1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
/** |
4
|
|
|
* @package s9e\TextFormatter |
5
|
|
|
* @copyright Copyright (c) 2010-2019 The s9e Authors |
6
|
|
|
* @license http://www.opensource.org/licenses/mit-license.php The MIT License |
7
|
|
|
*/ |
8
|
|
|
namespace s9e\TextFormatter\Configurator\Helpers\TemplateParser; |
9
|
|
|
|
10
|
|
|
use DOMDocument; |
11
|
|
|
use DOMElement; |
12
|
|
|
use DOMXPath; |
13
|
|
|
use RuntimeException; |
14
|
|
|
use s9e\TextFormatter\Configurator\Helpers\AVTHelper; |
15
|
|
|
use s9e\TextFormatter\Configurator\Helpers\TemplateLoader; |
16
|
|
|
|
17
|
|
|
class Parser extends IRProcessor |
18
|
|
|
{ |
19
|
|
|
/** |
20
|
|
|
* @var Normalizer |
21
|
|
|
*/ |
22
|
|
|
protected $normalizer; |
23
|
|
|
|
24
|
|
|
/** |
25
|
|
|
* @param Normalizer $normalizer |
26
|
|
|
* @return void |
27
|
|
|
*/ |
28
|
71 |
|
public function __construct(Normalizer $normalizer) |
29
|
|
|
{ |
30
|
71 |
|
$this->normalizer = $normalizer; |
31
|
|
|
} |
32
|
|
|
|
33
|
|
|
/** |
34
|
|
|
* Parse a template into an internal representation |
35
|
|
|
* |
36
|
|
|
* @param string $template Source template |
37
|
|
|
* @return DOMDocument Internal representation |
38
|
|
|
*/ |
39
|
71 |
|
public function parse($template) |
40
|
|
|
{ |
41
|
71 |
|
$dom = TemplateLoader::load($template); |
42
|
|
|
|
43
|
71 |
|
$ir = new DOMDocument; |
44
|
71 |
|
$ir->loadXML('<template/>'); |
45
|
|
|
|
46
|
71 |
|
$this->createXPath($dom); |
47
|
71 |
|
$this->parseChildren($ir->documentElement, $dom->documentElement); |
48
|
68 |
|
$this->normalizer->normalize($ir); |
49
|
|
|
|
50
|
68 |
|
return $ir; |
51
|
|
|
} |
52
|
|
|
|
53
|
|
|
/** |
54
|
|
|
* Append <output/> elements corresponding to given AVT |
55
|
|
|
* |
56
|
|
|
* @param DOMElement $parentNode Parent node |
57
|
|
|
* @param string $avt Attribute value template |
58
|
|
|
* @return void |
59
|
|
|
*/ |
60
|
9 |
|
protected function appendAVT(DOMElement $parentNode, $avt) |
61
|
|
|
{ |
62
|
9 |
|
foreach (AVTHelper::parse($avt) as $token) |
63
|
|
|
{ |
64
|
9 |
|
if ($token[0] === 'expression') |
65
|
|
|
{ |
66
|
3 |
|
$this->appendXPathOutput($parentNode, $token[1]); |
67
|
|
|
} |
68
|
|
|
else |
69
|
|
|
{ |
70
|
7 |
|
$this->appendLiteralOutput($parentNode, $token[1]); |
71
|
|
|
} |
72
|
|
|
} |
73
|
|
|
} |
74
|
|
|
|
75
|
|
|
/** |
76
|
|
|
* Append an <output/> element with literal content to given node |
77
|
|
|
* |
78
|
|
|
* @param DOMElement $parentNode Parent node |
79
|
|
|
* @param string $content Content to output |
80
|
|
|
* @return void |
81
|
|
|
*/ |
82
|
51 |
|
protected function appendLiteralOutput(DOMElement $parentNode, $content) |
83
|
|
|
{ |
84
|
51 |
|
if ($content === '') |
85
|
|
|
{ |
86
|
1 |
|
return; |
87
|
|
|
} |
88
|
|
|
|
89
|
50 |
|
$this->appendElement($parentNode, 'output', htmlspecialchars($content)) |
90
|
50 |
|
->setAttribute('type', 'literal'); |
