|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
namespace Openbuildings\Spiderling; |
|
4
|
|
|
|
|
5
|
|
|
use Symfony\Component\CssSelector; |
|
6
|
|
|
|
|
7
|
|
|
/** |
|
8
|
|
|
* Locator - converts varios locator formats into xpath |
|
9
|
|
|
* |
|
10
|
|
|
* @package Openbuildings\Spiderling |
|
11
|
|
|
* @author Ivan Kerin |
|
12
|
|
|
* @copyright (c) 2013 OpenBuildings Ltd. |
|
13
|
|
|
* @license http://spdx.org/licenses/BSD-3-Clause |
|
14
|
|
|
*/ |
|
15
|
|
|
class Locator { |
|
16
|
|
|
|
|
17
|
|
|
const DEFAULT_TYPE = 'css'; |
|
18
|
|
|
|
|
19
|
|
|
/** |
|
20
|
|
|
* Converted xpath cache |
|
21
|
|
|
* @var string |
|
22
|
|
|
*/ |
|
23
|
|
|
protected $_xpath; |
|
24
|
|
|
|
|
25
|
|
|
/** |
|
26
|
|
|
* The type of the locator, can be css, field, xpath, label, link or button |
|
27
|
|
|
* @var string |
|
28
|
|
|
*/ |
|
29
|
|
|
protected $_type; |
|
30
|
|
|
|
|
31
|
|
|
/** |
|
32
|
|
|
* The selector used to generate the xpath |
|
33
|
|
|
* @var string |
|
34
|
|
|
*/ |
|
35
|
|
|
protected $_selector; |
|
36
|
|
|
|
|
37
|
|
|
/** |
|
38
|
|
|
* Additional filters to apply after the xpath is found |
|
39
|
|
|
* @var array |
|
40
|
|
|
*/ |
|
41
|
|
|
protected $_filters; |
|
42
|
|
|
|
|
43
|
|
|
/** |
|
44
|
|
|
* @param string $type |
|
45
|
|
|
* @param string $selector |
|
46
|
|
|
* @param array $filters |
|
47
|
|
|
*/ |
|
48
|
67 |
|
function __construct($type, $selector, array $filters = array()) |
|
|
|
|
|
|
49
|
|
|
{ |
|
50
|
67 |
|
$this->_type = $type === NULL ? self::DEFAULT_TYPE : $type; |
|
51
|
67 |
|
$this->_selector = $selector; |
|
52
|
67 |
|
$this->_filters = $filters; |
|
53
|
67 |
|
} |
|
54
|
|
|
|
|
55
|
|
|
/** |
|
56
|
|
|
* Check if a Node item matches the current filters |
|
57
|
|
|
* @param Node $item |
|
58
|
|
|
* @param integer $index |
|
59
|
|
|
* @return boolean |
|
60
|
|
|
*/ |
|
61
|
3 |
|
public function is_filtered(Node $item, $index) |
|
62
|
1 |
|
{ |
|
63
|
3 |
|
foreach ($this->filters() as $filter => $value) |
|
64
|
|
|
{ |
|
65
|
|
|
try { |
|
66
|
|
|
switch ($filter) |
|
67
|
|
|
{ |
|
68
|
3 |
|
case 'at': |
|
69
|
2 |
|
$matches_filter = $this->filter_by_at($item, $index, $value); |
|
70
|
2 |
|
break; |
|
71
|
|
|
|
|
72
|
2 |
|
case 'value': |
|
73
|
2 |
|
$matches_filter = $this->filter_by_value($item, $index, $value); |
|
74
|
2 |
|
break; |
|
75
|
|
|
|
|
76
|
2 |
|
case 'text': |
|
77
|
1 |
|
$matches_filter = $this->filter_by_text($item, $index, $value); |
|
78
|
1 |
|
break; |
|
79
|
|
|
|
|
80
|
1 |
|
case 'visible': |
|
81
|
1 |
|
$matches_filter = $this->filter_by_visible($item, $index, $value); |
|
82
|
1 |
|
break; |
|
83
|
|
|
|
|
84
|
1 |
|
case 'attributes': |
|
85
|
1 |
|
$matches_filter = $this->filter_by_attributes($item, $index, $value); |
|
86
|
1 |
|
break; |
|
87
|
|
|
|
|
88
|
1 |
|
default: |
|
89
|
1 |
|
throw new Exception('Filter :filter does not exist', array(':filter' => $filter)); |
|
90
|
1 |
|
} |
|
91
|
|
|
|
|
92
|
3 |
|
if ( ! $matches_filter) |
|
93
|
3 |
|
return FALSE; |
|
94
|
|
|
|
|
95
|
3 |
|
} catch (Exception_Staleelement $e) { |
|
96
|
|
|
return FALSE; |
