| 1 |
lars |
1 |
<?php
|
|
|
2 |
/**
|
|
|
3 |
* base include file for SimpleTest
|
|
|
4 |
* @package SimpleTest
|
|
|
5 |
* @subpackage MockObjects
|
|
|
6 |
* @version $Id: parser.php 1532 2006-12-01 12:28:55Z xue $
|
|
|
7 |
*/
|
|
|
8 |
|
|
|
9 |
/**#@+
|
|
|
10 |
* Lexer mode stack constants
|
|
|
11 |
*/
|
|
|
12 |
if (! defined('LEXER_ENTER')) {
|
|
|
13 |
define('LEXER_ENTER', 1);
|
|
|
14 |
}
|
|
|
15 |
if (! defined('LEXER_MATCHED')) {
|
|
|
16 |
define('LEXER_MATCHED', 2);
|
|
|
17 |
}
|
|
|
18 |
if (! defined('LEXER_UNMATCHED')) {
|
|
|
19 |
define('LEXER_UNMATCHED', 3);
|
|
|
20 |
}
|
|
|
21 |
if (! defined('LEXER_EXIT')) {
|
|
|
22 |
define('LEXER_EXIT', 4);
|
|
|
23 |
}
|
|
|
24 |
if (! defined('LEXER_SPECIAL')) {
|
|
|
25 |
define('LEXER_SPECIAL', 5);
|
|
|
26 |
}
|
|
|
27 |
/**#@-*/
|
|
|
28 |
|
|
|
29 |
/**
|
|
|
30 |
* Compounded regular expression. Any of
|
|
|
31 |
* the contained patterns could match and
|
|
|
32 |
* when one does, it's label is returned.
|
|
|
33 |
* @package SimpleTest
|
|
|
34 |
* @subpackage WebTester
|
|
|
35 |
*/
|
|
|
36 |
class ParallelRegex {
|
|
|
37 |
protected $_patterns;
|
|
|
38 |
protected $_labels;
|
|
|
39 |
protected $_regex;
|
|
|
40 |
protected $_case;
|
|
|
41 |
|
|
|
42 |
/**
|
|
|
43 |
* Constructor. Starts with no patterns.
|
|
|
44 |
* @param boolean $case True for case sensitive, false
|
|
|
45 |
* for insensitive.
|
|
|
46 |
* @access public
|
|
|
47 |
*/
|
|
|
48 |
function ParallelRegex($case) {
|
|
|
49 |
$this->_case = $case;
|
|
|
50 |
$this->_patterns = array();
|
|
|
51 |
$this->_labels = array();
|
|
|
52 |
$this->_regex = null;
|
|
|
53 |
}
|
|
|
54 |
|
|
|
55 |
/**
|
|
|
56 |
* Adds a pattern with an optional label.
|
|
|
57 |
* @param string $pattern Perl style regex, but ( and )
|
|
|
58 |
* lose the usual meaning.
|
|
|
59 |
* @param string $label Label of regex to be returned
|
|
|
60 |
* on a match.
|
|
|
61 |
* @access public
|
|
|
62 |
*/
|
|
|
63 |
function addPattern($pattern, $label = true) {
|
|
|
64 |
$count = count($this->_patterns);
|
|
|
65 |
$this->_patterns[$count] = $pattern;
|
|
|
66 |
$this->_labels[$count] = $label;
|
|
|
67 |
$this->_regex = null;
|
|
|
68 |
}
|
|
|
69 |
|
|
|
70 |
/**
|
|
|
71 |
* Attempts to match all patterns at once against
|
|
|
72 |
* a string.
|
|
|
73 |
* @param string $subject String to match against.
|
|
|
74 |
* @param string $match First matched portion of
|
|
|
75 |
* subject.
|
|
|
76 |
* @return boolean True on success.
|
|
|
77 |
* @access public
|
|
|
78 |
*/
|
|
|
79 |
function match($subject, $match) {
|
|
|
80 |
if (count($this->_patterns) == 0) {
|
|
|
81 |
return false;
|
|
|
82 |
}
|
|
|
83 |
if (! preg_match($this->_getCompoundedRegex(), $subject, $matches)) {
|
|
|
84 |
$match = '';
|
|
|
85 |
return false;
|
|
|
86 |
}
|
|
|
87 |
$match = $matches[0];
|
|
|
88 |
for ($i = 1; $i < count($matches); $i++) {
|
|
|
89 |
if ($matches[$i]) {
|
|
|
90 |
return $this->_labels[$i - 1];
|
|
|
91 |
}
|
|
|
92 |
}
|
|
|
93 |
return true;
|
|
|
94 |
}
|
|
|
95 |
|
|
|
96 |
/**
|
|
|
97 |
* Compounds the patterns into a single
|
|
|
98 |
* regular expression separated with the
|
|
|
99 |
* "or" operator. Caches the regex.
|
|
|
100 |
* Will automatically escape (, ) and / tokens.
|
|
|
101 |
* @param array $patterns List of patterns in order.
