1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
/** |
4
|
|
|
* @package s9e\TextFormatter |
5
|
|
|
* @copyright Copyright (c) 2010-2018 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
|
|
|
|
16
|
|
|
class Parser extends IRProcessor |
17
|
|
|
{ |
18
|
|
|
/** |
19
|
|
|
* @var Normalizer |
20
|
|
|
*/ |
21
|
|
|
protected $normalizer; |
22
|
|
|
|
23
|
|
|
/** |
24
|
|
|
* @param Normalizer $normalizer |
25
|
|
|
* @return void |
26
|
|
|
*/ |
27
|
|
|
public function __construct(Normalizer $normalizer) |
28
|
|
|
{ |
29
|
|
|
$this->normalizer = $normalizer; |
30
|
|
|
} |
31
|
|
|
|
32
|
|
|
/** |
33
|
|
|
* Parse a template into an internal representation |
34
|
|
|
* |
35
|
|
|
* @param string $template Source template |
36
|
|
|
* @return DOMDocument Internal representation |
37
|
|
|
*/ |
38
|
|
|
public function parse($template) |
39
|
|
|
{ |
40
|
|
|
$xsl = '<xsl:template xmlns:xsl="' . self::XMLNS_XSL . '">' . $template . '</xsl:template>'; |
41
|
|
|
|
42
|
|
|
$dom = new DOMDocument; |
43
|
|
|
$dom->loadXML($xsl); |
44
|
|
|
|
45
|
|
|
$ir = new DOMDocument; |
46
|
|
|
$ir->loadXML('<template/>'); |
47
|
|
|
|
48
|
|
|
$this->createXPath($dom); |
49
|
|
|
$this->parseChildren($ir->documentElement, $dom->documentElement); |
50
|
|
|
$this->normalizer->normalize($ir); |
51
|
|
|
|
52
|
|
|
return $ir; |
53
|
|
|
} |
54
|
|
|
|
55
|
|
|
/** |
56
|
|
|
* Append an <output/> element to given node in the IR |
57
|
|
|
* |
58
|
|
|
* @param DOMElement $ir Parent node |
59
|
|
|
* @param string $type Either 'avt', 'literal' or 'xpath' |
60
|
|
|
* @param string $content Content to output |
61
|
|
|
* @return void |
62
|
|
|
*/ |
63
|
|
|
protected function appendOutput(DOMElement $ir, $type, $content) |
64
|
|
|
{ |
65
|
|
|
// Reparse AVTs and add them as separate xpath/literal outputs |
66
|
|
|
if ($type === 'avt') |
67
|
|
|
{ |
68
|
|
|
foreach (AVTHelper::parse($content) as $token) |
69
|
|
|
{ |
70
|
|
|
$type = ($token[0] === 'expression') ? 'xpath' : 'literal'; |
71
|
|
|
$this->appendOutput($ir, $type, $token[1]); |
72
|
|
|
} |
73
|
|
|
|
74
|
|
|
return; |
75
|
|
|
} |
76
|
|
|
|
77
|
|
|
if ($type === 'xpath') |
78
|
|
|
{ |
79
|
|
|
// Remove whitespace surrounding XPath expressions |
80
|
|
|
$content = trim($content); |
81
|
|
|
} |
82
|
|
|
|
83
|
|
|
if ($type === 'literal' && $content === '') |
84
|
|
|
{ |
85
|
|
|
// Don't add empty literals |
86
|
|
|
return; |
87
|
|
|
} |
88
|
|
|
|
89
|
|
|
$this->appendElement($ir, 'output', htmlspecialchars($content)) |
90
|
|
|
->setAttribute('type', $type); |
91
|
|
|
} |
92
|
|
|
|
93
|
|
|
/** |
94
|
|
|
* Parse all the children of a given element |
95
|
|
|
* |
96
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the parent node |
97
|
|
|
* @param DOMElement $parent Parent node |
98
|
|
|
* @return void |
99
|
|
|
*/ |
100
|
|
|
protected function parseChildren(DOMElement $ir, DOMElement $parent) |
101
|
|
|
{ |
102
|
|
|
foreach ($parent->childNodes as $child) |
103
|
|
|
{ |
104
|
|
|
switch ($child->nodeType) |
105
|
|
|
{ |
106
|
|
|
case XML_COMMENT_NODE: |
107
|
|
|
// Do nothing |
108
|
|
|
break; |
109
|
|
|
|
110
|
|
|
case XML_TEXT_NODE: |
111
|
|
|
if (trim($child->textContent) !== '') |
112
|
|
|
{ |
113
|
|
|