91
|
|
|
} |
92
|
|
|
|
93
|
|
|
/** |
94
|
|
|
* Append the structure for a <xsl:copy-of/> element to given node |
95
|
|
|
* |
96
|
|
|
* @param DOMElement $parentNode Parent node |
97
|
|
|
* @param string $expr Select expression, which is should only contain attributes |
98
|
|
|
* @return void |
99
|
|
|
*/ |
100
|
2 |
|
protected function appendConditionalAttributes(DOMElement $parentNode, $expr) |
101
|
|
|
{ |
102
|
2 |
|
preg_match_all('(@([-\\w]+))', $expr, $matches); |
103
|
2 |
|
foreach ($matches[1] as $attrName) |
104
|
|
|
{ |
105
|
|
|
// Create a switch element in the IR |
106
|
2 |
|
$switch = $this->appendElement($parentNode, 'switch'); |
107
|
2 |
|
$case = $this->appendElement($switch, 'case'); |
108
|
2 |
|
$case->setAttribute('test', '@' . $attrName); |
109
|
|
|
|
110
|
|
|
// Append an attribute element |
111
|
2 |
|
$attribute = $this->appendElement($case, 'attribute'); |
112
|
2 |
|
$attribute->setAttribute('name', $attrName); |
113
|
|
|
|
114
|
|
|
// Set the attribute's content, which is simply the copied attribute's value |
115
|
2 |
|
$this->appendXPathOutput($attribute, '@' . $attrName); |
116
|
|
|
} |
117
|
|
|
} |
118
|
|
|
|
119
|
|
|
/** |
120
|
|
|
* Append an <output/> element for given XPath expression to given node |
121
|
|
|
* |
122
|
|
|
* @param DOMElement $parentNode Parent node |
123
|
|
|
* @param string $expr XPath expression |
124
|
|
|
* @return void |
125
|
|
|
*/ |
126
|
10 |
|
protected function appendXPathOutput(DOMElement $parentNode, $expr) |
127
|
|
|
{ |
128
|
10 |
|
$this->appendElement($parentNode, 'output', htmlspecialchars(trim($expr))) |
129
|
10 |
|
->setAttribute('type', 'xpath'); |
130
|
|
|
} |
131
|
|
|
|
132
|
|
|
/** |
133
|
|
|
* Parse all the children of a given element |
134
|
|
|
* |
135
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the parent node |
136
|
|
|
* @param DOMElement $parent Parent node |
137
|
|
|
* @return void |
138
|
|
|
*/ |
139
|
71 |
|
protected function parseChildren(DOMElement $ir, DOMElement $parent) |
140
|
|
|
{ |
141
|
71 |
|
foreach ($parent->childNodes as $child) |
142
|
|
|
{ |
143
|
71 |
|
switch ($child->nodeType) |
144
|
|
|
{ |
145
|
71 |
|
case XML_COMMENT_NODE: |
146
|
|
|
// Do nothing |
147
|
3 |
|
break; |
148
|
|
|
|
149
|
71 |
|
case XML_TEXT_NODE: |
150
|
40 |
|
if (trim($child->textContent) !== '') |
151
|
|
|
{ |
152
|
40 |
|
$this->appendLiteralOutput($ir, $child->textContent); |
153
|
|
|
} |
154
|
40 |
|
break; |
155
|
|
|
|
156
|
70 |
|
case XML_ELEMENT_NODE: |
157
|
69 |
|
$this->parseNode($ir, $child); |
158
|
67 |
|
break; |
159
|
|
|
|
160
|
|
|
default: |
161
|
1 |
|
throw new RuntimeException("Cannot parse node '" . $child->nodeName . "''"); |
162
|
|
|
} |
163
|
|
|
} |
164
|
|
|
} |
165
|
|
|
|
166
|
|
|
/** |
167
|
|
|
* Parse a given node into the internal representation |
168
|
|
|
* |
169
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
