| 1 | <?php |
| 2 | |
| 3 | /** |
| 4 | * |
| 5 | * Baseline rule class for extension into a "real" parser component. |
| 6 | * |
| 7 | <<<<<<< Parse.php |
| 8 | * Text_Wiki_Parse_* classes do not stand on their own; they are called by a |
| 9 | * Text_Wiki object, typcially in the transform() method. Each rule class |
| 10 | * performs two separate activities: parse and process. |
| 11 | ======= |
| 12 | * @category Text |
| 13 | * |
| 14 | * @package Text_Wiki |
| 15 | * |
| 16 | * @author Paul M. Jones <pmjones@php.net> |
| 17 | * |
| 18 | * @license LGPL |
| 19 | * |
| 20 | */ |
| 21 | |
| 22 | /** |
| 23 | * |
| 24 | * Baseline rule class for extension into a "real" parser component. |
| 25 | * |
| 26 | * Text_Wiki_Rule classes do not stand on their own; they are called by a |
| 27 | * Text_Wiki object, typcially in the transform() method. Each rule class |
| 28 | * performs three main activities: parse, process, and render. |
| 29 | >>>>>>> 1.4 |
| 30 | * |
| 31 | * The parse() method takes a regex and applies it to the whole block of |
| 32 | * source text at one time. Each match is sent as $matches to the |
| 33 | * process() method. |
| 34 | * |
| 35 | * The process() method acts on the matched text from the source, and |
| 36 | * then processes the source text is some way. This may mean the |
| 37 | * creation of a delimited token using addToken(). In every case, the |
| 38 | * process() method returns the text that should replace the matched text |
| 39 | * from parse(). |
| 40 | * |
| 41 | * @category Text |
| 42 | * |
| 43 | * @package Text_Wiki |
| 44 | * |
| 45 | * @author Paul M. Jones <pmjones@php.net> |
| 46 | * |
| 47 | */ |
| 48 | |
| 49 | class Text_Wiki_Parse { |
| 50 | |
| 51 | |
| 52 | /** |
| 53 | * |
| 54 | * Configuration options for this parser rule. |
| 55 | * |
| 56 | * @access public |
| 57 | * |
| 58 | * @var string |
| 59 | * |
| 60 | */ |
| 61 | |
| 62 | var $conf = array(); |
| 63 | |
| 64 | |
| 65 | /** |
| 66 | * |
| 67 | * Regular expression to find matching text for this rule. |
| 68 | * |
| 69 | * @access public |
| 70 | * |
| 71 | * @var string |
| 72 | * |
| 73 | * @see parse() |
| 74 | * |
| 75 | */ |
| 76 | |
| 77 | var $regex = null; |
| 78 | |
| 79 | |
| 80 | /** |
| 81 | * |
| 82 | * The name of this rule for new token array elements. |
| 83 | * |
| 84 | * @access public |
| 85 | * |
| 86 | * @var string |
| 87 | * |
| 88 | */ |
| 89 | |
| 90 | var $rule = null; |
| 91 | |
| 92 | |
| 93 | /** |
| 94 | * |
| 95 | * A reference to the calling Text_Wiki object. |
| 96 | * |
| 97 | * This is needed so that each rule has access to the same source |
| 98 | * text, token set, URLs, interwiki maps, page names, etc. |
| 99 | * |
| 100 | * @access public |
| 101 | * |
| 102 | * @var object |
| 103 | */ |
| 104 | |
| 105 | var $wiki = null; |
| 106 | |
| 107 | |
| 108 | /** |
| 109 | * |
| 110 | * Constructor for this parser rule. |
| 111 | * |
| 112 | * @access public |
| 113 | * |
| 114 | * @param object &$obj The calling "parent" Text_Wiki object. |
| 115 | * |
| 116 | */ |
| 117 | |
| 118 | function Text_Wiki_Parse(&$obj) |
| 119 | { |
| 120 | // set the reference to the calling Text_Wiki object; |
| 121 | // this allows us access to the shared source text, token |
| 122 | // array, etc. |
| 123 | $this->wiki =& $obj; |
| 124 | |
| 125 | // set the name of this rule; generally used when adding |
| 126 | // to the tokens array. strip off the Text_Wiki_Parse_ portion. |
| 127 | // text_wiki_parse_ |
| 128 | // 0123456789012345 |
| 129 | $tmp = substr(get_class($this), 16); |
| 130 | $this->rule = ucwords(strtolower($tmp)); |
| 131 | |
| 132 | // override config options for the rule if specified |
| 133 | if (isset($this->wiki->parseConf[$this->rule]) && |
| 134 | is_array($this->wiki->parseConf[$this->rule])) { |
| 135 | |
| 136 | $this->conf = array_merge( |
| 137 | $this->conf, |
| 138 | $this->wiki->parseConf[$this->rule] |
| 139 | ); |
| 140 | |