$this->appendOutput($ir, 'literal', $child->textContent); |
114
|
|
|
} |
115
|
|
|
break; |
116
|
|
|
|
117
|
|
|
case XML_ELEMENT_NODE: |
118
|
|
|
$this->parseNode($ir, $child); |
119
|
|
|
break; |
120
|
|
|
|
121
|
|
|
default: |
122
|
|
|
throw new RuntimeException("Cannot parse node '" . $child->nodeName . "''"); |
123
|
|
|
} |
124
|
|
|
} |
125
|
|
|
} |
126
|
|
|
|
127
|
|
|
/** |
128
|
|
|
* Parse a given node into the internal representation |
129
|
|
|
* |
130
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
131
|
|
|
* @param DOMElement $node Node to parse |
132
|
|
|
* @return void |
133
|
|
|
*/ |
134
|
|
|
protected function parseNode(DOMElement $ir, DOMElement $node) |
135
|
|
|
{ |
136
|
|
|
// XSL elements are parsed by the corresponding parseXsl* method |
137
|
|
|
if ($node->namespaceURI === self::XMLNS_XSL) |
138
|
|
|
{ |
139
|
|
|
$methodName = 'parseXsl' . str_replace(' ', '', ucwords(str_replace('-', ' ', $node->localName))); |
140
|
|
|
if (!method_exists($this, $methodName)) |
141
|
|
|
{ |
142
|
|
|
throw new RuntimeException("Element '" . $node->nodeName . "' is not supported"); |
143
|
|
|
} |
144
|
|
|
|
145
|
|
|
return $this->$methodName($ir, $node); |
146
|
|
|
} |
147
|
|
|
|
148
|
|
|
// Create an <element/> with a name attribute equal to given node's name |
149
|
|
|
$element = $this->appendElement($ir, 'element'); |
150
|
|
|
$element->setAttribute('name', $node->nodeName); |
151
|
|
|
|
152
|
|
|
// Append an <attribute/> element for each namespace declaration |
153
|
|
|
$xpath = new DOMXPath($node->ownerDocument); |
154
|
|
|
foreach ($xpath->query('namespace::*', $node) as $ns) |
155
|
|
|
{ |
156
|
|
|
if ($node->hasAttribute($ns->nodeName)) |
157
|
|
|
{ |
158
|
|
|
$irAttribute = $this->appendElement($element, 'attribute'); |
159
|
|
|
$irAttribute->setAttribute('name', $ns->nodeName); |
160
|
|
|
$this->appendOutput($irAttribute, 'literal', $ns->nodeValue); |
161
|
|
|
} |
162
|
|
|
} |
163
|
|
|
|
164
|
|
|
// Append an <attribute/> element for each of this node's attribute |
165
|
|
|
foreach ($node->attributes as $attribute) |
166
|
|
|
{ |
167
|
|
|
$irAttribute = $this->appendElement($element, 'attribute'); |
168
|
|
|
$irAttribute->setAttribute('name', $attribute->nodeName); |
169
|
|
|
|
170
|
|
|
// Append an <output/> element to represent the attribute's value |
171
|
|
|
$this->appendOutput($irAttribute, 'avt', $attribute->value); |
172
|
|
|
} |
173
|
|
|
|
174
|
|
|
// Parse the content of this node |
175
|
|
|
$this->parseChildren($element, $node); |
176
|
|
|
} |
177
|
|
|
|
178
|
|
|
/** |
179
|
|
|
* Parse an <xsl:apply-templates/> node into the internal representation |
180
|
|
|
* |
181
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
182
|
|
|
* @param DOMElement $node <xsl:apply-templates/> node |
183
|
|
|
* @return void |
184
|
|
|
*/ |
185
|
|
|
protected function parseXslApplyTemplates(DOMElement $ir, DOMElement $node) |
186
|
|
|
{ |
187
|
|
|
$applyTemplates = $this->appendElement($ir, 'applyTemplates'); |
188
|
|
|
if ($node->hasAttribute('select')) |
189
|
|
|
{ |
190
|
|
|
$applyTemplates->setAttribute('select', $node->getAttribute('select')); |
191
|
|
|
} |
192
|
|
|
} |
193
|
|
|
|
194
|
|
|
/** |
195
|
|
|
* Parse an <xsl:attribute/> node into the internal representation |
