| 1 | <?php |
| 2 | |
| 3 | /** |
| 4 | * |
| 5 | * Parse for URLS in the source text. |
| 6 | * |
| 7 | * @category Text |
| 8 | * |
| 9 | * @package Text_Wiki |
| 10 | * |
| 11 | * @author Paul M. Jones <pmjones@php.net> |
| 12 | * |
| 13 | * @license LGPL |
| 14 | * |
| 15 | * @version $Id: Url.php,v 1.3 2005/02/23 17:38:29 pmjones Exp $ |
| 16 | * |
| 17 | */ |
| 18 | |
| 19 | /** |
| 20 | * |
| 21 | * Parse for URLS in the source text. |
| 22 | * |
| 23 | * Various URL markings are supported: inline (the URL by itself), |
| 24 | * numbered or footnote reference (where the URL is enclosed in square |
| 25 | * brackets), and named reference (where the URL is enclosed in square |
| 26 | * brackets and has a name included inside the brackets). E.g.: |
| 27 | * |
| 28 | * inline -- http://example.com |
| 29 | * numbered -- [http://example.com] |
| 30 | * described -- [http://example.com Example Description] |
| 31 | * |
| 32 | * When rendering a URL token, this will convert URLs pointing to a .gif, |
| 33 | * .jpg, or .png image into an inline <img /> tag (for the 'xhtml' |
| 34 | * format). |
| 35 | * |
| 36 | * Token options are: |
| 37 | * |
| 38 | * 'type' => ['inline'|'footnote'|'descr'] the type of URL |
| 39 | * |
| 40 | * 'href' => the URL link href portion |
| 41 | * |
| 42 | * 'text' => the displayed text of the URL link |
| 43 | * |
| 44 | * @category Text |
| 45 | * |
| 46 | * @package Text_Wiki |
| 47 | * |
| 48 | * @author Paul M. Jones <pmjones@php.net> |
| 49 | * |
| 50 | */ |
| 51 | |
| 52 | class Text_Wiki_Parse_Url extends Text_Wiki_Parse { |
| 53 | |
| 54 | |
| 55 | /** |
| 56 | * |
| 57 | * Keeps a running count of numbered-reference URLs. |
| 58 | * |
| 59 | * @access public |
| 60 | * |
| 61 | * @var int |
| 62 | * |
| 63 | */ |
| 64 | |
| 65 | var $footnoteCount = 0; |
| 66 | |
| 67 | |
| 68 | /** |
| 69 | * |
| 70 | * URL schemes recognized by this rule. |
| 71 | * |
| 72 | * @access public |
| 73 | * |
| 74 | * @var array |
| 75 | * |
| 76 | */ |
| 77 | |
| 78 | var $conf = array( |
| 79 | 'schemes' => array( |
| 80 | 'http://', |
| 81 | 'https://', |
| 82 | 'ftp://', |
| 83 | 'gopher://', |
| 84 | 'news://', |
| 85 | 'mailto:' |
| 86 | ) |
| 87 | ); |
| 88 | |
| 89 | |
| 90 | /** |
| 91 | * |
| 92 | * Constructor. |
| 93 | * |
| 94 | * We override the constructor so we can comment the regex nicely. |
| 95 | * |
| 96 | * @access public |
| 97 | * |
| 98 | */ |
| 99 | |
| 100 | function Text_Wiki_Parse_Url(&$obj) |
| 101 | { |
| 102 | parent::Text_Wiki_Parse($obj); |
| 103 | |
| 104 | // convert the list of recognized schemes to a regex-safe string, |
| 105 | // where the pattern delim is a slash |
| 106 | $tmp = array(); |
| 107 | $list = $this->getConf('schemes', array()); |
| 108 | foreach ($list as $val) { |
| 109 | $tmp[] = preg_quote($val, '/'); |
| 110 | } |
| 111 | $schemes = implode('|', $tmp); |
| 112 | |
| 113 | // build the regex |
| 114 | $this->regex = |
| 115 | "($schemes)" . // allowed schemes |
| 116 | "(" . // start pattern |
| 117 | "[^ \\/\"\'{$this->wiki->delim}]*\\/" . // no spaces, backslashes, slashes, double-quotes, single quotes, or delimiters; |
| 118 | ")*" . // end pattern |
| 119 | "[^ \\t\\n\\/\"\'{$this->wiki->delim}]*" . |
| 120 | "[A-Za-z0-9\\/?=&~_]"; |
| 121 | } |
| 122 | |
| 123 | |
| 124 | /** |
| 125 | * |
| 126 | * Find three different kinds of URLs in the source text. |
| 127 | * |
| 128 | * @access public |
| 129 | * |
| 130 | */ |
| 131 | |
| 132 | function parse() |
| 133 | { |
| 134 | // ------------------------------------------------------------- |
| 135 | // |
| 136 | // Described-reference (named) URLs. |
| 137 | // |
| 138 | |
| 139 | // the regular expression for this kind of URL |
| 140 | $tmp_regex = '/\[(' . $this->regex . ') ([^\]]+)\]/'; |
| 141 | |
| 142 | // use a custom callback processing method to generate |
| 143 | // the replacement text for matches. |
| 144 | $this->wiki->source = preg_replace_callback( |
| 145 | $tmp_regex, |
| 146 | array(&$this, 'processDescr'), |
| 147 | $this->wiki->source |
| 148 | ); |
| 149 | |
| 150 | |
