1 <?php |
|
2 // vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4: |
|
3 /** |
|
4 * Baseline rule class for extension into a "real" parser component. |
|
5 * |
|
6 * PHP versions 4 and 5 |
|
7 * |
|
8 * @category Text |
|
9 * @package Text_Wiki |
|
10 * @author Paul M. Jones <pmjones@php.net> |
|
11 * @license http://www.gnu.org/copyleft/lesser.html LGPL License 2.1 |
|
12 * @version CVS: $Id: Parse.php,v 1.5 2005/07/29 08:57:29 toggg Exp $ |
|
13 * @link http://pear.php.net/package/Text_Wiki |
|
14 */ |
|
15 |
|
16 /** |
|
17 * Baseline rule class for extension into a "real" parser component. |
|
18 * |
|
19 * Text_Wiki_Rule classes do not stand on their own; they are called by a |
|
20 * Text_Wiki object, typcially in the transform() method. Each rule class |
|
21 * performs three main activities: parse, process, and render. |
|
22 * |
|
23 * The parse() method takes a regex and applies it to the whole block of |
|
24 * source text at one time. Each match is sent as $matches to the |
|
25 * process() method. |
|
26 * |
|
27 * The process() method acts on the matched text from the source, and |
|
28 * then processes the source text is some way. This may mean the |
|
29 * creation of a delimited token using addToken(). In every case, the |
|
30 * process() method returns the text that should replace the matched text |
|
31 * from parse(). |
|
32 * |
|
33 * @category Text |
|
34 * @package Text_Wiki |
|
35 * @author Paul M. Jones <pmjones@php.net> |
|
36 * @license http://www.gnu.org/copyleft/lesser.html LGPL License 2.1 |
|
37 * @version Release: @package_version@ |
|
38 * @link http://pear.php.net/package/Text_Wiki |
|
39 */ |
|
40 class Text_Wiki_Parse { |
|
41 |
|
42 |
|
43 /** |
|
44 * |
|
45 * Configuration options for this parser rule. |
|
46 * |
|
47 * @access public |
|
48 * |
|
49 * @var string |
|
50 * |
|
51 */ |
|
52 |
|
53 var $conf = array(); |
|
54 |
|
55 |
|
56 /** |
|
57 * |
|
58 * Regular expression to find matching text for this rule. |
|
59 * |
|
60 * @access public |
|
61 * |
|
62 * @var string |
|
63 * |
|
64 * @see parse() |
|
65 * |
|
66 */ |
|
67 |
|
68 var $regex = null; |
|
69 |
|
70 |
|
71 /** |
|
72 * |
|
73 * The name of this rule for new token array elements. |
|
74 * |
|
75 * @access public |
|
76 * |
|
77 * @var string |
|
78 * |
|
79 */ |
|
80 |
|
81 var $rule = null; |
|
82 |
|
83 |
|
84 /** |
|
85 * |
|
86 * A reference to the calling Text_Wiki object. |
|
87 * |
|
88 * This is needed so that each rule has access to the same source |
|
89 * text, token set, URLs, interwiki maps, page names, etc. |
|
90 * |
|
91 * @access public |
|
92 * |
|
93 * @var object |
|
94 */ |
|
95 |
|
96 var $wiki = null; |
|
97 |
|
98 |
|
99 /** |
|
100 * |
|
101 * Constructor for this parser rule. |
|
102 * |
|
103 * @access public |
|
104 * |
|
105 * @param object &$obj The calling "parent" Text_Wiki object. |
|
106 * |
|
107 */ |
|
108 |
|
109 function Text_Wiki_Parse(&$obj) |
|
110 { |
|
111 // set the reference to the calling Text_Wiki object; |
|
112 // this allows us access to the shared source text, token |
|
113 // array, etc. |
|
114 $this->wiki =& $obj; |
|
115 |
|
116 // set the name of this rule; generally used when adding |
|
117 // to the tokens array. strip off the Text_Wiki_Parse_ portion. |
|
118 // text_wiki_parse_ |
|
119 // 0123456789012345 |
|
120 $tmp = substr(get_class($this), 16); |
|
121 $this->rule = ucwords(strtolower($tmp)); |
|
122 |
|
123 // override config options for the rule if specified |
|
124 if (isset($this->wiki->parseConf[$this->rule]) && |
|
125 is_array($this->wiki->parseConf[$this->rule])) { |
|
126 |
|
127 $this->conf = array_merge( |
|
128 $this->conf, |
|
129 $this->wiki->parseConf[$this->rule] |