|
97
|
|
|
} |
|
98
|
3 |
|
} |
|
99
|
|
|
|
|
100
|
3 |
|
return TRUE; |
|
101
|
|
|
} |
|
102
|
|
|
|
|
103
|
|
|
/** |
|
104
|
|
|
* Try matching "at" filter |
|
105
|
|
|
* |
|
106
|
|
|
* @param Node $item |
|
107
|
|
|
* @param integer $index |
|
108
|
|
|
* @param string $value |
|
109
|
|
|
* @return boolean |
|
110
|
|
|
*/ |
|
111
|
2 |
|
public function filter_by_at(Node $item, $index, $value) |
|
|
|
|
|
|
112
|
|
|
{ |
|
113
|
2 |
|
return $index == $value; |
|
114
|
|
|
} |
|
115
|
|
|
|
|
116
|
|
|
/** |
|
117
|
|
|
* Try matching "value" filter |
|
118
|
|
|
* |
|
119
|
|
|
* @param Node $item |
|
120
|
|
|
* @param integer $index |
|
121
|
|
|
* @param string $value |
|
122
|
|
|
* @return boolean |
|
123
|
|
|
*/ |
|
124
|
2 |
|
public function filter_by_value(Node $item, $index, $value) |
|
|
|
|
|
|
125
|
|
|
{ |
|
126
|
2 |
|
return $item->value() == $value; |
|
127
|
|
|
} |
|
128
|
|
|
|
|
129
|
|
|
/** |
|
130
|
|
|
* Try matching "text" filter |
|
131
|
|
|
* |
|
132
|
|
|
* @param Node $item |
|
133
|
|
|
* @param integer $index |
|
134
|
|
|
* @param string $value |
|
135
|
|
|
* @return boolean |
|
136
|
|
|
*/ |
|
137
|
2 |
|
public function filter_by_text(Node $item, $index, $value) |
|
|
|
|
|
|
138
|
|
|
{ |
|
139
|
2 |
|
$text = $item->text(); |
|
140
|
|
|
|
|
141
|
2 |
|
return ($text AND $value AND mb_stripos($text, $value) !== FALSE); |
|
142
|
|
|
} |
|
143
|
|
|
|
|
144
|
|
|
/** |
|
145
|
|
|
* Try matching "visible" filter |
|
146
|
|
|
* |
|
147
|
|
|
* @param Node $item |
|
148
|
|
|
* @param integer $index |
|
149
|
|
|
* @param boolean $value |
|
150
|
|
|
* @return boolean |
|
151
|
|
|
*/ |
|
152
|
1 |
|
public function filter_by_visible(Node $item, $index, $value) |
|
|
|
|
|
|
153
|
|
|
{ |
|
154
|
1 |
|
return $item->is_visible() === $value; |
|
155
|
|
|
} |
|
156
|
|
|
|
|
157
|
|
|
/** |
|
158
|
|
|
* Try matching "attributes" filter |
|
159
|
|
|
* |
|
160
|
|
|
* @param Node $item |
|
161
|
|
|
* @param integer $index |
|
162
|
|
|
* @param array $value |
|
163
|
|
|
* @return boolean |
|
164
|
|
|
*/ |
|
165
|
1 |
|
public function filter_by_attributes(Node $item, $index, array $value) |
|
|
|
|
|
|
166
|
|
|
{ |
|
167
|
1 |
|
foreach ($value as $attribute_name => $attribute_val) |
|
168
|
|
|
{ |
|
169
|
1 |
|
if ($item->attribute($attribute_name) != $attribute_val) |
|
170
|
1 |
|
return FALSE; |
|
171
|
1 |
|
} |
|
172
|
|
|
|
|
173
|
1 |
|
return TRUE; |
|
174
|
|
|
} |
|
175
|
|
|
|
|
176
|
|
|
/** |
|
177
|
|
|
* Return the xpath representaiton of the selector, using the appropriate method |
|
178
|
|
|
* @return string |
|
179
|
|
|
*/ |
|
180
|
49 |
|
public function xpath() |
|
181
|
|
|
{ |
|
182
|
49 |
|
if ( ! $this->_xpath) |
|
183
|
49 |
|
{ |
|
184
|
49 |
|
switch ($this->type()) |
|
185
|
|
|
{ |
|
186
|
49 |
|
case 'css': |
|
187
|
20 |
|
$this->_xpath = '//'.$this->convert_css_selector_to_xpath($this->selector()); |
|
188
|
20 |
|
break; |
|
189
|
|
|
|
|
190
|
35 |
|
case 'xpath': |
|
191
|
1 |
|
$this->_xpath = $this->selector(); |
|
192
|
1 |
|
break; |
|
193
|
|
|
|
|
194
|
34 |
|
case 'field': |
|
195
|
17 |
|