| 151 | // ------------------------------------------------------------- |
| 152 | // |
| 153 | // Numbered-reference (footnote-style) URLs. |
| 154 | // |
| 155 | |
| 156 | // the regular expression for this kind of URL |
| 157 | $tmp_regex = '/\[(' . $this->regex . ')\]/U'; |
| 158 | |
| 159 | // use a custom callback processing method to generate |
| 160 | // the replacement text for matches. |
| 161 | $this->wiki->source = preg_replace_callback( |
| 162 | $tmp_regex, |
| 163 | array(&$this, 'processFootnote'), |
| 164 | $this->wiki->source |
| 165 | ); |
| 166 | |
| 167 | |
| 168 | // ------------------------------------------------------------- |
| 169 | // |
| 170 | // Normal inline URLs. |
| 171 | // |
| 172 | |
| 173 | // the regular expression for this kind of URL |
| 174 | |
| 175 | $tmp_regex = '/(^|[^A-Za-z])(' . $this->regex . ')(.*?)/'; |
| 176 | |
| 177 | // use the standard callback for inline URLs |
| 178 | $this->wiki->source = preg_replace_callback( |
| 179 | $tmp_regex, |
| 180 | array(&$this, 'process'), |
| 181 | $this->wiki->source |
| 182 | ); |
| 183 | } |
| 184 | |
| 185 | |
| 186 | /** |
| 187 | * |
| 188 | * Process inline URLs. |
| 189 | * |
| 190 | * @param array &$matches |
| 191 | * |
| 192 | * @param array $matches An array of matches from the parse() method |
| 193 | * as generated by preg_replace_callback. $matches[0] is the full |
| 194 | * matched string, $matches[1] is the first matched pattern, |
| 195 | * $matches[2] is the second matched pattern, and so on. |
| 196 | * |
| 197 | * @return string The processed text replacement. |
| 198 | * |
| 199 | */ |
| 200 | |
| 201 | function process(&$matches) |
| 202 | { |
| 203 | // set options |
| 204 | $options = array( |
| 205 | 'type' => 'inline', |
| 206 | 'href' => $matches[2], |
| 207 | 'text' => $matches[2] |
| 208 | ); |
| 209 | |
| 210 | // tokenize |
| 211 | return $matches[1] . $this->wiki->addToken($this->rule, $options) . $matches[5]; |
| 212 | } |
| 213 | |
| 214 | |
| 215 | /** |
| 216 | * |
| 217 | * Process numbered (footnote) URLs. |
| 218 | * |
| 219 | * Token options are: |
| 220 | * @param array &$matches |
| 221 | * |
| 222 | * @param array $matches An array of matches from the parse() method |
| 223 | * as generated by preg_replace_callback. $matches[0] is the full |
| 224 | * matched string, $matches[1] is the first matched pattern, |
| 225 | * $matches[2] is the second matched pattern, and so on. |
| 226 | * |
| 227 | * @return string The processed text replacement. |
| 228 | * |
| 229 | */ |
| 230 | |
| 231 | function processFootnote(&$matches) |
| 232 | { |
| 233 | // keep a running count for footnotes |
| 234 | $this->footnoteCount++; |
| 235 | |
| 236 | // set options |
| 237 | $options = array( |
| 238 | 'type' => 'footnote', |
| 239 | 'href' => $matches[1], |
| 240 | 'text' => $this->footnoteCount |
| 241 | ); |
| 242 | |
| 243 | // tokenize |
| 244 | return $this->wiki->addToken($this->rule, $options); |
| 245 | } |
| 246 | |
| 247 | |
| 248 | /** |
| 249 | * |
| 250 | * Process described-reference (named-reference) URLs. |
| 251 | * |
| 252 | * Token options are: |
| 253 | * 'type' => ['inline'|'footnote'|'descr'] the type of URL |
| 254 | * 'href' => the URL link href portion |
| 255 | * 'text' => the displayed text of the URL link |
| 256 | * |
| 257 | * @param array &$matches |
| 258 | * |
| 259 | * @param array $matches An array of matches from the parse() method |
| 260 | * as generated by preg_replace_callback. $matches[0] is the full |
| 261 | * matched string, $matches[1] is the first matched pattern, |
| 262 | * $matches[2] is the second matched pattern, and so on. |
| 263 | * |
| 264 | * @return string The processed text replacement. |
| 265 | * |
| 266 | */ |
| 267 | |
| 268 | function processDescr(&$matches) |
| 269 | { |
| 270 | // set options |
| 271 | $options = array( |
| 272 | 'type' => 'descr', |
| 273 | 'href' => $matches[1], |
| 274 | 'text' => $matches[4] |
| 275 | ); |
| 276 | |
| 277 | // tokenize |
| 278 | return $this->wiki->addToken($this->rule, $options); |
| 279 | } |
| 280 | } |
| 281 | ?> |