170
|
|
|
* @param DOMElement $node Node to parse |
171
|
|
|
* @return void |
172
|
|
|
*/ |
173
|
69 |
|
protected function parseNode(DOMElement $ir, DOMElement $node) |
174
|
|
|
{ |
175
|
|
|
// XSL elements are parsed by the corresponding parseXsl* method |
176
|
69 |
|
if ($node->namespaceURI === self::XMLNS_XSL) |
177
|
|
|
{ |
178
|
60 |
|
$methodName = 'parseXsl' . str_replace(' ', '', ucwords(str_replace('-', ' ', $node->localName))); |
179
|
60 |
|
if (!method_exists($this, $methodName)) |
180
|
|
|
{ |
181
|
1 |
|
throw new RuntimeException("Element '" . $node->nodeName . "' is not supported"); |
182
|
|
|
} |
183
|
|
|
|
184
|
59 |
|
return $this->$methodName($ir, $node); |
185
|
|
|
} |
186
|
|
|
|
187
|
|
|
// Create an <element/> with a name attribute equal to given node's name |
188
|
36 |
|
$element = $this->appendElement($ir, 'element'); |
189
|
36 |
|
$element->setAttribute('name', $node->nodeName); |
190
|
|
|
|
191
|
|
|
// Append an <attribute/> element for each namespace declaration |
192
|
36 |
|
$xpath = new DOMXPath($node->ownerDocument); |
193
|
36 |
|
foreach ($xpath->query('namespace::*', $node) as $ns) |
194
|
|
|
{ |
195
|
36 |
|
if ($node->hasAttribute($ns->nodeName)) |
196
|
|
|
{ |
197
|
3 |
|
$irAttribute = $this->appendElement($element, 'attribute'); |
198
|
3 |
|
$irAttribute->setAttribute('name', $ns->nodeName); |
199
|
3 |
|
$this->appendLiteralOutput($irAttribute, $ns->nodeValue); |
200
|
|
|
} |
201
|
|
|
} |
202
|
|
|
|
203
|
|
|
// Append an <attribute/> element for each of this node's attribute |
204
|
36 |
|
foreach ($node->attributes as $attribute) |
205
|
|
|
{ |
206
|
9 |
|
$irAttribute = $this->appendElement($element, 'attribute'); |
207
|
9 |
|
$irAttribute->setAttribute('name', $attribute->nodeName); |
208
|
|
|
|
209
|
|
|
// Append an <output/> element to represent the attribute's value |
210
|
9 |
|
$this->appendAVT($irAttribute, $attribute->value); |
211
|
|
|
} |
212
|
|
|
|
213
|
|
|
// Parse the content of this node |
214
|
36 |
|
$this->parseChildren($element, $node); |
215
|
|
|
} |
216
|
|
|
|
217
|
|
|
/** |
218
|
|
|
* Parse an <xsl:apply-templates/> node into the internal representation |
219
|
|
|
* |
220
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
221
|
|
|
* @param DOMElement $node <xsl:apply-templates/> node |
222
|
|
|
* @return void |
223
|
|
|
*/ |
224
|
19 |
|
protected function parseXslApplyTemplates(DOMElement $ir, DOMElement $node) |
225
|
|
|
{ |
226
|
19 |
|
$applyTemplates = $this->appendElement($ir, 'applyTemplates'); |
227
|
19 |
|
if ($node->hasAttribute('select')) |
228
|
|
|
{ |
229
|
1 |
|
$applyTemplates->setAttribute('select', $node->getAttribute('select')); |
230
|
|
|
} |
231
|
|
|
} |
232
|
|
|
|
233
|
|
|
/** |
234
|
|
|
* Parse an <xsl:attribute/> node into the internal representation |
235
|
|
|
* |
236
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
237
|
|
