|
} |
372
|
7 |
|
return $this->dom; |
373
|
|
|
} |
374
|
|
|
} |
375
|
|
|
|