$this->_xpath = $this->field_to_xpath($this->selector()); |
|
196
|
17 |
|
break; |
|
197
|
|
|
|
|
198
|
21 |
|
case 'label': |
|
199
|
1 |
|
$this->_xpath = $this->label_to_xpath($this->selector()); |
|
200
|
1 |
|
break; |
|
201
|
|
|
|
|
202
|
20 |
|
case 'link': |
|
203
|
9 |
|
$this->_xpath = $this->link_to_xpath($this->selector()); |
|
204
|
9 |
|
break; |
|
205
|
|
|
|
|
206
|
14 |
|
case 'button': |
|
207
|
12 |
|
$this->_xpath = $this->button_to_xpath($this->selector()); |
|
208
|
12 |
|
break; |
|
209
|
|
|
|
|
210
|
2 |
|
default: |
|
211
|
2 |
|
throw new Exception('Locator type ":type" does not exist', array(':type' => $this->_type)); |
|
212
|
49 |
|
} |
|
213
|
47 |
|
} |
|
214
|
|
|
|
|
215
|
47 |
|
return $this->_xpath; |
|
216
|
|
|
} |
|
217
|
|
|
|
|
218
|
|
|
/** |
|
219
|
|
|
* Getter. Current type, one of css, field, xpath, label, link or button |
|
220
|
|
|
* @return string |
|
221
|
|
|
*/ |
|
222
|
58 |
|
public function type() |
|
223
|
|
|
{ |
|
224
|
58 |
|
return $this->_type; |
|
225
|
|
|
} |
|
226
|
|
|
|
|
227
|
|
|
/** |
|
228
|
|
|
* Getter. Current selector |
|
229
|
|
|
* @return string |
|
230
|
|
|
*/ |
|
231
|
49 |
|
public function selector() |
|
232
|
|
|
{ |
|
233
|
49 |
|
return $this->_selector; |
|
234
|
|
|
} |
|
235
|
|
|
|
|
236
|
|
|
/** |
|
237
|
|
|
* Getter. Current filters |
|
238
|
|
|
* @return array |
|
239
|
|
|
*/ |
|
240
|
18 |
|
public function filters() |
|
241
|
|
|
{ |
|
242
|
18 |
|
return $this->_filters; |
|
243
|
|
|
} |
|
244
|
|
|
|
|
245
|
|
|
/** |
|
246
|
|
|
* Convert field selector into xpath |
|
247
|
|
|
* |
|
248
|
|
|
* @param string $selector |
|
249
|
|
|
* @return string |
|
250
|
|
|
*/ |
|
251
|
17 |
|
public function field_to_xpath($selector) |
|
252
|
|
|
{ |
|
253
|
17 |
|
$type = "(self::input and (not(@type) or @type != 'submit')) or self::textarea or self::select"; |
|
254
|
|
|
|
|
255
|
17 |
|
$matchers['by name'] = "@name = '$selector'"; |
|
|
|
|
|
|
256
|
17 |
|
$matchers['by id'] = "@id = '$selector'"; |
|
257
|
17 |
|
$matchers['by placeholder'] = "@placeholder = '$selector'"; |
|
258
|
17 |
|
$matchers['by label for'] = "@id = //label[normalize-space() = '$selector']/@for"; |
|
259
|
17 |
|
$matchers['by option'] = "(self::select and ./option[(@value = \"\" or not(@value)) and contains(normalize-space(), \"$selector\")])"; |
|
260
|
|
|
|
|
261
|
17 |
|
return "//*[($type) and (".join(' or ', $matchers).")]"; |
|
262
|
|
|
} |
|
263
|
|
|
|
|
264
|
|
|
/** |
|
265
|
|
|
* Convert label selector to xpath |
|
266
|
|
|
* @param string $selector |
|
267
|
|
|
* @return string |
|
268
|
|
|
*/ |
|
269
|
1 |
|
public function label_to_xpath($selector) |
|
270
|
|
|
{ |
|
271
|
1 |
|
$type = "self::label"; |
|
272
|
|
|
|
|
273
|
1 |
|
$matchers['by id'] = "@id = '$selector'"; |
|
|
|
|
|
|
274
|
1 |
|
$matchers['by title'] = "contains(@title, '$selector')"; |
|
275
|
1 |
|
$matchers['by content text'] = "contains(normalize-space(), '$selector')"; |
|
276
|
1 |
|
$matchers['by img alt'] = "descendant::img[contains(@alt, '$selector')]"; |
|
277
|
|
|
|
|
278
|
1 |
|
return "//*[($type) and (".join(' or ', $matchers).")]"; |