|
* @param DOMElement $node <xsl:attribute/> node |
238
|
|
|
* @return void |
239
|
|
|
*/ |
240
|
17 |
|
protected function parseXslAttribute(DOMElement $ir, DOMElement $node) |
241
|
|
|
{ |
242
|
17 |
|
$attribute = $this->appendElement($ir, 'attribute'); |
243
|
17 |
|
$attribute->setAttribute('name', $node->getAttribute('name')); |
244
|
17 |
|
$this->parseChildren($attribute, $node); |
245
|
|
|
} |
246
|
|
|
|
247
|
|
|
/** |
248
|
|
|
* Parse an <xsl:choose/> node and its <xsl:when/> and <xsl:otherwise/> children into the |
249
|
|
|
* internal representation |
250
|
|
|
* |
251
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
252
|
|
|
* @param DOMElement $node <xsl:choose/> node |
253
|
|
|
* @return void |
254
|
|
|
*/ |
255
|
20 |
|
protected function parseXslChoose(DOMElement $ir, DOMElement $node) |
256
|
|
|
{ |
257
|
20 |
|
$switch = $this->appendElement($ir, 'switch'); |
258
|
20 |
|
foreach ($this->query('./xsl:when', $node) as $when) |
259
|
|
|
{ |
260
|
|
|
// Create a <case/> element with the original test condition in @test |
261
|
20 |
|
$case = $this->appendElement($switch, 'case'); |
262
|
20 |
|
$case->setAttribute('test', $when->getAttribute('test')); |
263
|
20 |
|
$this->parseChildren($case, $when); |
264
|
|
|
} |
265
|
|
|
|
266
|
|
|
// Add the default branch, which is presumed to be last |
267
|
20 |
|
foreach ($this->query('./xsl:otherwise', $node) as $otherwise) |
268
|
|
|
{ |
269
|
10 |
|
$case = $this->appendElement($switch, 'case'); |
270
|
10 |
|
$this->parseChildren($case, $otherwise); |
271
|
|
|
|
272
|
|
|
// There should be only one <xsl:otherwise/> but we'll break anyway |
273
|
10 |
|
break; |
274
|
|
|
} |
275
|
|
|
} |
276
|
|
|
|
277
|
|
|
/** |
278
|
|
|
* Parse an <xsl:comment/> node into the internal representation |
279
|
|
|
* |
280
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
281
|
|
|
* @param DOMElement $node <xsl:comment/> node |
282
|
|
|
* @return void |
283
|
|
|
*/ |
284
|
3 |
|
protected function parseXslComment(DOMElement $ir, DOMElement $node) |
285
|
|
|
{ |
286
|
3 |
|
$comment = $this->appendElement($ir, 'comment'); |
287
|
3 |
|
$this->parseChildren($comment, $node); |
288
|
|
|
} |
289
|
|
|
|
290
|
|
|
/** |
291
|
|
|
* Parse an <xsl:copy-of/> node into the internal representation |
292
|
|
|
* |
293
|
|
|
* NOTE: only attributes are supported |
294
|
|
|
* |
295
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
296
|
|
|
* @param DOMElement $node <xsl:copy-of/> node |
297
|
|
|
* @return void |
298
|
|
|
*/ |
299
|
4 |
|
protected function parseXslCopyOf(DOMElement $ir, DOMElement $node) |
300
|
|
|
{ |
301
|
4 |
|
$expr = $node->getAttribute('select'); |
302
|
4 |
|
if (preg_match('#^@[-\\w]+(?:\\s*\\|\\s*@[-\\w]+)*$#', $expr, $m)) |
303
|
|
|
{ |
304
|
|
|
// <xsl:copy-of select="@foo"/> |
305
|
2 |
|
$this->appendConditionalAttributes($ir, $expr); |
306
|
|
|
} |
307
|
2 |
|