|
|
|
102 |
* @access private
|
|
|
103 |
*/
|
|
|
104 |
function _getCompoundedRegex() {
|
|
|
105 |
if ($this->_regex == null) {
|
|
|
106 |
for ($i = 0, $count = count($this->_patterns); $i < $count; $i++) {
|
|
|
107 |
$this->_patterns[$i] = '(' . str_replace(
|
|
|
108 |
array('/', '(', ')'),
|
|
|
109 |
array('\/', '\(', '\)'),
|
|
|
110 |
$this->_patterns[$i]) . ')';
|
|
|
111 |
}
|
|
|
112 |
$this->_regex = "/" . implode("|", $this->_patterns) . "/" . $this->_getPerlMatchingFlags();
|
|
|
113 |
}
|
|
|
114 |
return $this->_regex;
|
|
|
115 |
}
|
|
|
116 |
|
|
|
117 |
/**
|
|
|
118 |
* Accessor for perl regex mode flags to use.
|
|
|
119 |
* @return string Perl regex flags.
|
|
|
120 |
* @access private
|
|
|
121 |
*/
|
|
|
122 |
function _getPerlMatchingFlags() {
|
|
|
123 |
return ($this->_case ? "msS" : "msSi");
|
|
|
124 |
}
|
|
|
125 |
}
|
|
|
126 |
|
|
|
127 |
/**
|
|
|
128 |
* States for a stack machine.
|
|
|
129 |
* @package SimpleTest
|
|
|
130 |
* @subpackage WebTester
|
|
|
131 |
*/
|
|
|
132 |
class SimpleStateStack {
|
|
|
133 |
protected $_stack;
|
|
|
134 |
|
|
|
135 |
/**
|
|
|
136 |
* Constructor. Starts in named state.
|
|
|
137 |
* @param string $start Starting state name.
|
|
|
138 |
* @access public
|
|
|
139 |
*/
|
|
|
140 |
function SimpleStateStack($start) {
|
|
|
141 |
$this->_stack = array($start);
|
|
|
142 |
}
|
|
|
143 |
|
|
|
144 |
/**
|
|
|
145 |
* Accessor for current state.
|
|
|
146 |
* @return string State.
|
|
|
147 |
* @access public
|
|
|
148 |
*/
|
|
|
149 |
function getCurrent() {
|
|
|
150 |
return $this->_stack[count($this->_stack) - 1];
|
|
|
151 |
}
|
|
|
152 |
|
|
|
153 |
/**
|
|
|
154 |
* Adds a state to the stack and sets it
|
|
|
155 |
* to be the current state.
|
|
|
156 |
* @param string $state New state.
|
|
|
157 |
* @access public
|
|
|
158 |
*/
|
|
|
159 |
function enter($state) {
|
|
|
160 |
array_push($this->_stack, $state);
|
|
|
161 |
}
|
|
|
162 |
|
|
|
163 |
/**
|
|
|
164 |
* Leaves the current state and reverts
|
|
|
165 |
* to the previous one.
|
|
|
166 |
* @return boolean False if we drop off
|
|
|
167 |
* the bottom of the list.
|
|
|
168 |
* @access public
|
|
|
169 |
*/
|
|
|
170 |
function leave() {
|
|
|
171 |
if (count($this->_stack) == 1) {
|
|
|
172 |
return false;
|
|
|
173 |
}
|
|
|
174 |
array_pop($this->_stack);
|
|
|
175 |
return true;
|
|
|
176 |
}
|
|
|
177 |
}
|
|
|
178 |
|
|
|
179 |
/**
|
|
|
180 |
* Accepts text and breaks it into tokens.
|
|
|
181 |
* Some optimisation to make the sure the
|
|
|
182 |
* content is only scanned by the PHP regex
|
|
|
183 |
* parser once. Lexer modes must not start
|
|
|
184 |
* with leading underscores.
|
|
|
185 |
* @package SimpleTest
|
|
|
186 |
* @subpackage WebTester
|
|
|
187 |
*/
|
|
|
188 |
class SimpleLexer {
|
|
|
189 |
protected $_regexes;
|
|
|
190 |
protected $_parser;
|
|
|
191 |
protected $_mode;
|
|
|
192 |
protected $_mode_handlers;
|
|
|
193 |
protected $_case;
|
|
|
194 |
|
|
|
195 |
/**
|
|
|
196 |
* Sets up the lexer in case insensitive matching
|
|
|
197 |
* by default.
|
|
|
198 |
* @param SimpleSaxParser $parser Handling strategy by
|
|
|
199 |
* reference.
|
|
|
200 |
* @param string $start Starting handler.
|
|
|
201 |
* @param boolean $case True for case sensitive.