|
130 ); |
|
131 |
|
132 } |
|
133 } |
|
134 |
|
135 |
|
136 /** |
|
137 * |
|
138 * Abstrct method to parse source text for matches. |
|
139 * |
|
140 * Applies the rule's regular expression to the source text, passes |
|
141 * every match to the process() method, and replaces the matched text |
|
142 * with the results of the processing. |
|
143 * |
|
144 * @access public |
|
145 * |
|
146 * @see Text_Wiki_Parse::process() |
|
147 * |
|
148 */ |
|
149 |
|
150 function parse() |
|
151 { |
|
152 $this->wiki->source = preg_replace_callback( |
|
153 $this->regex, |
|
154 array(&$this, 'process'), |
|
155 $this->wiki->source |
|
156 ); |
|
157 } |
|
158 |
|
159 |
|
160 /** |
|
161 * |
|
162 * Abstract method to generate replacements for matched text. |
|
163 * |
|
164 * @access public |
|
165 * |
|
166 * @param array $matches An array of matches from the parse() method |
|
167 * as generated by preg_replace_callback. $matches[0] is the full |
|
168 * matched string, $matches[1] is the first matched pattern, |
|
169 * $matches[2] is the second matched pattern, and so on. |
|
170 * |
|
171 * @return string The processed text replacement; defaults to the |
|
172 * full matched string (i.e., no changes to the text). |
|
173 * |
|
174 * @see Text_Wiki_Parse::parse() |
|
175 * |
|
176 */ |
|
177 |
|
178 function process(&$matches) |
|
179 { |
|
180 return $matches[0]; |
|
181 } |
|
182 |
|
183 |
|
184 /** |
|
185 * |
|
186 * Simple method to safely get configuration key values. |
|
187 * |
|
188 * @access public |
|
189 * |
|
190 * @param string $key The configuration key. |
|
191 * |
|
192 * @param mixed $default If the key does not exist, return this value |
|
193 * instead. |
|
194 * |
|
195 * @return mixed The configuration key value (if it exists) or the |
|
196 * default value (if not). |
|
197 * |
|
198 */ |
|
199 |
|
200 function getConf($key, $default = null) |
|
201 { |
|
202 if (isset($this->conf[$key])) { |
|
203 return $this->conf[$key]; |
|
204 } else { |
|
205 return $default; |
|
206 } |
|
207 } |
|
208 |
|
209 |
|
210 /** |
|
211 * |
|
212 * Extract 'attribute="value"' portions of wiki markup. |
|
213 * |
|
214 * This kind of markup is typically used only in macros, but is useful |
|
215 * anywhere. |
|
216 * |
|
217 * The syntax is pretty strict; there can be no spaces between the |
|
218 * option name, the equals, and the first double-quote; the value |
|
219 * must be surrounded by double-quotes. You can escape characters in |
|
220 * the value with a backslash, and the backslash will be stripped for |
|
221 * you. |
|
222 * |
|
223 * @access public |
|
224 * |
|
225 * @param string $text The "attributes" portion of markup. |
|
226 * |
|
227 * @return array An associative array of key-value pairs where the |
|
228 * key is the option name and the value is the option value. |
|
229 * |
|
230 */ |
|
231 |
|
232 function getAttrs($text) |
|
233 { |
|
234 // find the =" sections; |
|
235 $tmp = explode('="', trim($text)); |
|
236 |
|
237 // basic setup |
|
238 $k = count($tmp) - 1; |
|
239 $attrs = array(); |
|
240 $key = null; |
|
241 |
|
242 // loop through the sections |
|
243 foreach ($tmp as $i => $val) { |
|
244 |
|
245 // first element is always the first key |
|
246 if ($i == 0) { |
|
247 $key = trim($val); |
|
248 continue; |
|
249 } |
|
250 |
|
251 // find the last double-quote in the value. |
|
252 // the part to the left is the value for the last key, |
|
253 // the part to the right is the next key name |
|
254 $pos = strrpos($val, '"'); |
|
255 $attrs[$key] = stripslashes(substr($val, 0, $pos)); |
|
256 $key = trim(substr($val, $pos+1)); |
|
257 |
|
258 } |
|
259 |
|
260 return $attrs; |
|
261 |
|
262 } |
|
263 } |
|
264 ?> |
|