elseif ($expr === '@*') |
308
|
|
|
{ |
309
|
|
|
// <xsl:copy-of select="@*"/> |
310
|
1 |
|
$this->appendElement($ir, 'copyOfAttributes'); |
311
|
|
|
} |
312
|
|
|
else |
313
|
|
|
{ |
314
|
1 |
|
throw new RuntimeException("Unsupported <xsl:copy-of/> expression '" . $expr . "'"); |
315
|
|
|
} |
316
|
|
|
} |
317
|
|
|
|
318
|
|
|
/** |
319
|
|
|
* Parse an <xsl:element/> node into the internal representation |
320
|
|
|
* |
321
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
322
|
|
|
* @param DOMElement $node <xsl:element/> node |
323
|
|
|
* @return void |
324
|
|
|
*/ |
325
|
11 |
|
protected function parseXslElement(DOMElement $ir, DOMElement $node) |
326
|
|
|
{ |
327
|
11 |
|
$element = $this->appendElement($ir, 'element'); |
328
|
11 |
|
$element->setAttribute('name', $node->getAttribute('name')); |
329
|
11 |
|
$this->parseChildren($element, $node); |
330
|
|
|
} |
331
|
|
|
|
332
|
|
|
/** |
333
|
|
|
* Parse an <xsl:if/> node into the internal representation |
334
|
|
|
* |
335
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
336
|
|
|
* @param DOMElement $node <xsl:if/> node |
337
|
|
|
* @return void |
338
|
|
|
*/ |
339
|
3 |
|
protected function parseXslIf(DOMElement $ir, DOMElement $node) |
340
|
|
|
{ |
341
|
|
|
// An <xsl:if/> is represented by a <switch/> with only one <case/> |
342
|
3 |
|
$switch = $this->appendElement($ir, 'switch'); |
343
|
3 |
|
$case = $this->appendElement($switch, 'case'); |
344
|
3 |
|
$case->setAttribute('test', $node->getAttribute('test')); |
345
|
|
|
|
346
|
|
|
// Parse this branch's content |
347
|
3 |
|
$this->parseChildren($case, $node); |
348
|
|
|
} |
349
|
|
|
|
350
|
|
|
/** |
351
|
|
|
* Parse an <xsl:text/> node into the internal representation |
352
|
|
|
* |
353
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
354
|
|
|
* @param DOMElement $node <xsl:text/> node |
355
|
|
|
* @return void |
356
|
|
|
*/ |
357
|
10 |
|
protected function parseXslText(DOMElement $ir, DOMElement $node) |
358
|
|
|
{ |
359
|
10 |
|
$this->appendLiteralOutput($ir, $node->textContent); |
360
|
10 |
|
if ($node->getAttribute('disable-output-escaping') === 'yes') |
361
|
|
|
{ |
362
|
2 |
|
$ir->lastChild->setAttribute('disable-output-escaping', 'yes'); |
363
|
|
|
} |
364
|
|
|
} |
365
|
|
|
|
366
|
|
|
/** |
367
|
|
|
* Parse an <xsl:value-of/> node into the internal representation |
368
|
|
|
* |
369
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
370
|
|
|
* @param DOMElement $node <xsl:value-of/> node |
371
|
|
|
* @return void |
372
|
|
|
*/ |
373
|
6 |
|
protected function parseXslValueOf(DOMElement $ir, DOMElement $node) |
374
|
|
|
{ |
375
|
6 |
|
$this->appendXPathOutput($ir, $node->getAttribute('select')); |
376
|
6 |
|
if ($node->getAttribute('disable-output-escaping') === 'yes') |
377
|
|
|
{ |
378
|
1 |
|
$ir->lastChild->setAttribute('disable-output-escaping', 'yes'); |
379
|
|
|
} |
380
|
|
|
} |
381
|
|
|
} |