196
|
|
|
* |
197
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
198
|
|
|
* @param DOMElement $node <xsl:attribute/> node |
199
|
|
|
* @return void |
200
|
|
|
*/ |
201
|
|
|
protected function parseXslAttribute(DOMElement $ir, DOMElement $node) |
202
|
|
|
{ |
203
|
|
|
$attribute = $this->appendElement($ir, 'attribute'); |
204
|
|
|
$attribute->setAttribute('name', $node->getAttribute('name')); |
205
|
|
|
$this->parseChildren($attribute, $node); |
206
|
|
|
} |
207
|
|
|
|
208
|
|
|
/** |
209
|
|
|
* Parse an <xsl:choose/> node and its <xsl:when/> and <xsl:otherwise/> children into the |
210
|
|
|
* internal representation |
211
|
|
|
* |
212
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
213
|
|
|
* @param DOMElement $node <xsl:choose/> node |
214
|
|
|
* @return void |
215
|
|
|
*/ |
216
|
|
|
protected function parseXslChoose(DOMElement $ir, DOMElement $node) |
217
|
|
|
{ |
218
|
|
|
$switch = $this->appendElement($ir, 'switch'); |
219
|
|
|
foreach ($this->query('./xsl:when', $node) as $when) |
220
|
|
|
{ |
221
|
|
|
// Create a <case/> element with the original test condition in @test |
222
|
|
|
$case = $this->appendElement($switch, 'case'); |
223
|
|
|
$case->setAttribute('test', $when->getAttribute('test')); |
224
|
|
|
$this->parseChildren($case, $when); |
225
|
|
|
} |
226
|
|
|
|
227
|
|
|
// Add the default branch, which is presumed to be last |
228
|
|
|
foreach ($this->query('./xsl:otherwise', $node) as $otherwise) |
229
|
|
|
{ |
230
|
|
|
$case = $this->appendElement($switch, 'case'); |
231
|
|
|
$this->parseChildren($case, $otherwise); |
232
|
|
|
|
233
|
|
|
// There should be only one <xsl:otherwise/> but we'll break anyway |
234
|
|
|
break; |
235
|
|
|
} |
236
|
|
|
} |
237
|
|
|
|
238
|
|
|
/** |
239
|
|
|
* Parse an <xsl:comment/> node into the internal representation |
240
|
|
|
* |
241
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
242
|
|
|
* @param DOMElement $node <xsl:comment/> node |
243
|
|
|
* @return void |
244
|
|
|
*/ |
245
|
|
|
protected function parseXslComment(DOMElement $ir, DOMElement $node) |
246
|
|
|
{ |
247
|
|
|
$comment = $this->appendElement($ir, 'comment'); |
248
|
|
|
|
249
|
|
|
// Parse this branch's content |
250
|
|
|
$this->parseChildren($comment, $node); |
251
|
|
|
} |
252
|
|
|
|
253
|
|
|
/** |
254
|
|
|
* Parse an <xsl:copy-of/> node into the internal representation |
255
|
|
|
* |
256
|
|
|
* NOTE: only attributes are supported |
257
|
|
|
* |
258
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
259
|
|
|
* @param DOMElement $node <xsl:copy-of/> node |
260
|
|
|
* @return void |
261
|
|
|
*/ |
262
|
|
|
protected function parseXslCopyOf(DOMElement $ir, DOMElement $node) |
263
|
|
|
{ |
264
|
|
|
$expr = $node->getAttribute('select'); |
265
|
|
|
|
266
|
|
|
// <xsl:copy-of select="@foo"/> |
267
|
|
|
if (preg_match('#^@([-\\w]+)$#', $expr, $m)) |
268
|
|
|
{ |
269
|
|
|
// Create a switch element in the IR |
270
|
|
|
$switch = $this->appendElement($ir, 'switch'); |
271
|
|
|
$case = $this->appendElement($switch, 'case'); |
272
|
|
|
$case->setAttribute('test', $expr); |
273
|
|
|
|
274
|
|
|
// Append an attribute element |
275
|
|
|
$attribute = $this->appendElement($case, 'attribute'); |
276
|
|
|
$attribute->setAttribute('name', $m[1]); |