|
|
|
202 |
* @access public
|
|
|
203 |
*/
|
|
|
204 |
function SimpleLexer($parser, $start = "accept", $case = false) {
|
|
|
205 |
$this->_case = $case;
|
|
|
206 |
$this->_regexes = array();
|
|
|
207 |
$this->_parser = $parser;
|
|
|
208 |
$this->_mode = new SimpleStateStack($start);
|
|
|
209 |
$this->_mode_handlers = array($start => $start);
|
|
|
210 |
}
|
|
|
211 |
|
|
|
212 |
/**
|
|
|
213 |
* Adds a token search pattern for a particular
|
|
|
214 |
* parsing mode. The pattern does not change the
|
|
|
215 |
* current mode.
|
|
|
216 |
* @param string $pattern Perl style regex, but ( and )
|
|
|
217 |
* lose the usual meaning.
|
|
|
218 |
* @param string $mode Should only apply this
|
|
|
219 |
* pattern when dealing with
|
|
|
220 |
* this type of input.
|
|
|
221 |
* @access public
|
|
|
222 |
*/
|
|
|
223 |
function addPattern($pattern, $mode = "accept") {
|
|
|
224 |
if (! isset($this->_regexes[$mode])) {
|
|
|
225 |
$this->_regexes[$mode] = new ParallelRegex($this->_case);
|
|
|
226 |
}
|
|
|
227 |
$this->_regexes[$mode]->addPattern($pattern);
|
|
|
228 |
if (! isset($this->_mode_handlers[$mode])) {
|
|
|
229 |
$this->_mode_handlers[$mode] = $mode;
|
|
|
230 |
}
|
|
|
231 |
}
|
|
|
232 |
|
|
|
233 |
/**
|
|
|
234 |
* Adds a pattern that will enter a new parsing
|
|
|
235 |
* mode. Useful for entering parenthesis, strings,
|
|
|
236 |
* tags, etc.
|
|
|
237 |
* @param string $pattern Perl style regex, but ( and )
|
|
|
238 |
* lose the usual meaning.
|
|
|
239 |
* @param string $mode Should only apply this
|
|
|
240 |
* pattern when dealing with
|
|
|
241 |
* this type of input.
|
|
|
242 |
* @param string $new_mode Change parsing to this new
|
|
|
243 |
* nested mode.
|
|
|
244 |
* @access public
|
|
|
245 |
*/
|
|
|
246 |
function addEntryPattern($pattern, $mode, $new_mode) {
|
|
|
247 |
if (! isset($this->_regexes[$mode])) {
|
|
|
248 |
$this->_regexes[$mode] = new ParallelRegex($this->_case);
|
|
|
249 |
}
|
|
|
250 |
$this->_regexes[$mode]->addPattern($pattern, $new_mode);
|
|
|
251 |
if (! isset($this->_mode_handlers[$new_mode])) {
|
|
|
252 |
$this->_mode_handlers[$new_mode] = $new_mode;
|
|
|
253 |
}
|
|
|
254 |
}
|
|
|
255 |
|
|
|
256 |
/**
|
|
|
257 |
* Adds a pattern that will exit the current mode
|
|
|
258 |
* and re-enter the previous one.
|
|
|
259 |
* @param string $pattern Perl style regex, but ( and )
|
|
|
260 |
* lose the usual meaning.
|
|
|
261 |
* @param string $mode Mode to leave.
|
|
|
262 |
* @access public
|
|
|
263 |
*/
|
|
|
264 |
function addExitPattern($pattern, $mode) {
|
|
|
265 |
if (! isset($this->_regexes[$mode])) {
|
|
|
266 |
$this->_regexes[$mode] = new ParallelRegex($this->_case);
|
|
|
267 |
}
|
|
|
268 |
$this->_regexes[$mode]->addPattern($pattern, "__exit");
|
|
|
269 |
if (! isset($this->_mode_handlers[$mode])) {
|
|
|
270 |
$this->_mode_handlers[$mode] = $mode;
|
|
|
271 |
}
|
|
|
272 |
}
|
|
|
273 |
|
|
|
274 |
/**
|
|
|
275 |
* Adds a pattern that has a special mode. Acts as an entry
|
|
|
276 |
* and exit pattern in one go, effectively calling a special
|
|
|
277 |
* parser handler for this token only.
|
|
|
278 |
* @param string $pattern Perl style regex, but ( and )
|
|
|
279 |
* lose the usual meaning.
|
|
|
280 |
* @param string $mode Should only apply this
|
|
|
281 |
* pattern when dealing with
|
|
|
282 |
* this type of input.
|
|
|
283 |
* @param string $special Use this mode for this one token.
|
|
|
284 |
* @access public
|
|
|
285 |
*/
|
|
|
286 |
function addSpecialPattern($pattern, $mode, $special) {
|
|
|
287 |
if (! isset($this->_regexes[$mode])) {
|
|
|
288 |
$this->_regexes[$mode] = new ParallelRegex($this->_case);
|
|
|
289 |
}
|
|
|
290 |
$this->_regexes[$mode]->addPattern($pattern, "_$special");
|
|
|
291 |
if (! isset($this->_mode_handlers[$special])) {
|
|
|
292 |
$this->_mode_handlers[$special] = $special;
|
|
|
293 |
}
|
|
|
294 |
}
|
|
|
295 |
|
|
|
296 |
/**
|
|
|
297 |
* Adds a mapping from a mode to another handler.