|
279
|
|
|
} |
|
280
|
|
|
|
|
281
|
|
|
/** |
|
282
|
|
|
* Convert link selector to xpath |
|
283
|
|
|
* @param string $selector |
|
284
|
|
|
* @return string |
|
285
|
|
|
*/ |
|
286
|
9 |
|
public function link_to_xpath($selector) |
|
287
|
|
|
{ |
|
288
|
9 |
|
$matchers['by title'] = "contains(@title, '$selector')"; |
|
|
|
|
|
|
289
|
9 |
|
$matchers['by id'] = "@id = '$selector'"; |
|
290
|
9 |
|
$matchers['by content text'] = "contains(normalize-space(), '$selector')"; |
|
291
|
9 |
|
$matchers['by img alt'] = "descendant::img[contains(@alt, '$selector')]"; |
|
292
|
|
|
|
|
293
|
9 |
|
return "//a[".join(' or ', $matchers)."]"; |
|
294
|
|
|
} |
|
295
|
|
|
|
|
296
|
|
|
/** |
|
297
|
|
|
* Convert button selector to xpath |
|
298
|
|
|
* @param string $selector |
|
299
|
|
|
* @return string |
|
300
|
|
|
*/ |
|
301
|
12 |
|
public function button_to_xpath($selector) |
|
302
|
|
|
{ |
|
303
|
12 |
|
$type = "(self::input and @type = 'submit') or self::button"; |
|
304
|
|
|
|
|
305
|
12 |
|
$matchers['by title'] = "contains(@title, '$selector')"; |
|
|
|
|
|
|
306
|
12 |
|
$matchers['by id'] = "@id = '$selector'"; |
|
307
|
12 |
|
$matchers['by content text'] = "contains(normalize-space(), '$selector')"; |
|
308
|
12 |
|
$matchers['by img alt'] = "descendant::img[contains(@alt, '$selector')]"; |
|
309
|
12 |
|
$matchers['by value'] = "contains(@value, '$selector')"; |
|
310
|
12 |
|
$matchers['by name'] = "@name = '$selector'"; |
|
311
|
|
|
|
|
312
|
12 |
|
return "//*[($type) and (".join(' or ', $matchers).")]"; |
|
313
|
|
|
} |
|
314
|
|
|
|
|
315
|
|
|
/** |
|
316
|
|
|
* Return a pretty printed representation of the locator |
|
317
|
|
|
* @return string |
|
318
|
|
|
*/ |
|
319
|
2 |
|
public function __toString() |
|
320
|
|
|
{ |
|
321
|
2 |
|
if ($this->filters()) |
|
322
|
2 |
|
{ |
|
323
|
1 |
|
$filters = " filters: ".json_encode($this->filters()); |
|
324
|
1 |
|
} |
|
325
|
|
|
else |
|
326
|
|
|
{ |
|
327
|
1 |
|
$filters = ''; |
|
328
|
|
|
} |
|
329
|
|
|
|
|
330
|
2 |
|
return "Locator: ({$this->type()}) {$this->selector()}".$filters; |
|
331
|
|
|
} |
|
332
|
|
|
|
|
333
|
|
|
/** |
|
334
|
|
|
* Convert a CSS selector to XPath. |
|
335
|
|
|
* Uses Symfony CSS Selector 2.8 and above API if possible. |
|
336
|
|
|
* Otherwise fallback to pre-2.8 static interface. |
|
337
|
|
|
* |
|
338
|
|
|
* @param string $css_selector CSS selector |
|
339
|
|
|
* @return string XPath selector |
|
340
|
|
|
*/ |
|
341
|
20 |
|
private function convert_css_selector_to_xpath($css_selector) |
|
|
|
|
|
|
342
|
|
|
{ |
|
343
|
20 |
|
if (class_exists('Symfony\Component\CssSelector\CssSelectorConverter')) { |
|
344
|
|
|
$converter = new CssSelector\CssSelectorConverter(); |
|
345
|
|
|
return $converter->toXPath($this->selector()); |
|
346
|
|
|
} |
|
347
|
|
|
|
|
348
|
20 |
|
return CssSelector\CssSelector::toXPath($this->selector()); |
|
349
|
|
|
} |
|
350
|
|
|
} |
|
351
|
|
|
|
Adding explicit visibility (
private,protected, orpublic) is generally recommend to communicate to other developers how, and from where this method is intended to be used.