| 141 | } |
| 142 | } |
| 143 | |
| 144 | |
| 145 | /** |
| 146 | * |
| 147 | * Abstrct method to parse source text for matches. |
| 148 | * |
| 149 | * Applies the rule's regular expression to the source text, passes |
| 150 | * every match to the process() method, and replaces the matched text |
| 151 | * with the results of the processing. |
| 152 | * |
| 153 | * @access public |
| 154 | * |
| 155 | * @see Text_Wiki_Parse::process() |
| 156 | * |
| 157 | */ |
| 158 | |
| 159 | function parse() |
| 160 | { |
| 161 | $this->wiki->source = preg_replace_callback( |
| 162 | $this->regex, |
| 163 | array(&$this, 'process'), |
| 164 | $this->wiki->source |
| 165 | ); |
| 166 | } |
| 167 | |
| 168 | |
| 169 | /** |
| 170 | * |
| 171 | * Abstract method to generate replacements for matched text. |
| 172 | * |
| 173 | * @access public |
| 174 | * |
| 175 | * @param array $matches An array of matches from the parse() method |
| 176 | * as generated by preg_replace_callback. $matches[0] is the full |
| 177 | * matched string, $matches[1] is the first matched pattern, |
| 178 | * $matches[2] is the second matched pattern, and so on. |
| 179 | * |
| 180 | * @return string The processed text replacement; defaults to the |
| 181 | * full matched string (i.e., no changes to the text). |
| 182 | * |
| 183 | * @see Text_Wiki_Parse::parse() |
| 184 | * |
| 185 | */ |
| 186 | |
| 187 | function process(&$matches) |
| 188 | { |
| 189 | return $matches[0]; |
| 190 | } |
| 191 | |
| 192 | |
| 193 | /** |
| 194 | * |
| 195 | * Simple method to safely get configuration key values. |
| 196 | * |
| 197 | * @access public |
| 198 | * |
| 199 | * @param string $key The configuration key. |
| 200 | * |
| 201 | * @param mixed $default If the key does not exist, return this value |
| 202 | * instead. |
| 203 | * |
| 204 | * @return mixed The configuration key value (if it exists) or the |
| 205 | * default value (if not). |
| 206 | * |
| 207 | */ |
| 208 | |
| 209 | function getConf($key, $default = null) |
| 210 | { |
| 211 | if (isset($this->conf[$key])) { |
| 212 | return $this->conf[$key]; |
| 213 | } else { |
| 214 | return $default; |
| 215 | } |
| 216 | } |
| 217 | |
| 218 | |
| 219 | /** |
| 220 | * |
| 221 | * Extract 'attribute="value"' portions of wiki markup. |
| 222 | * |
| 223 | * This kind of markup is typically used only in macros, but is useful |
| 224 | * anywhere. |
| 225 | * |
| 226 | * The syntax is pretty strict; there can be no spaces between the |
| 227 | * option name, the equals, and the first double-quote; the value |
| 228 | * must be surrounded by double-quotes. You can escape characters in |
| 229 | * the value with a backslash, and the backslash will be stripped for |
| 230 | * you. |
| 231 | * |
| 232 | * @access public |
| 233 | * |
| 234 | * @param string $text The "attributes" portion of markup. |
| 235 | * |
| 236 | * @return array An associative array of key-value pairs where the |
| 237 | * key is the option name and the value is the option value. |
| 238 | * |
| 239 | */ |
| 240 | |
| 241 | function getAttrs($text) |
| 242 | { |
| 243 | // find the =" sections; |
| 244 | $tmp = explode('="', trim($text)); |
| 245 | |
| 246 | // basic setup |
| 247 | $k = count($tmp) - 1; |
| 248 | $attrs = array(); |
| 249 | $key = null; |
| 250 | |
| 251 | // loop through the sections |
| 252 | foreach ($tmp as $i => $val) { |
| 253 | |
| 254 | // first element is always the first key |
| 255 | if ($i == 0) { |
| 256 | $key = trim($val); |
| 257 | continue; |
| 258 | } |
| 259 | |
| 260 | // find the last double-quote in the value. |
| 261 | // the part to the left is the value for the last key, |
| 262 | // the part to the right is the next key name |
| 263 | $pos = strrpos($val, '"'); |
| 264 | $attrs[$key] = stripslashes(substr($val, 0, $pos)); |
| 265 | $key = trim(substr($val, $pos+1)); |
| 266 | |
| 267 | } |
| 268 | |
| 269 | return $attrs; |
| 270 | |
| 271 | } |
| 272 | } |
| 273 | ?> |