|
|
|
298 |
* @param string $mode Mode to be remapped.
|
|
|
299 |
* @param string $handler New target handler.
|
|
|
300 |
* @access public
|
|
|
301 |
*/
|
|
|
302 |
function mapHandler($mode, $handler) {
|
|
|
303 |
$this->_mode_handlers[$mode] = $handler;
|
|
|
304 |
}
|
|
|
305 |
|
|
|
306 |
/**
|
|
|
307 |
* Splits the page text into tokens. Will fail
|
|
|
308 |
* if the handlers report an error or if no
|
|
|
309 |
* content is consumed. If successful then each
|
|
|
310 |
* unparsed and parsed token invokes a call to the
|
|
|
311 |
* held listener.
|
|
|
312 |
* @param string $raw Raw HTML text.
|
|
|
313 |
* @return boolean True on success, else false.
|
|
|
314 |
* @access public
|
|
|
315 |
*/
|
|
|
316 |
function parse($raw) {
|
|
|
317 |
if (! isset($this->_parser)) {
|
|
|
318 |
return false;
|
|
|
319 |
}
|
|
|
320 |
$length = strlen($raw);
|
|
|
321 |
while (is_array($parsed = $this->_reduce($raw))) {
|
|
|
322 |
list($raw, $unmatched, $matched, $mode) = $parsed;
|
|
|
323 |
if (! $this->_dispatchTokens($unmatched, $matched, $mode)) {
|
|
|
324 |
return false;
|
|
|
325 |
}
|
|
|
326 |
if ($raw === '') {
|
|
|
327 |
return true;
|
|
|
328 |
}
|
|
|
329 |
if (strlen($raw) == $length) {
|
|
|
330 |
return false;
|
|
|
331 |
}
|
|
|
332 |
$length = strlen($raw);
|
|
|
333 |
}
|
|
|
334 |
if (! $parsed) {
|
|
|
335 |
return false;
|
|
|
336 |
}
|
|
|
337 |
return $this->_invokeParser($raw, LEXER_UNMATCHED);
|
|
|
338 |
}
|
|
|
339 |
|
|
|
340 |
/**
|
|
|
341 |
* Sends the matched token and any leading unmatched
|
|
|
342 |
* text to the parser changing the lexer to a new
|
|
|
343 |
* mode if one is listed.
|
|
|
344 |
* @param string $unmatched Unmatched leading portion.
|
|
|
345 |
* @param string $matched Actual token match.
|
|
|
346 |
* @param string $mode Mode after match. A boolean
|
|
|
347 |
* false mode causes no change.
|
|
|
348 |
* @return boolean False if there was any error
|
|
|
349 |
* from the parser.
|
|
|
350 |
* @access private
|
|
|
351 |
*/
|
|
|
352 |
function _dispatchTokens($unmatched, $matched, $mode = false) {
|
|
|
353 |
if (! $this->_invokeParser($unmatched, LEXER_UNMATCHED)) {
|
|
|
354 |
return false;
|
|
|
355 |
}
|
|
|
356 |
if (is_bool($mode)) {
|
|
|
357 |
return $this->_invokeParser($matched, LEXER_MATCHED);
|
|
|
358 |
}
|
|
|
359 |
if ($this->_isModeEnd($mode)) {
|
|
|
360 |
if (! $this->_invokeParser($matched, LEXER_EXIT)) {
|
|
|
361 |
return false;
|
|
|
362 |
}
|
|
|
363 |
return $this->_mode->leave();
|
|
|
364 |
}
|
|
|
365 |
if ($this->_isSpecialMode($mode)) {
|
|
|
366 |
$this->_mode->enter($this->_decodeSpecial($mode));
|
|
|
367 |
if (! $this->_invokeParser($matched, LEXER_SPECIAL)) {
|
|
|
368 |
return false;
|
|
|
369 |
}
|
|
|
370 |
return $this->_mode->leave();
|
|
|
371 |
}
|
|
|
372 |
$this->_mode->enter($mode);
|
|
|
373 |
return $this->_invokeParser($matched, LEXER_ENTER);
|
|
|
374 |
}
|
|
|
375 |
|
|
|
376 |
/**
|
|
|
377 |
* Tests to see if the new mode is actually to leave
|
|
|
378 |
* the current mode and pop an item from the matching
|
|
|
379 |
* mode stack.
|
|
|
380 |
* @param string $mode Mode to test.
|
|
|
381 |
* @return boolean True if this is the exit mode.
|
|
|
382 |
* @access private
|
|
|
383 |
*/
|
|
|
384 |
function _isModeEnd($mode) {
|
|
|
385 |
return ($mode === "__exit");
|
|
|
386 |
}
|
|
|
387 |
|
|
|
388 |
/**
|
|
|
389 |
* Test to see if the mode is one where this mode
|
|
|
390 |
* is entered for this token only and automatically
|
|
|
391 |
* leaves immediately afterwoods.