277
|
|
|
|
278
|
|
|
// Set the attribute's content, which is simply the copied attribute's value |
279
|
|
|
$this->appendOutput($attribute, 'xpath', $expr); |
280
|
|
|
|
281
|
|
|
return; |
282
|
|
|
} |
283
|
|
|
|
284
|
|
|
// <xsl:copy-of select="@*"/> |
285
|
|
|
if ($expr === '@*') |
286
|
|
|
{ |
287
|
|
|
$this->appendElement($ir, 'copyOfAttributes'); |
288
|
|
|
|
289
|
|
|
return; |
290
|
|
|
} |
291
|
|
|
|
292
|
|
|
throw new RuntimeException("Unsupported <xsl:copy-of/> expression '" . $expr . "'"); |
293
|
|
|
} |
294
|
|
|
|
295
|
|
|
/** |
296
|
|
|
* Parse an <xsl:element/> node into the internal representation |
297
|
|
|
* |
298
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
299
|
|
|
* @param DOMElement $node <xsl:element/> node |
300
|
|
|
* @return void |
301
|
|
|
*/ |
302
|
|
|
protected function parseXslElement(DOMElement $ir, DOMElement $node) |
303
|
|
|
{ |
304
|
|
|
$elName = $node->getAttribute('name'); |
305
|
|
|
|
306
|
|
|
if ($elName !== '') |
307
|
|
|
{ |
308
|
|
|
$element = $this->appendElement($ir, 'element'); |
309
|
|
|
|
310
|
|
|
// Copy this element's name |
311
|
|
|
$element->setAttribute('name', $elName); |
312
|
|
|
|
313
|
|
|
// Parse this element's content |
314
|
|
|
$this->parseChildren($element, $node); |
315
|
|
|
} |
316
|
|
|
} |
317
|
|
|
|
318
|
|
|
/** |
319
|
|
|
* Parse an <xsl:if/> 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:if/> node |
323
|
|
|
* @return void |
324
|
|
|
*/ |
325
|
|
|
protected function parseXslIf(DOMElement $ir, DOMElement $node) |
326
|
|
|
{ |
327
|
|
|
// An <xsl:if/> is represented by a <switch/> with only one <case/> |
328
|
|
|
$switch = $this->appendElement($ir, 'switch'); |
329
|
|
|
$case = $this->appendElement($switch, 'case'); |
330
|
|
|
$case->setAttribute('test', $node->getAttribute('test')); |
331
|
|
|
|
332
|
|
|
// Parse this branch's content |
333
|
|
|
$this->parseChildren($case, $node); |
334
|
|
|
} |
335
|
|
|
|
336
|
|
|
/** |
337
|
|
|
* Parse an <xsl:text/> node into the internal representation |
338
|
|
|
* |
339
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
340
|
|
|
* @param DOMElement $node <xsl:text/> node |
341
|
|
|
* @return void |
342
|
|
|
*/ |
343
|
|
|
protected function parseXslText(DOMElement $ir, DOMElement $node) |
344
|
|
|
{ |
345
|
|
|
$this->appendOutput($ir, 'literal', $node->textContent); |
346
|
|
|
if ($node->getAttribute('disable-output-escaping') === 'yes') |
347
|
|
|
{ |
348
|
|
|
$ir->lastChild->setAttribute('disable-output-escaping', 'yes'); |
349
|
|
|
} |
350
|
|
|
} |
351
|
|
|
|
352
|
|
|
/** |
353
|
|
|
* Parse an <xsl:value-of/> node into the internal representation |
354
|
|
|
* |
355
|
|
|
* @param DOMElement $ir Node in the internal representation that represents the node's parent |
356
|
|
|
* @param DOMElement $node <xsl:value-of/> node |
357
|
|
|
* @return void |
358
|
|
|
*/ |
359
|
|
|
protected function parseXslValueOf(DOMElement $ir, DOMElement $node) |
360
|
|
|
{ |
361
|
|
|
$this->appendOutput($ir, 'xpath', $node->getAttribute('select')); |
362
|
|
|
if ($node->getAttribute('disable-output-escaping') === 'yes') |
363
|
|
|
{ |
364
|
|
|
$ir->lastChild->setAttribute('disable-output-escaping', 'yes'); |
365
|
|
|
} |
366
|
|
|
} |
367
|
|
|
} |