|
|
|
392 |
* @param string $mode Mode to test.
|
|
|
393 |
* @return boolean True if this is the exit mode.
|
|
|
394 |
* @access private
|
|
|
395 |
*/
|
|
|
396 |
function _isSpecialMode($mode) {
|
|
|
397 |
return (strncmp($mode, "_", 1) == 0);
|
|
|
398 |
}
|
|
|
399 |
|
|
|
400 |
/**
|
|
|
401 |
* Strips the magic underscore marking single token
|
|
|
402 |
* modes.
|
|
|
403 |
* @param string $mode Mode to decode.
|
|
|
404 |
* @return string Underlying mode name.
|
|
|
405 |
* @access private
|
|
|
406 |
*/
|
|
|
407 |
function _decodeSpecial($mode) {
|
|
|
408 |
return substr($mode, 1);
|
|
|
409 |
}
|
|
|
410 |
|
|
|
411 |
/**
|
|
|
412 |
* Calls the parser method named after the current
|
|
|
413 |
* mode. Empty content will be ignored. The lexer
|
|
|
414 |
* has a parser handler for each mode in the lexer.
|
|
|
415 |
* @param string $content Text parsed.
|
|
|
416 |
* @param boolean $is_match Token is recognised rather
|
|
|
417 |
* than unparsed data.
|
|
|
418 |
* @access private
|
|
|
419 |
*/
|
|
|
420 |
function _invokeParser($content, $is_match) {
|
|
|
421 |
if (($content === '') || ($content === false)) {
|
|
|
422 |
return true;
|
|
|
423 |
}
|
|
|
424 |
$handler = $this->_mode_handlers[$this->_mode->getCurrent()];
|
|
|
425 |
return $this->_parser->$handler($content, $is_match);
|
|
|
426 |
}
|
|
|
427 |
|
|
|
428 |
/**
|
|
|
429 |
* Tries to match a chunk of text and if successful
|
|
|
430 |
* removes the recognised chunk and any leading
|
|
|
431 |
* unparsed data. Empty strings will not be matched.
|
|
|
432 |
* @param string $raw The subject to parse. This is the
|
|
|
433 |
* content that will be eaten.
|
|
|
434 |
* @return array/boolean Three item list of unparsed
|
|
|
435 |
* content followed by the
|
|
|
436 |
* recognised token and finally the
|
|
|
437 |
* action the parser is to take.
|
|
|
438 |
* True if no match, false if there
|
|
|
439 |
* is a parsing error.
|
|
|
440 |
* @access private
|
|
|
441 |
*/
|
|
|
442 |
function _reduce($raw) {
|
|
|
443 |
if ($action = $this->_regexes[$this->_mode->getCurrent()]->match($raw, $match)) {
|
|
|
444 |
$unparsed_character_count = strpos($raw, $match);
|
|
|
445 |
$unparsed = substr($raw, 0, $unparsed_character_count);
|
|
|
446 |
$raw = substr($raw, $unparsed_character_count + strlen($match));
|
|
|
447 |
return array($raw, $unparsed, $match, $action);
|
|
|
448 |
}
|
|
|
449 |
return true;
|
|
|
450 |
}
|
|
|
451 |
}
|
|
|
452 |
|
|
|
453 |
/**
|
|
|
454 |
* Breas HTML into SAX events.
|
|
|
455 |
* @package SimpleTest
|
|
|
456 |
* @subpackage WebTester
|
|
|
457 |
*/
|
|
|
458 |
class SimpleHtmlLexer extends SimpleLexer {
|
|
|
459 |
|
|
|
460 |
/**
|
|
|
461 |
* Sets up the lexer with case insensitive matching
|
|
|
462 |
* and adds the HTML handlers.
|
|
|
463 |
* @param SimpleSaxParser $parser Handling strategy by
|
|
|
464 |
* reference.
|
|
|
465 |
* @access public
|
|
|
466 |
*/
|
|
|
467 |
function SimpleHtmlLexer($parser) {
|
|
|
468 |
$this->SimpleLexer($parser, 'text');
|
|
|
469 |
$this->mapHandler('text', 'acceptTextToken');
|
|
|
470 |
$this->_addSkipping();
|
|
|
471 |
foreach ($this->_getParsedTags() as $tag) {
|
|
|
472 |
$this->_addTag($tag);
|
|
|
473 |
}
|
|
|
474 |
$this->_addInTagTokens();
|
|
|
475 |
}
|
|
|
476 |
|
|
|
477 |
/**
|
|
|
478 |
* List of parsed tags. Others are ignored.
|
|
|
479 |
* @return array List of searched for tags.
|
|
|
480 |
* @access private
|
|
|
481 |
*/
|
|
|
482 |
function _getParsedTags() {
|
|
|
483 |
return array('a', 'title', 'form', 'input', 'button', 'textarea', 'select',
|
|
|
484 |
'option', 'frameset', 'frame', 'label');
|
|
|
485 |
}
|
|
|
486 |
|
|
|
487 |
/**
|
|
|
488 |
* The lexer has to skip certain sections such
|
|
|
489 |
* as server code, client code and styles.
|
|
|
490 |
* @access private
|
|
|
491 |
*/
|
|
|
492 |
function _addSkipping() {
|
|
|
493 |
$this->mapHandler('css', 'ignore');
|
|
|
494 |
$this->addEntryPattern('<style', 'text', 'css');
|
|
|
495 |
$this->addExitPattern('</style>', 'css');
|
|
|
496 |
$this->mapHandler('js', 'ignore');
|
|
|
497 |
$this->addEntryPattern('<script', 'text', 'js');
|
|
|
498 |
$this->addExitPattern('</script>', 'js');
|
|
|
499 |
$this->mapHandler('comment', 'ignore');
|
|
|
500 |
$this->addEntryPattern('<!--', 'text', 'comment');
|
|
|
501 |
$this->addExitPattern('-->', 'comment');
|
|
|
502 |
}
|
|
|
503 |
|
|
|
504 |
/**
|
|
|
505 |
* Pattern matches to start and end a tag.
|
|
|
506 |
* @param string $tag Name of tag to scan for.
|
|
|
507 |
* @access private
|
|
|
508 |
*/
|
|
|
509 |
function _addTag($tag) {
|
|
|
510 |
$this->addSpecialPattern("</$tag>", 'text', 'acceptEndToken');
|
|
|
511 |
$this->addEntryPattern("<$tag", 'text', 'tag');
|
|
|
512 |
}
|
|
|
513 |
|
|
|
514 |
/**
|
|
|
515 |
* Pattern matches to parse the inside of a tag
|
|
|
516 |
* including the attributes and their quoting.
|
|
|
517 |
* @access private
|
|
|
518 |
*/
|
|
|
519 |
function _addInTagTokens() {
|
|
|
520 |
$this->mapHandler('tag', 'acceptStartToken');
|
|
|
521 |
$this->addSpecialPattern('\s+', 'tag', 'ignore');
|
|
|
522 |
$this->_addAttributeTokens();
|
|
|
523 |
$this->addExitPattern('/>', 'tag');
|
|
|
524 |
$this->addExitPattern('>', 'tag');
|
|
|
525 |
}
|
|
|
526 |
|
|
|
527 |
/**
|
|
|
528 |
* Matches attributes that are either single quoted,
|
|
|
529 |
* double quoted or unquoted.
|
|
|
530 |
* @access private
|
|
|
531 |
*/
|
|
|
532 |
function _addAttributeTokens() {
|
|
|
533 |
$this->mapHandler('dq_attribute', 'acceptAttributeToken');
|
|
|
534 |
$this->addEntryPattern('=\s*"', 'tag', 'dq_attribute');
|
|
|
535 |
$this->addPattern("\\\\\"", 'dq_attribute');
|
|
|
536 |
$this->addExitPattern('"', 'dq_attribute');
|
|
|
537 |
$this->mapHandler('sq_attribute', 'acceptAttributeToken');
|
|
|
538 |
$this->addEntryPattern("=\s*'", 'tag', 'sq_attribute');
|
|
|
539 |
$this->addPattern("\\\\'", 'sq_attribute');
|
|
|
540 |
$this->addExitPattern("'", 'sq_attribute');
|
|
|
541 |
$this->mapHandler('uq_attribute', 'acceptAttributeToken');
|
|
|
542 |
$this->addSpecialPattern('=\s*[^>\s]*', 'tag', 'uq_attribute');
|
|
|
543 |
}
|
|
|
544 |
}
|
|
|
545 |
|
|
|
546 |
/**
|
|
|
547 |
* Converts HTML tokens into selected SAX events.
|
|
|
548 |
* @package SimpleTest
|
|
|
549 |
* @subpackage WebTester
|
|
|
550 |
*/
|
|
|
551 |
class SimpleHtmlSaxParser {
|
|
|
552 |
protected $_lexer;
|
|
|
553 |
protected $_listener;
|
|
|
554 |
protected $_tag;
|
|
|
555 |
protected $_attributes;
|
|
|
556 |
protected $_current_attribute;
|
|
|
557 |
|
|
|
558 |
/**
|
|
|
559 |
* Sets the listener.
|
|
|
560 |
* @param SimpleSaxListener $listener SAX event handler.
|
|
|
561 |
* @access public
|
|
|
562 |
*/
|
|
|
563 |
function SimpleHtmlSaxParser($listener) {
|
|
|
564 |
$this->_listener = $listener;
|
|
|
565 |
$this->_lexer = $this->createLexer($this);
|
|
|
566 |
$this->_tag = '';
|
|
|
567 |
$this->_attributes = array();
|
|
|
568 |
$this->_current_attribute = '';
|
|
|
569 |
}
|
|
|
570 |
|
|
|
571 |
/**
|
|
|
572 |
* Runs the content through the lexer which
|
|
|
573 |
* should call back to the acceptors.
|
|
|
574 |
* @param string $raw Page text to parse.
|
|
|
575 |
* @return boolean False if parse error.
|
|
|
576 |
* @access public
|
|
|
577 |
*/
|
|
|
578 |
function parse($raw) {
|
|
|
579 |
return $this->_lexer->parse($raw);
|
|
|
580 |
}
|
|
|
581 |
|
|
|
582 |
/**
|
|
|
583 |
* Sets up the matching lexer. Starts in 'text' mode.
|
|
|
584 |
* @param SimpleSaxParser $parser Event generator, usually $self.
|
|
|
585 |
* @return SimpleLexer Lexer suitable for this parser.
|
|
|
586 |
* @access public
|
|
|
587 |
* @static
|
|
|
588 |
*/
|
|
|
589 |
static function &createLexer($parser) {
|
|
|
590 |
$lexer = new SimpleHtmlLexer($parser);
|
|
|
591 |
return $lexer;
|
|
|
592 |
}
|
|
|
593 |
|
|
|
594 |
/**
|
|
|
595 |
* Accepts a token from the tag mode. If the
|
|
|
596 |
* starting element completes then the element
|
|
|
597 |
* is dispatched and the current attributes
|
|
|
598 |
* set back to empty. The element or attribute
|
|
|
599 |
* name is converted to lower case.
|
|
|
600 |
* @param string $token Incoming characters.
|
|
|
601 |
* @param integer $event Lexer event type.
|
|
|
602 |
* @return boolean False if parse error.
|
|
|
603 |
* @access public
|
|
|
604 |
*/
|
|
|
605 |
function acceptStartToken($token, $event) {
|
|
|
606 |
if ($event == LEXER_ENTER) {
|
|
|
607 |
$this->_tag = strtolower(substr($token, 1));
|
|
|
608 |
return true;
|
|
|
609 |
}
|
|
|
610 |
if ($event == LEXER_EXIT) {
|
|
|
611 |
$success = $this->_listener->startElement(
|
|
|
612 |
$this->_tag,
|
|
|
613 |
$this->_attributes);
|
|
|
614 |
$this->_tag = '';
|
|
|
615 |
$this->_attributes = array();
|
|
|
616 |
return $success;
|
|
|
617 |
}
|
|
|
618 |
if ($token != '=') {
|
|
|
619 |
$this->_current_attribute = strtolower(SimpleHtmlSaxParser::decodeHtml($token));
|
|
|
620 |
$this->_attributes[$this->_current_attribute] = '';
|
|
|
621 |
}
|
|
|
622 |
return true;
|
|
|
623 |
}
|
|
|
624 |
|
|
|
625 |
/**
|
|
|
626 |
* Accepts a token from the end tag mode.
|
|
|
627 |
* The element name is converted to lower case.
|
|
|
628 |
* @param string $token Incoming characters.
|
|
|
629 |
* @param integer $event Lexer event type.
|
|
|
630 |
* @return boolean False if parse error.
|
|
|
631 |
* @access public
|
|
|
632 |
*/
|
|
|
633 |
function acceptEndToken($token, $event) {
|
|
|
634 |
if (! preg_match('/<\/(.*)>/', $token, $matches)) {
|
|
|
635 |
return false;
|
|
|
636 |
}
|
|
|
637 |
return $this->_listener->endElement(strtolower($matches[1]));
|
|
|
638 |
}
|
|
|
639 |
|
|
|
640 |
/**
|
|
|
641 |
* Part of the tag data.
|
|
|
642 |
* @param string $token Incoming characters.
|
|
|
643 |
* @param integer $event Lexer event type.
|
|
|
644 |
* @return boolean False if parse error.
|
|
|
645 |
* @access public
|
|
|
646 |
*/
|
|
|
647 |
function acceptAttributeToken($token, $event) {
|
|
|
648 |
if ($event == LEXER_UNMATCHED) {
|
|
|
649 |
$this->_attributes[$this->_current_attribute] .=
|
|
|
650 |
SimpleHtmlSaxParser::decodeHtml($token);
|
|
|
651 |
}
|
|
|
652 |
if ($event == LEXER_SPECIAL) {
|
|
|
653 |
$this->_attributes[$this->_current_attribute] .=
|
|
|
654 |
preg_replace('/^=\s*/' , '', SimpleHtmlSaxParser::decodeHtml($token));
|
|
|
655 |
}
|
|
|
656 |
return true;
|
|
|
657 |
}
|
|
|
658 |
|
|
|
659 |
/**
|
|
|
660 |
* A character entity.
|
|
|
661 |
* @param string $token Incoming characters.
|
|
|
662 |
* @param integer $event Lexer event type.
|
|
|
663 |
* @return boolean False if parse error.
|
|
|
664 |
* @access public
|
|
|
665 |
*/
|
|
|
666 |
function acceptEntityToken($token, $event) {
|
|
|
667 |
}
|
|
|
668 |
|
|
|
669 |
/**
|
|
|
670 |
* Character data between tags regarded as
|
|
|
671 |
* important.
|
|
|
672 |
* @param string $token Incoming characters.
|
|
|
673 |
* @param integer $event Lexer event type.
|
|
|
674 |
* @return boolean False if parse error.
|
|
|
675 |
* @access public
|
|
|
676 |
*/
|
|
|
677 |
function acceptTextToken($token, $event) {
|
|
|
678 |
return $this->_listener->addContent($token);
|
|
|
679 |
}
|
|
|
680 |
|
|
|
681 |
/**
|
|
|
682 |
* Incoming data to be ignored.
|
|
|
683 |
* @param string $token Incoming characters.
|
|
|
684 |
* @param integer $event Lexer event type.
|
|
|
685 |
* @return boolean False if parse error.
|
|
|
686 |
* @access public
|
|
|
687 |
*/
|
|
|
688 |
function ignore($token, $event) {
|
|
|
689 |
return true;
|
|
|
690 |
}
|
|
|
691 |
|
|
|
692 |
/**
|
|
|
693 |
* Decodes any HTML entities.
|
|
|
694 |
* @param string $html Incoming HTML.
|
|
|
695 |
* @return string Outgoing plain text.
|
|
|
696 |
* @access public
|
|
|
697 |
* @static
|
|
|
698 |
*/
|
|
|
699 |
static function decodeHtml($html) {
|
|
|
700 |
static $translations;
|
|
|
701 |
if (! isset($translations)) {
|
|
|
702 |
$translations = array_flip(get_html_translation_table(HTML_ENTITIES));
|
|
|
703 |
}
|
|
|
704 |
return strtr($html, $translations);
|
|
|
705 |
}
|
|
|
706 |
|
|
|
707 |
/**
|
|
|
708 |
* Turns HTML into text browser visible text. Images
|
|
|
709 |
* are converted to their alt text and tags are supressed.
|
|
|
710 |
* Entities are converted to their visible representation.
|
|
|
711 |
* @param string $html HTML to convert.
|
|
|
712 |
* @return string Plain text.
|
|
|
713 |
* @access public
|
|
|
714 |
* @static
|
|
|
715 |
*/
|
|
|
716 |
static function normalise($html) {
|
|
|
717 |
$text = preg_replace('|<!--.*?-->|', '', $html);
|
|
|
718 |
$text = preg_replace('|<img.*?alt\s*=\s*"(.*?)".*?>|', ' \1 ', $text);
|
|
|
719 |
$text = preg_replace('|<img.*?alt\s*=\s*\'(.*?)\'.*?>|', ' \1 ', $text);
|
|
|
720 |
$text = preg_replace('|<img.*?alt\s*=\s*([a-zA-Z_]+).*?>|', ' \1 ', $text);
|
|
|
721 |
$text = preg_replace('|<.*?>|', '', $text);
|
|
|
722 |
$text = SimpleHtmlSaxParser::decodeHtml($text);
|
|
|
723 |
$text = preg_replace('|\s+|', ' ', $text);
|
|
|
724 |
return trim($text);
|
|
|
725 |
}
|
|
|
726 |
}
|
|
|
727 |
|
|
|
728 |
/**
|
|
|
729 |
* SAX event handler.
|
|
|
730 |
* @package SimpleTest
|
|
|
731 |
* @subpackage WebTester
|
|
|
732 |
* @abstract
|
|
|
733 |
*/
|
|
|
734 |
class SimpleSaxListener {
|
|
|
735 |
|
|
|
736 |
/**
|
|
|
737 |
* Sets the document to write to.
|
|
|
738 |
* @access public
|
|
|
739 |
*/
|
|
|
740 |
function SimpleSaxListener() {
|
|
|
741 |
}
|
|
|
742 |
|
|
|
743 |
/**
|
|
|
744 |
* Start of element event.
|
|
|
745 |
* @param string $name Element name.
|
|
|
746 |
* @param hash $attributes Name value pairs.
|
|
|
747 |
* Attributes without content
|
|
|
748 |
* are marked as true.
|
|
|
749 |
* @return boolean False on parse error.
|
|
|
750 |
* @access public
|
|
|
751 |
*/
|
|
|
752 |
function startElement($name, $attributes) {
|
|
|
753 |
}
|
|
|
754 |
|
|
|
755 |
/**
|
|
|
756 |
* End of element event.
|
|
|
757 |
* @param string $name Element name.
|
|
|
758 |
* @return boolean False on parse error.
|
|
|
759 |
* @access public
|
|
|
760 |
*/
|
|
|
761 |
function endElement($name) {
|
|
|
762 |
}
|
|
|
763 |
|
|
|
764 |
/**
|
|
|
765 |
* Unparsed, but relevant data.
|
|
|
766 |
* @param string $text May include unparsed tags.
|
|
|
767 |
* @return boolean False on parse error.
|
|
|
768 |
* @access public
|
|
|
769 |
*/
|
|
|
770 |
function addContent($text) {
|
|
|
771 |
}
|
|
|
772 |
}
|
|
|
773 |
?>
|